,. . .. .
Mgmw
a}

A .21

. z 1.
s15: .
.

4% 3a....”

 

 

 

v:

:2.

32.1
r. .

_

.
.

u...

a

«Nahum
‘ .1,

33.5.5.1.

S £52,... £2
dhmflmmmzmr

mun“; “.54...“
an“): a.

..
a . .
.153...

g +1 a“

, ;

51

.3 xiii...)
.312

, 5.

4:3;

finfinflmmxy . 5n. B.

it? aigyauéaaamfiifi
afififlfiafi :
.r #3....“

., fig;

23;
favké.
um. £2 ..

ixLflNuht. 5.53:... 3... . 2.... 8.}
:z..z.flm_.. . si..¥fi?;fl:§¢§
)9 ‘- t1?»

.5. . y.

3.: :

intsnr-

5-1.1

 

 

 

 

 

 

MICHIGAN STATE UNIVERSITY LIBRARIES
m3 lllllllllllllllllllllllllllllllllllllllllllllllllllll
3 1293 02079 9288

K)
\\

This is to certify that the

dissertation entitled

Optimizing Across the Value Chain:
External Competitive Threats and the Use
of Cost Information in Buyer—Supplier Negotiations

presented by

Andrea R. Drake

has been accepted towards fulfillment
ofthe requirements for

._P_h;1_)'____ degree in M

‘S/IQ’IZIw/‘r /7['//ak Q...’

Major professor

Date /2’//L/[ 96/

MSU is an Affirnnuiw Aru'un/Equul Oppnrlunily Instilutiml 0‘ 12771

 

 

 

 

LIBRARY
Michigan State-
University

 

 

 

OPTIMIZING ACROSS THE VALUE CHAIN:
EXTERNAL COMPETITIVE THREATS AND THE USE OF COST INFORMATION
IN BUYER-SUPPLIER NEGOTIATIONS
By

Andrea R. Drake

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Department of Accounting

1 999

ABSTRACT

OPTIMIZING ACROSS THE VALUE CHAIN:
EXTERNAL COMPETITIVE THREATS AND THE USE OF COST INFORMATION
IN BUYER-SUPPLIER NEGOTIATIONS
By

Andrea R. Drake

The opportunity to minimize costs across a greater portion of the value chain has
prompted many firms to strive for collaborative relations with their customers and
suppliers. A key to the success of such efforts is the sharing of cost information that can
be used to discover the most economical production strategies across firms. The detailed
cost driver information provided by activity-based costing (ABC) systems can be used to
discover a wider variety Of cost minimizing choices than can the less detailed
information provided by volume-based cost systems. However, the results reported here
Show that individuals are more reluctant to share detailed ABC information as compared
to less detailed information. Consequently, when detailed cost information is not shared,
cost minimizing options are often foregone.

In addition, the study hypothesized that the prospect of losses induced by a strong
competitive threat impacts the decision to share cost information. Prospect theory
predicts that individuals expecting losses will be more willing to share cost information
than those expecting profits. However, they will still be relatively less willing to share
detailed ABC information, regardless of the competitive environment. The results are

consistent with these predictions, and although fewer subject pairs shared ABC

information, the agreements reached by those that did were closer to the optimal than the
agreements reached by pairs where either no information or less detailed cost information
was shared. In addition, the sharing Of either more or less detailed cost information lead
to a greater sense of cooperation among negotiators and greater satisfaction with
outcomes.

The study highlights an important conflict related to the use of ABC information
in a negotiation context where the bargainers have both cooperative and competitive
motivations. The cooperative motive results from the integrative potential Of the
negotiation, whereby the total amount of joint profits can be increased if cost minimizing
Options can be discovered and agreed upon. The Sharing of ABC information increases
the chances that these Options will be found and total joint profits maximized. However,
a competitive motive arises because both negotiators have an incentive to maximize their
individual profits. The focus on individual rather than joint profits can result in a
reluctance to share ABC information, resulting in foregone cost savings. Further research
is needed on how to overcome this reluctance so that the integrative potential of a
negotiation can be realized, while concurrently ensuring that the resulting joint profits are

divided equitably among the parties.

Copyright by
Andrea Ranval Drake
1 999

ACKNOWLEDGEMENTS

I would like to wholeheartedly thank my committee members, Sue Haka, Joan
Lufi, Matt Anderson and Jack Hunter for their support, guidance and patience during the
many stages Of this project. Their thoughtful insights, suggestions and attention to detail
have much improved the manuscript. I am especially grateful to my chairperson, Sue
Haka, who throughout my doctoral career has been a wonderful source of ideas,
encouragement and general hints on how to survive academia. I could not have asked for
a greater mentor.

I am also indebted to my fellow doctoral students at Michigan State University,
all of whom never failed to volunteer to help with various experimental tasks, often on
very Short notice. I am especially grateful to Ola Smith, Donna Booker, Elizabeth
Connors and Dan Heitger for helping me run the many experimental sessions. Without
their help and friendship, I doubt this study would have gotten to this point. A special
note of thanks also goes to Sue Ravenscroft, who provided many useful comments on the
dissertation and many humorous notes of encouragement along the way.

I am also extremely grateful to my family for their encouragement and support. I
owe many thanks to my parents who unfailingly conveyed to me their belief that I could
accomplish what I set my mind to. The security and love they provided gave me the
courage to pursue goals I might not otherwise have considered. I am also indebted to my
sister, Lori, who helped me through some of the harder parts of this pursuit. She is truly a

beautiful spirit that has always inspired me.

Finally, I owe deep debt of gratitude to my husband, Bob, for being an ever-
present source of love, laughter, and lost keys. His remarkable ability to keep our life in
some semblance of order under the most chaotic of circumstances made it possible for me

to finish this dissertation.

Vi

TABLE OF CONTENTS

List of Tables ................................................................................................ ix
LiSt of Figures ............................................................................................... x
Chapter 1
Introduction ................................................................................................... 1
Overview .................................................................................................. 1
Importance Of Research Issues ................................................................. 7
Contribution Of Research ......................................................................... 8
Chapter 2
Literature Review .......................................................................................... 10
Buyer—Supplier Relations ......................................................................... 10
Alternative Cost Systems and Cost Minimization Within
and Across Firms ................................................................................... 12
Prospect Theory and Expected Utility Theory ......................................... 17
Implications Of Prospect Theory for Buyer-Supplier Relations
and Negotiations .................................................................................... 20
Chapter 3
Hypotheses .................................................................................................... 26
Competitive Threats and Cost Information Sharing ................................ 26
Sharing More vs. Less Detailed Cost Information ................................... 27
Cost Information Sharing and the Optimality Of Agreements ................. 31
Distributive vs. Integrative Agreements ............................................. 31
Cost Information Sharing and the Efficiency of Negotiation .................. 33
Cost Information Sharing and Negotiator Attitudes ................................ 35
Chapter 4
Experimental Manipulations and Procedures ............................................... 37
Overview .................................................................................................. 37
Experimental Manipulations .................................................................... 38
Experimental Procedures ......................................................................... 41
Hypothetical Valuation of Experimental Prospects ................................. 43

vii

Chapter 5

Experimental Results and Discussion ........................................................... 46
Hypothesis 1 ............................................................................................. 46
Hypothesis 2 ............................................................................................. 47
Hypothesis 3 ............................................................................................. 47
Hypotheses 4a, 4b, and 4c ........................................................................ 50
Hypotheses 5a, 5b, and 5c ........................................................................ 52
Hypothesis 6 ............................................................................................. 55
Hypothesis 7 ............................................................................................. 56
Manipulation Checks ............................................................................... 57
Control Variables ..................................................................................... 58

Chapter 6

Conclusion .................................................................................................... 60
Limitations and Future Research ............................................................. 60
Concluding Discussion ............................................................................ 62

Figures ........................................................................................................... 65

Tables ............................................................................................................ 69

Appendices .................................................................................................... 90
A - Informed Consent Declaration .......................................................... 90
B - Introduction Script - Casing Seller ................................................... 91
C - Introduction Script - Casing Buyer ................................................... 96
D - Sample Negotiation Script - Threat Condition ................................. 101
E - Sample Negotiation Script - No Threat Condition ............................ 102
F - Profit/Loss Calculation Sheet - Seller ............................................... 103
G - Profit/Loss Calculation Sheet - Buyer .............................................. 104
H - Standard Casing Production Cost Report - ABC Format ................. 105
I - Standard Casing Production Cost Report - STC Format .................... 106
J - Standard Casing Usage Cost Report - ABC Format .......................... 107
K - Standard Casing Usage Cost Report - STC Format .......................... 108
L - Contract Proposal Form .................................................................... 109
M - Pre-negotiation Questionnaire .......................................................... 1 10
N - Exit Questionnaire ............................................................................ 112
O - Agreement Verification Form ........................................................... 1 16

References ..................................................................................................... 1 l8

viii

LIST OF TABLES

Table 1 — Standard Casing Profit Calculations

 

Table 2 — Possible Payoff Ranges from Cost Report Sharing ..................................

Table 3 — Hypothetical Valuation of Experimental Prospects ..................................

 

Table 4 — Proportion of Pairs that Shared RepOIts

 

Table 5 — Optimality of Agreements

Table 6 — Minutes to Reach Near Optimal Agreement

 

Table 7 — Number of Proposals Needed to Reach
Near Optimal Agreement

 

 

Table 8 — Perceived Level of Cooperation

Table 9 — Satisfaction with Outcome

 

Table 10 — Manipulation Checks

 

Table 11 — ANOVA Results for Control Variables

 

69

71

72

73

75

77

79

81

83

85

87

LIST OF FIGURES

Figure 1 — Overview of Hypotheses ......................................................................... 65
Figure 2 -— Illustration of Integrative Potential .......................................................... 66
Figure 3 — Hypothetical Value Function ................................................................... 68

CHAPTER 1
INTRODUCTION
Overview
During the 1980’s, research showed that a primary factor contributing to the
competitive strength of many Pacific rim firms was the prevalence of collaborative, long-
term relationships between manufacturers, customers, and component suppliers (Cole and
Yakushiji, 1984). Such collaboration allowed these firms to optimize processes across a
larger portion of the value chain. This was posSible because transactions between buyers
and suppliers often involved integrative potential, whereby the participants could increase
the total available profits by actively working together to discover cost saving
innovations. One key to discovering such innovations was the explicit sharing of
information between firms.
Recognition Of the potential benefits of collaboration has prompted many US.
firms to rethink their own supplier relations (Bensaou and Venkatraman, 1995;
McMillan, 1990; Cole and Yakushiji, 1984). Efforts to increase the level Of cooperation
between buyers and suppliers in the US. have met with mixed success, however. Often,
a critical element missing is a good-faith effort by both parties to share information and
search for mutually beneficial solutions (Richardson, 1993; Dyer, 1996a, 1996b). Given
that cost reduction has been found to be a primary motivation behind establishing more
cooperative buyer-supplier relations (Ellram 1995), the sharing of cost information is the
focus of the current study. The sharing of cost information is necessary to minimize costs
across firms because the choices made by one firm often impact the costs of the other.

Based on the goal of minimizing costs across firms, the primary objectives of this

experimental study are to: I) examine the impact of sharing more vs. less detailed cost
driver information on the overall profitability of negotiated agreements; 2) determine the
relative willingness of negotiators to share more vs. less detailed cost information, and; 3)
investigate whether the external competitive environment influences the relative
willingness of negotiators to share more vs. less detailed cost information.

The distinction between more vs. less detailed cost information is important
because the level of detail contained in the cost information shared is likely to affect the
extent of its usefulness in identifying cost minimizing Options. In the current context, the
level of detail refers to the number of cost drivers (and related costs) that are explicitly
displayed in a cost report that can be shared with another negotiator. A “less” detailed
report has information on only a limited number of cost drivers related to the production
of a component. In contrast, a “more” detailed report contains information on a greater
number of relevant cost drivers. This distinction parallels a general difference between
modern cost accounting systems, such as activity-based costing (ABC), and more
traditional, volume-based cost systems. The focus of volume-based cost systems is often
on the costs of labor and materials, while the drivers of overhead costs are largely
ignored. In contrast, ABC systems provide managers with information regarding
overhead cost drivers, in addition to the drivers of labor and material costs. The additional
overhead cost driver information provided by ABC systems can be used by managers to
optimize production processes involving overhead costs within their firms, in addition to
processes involving materials and labor. However, such cost information also has the
potential to help managers optimize processes across firms, particularly when the

production choices of one firm impact the costs of other firms.

Buyer-supplier negotiations represent a suitable setting in which to study cross-
firrn optimization because there is an opportunity for both parties to suggest changes to
how a component is produced. Shared cost information can be used to identify which
suggestions are most cost effective across both firms. Given the potential role of shared
cost information in maximizing cross-firm profits, identifying and examining factors that
impact whether or not negotiators share cost information is critical.

The underlying competitive environment in which buyer-supplier negotiations
take place is one potentially important variable related to the sharing of cost information.
For example, a common factor many successful buyer-supplier partnerships share is that
one or both parties recognized that they were facing a “competitive threat,” either
individually or as an industry, that put the long-run Viability of their firm(s) at risk
(Ellram and Edis, 1996; Johnston and Lawrence, 1988). Examples of such competitive
threats included falling profits, declining market share, or the prospect of more efficient
extemal competitors. The presence of one or more of these competitive threats seemed
to induce some buyers and suppliers to work with, rather than against, each other. Again,
a vital factor contributing to successful buyer-supplier collaborations was the sharing Of
information.

For simplicity, this study defines the presence or absence of a competitive threat
in terms of the combined payoff earned by a buyer-supplier pair adopting a status quo
agreement involving the sale of an intermediate component. In the presence of the
competitive threat, adopting the status quo agreement leads to financial losses to both
parties. In the absence of the competitive threat, adopting the status quo agreement leads

to profitable outcomes to both parties. The external competitive environment induces the

profits or losses resulting from the status quo agreement by impacting the market price of
the final product produced by the buyer, and the market price of the intermediate
component produced by the seller. The study presents an argument based on prospect
theory for why managers facing a competitive threat (losses) are relatively more willing
to Share cost information than managers facing no such threat (profits).

Willingness to Share cost information is an important area of study because
despite the logic that sharing cost information may lead to cost reductions, firms in the
US. are in general very reluctant to share this information (Helper, 1991). One reason
firms hold their cost information proprietary is that revealing it unveils their profits from
ensuing deals. Due to the traditional competitive nature of buyer-supplier relations, there
is a fear that the other party will use the information to extract a higher share of the joint
profits. This fear is justified given numerous studies of bargaining behavior which Show
that the revelation of individual profits during a negotiation where the parties are
motivated to maximize their own outcomes can have a dramatic impact on final
agreements and profit splits (Roth, 1995; Ochs and Roth, 1989; Roth and Mumighan,
1982; Roth and Malouf, 1979).

As noted above, the competitive environment can have an impact on the decision
to share cost information (that also reveals profit information) by inducing a profit or loss
frame of reference in which the negotiation is taking place. A considerable amount of
research has shown that individuals behave differently toward risky decisions when
facing potential losses than when facing potential gains (Kahneman and Tversky, 1979;
Kahneman, Knetsch and Thaler, 1991; Tversky and Kahneman, 1986, 1991, 1992;

Tversky, Slovic, and Kahneman, 1990). In the current experimental context, the sharing

of cost information is designated as a risky choice relative to not sharing because of the
greater uncertainty of the consequences. A negotiator that chooses to share his/her cost
information could realize increased profits if the information is used to discover cost
saving options and the negotiator receives a share of the increased profits. However,
lower profits could also be realized if the information is used by the other negotiator to
garner a greater share of the joint profits. Not sharing cost information is designated as
less risky because it will likely result in a profit level that is near to a known, standard
amount. The influence of loss aversion, specified within the tenets of prospect theory,
predicts that managers will be relatively more willing to choose a given risky option (i.e.
sharing cost information) when facing potential losses than when facing potential gains of
similar magnitude. In addition, loss aversion is used as the basis for predicting that
managers will be relatively less willing to share more detailed ABC information than they
will less detailed cost driver information. Furthermore, theory predicts that this relative
unwillingness to share more detailed cost driver information will remain, regardless of
which of the gain or loss frames of the current context is present.

In summary, the overall objective of this study is to examine the decision of
whether or not to share cost information within the context of a buyer-supplier
negotiation. The goal is to investigate how managers react to the potential risk and
rewards of sharing more vs. less detailed cost driver information, and how the external
environment can affect this reaction. The amount of detail contained in the cost
information that can be shared is manipulated via an “official” cost report that can be
given to the other negotiator. The “more detailed” report, denoted as the ABC report,

provides the costs and cost drivers of all labor, material and overhead costs. The “less

 

detailed” (STC) report provides cost driver information on only labor and materials
costs. Overhead costs are shown in the STC report simply as the lump sum that would be
allocated to the component if it was produced according to the status quo agreement.
Parties that choose to share ABC reports face the greatest opportunity for discovering
cost reductions, but also leave themselves more vulnerable by exposing a more complete
payoff structure to their opponent. Sharing the less detailed STC report may reduce the
chances of discovering all possible cost reductions, but allows the sharer to retain private
information regarding her payoff structure. In all conditions, negotiators are aware of
their own costs and drivers and may verbally communicate this information rather than
explicitly sharing the cost report. However, only information contained in the cost
reports can be considered credible, or “true.” Again, the principles of prospect theory
predict that negotiators are relatively more willing to share less detailed rather than more
detailed cost information, regardless of the competitive environment.

The explicit sharing of cost information is hypothesized to lead to agreements that
are closer to the optimal than when no cost information is shared. However, it is also
predicted that sharing ABC information will lead to agreements that are relatively closer
to the Optimal than agreements where less detailed information is shared. Similarly,
Sharing either type of cost information is predicted to lead to more efficient negotiations,
where less time and effort are required to reach agreement. However, sharing ABC
information is predicted to lead to more efficient negotiations than sharing the less
detailed STC information.

A final objective of the study is to investigate post-negotiation attitudes among

buyer-supplier pairs. It is hypothesized that pairs that share cost information exhibit

greater feelings of cooperation and are more satisfied with negotiation outcomes. A post-
experiment questionnaire is used to assess these perceptions among subjects. The
examination of post negotiation attitudes is important because research has indicated that
the attitudes negotiators hold toward one another may be associated with future

collaborative behavior (Brodt, 1994; Thompson and Hastie, 1990).

Importance of Research Issues

The issues examined in the study are important for several reasons. First and
foremost, the minimization of costs across firms has come to represent a competitive
advantage in some industries and a necessity for survival in others. Factors that affect the
efforts of firms to accomplish this goal can help explain their resulting success or failure.
The sharing of cost information within the context of a buyer—supplier negotiation is the
factor investigated in the current study. Both the competitive environment and the
behavioral influence of loss aversion are hypothesized to affect the decision to share cost
information.

Secondly, ABC systems have been used to minimize costs and streamline
processes within firms. The next important step is to investigate whether ABC systems
are useful in minimizing costs across firms. For example, the activities of one firm, such
as a supplier, can often impact the costs of a subsequent activity performed by the
customer. The choice or option that minimizes costs within the supplying firm may not
minimize costs in the buying firm, and vice versa. The explicit sharing of detailed cost
driver information can be essential to attain overall cost minimization across both firms.

Thus, factors that impact whether and how detailed cost driver information is used across

firms warrants further research. The factor focused on in the current study is the relative
willingness of buyers and suppliers to share more versus less detailed cost driver
information. If negotiators are hesitant to share more detailed cost information, cost
saving opportunities may be foregone.

Contribution of Research

The study contributes to the literature in several ways. First, for buyer-supplier
relations, the outcomes reported here Show the importance of sharing cost information in
the context of negotiations with integrative potential. The tenets of prospect theory are
used to illustrate that the competitive environment and the influence of loss aversion can
affect decisions to share this information. Although the importance of information
sharing has been noted in previous research on buyer-supplier relations, little research has
been done on what factors impact these decisions. In addition, the study provides
supporting evidence that satisfaction with outcomes is higher when cost information is
shared in a negotiation. This result is in contrast to the belief that sharing such
information would lead to dissatisfaction with outcomes, due to the other party using the
information to garner a higher share Ofthejoint profits.

Secondly, the study has implications for firms adopting more sophisticated cost
systems, such as activity-based costing, that yield more detailed cost driver information.
This information, if used in a collaborative manner, can lead to substantial cost reductions
across firms. However, the reluctance of managers to share highly detailed information
implies that some of the benefits of these cost systems may be forfeited.

Third, the study contributes to research on integrative bargaining by investigating

how the explicit sharing of cost information impacts the outcomes of a buyer-supplier

negotiation. The study Shows that agreements are significantly closer to the optimal
when cost driver information is shared. Further, the sharing of more detailed information
leads to better outcomes than the sharing of less detailed information. This is contrary to
the subset of bargaining literature that suggests that negotiation itself should be sufficient
for the parties to discover and adopt the Optimal agreement because preference
information is revealed through the process of offers and counter-offers.

Fourth, the results reported here add confirming evidence that the predicted
effects of loss aversion from prospect theory hold in a complex, integrative bargaining
context. The context requires the negotiating managers to make a choice between a
relatively certain option (i.e. not sharing their cost reports) and an option involving
uncertain outcomes with unknown probabilities (i.e. sharing their cost reports). In
addition, although prior studies have used the tenets of prospect theory to predict the
decision making behavior of managers regarding risky and riskless choices, this is one of
the first to use the decision to share cost information as the risky option.

The next chapter provides a review of relevant literature related to buyer-supplier
relations, alternative costing systems and prospect theory. Chapter three develops the
formal hypotheses while the experimental procedures are outlined in chapter four. The
results of the study are discussed in chapter five. Chapter six acknowledges the
limitations of the study, suggests areas for future research and contains the concluding

discussion.

CHAPTER 2
LITERATURE REVIEW

Bu er- u lier Relations

Cooperative buyer-supplier relationships have been repeatedly cited as a
significant factor contributing to the competitive strength of many Japanese
manufacturing firms (Bensaou 1997, Dyer 1996a, 1996b, Richardson 1993). Compared
to their Western counterparts, Japanese buyer-supplier relations are characterized as long-
term, highly collaborative in nature, and mutually beneficial. Relatively few suppliers are
used and buyers and suppliers often consider each other partners and work together to
solve manufacturing problems and increase productive efficiency across a wide range of
activities. The sharing of cost and other strategic information is routine in many Japanese
partnerships (Dyer 1996a, McMillan 1990). Such collaboration and information sharing
has Often led to the discovery of incremental cost saving measures which have resulted in
a competitive advantage (Dyer 1996a, 1996b, Cusumano and Takeishi 1991, Cole and
Yakushiji 1984).

In contrast, traditional buyer-supplier relations in the US. and Europe are
described as tenuous, short-tenn, non-cooperative, and often hostile. Western
manufacturers have tended to maintain numerous suppliers with whom they engage in
Short-term, arms length transactions. There is Often little information sharing coupled
with feelings of mistrust between buyers and suppliers. This traditional View is
highlighted by Ashkenas (1990) who notes that the competitive nature of buyer-supplier

relations has “led to dynamics which unintentionally result in minimization or destruction

10

of competitive value (p. 386).” Limited information sharing and a lack of joint problem
solving are noted as consequences to this competitive orientation:
“Companies in the value chain tend to withhold information from each
other such as real costs of materials, in-process quality problems, profit
margins, and all manner of problems that are viewed as ‘dirty linen’.”
(Ashkenas 1990 p. 386).“

To reap the benefits of collaboration, recent research has suggested that firms
make an effort to increase the level of cooperation in their inter-firm relations (Bensaou,
1997; Bensaou and Venkatraman, 1995; Cusumano and Takeishi, 1991; Helper, 1991;
McMillan, 1990; Ashkenas, 1990). Surveys indicate that although numerous firms in the
US. are attempting to establish more cooperative buyer-supplier relations, progress and
satisfaction vary widely (Edelman, 1997; Ellram, 1995; Richardson,'1993). As indicated
by the above quotation from Ashkenas (1990), a high level of information exchange has
been repeatedly cited as a critical factor determining the quality and success of buyer-
supplier partnerships (Ellram, 1995; Richardson, 1993). For example, in a survey of 80
buyer-supplier pairs that identified themselves as being part of a mutual “partnering”
relationship, Ellram (1995) notes that buyers ranked two-way information sharing as the
most important factor determining the success of the partnership. For suppliers, the
importance Of two-way information sharing was second only to top management support
(Ellram 1995, p. 41). Similarly, poor communication was cited by both buyers and
suppliers as the number one factor contributing to the failure of past partnering efforts.

Successful collaborative efforts among US. firms have often resulted in

substantial cost savings and process improvements. For example, Chapman et a1. (1998)

11

present the experiences of several firms who chose to work closely with a few select
suppliers. They cite a fast food chain that was successful in their efforts:

"The result has been not only annual cost reductions of 4 to 5 percent over several
years, but faster product development and innovative packaging...(p. 64)"

Also cited is an automotive manufacturer that began integrating suppliers into the
development process with substantial benefits:
"One development team that collaborated with a supplier to redesign an

instrument panel was able to cut the part count by 30 percent, halve the number of
assembly steps and materials specifications, and shrink development time from

In

years to months (p. 64) .

Given the importance of information sharing on successful collaborative efforts,
this study focuses specifically on the sharing of cost information that can be used to
discover cost minimizing options involving more than one firm. The next section briefly
reviews a subset of the cost system literature highlighting the differences in the
information provided by traditional and modern cost systems. The section also describes
how the information from different cost systems can be used to minimize costs across
firms.

Alternative Cost Systems and Cost Minimization Within and Across Firms
During the 1980’s, research revealed that traditional cost systems that focused on
direct material and direct labor costs were inadequate to manage operations that were

becoming increasingly overhead intense (Johnson and Kaplan, 1987; Anderson, 1995;

 

' Chapman et a1. (1998) also note that limiting the number of suppliers may not always
result in cost savings. The opportunities for suppliers to take advantage of their "single-
source" status may result in higher costs. They describe the experience of a pulp and
paper company which selected a single supplier to provide all of its miscellaneous parts.
The supplier charged the firm prices that were up to 70 percent higher than the market

price.

12

Swenson, 1995; Argyris and Kaplan, 1994). Traditional, volume-based cost systems
tended to allocate overhead to products by aggregating overhead costs and then assigning
it to products based on a volume-based measure, such as units produced or direct labor
hours. In an effort to measure and manage costs more effectively, more detailed cost
systems, such as activity-based costing (ABC) were advocated.

ABC systems focus on providing more detailed overhead cost information by
identifying the activities related to overhead costs and then determining a cost driver for
each activity. Once a cost per unit of each cost driver is calculated, overhead costs can be
assigned to products based on activity usage. For example, suppose the main activity of
the purchasing department is placing orders, and the related cost driver is determined to
be the number of orders placed. Once a cost per order is determined, purchasing costs
can be assigned to individual products based on the number of purchase orders that are
required. The detailed information regarding the cost driver of purchasing costs can be
used by the product manager to determine an optimal number of purchase orders for the
product. In contrast, under a traditional system the costs of a firm’s purchasing
department might be added to an aggregate overhead cost pool and then allocated to
products based on direct labor hours. The system provides little information that a
product manager can use to optimize the purchasing costs of a particular product, unless
purchasing costs are truly related to direct labor hours.

Studies of successful implementations of ABC systems have highlighted that the
more detailed overhead cost information was used by managers to make better pricing
and product mix decisions (by having a better picture of the true costs of various

products) and to manage activities better (by having more information on what drives

13

costs) (Dedera, 1996; Ness and Cucuzza, 1995; Drumheller, 1993). Drumheller (1993) in
describing how ABC was successfully implemented at a small manufacturing firm notes
that an important function of the new ABC system was to aid in the minimization of
overhead process costs:

“Determining (overhead) process costs made it clear what the cost of adding a

new part number, a new vendor, or additional manufactured parts would be.

Process costing made it possible to forecast how overhead costs would be affected

by change. Decisions about designing or redesigning a product could thus be

made to minimize overhead costs and not simply direct labor (Drumheller, 1993,

p. 23).”

A specific example of how ABC information was used to solve an optimization
problem within a production process is highlighted by Ness and Cucuzza (1995 p. 137) in
a study describing the successful implementation of ABC at Chrysler. The ABC
information was used to address the problem of how many wiring harnesses were needed
in the production of a new minivan. The options ranged from one to nine harnesses, with
different parties advocating different numbers. The design engineers, who were
evaluated on how well they kept material costs down, wanted nine harnesses so that the
minimum necessary amount of materials would be put onto each vehicle. However,
producing nine harnesses increased production complexity. Manufacturing plant
managers advocated using only one harness to minimize inventory and direct labor costs.
The key to finding the Optimal solution of two harnesses was using the ABC information
to figure out which number minimized total costs.

Similarly, other studies of ABC implementations note that success of the new

system is critically linked with better decision outcomes regarding cost minimization

(Swenson, 1995; Shields and McEwen, 1996). However, for the ABC information to be

14

used effectively, it must often be shared with other parties because the costs are often
controlled by many individuals (Drake, Haka and Ravenscrofi, 1999). A study by Drake,
Haka and Ravenscroft (1999) provided participants with ABC information and found that
if the information was not used as part of a coordinated effort across multiple individuals,
cost reductions pinpointed by the ABC information were not undertaken. The presence
of competitive incentives was found to be a major hindrance to coordination efforts. In
the current context of a buyer-supplier negotiation, a competitive orientation exists
because both parties are trying to maximize their own payoffs. This competitive element
may prevent negotiators from sharing the ABC information. Without the relevant
information, cross-firm cost savings may not be identified, much less adopted.

Cross-firm cost minimization becomes more important as firms become less
vertically integrated. These firms need to take a value chain perspective and look across
their organizational boundaries for cost saving opportunities (Cooper and Slagmulder
1998, Shank and Govindarajan, 1992). Cooper and Slagmulder (1998) note:

“For the majority of the 20lh Century, management accounting practice has limited

its scope to the boundaries of the firm. This limitation makes it difficult for the

firm to take advantage of any cost-reduction synergies that exist across the supply
chain. Such synergies can only be achieved by coordinating the cost-reduction

activities of multiple firms (p. 18).”

The problems caused by a lack of coordination between firms is highlighted in a
case described in Shank and Govindaraj an (1992, p. 20). The case involved a large US.
car manufacturer that determined it could reduce its costs at assembly plants by
implementing a just-in-time approach to the inventory provided by suppliers. Reducing

inventory buffer stocks did reduce costs at the assembly plants, but the resulting

scheduling variability imposed on the suppliers resulted in increased costs at the suppliers

15

end. The increase in supplier costs more than outweighed the cost savings experienced
by the assembly plants, resulting in higher total costs across the value chain.

In contrast, Roberts and Silvester (1996) report a case where ABC information
was used successfully to improve overall profits between a firm and its customers. Prior
to the implementation of ABC, the firm’s managers believed that the cost of placing
printed numbers and letters on the printed circuit boards it manufactured was immaterial.
The ABC information revealed that the process cost was not immaterial, and through
working with its customers management, was eventually able to eliminate it.
Importantly, the cost savings were shared with the customers.

Although the benefits to coordination between firms seem intuitive, the actual
practice of coordination and the sharing of cost and other information across firms is
difficult due to the traditionally competitive nature of buyer-supplier relations (Helper,
1991). In describing the overall cost savings that can be discovered by cross-firm
interactions, Cooper and Slagmulder ( 1998) note:

“To achieve these savings, all of the firms must be willing to cooperate, share

information, and work for the common good. Unfortunately, the adversarial

character of many buyer-supplier relationships in Western firms is not conducive

to these types ofinteractions (p. 19).“

Factors that dampen the competitive aspect of a negotiation and encourage the
parties to work together to maximize total joint profits represent an interesting avenue of
research. This study investigates whether the prospect of losses induced by a strong

competitive threat to both negotiating parties, induces them to take the risky option of

sharing their cost information and working together to find optimal production solutions.

The next section of the paper reviews literature related to prospect theory that
provides a basis for predicting that losses associated with competitive threats can induce a
greater level of cost information sharing.

Prospect Theog and Expected Utility Theog

Since the seminal work of Von Neumann and Morganstem (1944), expected
utility theory has reigned as the predominant normative model of choice under risk and
uncertainty (Kahneman and Tversky, 1979; Tversky and Kahneman, 1992). Based on a
set of plausible axioms, it implies that there exists a utility function such that an
individual’s preferences over a set of risky prospects can be ordered according to the
expected utility of the prospects (Varian, 1992). A key characteristic of expected utility
theory is that the value of a prospect is based on final wealth, regardless of the initial
wealth position of the decision maker”. This suggests that different representations of an
objectively identical choice problem (in terms of final wealth) should not affect a
subjects’ preference order (Kahneman and Tversky, 1979).

Although normatively appealing, research has uncovered systematic violations of
the predictions of expected utility theory (Kahneman and Tversky, 1979; Tversky and
Kahneman, 1986, 1991, 1992; Kahneman, Knetsch, and Thaler, 1991; Tversky and
Simonson, 1993). The results of the following two experiments in Kahneman and

Tversky (1979) implies that the carriers of value are gains and losses relative to a given

 

2 Kahneman and Tversky (1979) describe this characteristic as “asset integration,”
meaning that “a prospect is acceptable if the utility resulting from integrating the prospect
with one’s assets exceeds the utility of those assets alone. Thus, the domain of the utility
function is final states (which include one’s asset position) rather than gains or losses (p.
264)”

17

reference point, not the final amount of wealth or assets as implied by expected utility
theory.
Experiment 1: Imagine you have been given $1,000 and must choose between

A) a 50% chance of winning $1,000 and a 50% of winning $0 and .

B) winning $500 for sure.
The majority of participants (84%) chose B, the risk averse option.
Experiment 2: Imagine you have been given $2,000 and must choose between

C) a 50% chance of losing $1,000 and a 50% chance of losing $0 and

D) losing $500 for sure.

The majority of participants (69%) chose C, the risky option. However, A and C
are identical in that they involve a 50% chance of walking away with $2,000 and a 50%
chance of walking away with $1,000. Similarly B and D are identical to the certainty of
walking away with $1,500. Based on expected utility theory, if a subject prefers B to A,
she should also prefer D to C because the choice problem is the same in experiments 1
and 2 in terms of final wealth. The results of these experiments are relevant to the current
study because the presence or absence of a competitive threat induces similar frameworks
within which negotiators must make the decision to share or not share their cost
information.

Another key component to describing choice under risk and uncertainty is the

proposed shape of the utility function, v('). Although expected utility theory does not
place restrictions on the shape of the function per se, common applications of the theory

have imposed that the utility function is concave over the entire wealth domain (i.e. the

18

second derivative, v' , of the function is less than zero). There are two significant
implications of this proposed shape. First, it implies diminishing marginal utility of
wealth, such that the change in utility of receiving an additional $100 is greater at a
starting point of $10 than $10,000. Second, it implies risk aversion over the entire wealth
domain.

In contrast, prospect theory implies that the utility/value function is concave (V' <

0) for amounts above the reference point and convex (v. > 0) for amounts below the
reference point. The implication is that risk averse behavior is predicted for situations
involving amounts greater than the reference point, but risk-seeking behavior is predicted
for amounts less than the reference point (Kahneman and Tversky, 1979; Tversky and
Kahneman, 1986). This prediction is consistent with the results given above for
Experiments 1 and 2 from Kahneman and Tversky (1979). The proposed shape of the
value function also implies that the marginal utility of a change in wealth decreases with
the distance from the reference point, in either direction. Thus, the impact on utility of a
$100 less will be less at an initial loss position of-$10,000 than at an initial position of
-$10.

An additional critical aspect of prospect theory is the tenet of loss aversion. Loss
aversion implies that, relative to a given reference point, the pain of losing $100 is greater
than the pain of a foregone gain of $100. Thus, the value function is “steeper” in the
domain of losses than in the domain of gains. Overall, prospect theory predicts a non-

linear, S-Shaped value function that is asymmetric relative to a neutral reference point.

19

Prospect theory is used in this study for two reasons. First, a sample of reported
cases where buyer-supplier relations have involved the “risky” practice of information
sharing and collaborative problem solving reveals that the firms involved were often
facing strong competitive threats (Ellram and Edis, 1996; Richardson, 1993; Johnston and
Lawrence, 1988). Such threats indicate that the decision to share information with a
buyer-or supplier was made in the context of potential losses to the participants. Thus,
the risky decision to Share information would be consistent with the risk seeking behavior
under losses implied by prospect theory. In addition, prospect theory can also be used to
describe the relative willingness of subjects to share less detailed versus more detailed
cost information under the contexts of both profits and losses.

The following section presents a review of literature tying the predictions of
prospect theory to the context Of a buyer-supplier negotiation in which integrative
potential exists.

Implications of Prospect Theory for Buyer-Supplier Relations and Negotiations

Ellram and Edis (1996) describe how Kodak established a successful business
partnership with its suppliers. At the time of establishing the partnership, Kodak was
facing extremely stiff competition and price erosion. One strategy Kodak used to remain
competitive in the imaging industry was to reduce costs. An area identified as a key
source of possible cost reductions was the set of transactions and processes involving
Kodak’s suppliers. To realize these potential cost savings, Kodak engaged in an
extensive effort to change the nature of its relations with its suppliers. Instead of
engaging in a multitude of arms-length, competitively—oriented transactions with

numerous suppliers, it strove to establish more cooperative relationships with a few key

20

suppliers. The resulting “partnerships” involved a higher level of communication and
early involvement of suppliers in the product development stage. Through greater
collaboration with its suppliers, substantial cost and time savings were realized. These
improvements have helped Kodak remain profitable and competitive in the imaging
industry.

Similarly, Johnston and Lawrence (1988) describe the case of McKesson, a
wholesale health care product distributor that recognized that the independent drugstores
it supplied were in danger of being put out of business by large chain stores, such as
Kmart and Walmart. Realizing that its own future depended on the survival of its
customers, McKesson began to search for ways to help the independents stay
competitive. By actively working with its customers, several innovations and process
improvements were devised that resulted in cost savings for both itself and its customers.
McKesson and its customers became what Johnston and Lawrence (1988) describe as a
“value-adding partnership” defined as a “set of independent companies that work closely
together to manage the flow of goods and services along the entire value-added chain (p.
94).”

As indicated by the above cases, some successful buyer-supplier partnerships
have in common the presence of a competitive threat, which has had (or is predicted to
have) a significant detrimental impact on the financial condition of the involved firm(s).
In general, firms facing such threats are likely to be facing actual or potential losses for
the near future. The managers of such firms are similarly facing “losses” in the form of
lower pay if their compensation is tied to firm performance.

The competitive threats to Kodak and McKesson prompted an effort to switch

21

The competitive threats to Kodak and McKesson prompted an effort to switch
from the traditional, competitive buyer-supplier paradigm to more collaborative
relationships involving greater communication and problem solving activities. A key
aspect of such collabOrative relationships is the sharing Of information that can be used by
the parties to discover cost saving opportunities. This experimental study investigates
the decision to share cost information with a buyer or supplier during the negotiation of
the sale of a component. The sharing of cost information is critical because the context
examined involves inter-related costs and integrative potential. The costs are inter-related
because the agreed upon production options impact the costs of both firms. Integrative
potential exists because total joint profits can be increased if the optimal set of options
can be discovered and agreed upon. The combined cost information from both firms can
be used to determine which set of options minimizes total costs3. However, the sharing of
cost information is assumed to represent a “risky” Option relative to not sharing. As
detailed in the paragraphs below, the sharing of cost information is risky, due to the
negative consequences that are possible in a context where negotiators have an incentive
to maximize their own gains. The tenets of prospect theory are applied in the current
study to predict that negotiators are relatively more willing to engage in the risky action
of sharing cost information when they are faced with a competitive threat that induces a

“loss” frame of reference.

 

3 It is technically sufficient to discover the optimal set of options if only one of the parties
shares their cost information. The party with both their own and the others’ cost
information has all the information necessary to figure out the optimal set.

22

The sharing of credible cost information is considered to be the ‘risky’ choice,
relative to the status quo of keeping it confidential, because the potential profits or losses
to the sharer could exhibit a greater variance than if the information is kept private. The
assumption that sharing information is considered risky is based on past negotiation
research which notes the general hesitation of bargainers to share information revealing
their payoffs (Pruitt and Carnevale, 1993; Lax and Sebenius, 1986). Lax and Sebenius
(1986) note that negotiators realize that sharing information could lead to higher joint
gains, but fear that their share of the joint gains can be reduced if the other party is able to
use the information to negotiate a largershare of those gains.

Negotiators can use the other party’s cost information, in conjunction with their
own, to estimate the payoffs that would accrue to both parties as the result of a given
offer involving an exchange price (and any related terms and conditions). Experimental
studies of bilateral bargaining have shown that the distribution of payoff knowledge can
have a dramatic impact on resulting profit splits4 (Roth and Malouf, 1979; Roth and
Mumighan, 1982). For example, an early study by Roth and Malouf (1979) involving
binary lotteries showed that when negotiators were informed of only their own payoffs,
they tended to split a given amount of lottery tickets equally, affording each an equal
chance of winning his/her prize. However, when both negotiators knew the others’ prize,
they tended to split the lottery tickets such that the negotiator with the larger prize
received a smaller share of tickets. Thus, under full payoff information, both negotiators
ended up with approximately equal expected values. For the party with the larger prize,

the sharing of payoff information resulted in a lower expected value, while the party with

23

the smaller prize faced a greater expected value. Such experimental evidence provides
justification for the aforementioned view that sharing cost information with a buyer or
supplier in the context of negotiation is “risky” because of the potential impact on final
payoffs. Within the context of a buyer-supplier negotiation, each manager faces the risk
that if they share their cost information, they may be forced to accept a lower profit
agreement than if they had kept the information private.

The idea of applying the predictions of prospect theory to managerial decision
making was proposed by Bowman (1980, 1982) after noting an interesting paradox that
seemed contrary to economic theory. Using after-tax return on equity and its variance as
measures of return and risk, respectively, Bowman (1980) examined 85 industries over
the period 1968-1976. The results showed a negative risk/retum relation within 56 (68%)
of the industries and a positive relation for only 21 (25%). The remaining 8 industries
were indeterminate. The overall conclusion was that “in the majority of industries
studied, higher average-profit companies tended to have lower risk, i.e., variance, over
time (p. 19).” The findings appeared contrary to the standard economic argument that

risk and return should be positively correlated?

 

4 For a more extensive review of this literature see Roth, 1995.

5 Bowman (1980) notes that the firm level paradox documented in his study may be
eliminated in shareholder markets through the pricing of securities. In a large, efficient
capital market, firms with high returns and low risk will be higher priced, affording the
shareholder a lower return for a lower risk security.

24

A follow-up study (Bowman 1982) directly examined the hypothesis that troubled
companies, via their managers, take more risks than their more successful counterparts by
using content analysis of annual report data. Again, the results supported the predictions
of prospect theory. The managers of firms with lower and more variable profits took on a
greater number of risky activities, such as acquisitions, divestitures, and new projects
(Bowman, 1982). The findings were also consistent with experimental results reported
by Laughhunn, Payne, and Crum (1980) which showed that when facing below target
returns, managers preferred risk seeking behavior as opposed to risk averse behavior.
Since these early studies, the negative risk/return relation and its consistency with the
predictions of prospect theory have been replicated by several authors (Bettis and
Mahajan, 1985; Singh, 1986; Fiegenbaum and Thomas, 1988; Miller and Bromiley,
1990; Lee, 1997).

In summary, research illustrates that the threat of losses due to strong competitive
threats induce managers to engage in riskier behavior than they would under more
profitable conditions. The sharing of cost information during a negotiation represents an
example of such “risky” behavior. Based on prospect theory, the next section of the paper
hypothesizes that the proportion of managers who choose to share their cost information
will be greater under the specified loss context than under the profitable context. The
theory is also used to predict that under both contexts, the proportion of managers
choosing to share more detailed ABC information will be less than the proportion
choosing to share less detailed information. Hypotheses regarding the impact of sharing

both types of cost information on negotiation outcomes are also developed.

25

 

CHAPTER 3
HYPOTHESES
Competitive Threats and Cost Information Sharing

Given that reducing costs is a primary objective of buyers wishing to establish
collaborative relationships with suppliers (Ellram, 1995), the sharing of cost information
is the focus of the current study. Such information can be used by the parties to increase
the total available surplus and/or it can be used by one party to extract a greater share of v
any surplus from the other party. The uncertainty regarding how the other negotiator
might “use” shared cost information creates risk for the partner considering sharing, due
to the potential for both positive or negative consequences. In the current study, the
status quo of not sharing information is considered less risky because agreements are
expected to center closely around a known standard sales arrangement. Past negotiation
studies have shown that when information regarding individual preferences and payoffs is
not shared between negotiators, simple compromises are often the result (Camevale and
Pruitt, 1992). Concordant with this finding, the standard sales arrangement consists of
the compromise solution to each ofthe negotiable issues.

In accordance with prospect theory, the first hypothesis predicts that a greater
proportion of negotiators will engage in the risky behavior of sharing their cost reports
when faced with a competitive threat. The competitive threat induces a “loss” frame,
whereas the absence of such a threat results in a profitable or “gain” frame. In this
experimental study a dummy variable was used to record for each pair whether one or
both negotiators shared their cost report with the other negotiator. A pair was classified

as sharing if either one or both negotiators gave their “official” cost report to the other.

26

 

For Hypothesis 1, sharing is thus measured as the proportion of negotiating pairs in which
one or both pair members chose to share their cost report.
H1: The proportion of negotiating pairs that choose to share their cost

reports is greater under the presence of an external competitive threat than
in the absence of such a threat.

Sharing More vs. Less Detailed Cost Information

A second important question concerns the amount of information bargainers
share. Are negotiators more willing to share some cost information, rather than all? The
current study investigates this question by manipulating the level of detail contained in
the “official” cost reports that can be shared. Under all conditions, negotiators are
furnished with complete, private information regarding their own costs and cost drivers at
the start of the experiment. During the negotiation, they are given the opportunity to
provide the other party with a copy of their cost report which, depending on the
experimental condition, will have one of two possible formats. One cost report format
(ABC), provides a detailed description of all costs and drivers, while a second format
(STC) describes only the costs and drivers of materials and labor, leaving overhead as an
allocated lump-sum (on a per unit basis). The ABC format is modeled after activity-
based costing methods, while the STC fomiat is modeled after more traditional costing

methods. The content of the reports is such that all information contained in the STC

27

report is contained in the ABC report, however the ABC report contains additional
information on the costs and drivers of overheadé.

The significance of the level of detail present in the two cost reports is that it
changes the magnitude of the range of potential gains and losses that might be expected
from sharing the cost report, but does not change the midpoint of the range. Relative to
the status quo of not sharing any cost information, sharing a more detailed cost report
offers the opportunity to discover more cost saving options than Sharing a less detailed
report. Thus, the potential gains from sharing a more detailed report are expected to be
greater than those related to sharing a less detailed report. However, the magnitude of i
potential losses due to opportunism by the other negotiator is also greater (for a detailed
numerical example see the Chapter entitled Experimental Manipulations and Procedures).
Thus, the sharing of more detailed cost information has a greater outcome variance than
the sharing of less detailed information.

Given the prediction that individuals engage in riskier behavior when faced with
losses, it might be posited that managers will share more rather than less detailed cost
information when faced with a competitive threat. However, although prospect theory
predicts risk seeking behavior under losses/threats it does not necessarily imply that a
prospect with a greater symmetric range of outcomes will be valued higher than a

prospect with a smaller range with the same midpoint. Like expected utility theory, it

 

° The ABC report can be described as mathematically finer than the STC report in that it
contains all the information of the STC report and (possibly) more (Newman, 1980).
This distinction is important because any observed differences in behavior across the two
cost report formats can be attributed to more (less) information being present in the ABC
(STC) report rather than the alternative explanation that the two reports contain
completely different information.

28

supports the prediction that when faced with two gambles involving equal expected
values but with different variances, the gamble with the lower variance will be preferred
(Varian, 1992). Kalmeman and Tversky (1979) supports this prediction based on the
principle of loss aversion, meaning that losses involve a greater change in utility than that
associated with gains of equal magnitude. They note that people find symmetric bets
with a 50% chance of winning x and a 50% chance of losing x, “distinctly unattractive.”
Furthermore they note that “the aversiveness of symmetric fair bets generally increases

with the size of the stake (p. 279).” Thus, if outcome x is greater than outcome y, a fair

v'”

chance of winning or losing y is preferred to a fair chance of winning or losing x.

In the current study, subjects are informed that maintaining the status quo of not
sharing their cost report will likely lead to agreements that are at or near the standard
sales arrangement, which yields a known profit/loss amount. In considering whether or
not to move away from the status quo, potential advantages and disadvantages will be
evaluated relative to the status quo outcome. Sharing less detailed (STC) cost
information has the potential to lead to a smaller symmetric range of advantages and
disadvantages, relative to the status quo, than sharing more detailed (ABC) information7.
Thus, a relatively greater proportion of negotiating pairs can be predicted to Share STC
cost information, regardless of the presence or absence of a competitive threat. The
presence or absence of a competitive threat serves only to change the status quo reference

outcome from a negative to a positive amount, respectively. The variance, in terms of

 

7 For a more detailed description of the range of possible advantages and disadvantages to
sharing more/less detailed cost information, see the Experimental Manipulations and
Procedures Chapter and Table 2.

29

potential gains and losses relative to that status quo amount, remains the same under
either threat condition

Recall that negotiators in the current study are furnished with only one of the two
types of cost reports. Thus, it is predicted that the proportion of negotiating pairs that
choose to Share their cost reports is smaller in the group with more detailed (ABC) cost
reports than in the group with less detailed (STC) cost reports.

H2: The proportion of negotiating pairs that choose to share their cost

reports is greater among those that have the option to share an STC cost

report than among those that have the option to share an ABC cost report.

The next hypothesis addresses the question of how the predictions of H1 and H2
can be combined. Recall that H1 predicts that a greater proportion of negotiating pairs
will share cost information under the threat condition (losses) than under the no threat
condition (gains). However the principle of loss aversion suggests that, regardless of the
threat condition, a greater proportion of negotiating pairs will share under the STC cost
report condition. Thus, the overall prediction is that there will be no interaction between
the threat condition and the cost report format. If only main effects hold, the following
hypothesis can be expected:

H3: The highest proportion of negotiating pairs sharing their cost reports

occurs under the threat condition combined with the STC report format; the

lowest proportion of pairs sharing their cost reports occurs under the no

threat condition combined with the ABC report format.

Overall, under loss conditions a greater proportion of pairs share information in
general, but the proportion that Share less detailed (STC) information is greater than the

proportion that share more detailed (ABC) information. Under gain conditions, the

proportion of pairs that share is lower, but the proportion that share ABC information is

30

v”

even less than the proportion that share STC information. The importance of this result is
highlighted in the following section which describes the impact of sharing the two types

of cost reports on the optimality of the resulting agreements.

Cost Information Sharing and the thimalig of Agreements

Distributive VS. Integrative Agreements

The majority of research on negotiation has focused on distributive settings, as
opposed to those with integrative potential (Roth, 1995; Pruitt-and Camevale, 1993). A
negotiation is distributive if bargainers have a fixed amount of resources to distribute
among themselves, such that one bargainer’s gain, must result in a loss to the other. In ii
contrast, negotiations have integrative potential if the amount of resources to be
distributed can be expanded, or the total utility can be increased, with some effort or
thought on the part of the negotiators (Pruitt and Camevale, 1993). Research has shown
that negotiators often miss the integrative potential of a given situation, and settle on less
than optimal agreements (Thompson and Hastie,1990; Bazerrnan, Magliozzi and Neale,
1985). A classic example was first described by Follett in 1940. Two sisters have one
orange which they both want. They eventually agree to a simple compromise - they split
the orange in half. However, one sister wanted the peel for a cake and the other wanted
the juice to drink. If they had shared information regarding their needs and preferences,
they would have discovered they both could have been happier if one sister received all
the peel, and the other sister received all the juice.

Negotiations with integrative potential often involve numerous issues that are of
differing importance to each party, as is true in this study. As seen in Table 1 and

Appendices F and G, this study presents a multiple-issue negotiation with integrative

31

potential in that the total amount of joint profits can be expanded relative to a status quo
agreement. For example, the status quo or standard casing agreement specifies that a
casing be made of plastic. The seller’s cost of making the casing out of plastic is $15 and
the cost to the buyer of using a plastic casing is $25, for a total cost of $40. Thistotal
cost can be reduced to $30 if the negotiators can agree to make the casing out of steel. It
costs the seller $25 to make the casing out of steel, but the costs to the buyer of using a
steel casing are only $5. Thus, the joint profit can be increased by $10. Figure 2
illustrates how the total joint profits of the bargainers can be increased if integrative
choices are made on all of the six negotiable items.

Information sharing has been found to be significant factor in the creation of
integrative agreements (Pruitt and Camevale, 1993; Thompson, 1991; Thompson and
Hastie, 1990). Past studies have revealed that bargainers often go into a negotiation with
the idea that there is a fixed amount of resources to be shared and in the case of multiple
issues, they often believe the other negotiator’s preferences are in direct opposition to
their own (Brodt, 1994; Thompson. 1991; Thompson and Hastie, 1990). These beliefs
bias the parties against searching for integrative agreements. Unless information is
revealed during the negotiation, it is unlikely that the parties will reach an integrative
agreement that maximizes their joint gain. Therefore, it is hypothesized that pairs that
share their cost reports reach agreements that are closer to the optimal, integrative
agreement, than pairs that do not share. Again, a pair is considered to have shared if one
or both negotiators gives their respective cost reports to the other.

H4a: Negotiating pairs that share their cost reports reach agreements that
are closer to the optimal than pairs that do not share.

32

'-

In the current experiment, both members of each negotiating pair have the same
type of cost report that can be shared, either ABC or STC. Negotiators can reveal some
(all) of their private infOrmation if they choose to share their STC (ABC) cost report.
Sharing an ABC report enables one or both parties to discover the optimal cost
minimizing choice for each of the six negotiable casing features. Sharing an STC report
enables only the discovery of the optimal choices for three of the six features. Therefore,
it is predicted that the optimality of agreements is greater when negotiators share ABC
reports than when STC reports are shared.

H4b: Negotiating pairs that share ABC reports reach agreements that are
closer to the optimal than those that share STC reports.

In the absence of cost report sharing, it is predicted that neither the cost report type, nor
the threat condition significantly impact the level of Optimality of the agreements
reached.

H4c: There are no significant differences across either the report type or the

presence or absence of a threat in the optimality level reached by negotiating
pairs that do not share cost reports.

Cost Information Sharing and the Efficiency of Negotiation

 

Most studies of bargaining recognize that there are non-negligible costs to
continuing a negotiation, either through a longer period of time or through the exchange
of more proposals (Rapoport, Erev, and Zwick, 1995; Rubinstein, 1982). It is therefore
useful to examine the amount of time and the number of proposals necessary to reach
agreement under the various experimental conditions.

Negotiators can often implicitly share information through their offers and

counter-offers (Pruitt and Camevale, 1993; Thompson and Hastie, 1990). Thus, it is

33

r

possible that negotiators may eventually agree on a near-optimal agreement even without
explicitly sharing their cost reports. However, it is likely to take longer and/or require
more proposals than if they had explicitly shared their cost information. Thus, the
following hypothesis is proposed:

H5a: Negotiating pairs that share cost reports need less time/fewer proposals

to reach near optimal agreements than pairs that do not share their cost

reports.

In addition, it is hypothesized that negotiators that share more detailed ABC
information, and reach near optimal agreements, will need less time/ fewer proposals than
will pairs that share less detailed STC information. Negotiators may still engage in back
and forth offers and counter-offers in an attempt to discover the Optimal choices for the
overhead costs not disclosed in the STC report. This leads to the prediction of relatively
less efficient negotiations when STC reports are shared than when ABC reports are
shared.

HSb: Negotiating pairs that share ABC cost reports need less time/fewer

proposals to reach near optimal agreements than pairs that share STC cost

reports.
In the absence of cost report sharing. no differences in negotiating time or number of
proposals are predicted for pairs to reach near optimal agreements for either the report
type or the presence or absence of an external threat.

H5c: For negotiating pairs that do not share cost reports, there are no

significant differences across either the report type or the presence or

absence of a threat in the time/number of proposals necessary to reach near
optimal agreements.

34

'7

Cost lnforrnation Sharing and Negotiator Attitudes

As previously noted, information sharing has been associated with more
cooperative, trusting relationships between buyers and suppliers. The link between
information sharing and cooperative attitudes and outcome satisfaction is also supported
in experimental studies of negotiation (Brodt, 1994; Thompson and Hastie, 1990). The
discovery and adoption of integrative, mutually beneficial agreements has been found to
lead to more cohesive relations and higher satisfaction with outcomes (Bazerman et a].
1985; Bazerrnan and Neale, 1985).

The current study examines the link between explicit information sharing and
buyer-supplier relations by measuring the attitudes of negotiators toward one another
after agreements have been reached. The measures include the perceived level of
cooperation that took place during the negotiation, and overall satisfaction with the
outcome of any agreements. Given the assumption that the act of sharing one’s private
information can be considered relatively more cooperative than keeping the information
private, the following hypothesis is proposed:

H6: Negotiating pairs that share their cost reports exhibit a higher perceived
level of cooperation than pairs that don’t share their cost reports.

35

Recall that Hypothesis 4a predicts that pairs that share cost reports will reach
agreements that are closer to the optimal than pairs that don’t share their reports. If
negotiators are relatively more satisfied with agreements that are closer to the optimal, the
following can be predicted:

H7: Negotiating pairs that share their cost reports exhibit greater
satisfaction with outcomes than pairs that don’t share their reports.

The next section of the paper describes the experimental manipulations and procedures

followed in testing the above hypotheses.

 

36

CHAPTER 4
EXPERIMENTAL MANIPULATIONS AND PROCEDURES

Experimental Overview

Subjects randomly assigned to the role of either buyer or seller, negotiated the
price and physical characteristics of an intermediate component during a face-to-face
meeting. The component, or casing, was used by the buyer to produce a final product
that was sold at a fixed price. Both buyers and sellers incurred materials, labor, and
overhead costs which varied depending on the agreed upon features of the casing. In all
conditions, buyers and sellers were aware of all of their own costs and cost drivers.

Prior to the negotiation, subjects were informed of the availability of a standard
sales arrangement, consisting of a selling price and a given set of casing characteristics.
The standard sales arrangement provided a reference point by which alternative offers
could be compared. The specific payoffs to both parties from the standard sales
arrangement depended on whether or not an external “competitive threat” existed (see
Table 1). In the presence of a competitive threat, the standard arrangement resulted in
moderate losses to both parties. In the absence of a competitive threat, the standard
arrangement yielded moderate profits to both parties. Profits and losses were
manipulated via the selling prices of both the standard casing and the final product
produced by the buyer. The external threat had the effect of driving these prices down to
below profitable levels.

Throughout all experimental conditions, both parties were aware of how their
costs changed depending on the features of the casing (see Appendices F and G). In

addition, each negotiator had an “official” cost report which they could choose to share

37

r

with their partner or keep to themselves (see Appendices H, I, J and K). The level of
detail contained in the cost report varied depending on the experimental condition. In the
activity-based costing (ABC) condition, the drivers of material, labor and overhead costs
were explicitly described. In the standard costing (STC) condition, materials and labor
costs were described as in the ABC condition, but overhead costs were combined and not
explicitly described. Thus, the two cost system conditions varied the amount of detailed
information contained in the official cost report which could be shared between
negotiators. Regardless of the report format, sharing it revealed the giver’s payoff
resulting from the standard sales arrangement. In addition, if an ABC format was shared,
the receiver could calculate the giver’s payoffs to any agreement. The potential
advantage to sharing cost reports is the discovery of mutually beneficial changes to the
standard casing. As shown in Table 1, the total costs incurred across the two firms could
be reduced by $40 (15%) if the negotiators agreed on the optimal combination of easing
characteristics. Both negotiators could verbally communicate any information they
wished, however, they were informed that only information contained in the cost reports
could be taken as credible.

The primary measures taken in the study were whether subjects chose to share
their cost reports and the optimality of agreements reached. Also measured were the
efficiency of the negotiation, the level of cooperation felt between the negotiators, and the
resulting satisfaction with outcomes.

Experimental Manipulations
One of the two primary experimental manipulations is the presence or absence of

an external competitive threat. This is manipulated through both the payoffs from the

38

‘g‘rjfliiililtern'.

 

 

standard sales arrangement and by differences in the description of market conditions
given to subjects prior to negotiations. In the absence Of an external threat, profits accrue
to both parties if they adopt the standard sales arrangement. In the presence of an
external threat, losses accrue to both parties if they adopt this agreement. Table 1 and
Figure 2 contain the profits (losses) of each party under the standard sales agreement
given the presence or absence of the external threat. The descriptions of market
conditions given to subjects under the alternatives are contained in Appendices B and C.

The second of the two experimental manipulations is the ability of negotiators to
Share more versus less detailed cost information with their partners. This manipulation
was accomplished via the format of each parties’ official cost report. The ABC format
contains detailed information concerning the costs and drivers of materials, labor and
overhead as shown in Appendices H and J. The STC format, shown in Appendices I and
K, presents the cost drivers and related costs for only labor and materials, while leaving
overhead as an allocated (per unit), lump-sum cost. Thus, parties who choose to share an
ABC report reveal all information regarding production or usage costs. In contrast, those
who share an STC report can retain specific information regarding overhead costs and
drivers. It is important to note that under all experimental conditions, buyers and sellers
knew all of their own costs and drivers and used the profit calculation sheets shown in
Appendices F and G to calculate their profits/ losses from alternative proposals.

As part ofthe cost report manipulation, subjects in the ABC (STC) condition were
told that sharing their cost report was a riskier (less risky) option relative to Sharing a
report that contains partial (full) information (see Appendices B and C). Subjects were

informed that greater risk is associated with a greater range of possible payoffs. Table 2

39

shows an example of how the range of payoffs could vary depending on more detailed
(ABC) or less detailed (STC) information is shared. Take for example the casing seller
under the no threat/STC condition. If she chose to share her cost report, one possible
“best” outcome would be if she could convince the buyer to agree to the cost minimizing
choices for her labor and materials costs (i.e. tin base, red blocks, and unassembled
casings). If overhead costs remained unchanged (i.e. $65) her total costs could be
reduced to $80 yielding a profit of $65, assuming the standard casing selling price of
$145. In contrast, a possible “worst” outcome would be if the buyer forced her to accept
the cost maximizing options for labor and materials (i.e. steel base, blue blocks, and fully
assembled casings). Again, assuming overhead costs remained unchanged at $65, her
total costs could increase to $180, yielding a loss of ($35) if the selling price were $145.
Thus, the range Of payoffs to sharing an STC report under this example would be ($35) to
$65.

A similar logic can be applied to casing sellers under the no threat/ABC
condition. In this case, the “best” outcome would involve an agreement that minimizes
all of the seller’s costs, including overhead. Under such an agreement, the seller’s total
costs could be reduced to $30, yielding a profit of$115 assuming the standard selling
price of $145. The “worst” outcome would be an agreement that maximized all of the
seller’s costs resulting in total costs of $230. This agreement would yield the seller a loss
of ($85) given the standard selling price of $145. Thus, the range of possible payoffs to
sharing an ABC report is ($85) to $1 15, which is larger than the range of sharing an STC

report.

40

v

A key aspect of the environment is that the negotiable casing features involve
conflicting preferences, but only one cost-minimizing combination. For example, the
casing producer prefers that the base be made of tin, while the casing buyer prefers that
the base be make of steel. However, total costs across both buyer and seller are
minimized if they agree on a steel base.

Overall, costs can be reduced by approximately 15% ($40) if negotiators can
agree to various changes to the standard sales arrangement. The optimal agreement
involves casings made with steel bases and large red blocks that are shipped to the buyer
unassembled, double-bagged, and in three weeks from the date of agreement. Total costs
of the standard sales arrangement equal $260, while the costs of the optimal agreement
equal$220.

Experimental Procedures

To begin the experiment, subjects were randomly assigned to the role of either
buyer or seller and asked to sign the consent form shown in Appendix A. Buyers and
sellers were then separated and given instructions according to the scripts contained in
Appendices B and C. After these instructions. subjects completed the “pre-
questionnaire” shown in Appendix M to measure their perceptions of the experimental
manipulations (threat and report format) and the riskiness of sharing their cost report.

Following this initial instruction period, all subjects were brought together and
shown a scripted demonstration of a possible negotiation (see Appendices D and E). The
purpose of the demonstration was to ensure that subjects understood how to use the
proposal form shown in Appendix L and how to go about making offers and counter-

offers. Once the demonstration was completed, subjects were taken to separate rooms to

41

negotiate. Once an agreement was reached, the experimenter was notified to verify the
conditions of the agreement and the resulting profits accruing to each party.

To conclude the experiment, subjects completed the exit questionnaire shown in
Appendix N. The questions measure factors such as subjects’ attitudes toward the other
negotiator and satisfaction with outcomes. To ensure that subjects could not “lose”
money overall, they received an initial endowment of dollars at the beginning of the
experiment ($7 under the no threat condition and $9 under the threat condition).
Depending on their performance in the negotiation, subjects could increase or decrease
this pool of dollars. The amount of money that subjects received (or were required to
“pay” the experimenter) was equal to 7% of the profits or losses earned in the negotiation.
The initial dollar amounts were calculated such that the total wealth of participants was
approximately equal in both the threat and no threat conditions, taking into account the
profit or loss from the standard agreement. As noted in Appendices B and C, subjects
were informed that it was likely that their outcomes would be near the standard
agreement if they did not share their cost reports. Thus, subjects in the threat condition
would be facing a choice between not sharing their report and earning approximately $8
($9 - .07($15)) and sharing their report and facing the potential gains and losses noted in
Table 2. Similarly, subjects in the no threat condition would be facing a choice between
not sharing their report and earning approximately $8 ($7 + .07($15) and sharing their
report and facing the risky outcomes. Thus, subjects in both the threat (loss) and no threat
(gain) conditions are choosing between $8 with relative certainty (by not sharing) and an
uncertain amount (by sharing). Based on expected utility theory, the choice of whether to

share or not should be independent of the threat/no threat condition, since the absolute

42

wealth associated with the certain outcome is approximately the same in both cases. The
next section provides an example of how such a choice would be evaluated differently
based on prospect theory.
Hypothetical Valuation of Experimental Prospects
As an example8 of the predictions made by Hypotheses 1 and 2 as applied to the
current experimental context, consider a hypothetical individual whose value function,
v(x), is described by the following:
v(x) = x“4 for x 2 0 Equation 1
v(x) =x“3 forx <0
Where x is a change in wealth relative to a given reference point.
Figure 3 contains a simplified graph of the above value function. Consistent with
prospect theory, the function has the following key properties:
i) it is concave with respect to gains (v'(x) < 0, for x > 0).
ii) it is convex with respect to losses (V'(x) > O, for x <0).
iii) it is steeper for losses than for gains (v(x) < V'(-x)).
Assume also that the individual has a probability weighting function, Tt(p), defined over
the domain of objective probability, p, as follows:
7r(p) = .1 + .9(p2) for 0 < p < 1 Equation 2
7t(p)= 0 forp=0

n(p)= l forp=l

\

8 The hypothetical value and probability weighting functions are taken (in part) from
NeWman (1980).

43

 

 

The above weighting function satisfies the properties that very small probabilities will be
overweighted and moderate and large probabilities will be underweighted.

The value of a regular9 prospect can be calculated by combining the value and
probability weighting functions as follows (Kahneman and Tversky, 1979):

V(x,, p,; x2, p2) = Tt( p,)V(x,) + II( p2)v(x2) Equation 3

Where: x1 and x2 represent changes in wealth that will occur with probabilities p1
and p2 , respectively and either x13 0 S x2 , or xI 2 0 2 x2 , or pI + p2 < 1.

Recall that at the start of the experiment subjects are informed that their firms’ are
experiencing either profits or losses based on the absence or presence of a competitive
threat (see Appendices B and C). In the presence of the threat, the standard agreement
results in a loss of $15; in its absence, the standard agreement results in a profit of $15.
Subjects are informed that if they don’t share their cost information in the upcoming
negotiation, the outcome is likely to be near the standard agreement. Thus, a subject in
the threat condition is faced with earning a loss of $15 with relative certainty if they
choose to not share their cost information. A subject in the no-threat condition is faced
with earning a profit of$15 with relative certainty if they choose not to share their cost
information. As shown in Table 3. a subject with the value function described in
equation 1 would value a certain gain of$15 at 1.97 and a certain loss of$15 at -2.47.

Recall from Table 2, that the range of possible gains and losses to sharing either
cost report is symmetric around the profit or loss from the standard agreement, regardless

\

9 Kahneman and Tversky (1979) define a prospect as regular “if it is neither strictly
Positive nor strictly negative” (p. 276) meaning that either xl _<_ O 5 x2 , or x1 2 0 2 x2 , or
pr + I)2 < l.

44

 

of the threat/no threat condition. For example, if an STC report is shared, the most
favorable outcome results in a $50 gain relative to the standard agreement and the least
favorable outcome results in a $50 loss. The range associated with sharing an ABC
report is twice as large; the most favorable outcome results in a $100 gain relative to the
standard agreement and the least favorable outcome results in a $100 loss.

For simplicity, assume that the probability of the best and worst outcomes are
considered equally likely so that p = .5 in all cases. From equation 2, this moderate value
of p yields a probability weight equal to .325“). Applying Equation 3 yields a value of -
.33 to the prospect of sharing an STC report and a value of -.48 to the prospect of sharing
an ABC report, as shown in Table 3. The example supports the prediction of Hypothesis
1 that under the threat condition, the value to sharing either cost report is greater than not
sharing. Similarly, the value to not sharing either cost report is greater under the no threat
condition than sharing.

The example also supports the prediction of Hypothesis 2. Under either threat
condition, a higher (less negative) value is placed on sharing an STC report than an ABC

report. The following chapter presents the results of the experiment.

 

‘0 For moderate values of p, the weighting function will exhibit subcertainty such that
n( p) + 1t( l-p) < 1 (Kahneman and Tversky,]979).

45

CHAPTER 5
EXPERIMENTAL RESULTS AND DISCUSSION

The experiment was conducted on 53 negotiating pairs, consisting primarily of
MBA, masters and upper division undergraduate students enrolled at Michigan State
University. Each of the several experimental sessions took approximately 60 to 75
minutes. Subjects were compensated for participating in the experiment based on their
performance and the average subject earned approximately $12.30.
Hypothesis 1

Hypothesis I predicted that the proportion of negotiating pairs that chose to share
their official cost report would be higher for the group facing losses (as a result of a
competitive threat) than for the group facing profitable conditions. Table 4 Panel A
shows the percentage of pairs that chose to share their cost reports within the four
experimental conditions. Of the 26 pairs randomly assigned to the competitive threat
condition, 19 (73%) chose to share their reports. In contrast, of the 27 pairs assigned to
the profitable, no threat condition, only 10 (37%) chose to share. It is interesting to note
that of the pairs who Chose to share, all involved a mutual sharing of reports. This result
is consistent with Thompson and Hastie (1990) who also noted a reciprocity of
information sharing within a bilateral bargaining experiment with integrative potential.

Table 4 Panel B presents the ANOVA results testing the Significance of
differences in the percentage of pairs that shared cost reports across the experimental
conditions. The main effect of the threat/no threat condition was significant, yielding an
F-statistic of 8.80 with a significance level of .005. This result, combined with the lack

of a significant interaction between the threat and cost report format conditions provide

46

 

support for Hypothesis 1. The prospect of losses resulted in a significantly higher
percentage of pairs sharing their cost reports than did the prospect of profits.
Hypotheses 2

Hypothesis 2 predicted that the proportion of pairs that shared their cost reports
would be higher for the group with the option to share a less detailed (STC) cost report
than for the group with the option to share a more detailed (ABC) report. Panel A of
Table 4 shows that of the 25 pairs with the Option to share an STC report, 18 (72%) chose

to share their reports. Moreover, only 11 (39%) of the 28 pairs with the option to share

 

an ABC report chose to share. The ANOVA results shown in Panel B of Table 4 reveal
that this difference was significant (F-statistic: 7.29; p-value: .009). As with Hypothesis
1, this Significant main effect for cost report format, combined with a lack of a significant
interaction term, yielded support for Hypothesis 2.
Hypothesis 3

Hypothesis 3 predicts that the effects of the threat condition and report type will
combine in an additive manner such that negotiators facing losses and having STC
reports share those reports with the greatest frequency, while negotiators facing profits
and having ABC reports share with the lowest frequency. Panel A of Table 4 reveals
results consistent with this prediction. The highest incidence of sharing occurred under
the STC/threat condition where l l of the 12 pairs (92%) chose to share their reports.
Under the ABC/no threat condition, only 3 of the 14 pairs (21%) chose to share.

To test the significance of Hypothesis 3, a contrast coded model was applied.
Contrast coding is a method used to test the significance of a specific pattern of means

among a set of experimental conditions (Rosenthal and Rosnow 1985, Buckless and

47

Ravenscrofi 1990). By specifying the pattern of means, a more powerful test is created
which uses only one degree of freedom". To apply a contrast model, the set of
experimental means is weighted by a corresponding set of contrast codes that matches the
mean pattern to be tested. The coding must be chosen such that the set adds up to zero.
For Hypothesis 3 it is predicted that the mean in the threat/STC condition is the highest,
the mean in the no threat/ABC condition is the lowest, and the remaining means fall in
the middle of these extremes. Thus, the following contrast codes (in parentheses) were

applied”:

 

(l) Threat/STC; (-1) No Threat/ABC; (O) Threat/ABC; (0) No Threat/STC.
The results of testing the model are shown at the bottom of Panel B in Table 4 and
provide strong support for Hypothesis 3 (F-statistic, 16.38; p-Value < .001 )'3.

To test the relative significance of a particular contrast model compared to other
possible models, Rosenthal and Rosnow (1985) suggest applying the “net benefit test”.
The purpose of the net benefit test is to compare the proportion of variance accounted for
by the model with the proportion of variance expected from an average, or randomly
chosen contrast model. The difference between these two variance proportions is

considered the “net benefit” Of using the specific contrast model. The proportion of

 

" The application of contrast coding represents a more powerful test because it tests for a
specific ordering among a set of means. This is in contrast to “omnibus” F-tests that test
for any differences among a set of means, regardless of order.

'3 Other sets of contrast weights that follow the basic pattern can also be used, as long as
they sum to zero (6g. 2, -2, O, 0). However, the contrast weights applied (1, -l, 0, 0)
provide a conservative test.

'3 The total explained sum of squares, degrees of freedom, and residual mean square from
the standard ANOVA contained in Panel B of Table 4 are used to calculate the model
sum of squares and F-statistic. Exact calculations can be found in Rosenthal and
Rosnow, 1985.

48

variance accounted for by a given contrast model is equal to the squared correlation
between the contrast weights and the related cell means (r2). The proportion of variance
expected from a random contrast is simply calculated as the inverse of the degrees of
freedom (1/df) associated with the total explained sum of squares in the standard
ANOVA. For Hypothesis 3, the squared correlation between the chosen contrast weights
and the cell means was equal to .998. A random contrast applied to the current study

would be expected to account for 1/3 or .333 of the variance. Thus, the net benefit of the

 

applied contrast is equal to .998 - .333, or .665.

The net benefit calculated above can then be compared to the maximum possible
net benefit given a particular experimental design. For the current 2 x 2 design, the
maximum net benefit is calculated as (df - l)/df, or 2/3. Thus, the obtained net benefit of
.665 is over 99% of the maximum possible net benefit of .667. This suggests that there is
little chance of another contrast model (with a different mean pattern) performing as well
or better than the chosen model. To test the significance of the net benefit resulting from
a contrast model an F-statistic can also be calculated as r2(df - l)/(1 - r2), which is
distributed as F(1, df - 1) (Rosenthal and Rosnow 1985, p. 88). The F-Statistic related to
the net benefit of the current contrast model is 998, which is significant (p-value = .001).

In summary, the results provide considerable support for Hypothesis 3. The
applied contrast model accounts for a significant proportion of the obtained variance in
the standard ANOVA. In addition, the model accounts for a significant proportion of

variance as compared to other possible models.

49

H theses 4a 4b and 4c

Hypothesis 4a predicted that negotiating pairs that shared their cost reports would
reach agreements that were closer to the optimal agreement than would pairs that didn’t
share. Table 5 shows the average percentage of the maximum possible cost reduction,
relative to the standard agreement, that pairs obtained via their agreements. Recall that
the standard agreement resulted in total costs of $260 across both negotiators. The
optimal agreement would result in total costs of $220, so the maximum possible cost
reduction is $40. The percentages listed in Table 5 are the obtained cost reduction
divided by the $40 maximum. Thus, if a pair reached the optimal agreement they would
be at 100% of Optimal, whereas a pair that agreed to the standard arrangement would be
at 0% of Optimal.

Panel A of Table 5 shows that on average, pairs that did not Share cost reports
reached agreements that were only 33% of optimal. In contrast, Panel B of Table 5
shows that pairs that shared cost reports reached agreements that were 82% of optimal.
The ANOVA results presented in Panel C of Table 5 provide support for Hypothesis 4a
by revealing a Significant main effect for pairs that shared vs. pairs that didn’t (F-statistic
= 12.9; p—value = .001) and a lack of significant interaction effects.

Hypothesis 4b predicted that pairs that shared more detailed (ABC) cost
information would reach agreements that were closer to the optimal than pairs that shared
less detailed (STC) cost information. The mean optimality levels contained in Panel B of
Table 5 are consistent with this prediction. On average, pairs that shared ABC reports
reached agreements that were 92% of optimal, while pairs that shared STC reports

reached a cements that were only 76% of o timal. Also consistent with the rediction is
gr P P

50

the reduction in the standard deviation of agreements among pairs that shared ABC
information compared to pairs that shared STC information. Although the optimality
levels are in the predicted direction, a t-test revealed the difference to be insignificant (t-
statistic = 1.41; p-value = .09 (1-tailed)). However, given that the test was performed on
a substantially reduced sample size (n = 27), the failure may be due to the reduced power
of the test”. It is interesting to note that of the 11 pairs that shared ABC reports, 10

(91%) reached the optimal agreement. In contrast, only 10 out of 18 pairs (56%) that

 

shared STC reports reached the optimal agreement.

Hypothesis 4c predicted that there would be no significant differences in

 

optimality levels across the experimental conditions for pairs that did not share cost
reports. The mean optimality levels for three of the four experimental conditions shown
in Panel A of Table 5 are consistent with this prediction. The condition that produced a
noticeably different optimality level was the ABC/No Threat combination. Pairs in this
cell reached agreements that were only 10% of optimal, while pairs in the other cells
reached agreements that were roughly 52% of optimal. However, an ANOVA revealed
that there were no significant main effects or interactions for the experimental conditions,
providing support for Hypothesis 4c15 . Again, this result should be interpreted cautiously
given the reduced sample size (n=24).

A contrast coded model was employed to test the combined predictions of H4a,

H4b and H4c. As noted in the test of Hypothesis 3, such a model has the advantage of

 

'4 The power of the test of H4b was estimated at 25% using the Pearson-Hartley Power
Charts contained in Keppel (1991, p. 509-518).

51

using the entire sample size to test for a specific predicted pattern of means across all
eight possible combinations of sharing, threat, and cost report format. The following
contrast codes (in parentheses) were applied to the optimality levels shown in Panels A
and B of Table 5:

(-2) Did not share/ABC/Threat; (3) Shared/ABC/Threat

(-2) Did not share/ABC/No Threat; (3) Shared/ABC/No Threat

(-2) Did not share/STC/Threat; (l) Shared/STC/Threat

(-2) Did not share/STC/No Threat; (1) Shared/STC/No Threat

 

The results of testing the contrast model are Shown at the bottom of Panel C of Table 5.
The model was highly significant (F -statistic = 13.8; p-value = .001), providing overall
support for the combined predictions of H4a, H4b, and H4c. In addition, the net benefit
of the contrast model was calculated as the difference between the proportion of variance
accounted for by the model (.731) and the expected proportion of variance for an average,
or random model, .143). The net benefit was then compared to the maximum possible
net benefit of .86. The net benefit from the applied model represents 68.4% of the
maximum possible net benefit. The associated F -test indicates that this proportion is
significant (F-statistic = 16.3, p-value = .007).
Hypotheses 5a, 5b, and 5c

In addition to leading to optimal agreements, Hypothesis 5a predicted that

negotiators that shared their cost reports would reach near optimal agreements in less

 

'5 A contrast coded model designed to test for a significantly lower optimality level
within the ABC/No Threat cell (among only pairs that did not share cost reports) was also
insignificant.

52

time/with fewer proposals than pairs that did not share. Likewise, Hypothesis 5b
predicted that pairs that shared more detailed (ABC) information would require less
time/fewer proposals to reach near optimal agreements than pairs that shared less detailed
(STC) information. In the absence of sharing, no differences in the amount of time or
number of proposals were predicted in Hypothesis 5c. Table 6 contains the average
number of minutes necessary to reach near optimal agreements, while Table 7 contains
the average number of proposals. Near optimal agreements were defined as those that
achieved 50% or more of the total available cost savings, relative to the standard
agreement. Only 34 of the 53 pairs met this standard.

Contrary to Hypotheses 5a, Table 6 and Table 7 show little difference between
pairs that shared and pairs that did not in the number of minutes or number of proposals
needed to reach near optimal agreements. Table 6 shows that pairs that did not share took
an average of 25.5 minutes, while pairs that did share took 24.4 minutes. An ANOVA
revealed no significant effect for sharing (F-statistic=.20, p-value=.66), report type (F-
statistic=.18, p-value=.68) or threat condition (F-statistic = 1.23, p-value=.27) on the
number of minutes needed to reach near optimal agreements. Similarly, an ANOVA run
on only pairs that Shared reports failed to find a significant effect for report type (F-
statistic = 1.13, p-value = .30), thus there is no evidence to support either Hypothesis 5a
or 5b. Although an ANOVA run on only pairs that did not share showed no significant
differences in time needed to reach across the experimental conditions, the support for

Hypothesis 5c is meaningless in the absence of support for Hypotheses 5a and 5b. To test

53

the combined predictions of Hypotheses 5a, 5b and 50, a contrast coded model specified
as follows was applied:

(2) Did not share/ABC/Threat; (-3) Shared/ABC/Threat

(2) Did not share/ABC/No Threat; (-3) Shared/ABC/No Threat

(2) Did not share/STC/Threat; (-l) Shared/STC/Threat

(2) Did not share/STC/No Threat; (-1) Shared/STC/No Threat
Even with the increased power of using the contrast coded model, the predictions failed
to be supported. The model yielded an F-statistic of only .65 (p-value = .42).

Table 7 Shows that pairs that did not share reports took an average of 3.6
proposals to reach near optimal agreements, while pairs that shared took only 3.0
proposals. Although in the direction predicted by Hypothesis 5a, an ANOVA again
revealed no significant effect for sharing (F-statistic = 1.0, p-value=.32). An ANOVA
run on only pairs that shared found no support for the prediction of Hypothesis 5b that
sharing ABC reports would result in fewer proposals needed than sharing STC reports (F-
statistic = 1.22, p-value = .28). As in the case of minutes needed, support for Hypothesis
5c is moot. As a final test, a contrast coded model similar to that applied to the number
of minutes was used to test the combined predictions Of Hypotheses 5a, 5b and 5c. The
contrast model was also insignificant (F-statistic = 2.2, p-value = .14).

One explanation for the lack of support for Hypotheses 5a and 5b is that the
experiment did not measure at what time during the negotiation pairs actually exchanged
cost reports. If pairs exchanged early in the negotiation, they may have been able to use
the information to reduce the time and/or number of proposals necessary to reach

agreement. However, if pairs waited until late in the negotiation, they may not have been

54

 

able to significantly reduce these measures due to the 30 minute time constraint. Another
explanation may be that analyzing the information contained in the cost reports and
“figuring out” the optimal casing configuration took a significant amount of time itself.
Thus, minutes that would have been spent “haggling” had the reports not been shared
were instead spent on problem solving once the reports were shared. However, once
reports were shared, the exchange of proposals may have been suspended until the
partners finished analyzing the cost information. This combined scenario may explain

why the number of proposals came closer to matching the predictions than did the

 

number of minutes.

 

Hypothesis 6 - Perceived Cooperation

Hypothesis 6 predicts that pairs that share their cost reports exhibit a higher
perceived level of cooperation than pairs that don’t share. Table 8 shows the average
summed responses per pair to three exit questionnaire items pertaining to cooperation and
trust. The overall means shown in Panels A and B are consistent with the prediction —
pairs that shared had a mean score of 17.7 while those that didn’t had a mean of only
12.8. Panel C of Table 8 shows the difference to be significant, providing support for
Hypothesis 6a (F-statistic = 32.5, p-value < .001). In addition, it is interesting to note that
there was also a significant main effect for the threat condition. Pairs facing the threat
had a substantially lower perception of the level of cooperation that occurred during the
negotiation than pairs that faced no threat (F-statistic=6.86, p-value=.01), regardless of

whether reports were shared or not.

55

Hypothesis 7 - Outcome Satisfaction

Hypothesis 7 predicted that pairs that Shared cost reports would be more satisfied
with the outcome of the negotiation than pairs that did not share. Table 9 shows the
average summed responses per pair to three exit questions related to outcome satisfaction.
An examination of the overall means support the hypotheses. Pairs that shared reports
had a mean satisfaction score of 16.5, while pairs that didn’t share had a mean score of
15.0. The ANOVA results presented in Panel C of Table 9 confirm that the effect was

significant (F-statistic=9.28, p-value=.004). In addition, there were also significant main

 

effects for report type (F-statistic=5.73, p-Value=.021) and threat condition (F-
statistic=4.38, p-Value=.042). Thus, pairs that faced the threat condition were relatively
less satisfied with their outcomes than pairs that faced no threat, while pairs with ABC
reports were relatively more satisfied with their outcomes than pairs with STC reports.
The lower level of earnings to be made strictly from the negotiation in the threat
condition would explain the somewhat lower level of satisfaction across both report type
and sharing conditions. Subjects would likely be happier earning a greater absolute
amount on the negotiation than less. The effect for report type is likely be driven by the
pairs that shared reports. Given the previous support for Hypotheses 4a, 4b and 4c. it is
likely that pairs that shared ABC reports were able to garner more of the available profits,
by getting closer to the optimal agreement. than pairs that shared STC reports. An
ANOVA run on only pairs that shared reports found a marginally significant result for
report type (F-statistic=3.67, p-value=.067), while an ANOVA on pairs that didn’t share

found relatively less support for report type (F -statistic=2.6, p-value=.12).

56

Manipulation Checks

At the beginning of each experimental session, buyers and sellers were given
separate training according to the scripts contained in Appendices A and B. Immediately
following this training, all subjects were asked to complete the Pre-questionnaire shown
in Appendix M. This questionnaire were designed to assess the effectiveness of the script
in making subjects aware of the experimental manipulations and of the risks associated
with sharing their cost reports. Panels A and B of Table 10 show that subjects in all
conditions knew how to calculate their own costs and understood that they had the Option
of sharing their official cost report during the negotiation. ANOVA results indicated no
significant differences across experimental conditions for either of these two items.

Panel C of Table 10 shows the average summed responses within pairs to two
questions pertaining to the presence or absence of the competitive threat (i.e. losses). The
mean scores are in the predicted direction and an ANOVA revealed a highly significant
difference in scores between the threat and no threat conditions. Also, there was no
significant difference in scores across cost report types and no significant interaction
effect. Overall, the threat/no threat manipulation appears to have been successful.

Panel D of Table 10 contains a Similar analysis ofthe Official cost report
manipulation. Based on the average scores for the two items measuring this condition,
subjects correctly perceived whether they had an ABC or STC report to share. The
ANOVA revealed a significant effect for cost report type along with no significant effect
for threat condition, or for the interaction of report type and threat condition.

Panels E and F of Table 10 present the average scores for items pertaining to the

possible consequences and risk of sharing official cost reports. Appendices B and C

57

 

show that the training scripts for all conditions informed subjects that if they shared their
report, the other manager “could use this information to obtain a better deal for himself,
thus, reducing your profits.” Panel E Shows the average scores related to the item
designed to assess whether subjects realized that sharing their report (regardless of type or
threat condition) could result in negative consequences. The average scores in all
conditions are high, indicating that subjects agreed that there could be negative
consequences to sharing their reports. Moreover, an ANOVA revealed no significant

differences in responses across the experimental conditions.

 

The item analyzed in Panel F of Table 10 was designed to measure subjects
perception of the degree of risk to sharing their cost reports. Unlike the item in Panel E,
the related sections from the scripts in Appendices B and C differed across report type.
Subjects with ABC reports were instructed that it was relatively more risky to share their
full information reports than to share reports with partial information. Subjects with STC
reports were instructed that sharing their reports was moderately risky compared to not
sharing, or sharing a report that revealed full information. Thus, when asked how risky
they felt it would be to share their cost reports, subjects in the ABC condition should
respond with higher scores than subjects in the STC condition. The results in Panel F
support this prediction, and an ANOVA revealed that there was indeed a significant effect
for report type, but no significant effect for threat condition or for the interaction between
report type and threat condition.

Control Variables

Several possible confounding variables were measured in the exit questionnaire

shown in Appendix N. Gender, degree program, grade point average, national origin, and

58

whether participants had ever held a job involving negotiation were all assessed with
direct questions. In addition, a measure of risk aversion was obtained by having subjects
make a series of choices between a sure “win” of a fixed amount and various “gambles.”
As shown in Panels A through F of Table 11, an analysis of these variables
revealed no significant differences across the experimental conditions for grade point
average, national origin, or risk aversion. There were however, significantly more females
within the ABC report condition, as shown in Panel G of Table 11. Using gender as a

separate factor in testing the hypotheses failed to result in any significant effects for

 

gender and did not materially change the main results reported in the analysis section.
Likewise, as noted in Panel H, there were significantly more MBA students within the
STC report condition. However, using MBA status as a covariate in the analyses did not
result in any significant effects and the results for the hypotheses tests were essentially
unchanged. Finally, there were more subjects who reported that they had held a job
involving negotiation within the no threat condition (see Panel 1 of Table 11). Using this
factor as a covariate in the analyses again failed to find significance for this “factor” and

the main results of the hypotheses tests were not significantly affected.

59

CHAPTER 6
LIMITATIONS, FUTURE AREAS OF RESEARCH AND CONCLUDING
DISCUSSION
Limitations and Future Areas of Research

The experimental nature of the study results in several limitations. Only a limited
number of factors could be examined concurrently. The impact of other factors, such as
the relative power of the negotiators, the size of the transaction, past negotiation history,
etc. warrants further investigation. In addition, the experimental setting is very simplified
and further research is necessary to determine whether the results can be generalized to
less definitive settings.

Another limitation of the study also presents a future research opportunity. Due
to time constraints, the study only involved one negotiating round. An obvious extension
is to perform the experiment for several rounds to determine if the results reported
represent equilibrium behavior, or whether subsequent rounds would result in changing
behaviors, perhaps toward the sharing of cost information. Although the behavior
reported in the current study may not be equilibrium behavior, the results are still
important because the purpose of the study was to investigate factors that can impact
whether cost information is shared in a bargaining situation where the status quo has been
not sharing the information. The results may indicate the relative timing ofinformation
sharing under different contexts.

In the current experiment, both buyers and sellers had the option to share the same
type of cost report, either aggregated or disaggregated. In many situations, one firm may

possess disaggregated cost information, while the other firm only has aggregated

6O

information. An interesting question is how sharing behavior is affected when one party
is unable to reciprocate in an equivalent manner.

Another factor that may impact the relative willingness of managers to share cost
information is the relative complexity of the transaction. The task in the current study
was relatively complex, involving six cost drivers with three different levels for each
driver. There were also cost drivers that were more or less important to each negotiator.
In addition, these drivers resulted in different degrees of cost for the other negotiator. For
example, to the buyers, minimizing installation costs was important because each level of
the cost driver (material used) meant a cost saving of $20. Other items were less
important as each level of the cost driver resulted in lower cost reductions. To the seller,
the cost related to the use of different materials was relatively unimportant, because
different materials resulted in cost changes Of only $10. Thus, without having exchanged
any cost information, it would seem relatively easy for the pair to agree on the optimal
choice (steel) since it saved the buyer a relatively large amount and cost the seller a
relatively small amount. In contrast, without Sharing cost information, it may be
relatively harder for negotiators to agree on cost minimizing options where the
differences between the buyer and seller costs are closer together. Thus an interesting
hypotheses is whether more or less information sharing would be observed during the
negotiation of components of greater complexity (i.e. more cost drivers) and with smaller
possible incremental cost savings. Relatively less information sharing might be observed
for Simple components and for components where the cost saving options can be

discovered easily through the negotiating process. In contrast, more information sharing

61

might be expected for complex components and for components where there are only
small incremental cost saving opportunities possible.
Concluding Disgugsion

The objective of this study is to examine factors which influence a manager’s
decision to share credible cost information with an outside customer or supplier. Using
the assumption that managers consider the sharing of such information to be risky, the
study hypothesizes that they act in accordance with prospect theory during a bilateral
negotiation. It is predicted that managers are more willing to share partial, aggregated
information than complete, disaggregated information, and managers facing losses
brought on by a competitive threat are more willing to share information than managers
facing gains. The results support these predictions and indicate that information sharing
leads to more integrative agreements and supports improved relations between buyers and
suppliers.

The results of this study suggest that adverse conditions present an opportunity for
firms to break out of traditional, competitive interactions with customers and suppliers.
The presence of an external competitive threat induces greater information sharing and
problem solving, with the result being the optimization of processes across a wider
portion of the value chain.

The results also indicate that managers are relatively more willing to share cost
reports that are more aggregated and less detailed in terms of costs and drivers. Thus,
firms with highly refined cost systems may not realize all of the gains from buyer-

supplier transactions due to the reluctance of managers to share the information necessary

62

for inter-firm cost minimization. For firms with such refined cost systems, efforts to
encourage information sharing may be necessary, especially during profitable times.

The study contributes to the literature on the behavioral impacts of alternative cost
systems by demonstrating that a reluctance to share more detailed information can result
in less than optimal returns from implementing highly refined, disaggregated cost
systems. The experiment also contributes to the study of prospect theory by
demonstrating that its predictions can be applied in a negotiation context where the
“risky” decision involves the sharing of cost information. To the study of integrative
bargaining, it provides further evidence that the option to Share credible information
regarding preferences and payoffs with differing levels of detail has a significant impact
on the optimality of agreements.

An underlying goal of this study is to bring attention to the role that management
accounting information can play in the area of buyer-supplier relations. As the study
shows, cost information can be used to optimize processes that span more than one firm’s
boundaries. The key is to induce managers to share this cost information so it can be
used in a collaborative, problem solving manner to reduce the total costs of cross-firm
processes.

This is a particularly interesting time for studying the field ofbuyer-supplier
relations due to the changes that are occurring in how these relationships are Viewed.
Efforts to change from the traditional, competitive paradigm to one of cooperation is
likely impacted by the inherently mixed-motive nature ofbuyers-supplier interactions.
Being from separate firms, negotiating managers have a competitive motive to try to

garner as large a share of any joint profit as possible. However, there is also a

63

 

cooperative element in that joint profits may be increased if the competitive element can
be put aside in favor of collaboration and joint problem solving. The study of factors that
affect which of these two motivations takes precedence in managerial decision making

during buyer-supplier interactions represents a exciting area of future research.

 

64

 

FIGURES

H 4a, b, c Optimality of

Agreements
Presence of an H1, H3 /
External Threat , H5a, b, c Efficiency of
Inforrnatron N . .
, e egotratron
Sharing

H6 _
Level of detail / \ Percerved
H2, H3 H7

 

. . Coo eration
contarned m p

officral cost report Outcome

Satisfaction

Figure 1

Overview of Hypotheses

65

Panel A: No Threat Condition

Buyer’s profit $
$70

(.13

$30

$15

 

 

- $90 $0 $15 $30 $70 Seller’s

profit S

- $90

 

0 Point A represents the profits accruing to the buyer, $15, and seller,
$ 15, if they agree to exchange a standard casing at the standard price.

0 Points B, C, and D represent possible profit splits if they agree to
exchange a casing with the optimal combination of characteristics.

0 Point E represents the no agreement payoffs to each party, - $90.

Figure 2

Illustration of Integrative Potential

66

Panel B: Threat Condition

— $120

0E

Buyer’s Profit $

$10

-$30 -$15/
f//

 

 

 

'1 0B

.\‘ C
..\
\.
.\

$0 $10

> -$15

- $30

—$120

Seller’s Profit $

° POint A represents the losses accruing to the buyer, ($15), and seller, ($15), if they

agree to exchange a standard casing at the standard price.

' POints B, C, and D represent possible profit splits ifthey agree to exchange a casing
Wlth the optimal combination of characteristics.

. POint E represents the no agreement losses to each party, ($120).

Figure 2 (cont’d)

67

 

v(x) A

V=Xl M

A
It

$x

 

 

Where:

x is defined as changes in wealth (in dollars), relative to a
given reference point.

v is defined as the value placed on a change in wealth, x; and
v(x)=xH forx20

v(x)=xl3 forx<0

Figure 3

Hypothetical Value Function

68

TABLES

Table 1
Standard Casing Profit Calculations
Panel A: No External Competitive Threat

Beta’s Final Product selling price = $290
Alpha’s Standard Casing selling price = $145

Calculation of total available surplus:

Standard Casing

F inal Product selling price $290
Buyer’s cost of using the standard casing (130)
Seller’s cost of producing the standard casing (_ISQ)
Total surplus available $30
thimal Casing
Final Product selling price $290
Buyer’s cost of using the optimal casing (110)
Seller’s cost of producing the optimal casing (LO)
Total surplus available $70

Calculation of individual profits under the standard casing sales arrangement:

 

Buyer Seller

Revenues $290 Revenues $145
(:35ng cost (145) Production cost m
PTOduction cost I 130) Seller profit $15
Buyer profit $15

No agreement payoffs:
Seller: $90 loss

Buyer: $90 loss

69

Table 1 (cont’d)
Panel B: External Competitive Threat Present

Beta’s Final Product selling price = $230
Alpha’s Standard Casing selling price = $115

Calculation oftotal available surplus:

Standard Casing

F inal Product selling price

Buyer’s cost of using the standard casing

Seller’s cost of producing the standard casing
Total surplus available

Optimal Casing

Final Product selling price

Buyer’s cost of using the optimal casing

Seller’s cost ofproducing the optimal casing
Total surplus available

$230
(130)
rm)
$(30)

$230
(110)

1110)
$10

Calculation of individual losses under the standard casing sales arrangement.

 

M” Seller
Revenues $230 Revenues
Casing cost (1 15) Production cost
ProdUCtion cost ($1 Seller profit
BUyerpmm sun

N0 agreement payoffsr
Seller: $120 loss

BUyer: $120 loss

70

$115

1130)
$(15)

 

 

 

Table 2

Possible Payoff Ranges from Cost Report Sharing'

 

No Threat Threat
Range Range
Seller Splig
STC $(35) - $65 STC ($65) - $35
ABC $(85) - $115 ABC ($115) - $85
Buyer Buyer
STC $(35) - $65 STC ($65) - $35
ABC $(85)-$115 ABC ($115)-$85

 

*Shows that the possible payoff ranges to negotiators are relatively greater when an ABC
report is shared than when an STC report is shared. The ranges are calculated based on
each negotiators best and worst outcome agreement given the type of report shared and
the standard casing price of$115 (threat) or $145 (no threat).

71

Table 3

Hypothetical Valuation of Experimental Prospects

 

 

 

 

Threat No Threat
Share Don’t Share Share Don’t Share
STC -.33 -2.47 -.33 1.96
ABC -.48 -2.47 -.48 1.96 7

 

 

 

 

 

 

 

Cells contain the calculated value, V(x,, p,; x2, p2) of the associated prospect
or outcome and is based on a hypothetical individual whose value, v(x) and
probability weighting, Tt(p) functions are as follows:

v(x) = x"4 for x 2 0
v(x) = x"3 for x < 0
Where x is a change in wealth relative to a given reference point.

rt(p)=.l +.9(p3) for0<p<l
rt(p)= 0 forp=0
rt(p)=l forp=1
Where p = probability
and,

V(X,, pl; X3, P3) = 7t( pI)V(XI) + 7“ p2)V(X2)

Where: XI and x2 represent changes in wealth that will occur with
probabilities p1 and p3 , respectively

72

Table 4

Proportion of Pairs that Shared Reports'

Panel A: Summary Statistics2

 

 

 

 

 

 

Threat No Threat marginal
means
57% [8/14] 21% [3/14] 39% [11/28]
ABC (.51) (.43) (.50)
n=1 4 n=14 n=28
92% [11/12] 54% [7/13] 72% [18/25]
STC (.29) (.52) (.46)
n=12 n=13 n=25
marginal 73% [19/26] 37% [IO/27]
means (.45) (.49)
n=26 n=27

‘Used to test H1, H2, and H3.

H 1 : The proportion of pairs that chose to share their cost reports is greater in the threat
(loss) condition than in the no threat (gain) condition.

H2: The proportion of pairs that chose to share their cost reports is higher in the STC
(less detailed) condition than in the ABC (more detailed) condition.

H3: The highest percentage of pairs sharing their cost reports will is in the threat/STC
Condition, while the lowest percentage is in the no threat/ABC condition.

2Cells contain the percentage of pairs that shared [number of pairs that shared/total pairs
In condition], (standard deviation) and n=number of pairs per condition.

73

 

 

 

lane] B: AN OVA Results

Table 4 (cont’d)

 

 

 

/ Source Sum of df Mean F-statistic Significance
Squares Square of F

Report 1.48 1 1.48 7.29 .009

Threat 1.78 1 1.78 8.80 .005

Report x Threat .001 1 .001 .01 .93

Explained 3 .20 3 1.07 5.26 .003

Error 9.93 49 .20

Total 13.13 52 .25

Contrast Model 3.27 1 3.27 16.38 .000

 

74

Table 5
Optimality of Agreements‘

Panel A: Pairs that DID NOT Share2

 

 

 

 

 

 

 

Threat No Threat marginal
means
46% 10% 23 %
ABC (.34) (.30) (.35)
IF n=1 1 n=1 7
50% 58% 5 7%
STC (.00) (.47) (.43)
n=l n= n=7
Overall
marginal 46% 27% 33%
means (.31) ( .43) (.40)
n=7 n= 1 7 n=24

lUsed to test H4a, H4b and H4c.

H4a: Pairs that shared their cost reports will reach agreements that are closer to the
Optimal than pairs that did not share their cost reports.

H4b: Of the pairs that shared cost reports, those in the ABC (disaggregated) condition
Will reach agreements closer to the optimal than those in the STC (aggregated) condition.
H4c: Of the pairs that did not share cost reports, there will be no significant differences
in the optimality level of agreements across either the report type or threat condition.
2Cells contain the average percentage of the maximum possible cost reduction (relative to
the standard agreement) that pairs achieved, (standard deviation) and n = number of pairs
Per condition.

75

Panel B: Pairs that DID Share

Table 5 (cont’d)

 

 

 

 

 

 

 

 

 

Threat No Threat marginal
means
89% 100% 92%
ABC (.31) (.00) (.26)
n=8 n=3 n=1 1
69% 86% 76%
STC (.40) (.28) (.36)
n=1 1 n=7 n=18
Overall
marginal 78% 90% 82%
means (.37) (.24) (.33)
n=19 n=10 n=29
Panel C: ANOVA Results
Source Sum of df Mean F-statistic Significance
Squares Square of F
Share 1.53 l 1.53 12.90 .001
Report .02 l .02 .13 .717
Threat .00 l .00 .00 .999
Share x Report .35 l .35 2.97 .091
Share x Threat .14 l .14 1.19 .281
Report x Threat .12 l .12 .97 .329
Share x Report x .07 l .07 .59 .446
Threat
Explained 4.57 7 .65 5.52 .000
Error 5.32 45 .12
Total 9.90 52 .19
Contrast Model 1.63 l 1.63 13.58 .001

 

76

 

 

Table 6
Minutes to Reach Near Optimal Agreement" 2

Panel A: Pairs that DID NOT Share3

 

 

 

 

 

 

Threat No Threat marginal
means
30.0 21.5 26.6
ABC (0.0) (4.9) (5.3)
n=3 =2 n=5
25.0 24.3 24.4
STC (NA) (10.3) (8.9)
n=l n=4 n=5
Overall
marginal 28.8 23.3 25.5
means (2.5) (8.4) (7.0)
n=4 n=6 n=10

 

lNear optimal agreements are those in which 50% or more of the available cost savings
are achieved (relative to the standard agreement).

2Used to test Hypotheses 5a, 5b and 5c.

H5a: Pairs that shared their cost reports required less time to reach near optimal
agreements than pairs that did not share their reports.

HSb: Of the pairs that Shared cost reports, those that shared ABC reports required less
time to reach near Optimal agreements than pairs that shared STC reports.

H5c: For negotiating pairs that do not share cost reports, there will be no significant
differences across either the report type or the presence or absence of a threat in the time
necessary to reach near optimal agreements.

3Cells contain the average number of minutes needed to reach a near optimal agreement,
(standard deviation) and n=number of pairs per condition.

77

Table 6 (cont’d)

Panel B: Pairs that DID Share

 

 

 

 

 

 

 

 

 

Threat No Threat marginal
means
23.6 19.3 22.3
ABC (6.2) (2.1) (5.6)
n=7 n=3 n=10
27.1 24.2 25.9
STC (12.0) (8.6) (10.4)
n=8 n=6 n=14
Overall
marginal 25.5 22.6 24.4
means (9.6) (7.3) (8.8)
n=15 n= n=24
Panel C: ANOVA Results
Source Sum of df Mean F -statistic Significance
Squares Square of F
Share 15.06 1 15.06 .20 .659
Report 13.21 1 13.21 .18 .679
Threat 94.87 1 94.87 1.23 .272
Share x Report 3968 1 39.68 .53 .475
Share x Threat 1.48 l 1.48 .02 .890
Report x Threat 28.60 1 28.60 .38 .543
Share x Report x 14.68 1 14.68 .20 .663
Threat
Explained 249.72 7 35.67 .47 .845
Error 1961.34 26 75.44
Total 2211.06 33 67.00
Contrast Model 48.99 1 48.99 .65 .425

 

78

 

 

 

Table 7
Number of Proposals Needed to Reach Near Optimal Agreements“ 3

Panel A: Pairs that DID NOT Share3

 

 

 

 

 

 

Threat No Threat marginal
means
2.7 3.5 3 .0
ABC (1.5) (2.1) (1.6)
n=3 n=2 n=5
4.0 4.3 4.2
STC (NA) (1.7) 1.5)
n=l n= n=5
Overall
marginal 3.0 4.0 3.6
means (1.41) (1.7) (1.6)
n=4 n=6 n=10

'Near optimal agreements are those in which 50% or more of the available cost savings
are achieved (relative to the standard agreement).

2Used to test Hypotheses 5a, 5b and 5c.

H5a: Pairs that shared their cost reports required fewer proposals to reach near optimal
agreements than pairs that did not share their reports.

H5b: Of the pairs that shared cost reports, those that shared ABC reports required fewer
proposals to reach near optimal agreements than pairs that shared STC reports.

H5c: For negotiating pairs that do not share cost reports, there will be no significant
differences across either the report type or the presence or absence of a threat in the
number of proposals necessary to reach near optimal agreements.

3Cells contain the average number of proposals needed to reach a near optimal agreement,
(standard deviation) and n=number ofpairs per condition.

79

 

Panel B: Pairs that DID Share

Table 7 (cont’d)

 

 

 

 

 

 

 

 

 

Threat No Threat marginal
means
3.1 1.3 2.6
ABC (2.7) (.56) (2.4)
n=7 n= n=10
3.5 3.0 3.3
STC (2.0) (1.7) (1.8)
n=8 n=6 n=14
Overall
marginal 3.3 2.4 3.0
means (2.3) (1.6) (2.1)
n=15 n=9 n=24
Panel C: ANOVA Results
Source Sum of df Mean F-statistic Significance
Squares Square of F
Share 4.2 1 4.2 1.02 .321
Report 5.92 l 5.92 1.46 .238
Threat .53 l .53 .13 .721
Share x Report .00 1 .00 .00 .986
Share x Threat 4.04 l 4.04 .97 .328
Report x Threat .19 l .19 .05 .833
Share x Report x 1.26 l 1.26 .31 .583
Threat
Explained 17.50 7 2.50 .62 .737
Error 105.44 26 4.06
Total 122.94 33 3.73
Contrast Model 9.08 l 9.08 2.24 .142

 

8O

 

 

Table 8
Perceived Level of Cooperation"2

Panel A: Pairs that DID NOT Share3

 

 

 

 

 

 

marginal
Threat No Threat means
10.5 13.9 12.7
ABC (3.5) (2.9) (3.5)
n=6 n=1 1 n=1 7
7.5 13.8 12.9
STC (.00) (4.1) (4.5)
n=l n=6 n=7
Overall
marginal 10.1 13.9 12.8
means (3.4) (3.3) (3.7)
n=7 n=l7 n=24

lAverage summed responses per pair to the following questions (Cronbach’s Alpha =
.77):

l. I think the other negotiator and I cooperated as a “team” to find the most cost
effective way of

making casings.

2. During the negotiation I felt like I was competing against the other negotiator
(reverse scored).

3. Overall, I didn’t really trust the other negotiator (reverse scored).
2Used to test H6: Negotiating pairs that share their cost reports will exhibit a higher
perceived level of cooperation than pairs that did not share their cost reports.
3Cells contain the average level ofcoopcration as measured by the scale described above,
(standard deviation), and n=number ofpairs per condition.

81

 

 

Panel B: Pairs that DID Share

Table 8 (cont’d)

 

 

 

 

 

 

 

 

 

marginal
Threat No Threat means
18.2 18.0 18.1
ABC (2.6) (1.5) (2.3)
n=8 n=3 n=11
16.5 18.9 17.4
STC (3.0) (3.2) (3.2)
n=ll n=7 n=18
Overall
marginal 17.2 18.6 17.7
means (2.9) (2.8) (2.9)
n=l9 n=10 n=29
Panel C: ANOVA Results
Source Sum of df Mean F-statistic Significance
Squares Square ofF
Share 315.69 1 315.69 32.52 .000
Report 7.04 l 7.04 .73 .399
Threat 66.54 1 66.54 6.86 .012
Share x Report 2.48 1 2.48 .26 .616
Share x Threat 27.42 1 27.42 2.83 .100
Report x Threat 13.90 1 13.90 1.43 .238
Share x Report x .09 l .09 .01 .926
Threat
Explained 425.91 7 60.85 6.27 .000
Error 436.80 45 9.71
Total 862.71 52 16.59

 

 

82

 

Table 9

Satisfaction with Outcome"2

Panel A: Pairs that DID NOT Share3

 

 

 

 

 

 

 

Threat No Threat marginal
means
i 14.9 15.5 15.3
ABC (3.4) (2.4) (2.7)
n= n=ll n=1 7
10.5 15.0 14.4
STC (.00) (1.8) (2.4)
n=1 n= n=7
Overall
marginal 14.3 15.4 15.04
means (3.5) (2.2) (2.6)
n=7 n=17 n=24

lAverage summed responses per pair to the following questions (Cronbach’s Alpha =
.62):

l. I am satisfied with the amount I earned on the negotiation.

2. I believe the arrangement we agreed upon was the most cost effective way to produce
casings.

3. If I have to negotiate again, I would prefer to negotiate with the same person.
2Used to test H7: Negotiating pairs that share their cost reports will exhibit greater
satisfaction with outcomes than pairs that did not share their cost reports.
3Cells contain the average level of satisfaction as measured by the scale described above,
(standard deviation), n=number Of pairs per condition.

83

Panel B: Pairs that DID Share

Table 9 (cont’d)

 

 

 

 

 

 

 

 

 

Threat No Threat marginal
means
17.3 19.0 17.7
ABC (3.2) (2.3) (3.0)
n=8 n=3 n=1 1
15.2 16.6 15.7
STC (3.0) (2.1) (2.7)
n=11 n=7 n=18
Overall
marginal 16.1 17.3 16.5
means (3.2) (2.4) (2.9)
n=19 n=10 n=29
Panel C: ANOVA Results
Source Sum of df Mean F-statistic Significance
Squares Square ofF
Share 68.51 1 68.51 9.28 .004
Report 42.28 1 42.28 5.73 .021
Threat 32.30 1 32.30 4.38 .042
Share x Report .10 l .10 .01 .907
Share x Threat 1.87 l 1.87 .25 .617
Report x Threat 5.82 1 5.82 .79 .379
Share x Report x 8.46 l 8.46 1.15 .290
Threat
Explained 93.19 7 13.31 1.80 .110
Error 332.29 45 7.38
Total 425.47 52 8.18

 

84

 

 

Table 10
Manipulation Checks

All questions were measured on a 7 point Likert scale with 1 representing strongly
disagree and 7 representing strongly agree.

 

Panel A: Ability to calculate costs. Expectation: high scores in all conditions.

Q: During the negotiation I can calculate all of my own costs.

STC ABC
Threat 6.3 6.4
No Threat 6.7 6.3

Overall mean: 6.4. No significant differences between conditions.

 

Panel B: Ability to share report. Expectation: high scores in all conditions.

Q: During the negotiation, I have the option of giving my official cost report to the other
negotiator or keeping it to myself.

STC ABC
Threat 6.6 6.5
No Threat 6.9 6.7

Overall mean: 6.7. No significant differences between conditions.

 

Panel C: Understanding ofthreat/no threat condition (sum ofthe following 2
questions). Expectation: significantly higher scores under the threat condition.

Q: Ifthe other negotiator and 1 agree to exchange a standard casing at the standard price,
I will lose money.

Q: The industry in which the other negotiator and I operate in is currently facing strong
competition which as driven prices below profitable levels.

STC ABC
Threat 12.8 12.3
No Threat 3.4 3.6
Threat Condition: overall mean = l 1.9

No Threat Condition: overall mean = 3.1
Significantly higher scores in the threat condition (F-stat.=804.0, p-value < .001).

 

85

 

Table 10 (cont’d)

 

Panel D: Understanding of ABC/STC cost report format (sum of the following 2
questions). Expectation: significantly higher scores under the ABC condition.

Q: If I give my official cost report to the other negotiator, they will know what my labor
and materials costs are (and what drives them) but NOT what drives my overhead costs
(reverse scored).

Q: If I give my official cost report to the other negotiator, they will know what
determines ALL of my costs and what those cost amounts are.

STC ABC
Threat 3.8 11.6
No Threat 2.8 12.1

 

STC Condition: overall mean = 3.3
ABC Condition: overall mean = 11.8
Significantly higher scores under the ABC condition (F-stat.=547.8, p-value<.001).

 

Panel E: Consequences of sharing. Expectation: high scores in all conditions.

Q: If I give my Official Cost Report to the other negotiator, they may be able to use the
information to increase their own profits at my expense.

STC ABC
Threat 6.3 6.3
No Threat 5.8 6.3

Overall mean: 6.2. No significant differences between conditions.

 

Panel F: Perceived risk of sharing official cost report. Expectation: significantly
higher scores under the ABC condition.

 

Q: Please rate how risky you feel it would be to share your Official Cost Report with the
other negotiator (l=not risky at all; 7=very risky).

STC ABC
Threat 4.9 5.2
No Threat 4.3 5.5

STC Condition: overall mean = 4.6
ABC Condition: overall mean = 5.3
Significantly higher scores under the ABC condition (F-stat.=8.7, p-value = .005).

 

86

Table 1 1

ANOVA Results for Control Variables

Panel A: Grade Point Average

 

 

 

 

 

 

 

 

 

 

 

Source Sum of Df Mean F-statistic Significance
Squares Square of F

Report .20 1 .20 2.15 .19

Threat .12 1 .12 1.31 .26

Report x Threat .09 1 .09 .94 .34

Explained .41 3 .14 1.48 .23

Error 4.35 47 .09

Total 4.76 50 .10

Panel B: Subjects of US. or Canadian Origin

Source Sum of (If Mean F -statistic Significance
Squares Square of F

Report .00 l .00 .001 .98

Threat .01 1 .01 .074 .79

Report x Threat .01 1 .01 .074 .79

Explained .02 3 .01 .046 .99

Error 6.53 49 .13

Total 6.55 52 .13

Panel C: Subjects of Asian Origin

Source Sum of (if Mean F-statistic Significance
Squares Square ofF

Report .00 l .00 .01 .93

Threat .03 l .03 .28 .60

Report x Threat .03 1 .03 .28 .60

Explained .06 3 .02 .18 .91

Error 5.75 49 .12

Total 5.81 52 .1 1

 

87

 

 

 

 

 

Table 1 l (cont’d)

Panel D: Subjects of European Origin

 

 

 

Source ' Sum of (If Mean F-statistic Significance
Squares Square of F
Report .00 1 .00 .21 .65
Threat .01 1 .01 .41 .53
Report x Threat .01 1 .01 .41 .53
Explained .01 3 .01 .33 .80
Error .69 49 .01
Total .71 52 .01

 

 

 

Panel B: Subjects of Latin American Origin

 

 

 

Source Sum of (If Mean F-statistic Significance
Squares Square of F
Report .00 1 .00 .01 .91
Threat .02 l .02 2.10 .15
Report x Threat .00 1 .00 .01 .91
Explained .02 3 .01 .70 .56
Error .46 49 .01
Total .48 52 .01

 

 

Panel F: Risk Aversion

 

 

 

 

Source Sum of df Mean F—statistic Significance
Squares Square of F
Report .31 l .31 .17 .69
Threat 2.98 l 2.98 1.58 .22
Report x Threat 1.46 l 1.46 .77 .38
Explained 4.60 3 1.53 .81 .49
Error 92.38 49 1.89
Total 96.97 52 1.87

 

 

88

Table 11 (cont’d)
Panel G: Gender

 

 

 

 

Source Sum of df Mean F-statistic Significance
Squares Square of F

Report .53 l .53 4.97 .03.

Threat .02 1 .02 .14 .71

Report x Threat .06 1 .06 .59 .45

Explained .61 3 .21 1.93 . 14

Error 5.21 49 .11

Total 5.82 52 .11

'Note: Significantly more females within the ABC report condition

 

Panel H: Degree Program

 

 

 

 

Source Sum of (if Mean F-statistic Significance
Squares Square of F

Report 1.06 l 1.06 5.35 .03‘

Threat .03 l .03 .17 .69

Report x Threat .00 l .00 .01 .91

Explained 1.09 3 .36 1.93 .14

Error 9.71 49 .20

Total 10.79 52 .21

‘Note: Significantly more MBA/masters students within the STC report condition

 

 

Panel 1: Job Involving Negotiation

 

 

 

 

Source Sum of df Mean F-statistic Significance
Squares Square of F

Report .25 1 .25 2.85 .10

Threat .64 l .64 7.37 .01'

Report x Threat .17 1 .17 1.94 .17

Explained 1.05 3 .35 4.01 .01

Error 4.26 49 .09

Total 5.30 52 .10

'Note: Significantly more subjects with jobs involving negotiation within the no threat

condition

 

 

89

 

APPENDICES

 

Appendix A
Informed Consent Declaration

The purpose of this research is to investigate the processes and outcomes of
negotiation. In order to comply with University rules, it is necessary that you read and
Sign this form.

This research requires you to participate in several rounds of negotiation over the price
at which an intermediate product will be sold to an outside firm. In addition, you will
be asked some questions regarding how you went about negotiating and how you were
paid. All the information you supply will be kept confidential by the researchers and
will be seen only by them. You do not give your name on any form containing
research results. The only place you write your name is on a sheet verifying receipt of
your pay and possibly on a sign-up sheet if you would like to receive the results of the
experiment when it is completed. The researchers have no way to connect your name
on that list to any forms you fill out during the experiment or to the questionnaire you
will complete at the end.

You will be paid US. dollars for participating in this experiment, but the exact amount
will depend upon your performance. The details of how you'll be paid will be
explained shortly during a training session that precedes the actual negotiations.

You may stop at any time during the session, but if you choose to do so, you will not
be eligible for any compensation.

Along with these rights, you do incur some responsibilities by agreeing to participate.
Your primary responsibility is to perform the negotiation task to the best of your ability
and to answer honestly the questionnaire at the end of the experiment. We also ask
that you do not discuss this experiment with students outside of this group.

If you have any questions or comments concerning this research, you may contact Sue
Haka at 355-3388 or David Wright at 355-2180.

I certify that I have read and understood the rights and responsibilities 1 have incurred
as a subject in this research. Given this understanding, I voluntarily agree to
participate.

 

Print your name, please

 

 

Signature Date

90

Appendix B
Introduction Script - Casing Seller

Good afternoon and welcome to The Alpha Casing Company. I'm happy to inform
you that you've all been hired as Managers for Alpha. I'm going to spend a few
minutes now to:

1. Explain what Alpha Casing makes and how it is used by our customers.

2. Tell you about what your job is going to be at Alpha as well as the current market
conditions Alpha is facing.

3. Explain how you’ll be paid and how it relates to your performance, and finally

4. I’ll show you a demonstration of how you might perform your task to familiarize
you with the decisions you’ll be making.

So to start, what does Alpha make? (Show basic casing). Alpha Casing makes casings
which hold and protect computer chips which are manufactured by our customers
(Show chips). A standard casing is composed of a base upon which various blocks are
added. The open center Space is where our customers install their chips. Once the
casings are produced, they are packed in bags and shipped.

Your job today will be to sell casings to a computer chip manufacturer. Alpha has
several plants which operate independently, so you'll each be responsible for your own
plant and your own performance and each of you will deal with your own customers.

 

Market Condition - Threat

Currently, the industry that Alpha operates in is facing very stiff competition from
overseas producers. As a result, the selling prices of both Alpha’s casings and our
customers products have fallen to below profitable levels. However, Alpha has
decided to remain in business for the near future and has sufficient resources to
continue operating at the present loss level for the next several periods.

Market Condition - No Threat

Currently, the industry that Alpha operates in is facing no serious competition. As a
result, the selling prices of both Alpha’s casings and our customers products are at
profitable levels. Alpha has decided to remain in business for the near future and has
sufficient resources to continue operating at the present profit level for the next several
periods.

 

91

Now, let me explain how you'll be paid...

I’m going to start off by giving you each $9 (threat) or $7 (no threat), (give each
participant $9 or $7). Now, depending on how successful you are at selling casings,
you may increase or decrease this amount. If you earn a profit, I’ll pay you more
money. If you earn a loss, you’ll have to pay back some or all of the $9 (or $7) to me.
Alpha is a for profit company, and you should try to earn a profit and make as much
money as you can.

Your profits or losses are determined by the price you sell the casings for, minus the
costs of producing them. You will meet face to face with a customer and negotiate the
selling price and conditions of the sales arrangement.

Based on current market conditions, I'm going to Show you how the profits/ losses from
a sale of a Standard Casing, such as you see here, would be calculated.

 

(Hand out profit/loss calculation Sheets - Show overhead)

Please turn to the first inside page. This sheet is a sample profit calculation sheet. The
column marked proposals is where you will evaluate trial proposals that haven't been
agreed to yet. The column marked agreement is where you'll fill in your actual profit
calculation based on your final agreement.

 

 

Market Condition - Threat

Based on recent past sales, a standard casing sells for an average of $115. So suppose
you meet with the customer and you both agree to exchange a standard casing for that
price. The $115 would represent your revenue.

Market Condition - No Threat
Based on recent past sales, a standard casing sells for an average of $145. So suppose

you meet with the customer and you both agree to exchange a standard casing for that
price. The $145 would represent your revenue.

 

The cost of producing casings is based on six things: two different materials costs,
labor costs, and three types of overhead costs.

Casings are made up of two materials... a base and blocks. The cost of the base
depends on the type of material used. If the base is made of tin, it costs $5. If it is
made of plastic, it costs $15 and if it’s made of steel the cost is $25. Standard casings
are made of plastic, so the base cost would be $15.

92

Block costs are determined by what color blocks are used. If the blocks are red, the
cost is $5. If the blocks are yellow the cost is $25, and if they are blue the cost is $45.
Standard casings are made of yellow blocks, so the cost is $25.

Labor costs are based on how completely the casings are assembled prior to shipping.
If the casing is completely unassembled, labor costs are $5. If the casings are partially
assembled, meaning the caps are left off, the cost is $25. If the casing is fully
assembled the labor cost is $45. Standard casings are partially assembled so the cost is
$25.

Overhead consists of setup, supervision and handling costs. Setup costs are determined
by the size of blocks used. Setup costs are $5 if large blocks are used, $20 if medium
blocks are used and $35 if small blocks are used. Standard casings are made of
medium blocks, so setup costs are $20.

Supervision costs are based on whether the casings are made within one, two, or three
weeks. So supervision costs would be $5 if they’re made in three weeks, $25 if they’re
made in two weeks and $45 if they’re made in one week. Standard casings are made in
two weeks, so supervision costs are $25.

 

Finally, handling costs are based on whether casings are shipped unbagged, single-
bagged, or double-bagged. The cost is $5 if they are unbagged, $20 if they’re single
bagged, and $35 if they’re double-bagged. Standard casings are single-bagged, so
handling costs are $20.

 

Market Condition - Threat

Thus, the total costs of producing a basic casing would be $130 and your loss on the
sale would be $115 - $130 or ($15). Now as far as your pay goes, I’ll give you 7% of
any profits you make, but you’ll have to give me 7% of any losses. So on this deal,
you’d owe me 7% of $15, or $1.05.

 

Market Condition - No Threat

Thus, the total costs of producing a basic casing would be $130 and your profit on the
sale would be $145 - $130 or $15. Now as far as your pay goes, I’ll give you 7% of
any profits you make, but you’ll have to give me 7% of any losses. So on this deal, I’d
pay you 7% of$15, or $1.05.

 

Are there any questions about how profits or losses are calculated?
Now underneath this page are more blank profit calculation sheets you can use in each

of the meetings today. Does everyone have a pencil with an eraser and a calculator?
Again, the column marked proposals is for you to evaluate proposals during a given

93

meeting, sort of like scratch paper. The final column marked agreement is where you
should calculate your actual profit or loss from the final agreed upon arrangement.
The next thing I’m going to do is explain a little about our customers.

 

Market Condition - Threat

Beta Chips is Alpha’s primary customer. Their product, which consists of a computer
chip enclosed in a casing, sells for a fixed price of $230. (Show supplemental
information on bottom of profit/loss calculation sheet). Their profits are determined by
taking that $230 and subtracting what they pay for our casing and what it costs to insert
their chips in the casing.

If you can’t come to an agreement with Beta in the allotted time, you will both earn an
automatic loss of $120, and you’ll each have to pay me 7% of that or $8.40.

Market Condition - No Threat

 

Beta Chips is Alpha’s primary customer. Their product, which consists of a computer
chip enclosed in a casing, sells for a fixed price of $290. (Show supplemental
information on bottom of profit/ loss calculation sheet). Their profits are determined by
taking that $290 and subtracting what they pay for our casing and what it costs to insert
their chips in the casing.

If you can’t come to an agreement with Beta in the allotted time, you will both earn an
automatic loss of $90, and you’ll each have to pay me 7% of that or $6.30.

 

 

When you go to negotiate with Beta you have the option of giving them Alpha’s official
cost report (show on overhead).

 

Cost Report Format - STC

It shows Alpha’s materials and labor costs and what drives them. Overhead is shown
as just the lump sum from producing a standard casing. So if you give Beta this
information they will know exactly what your labor and materials costs are, but they
won’t know what drives your overhead costs.

Cost Report Format — ABC

It shows Alpha’s materials, labor and overhead costs and what drives them. So if you
give Beta this information they will know exactly what all of your costs are.

 

There are both advantages and disadvantages from giving the cost report to Beta. One
advantage is that it may help you and the Beta manager figure out cheaper ways to

94

produce casings, because our costs may be linked to theirs. Thus, you may be able to
increase overall profits and your own earnings. In addition, you only have 30 minutes
in which to reach an agreement, and sharing the cost report may help you reach
agreement quicker.

The disadvantage is that it will partially (STC) fully (ABC) reveal to Beta what you’ll
cam on any potential deal because they’ll know the price they’re offering you and some
(STC) all (ABC) of your costs. The Beta manager could use this information to obtain
a better deal for himself, thus, reducing your profits.

In past negotiations, the cost report has been kept confidential and the standard casing
agreement has been the result. It is predicted that if you don’t share the cost report,
your agreements and payoffs will be very close to the standard arrangement.

 

Cost Report Format - STC

Choosing to share the cost report is considered riskier than keeping it confidential
because it’s expected that if you share, your payoff may be higher or lower than if you
had not shared. However, since the report only contains partial cost information,
sharing it is less risky than if full cost information had been revealed. In summary, it is
riskier to share the report than keep it confidential, but it’s not as risky as if all the cost
information were revealed.

Cost Report Format - ABC

Choosing to share the cost report is considered much riskier than keeping it
confidential because it’s expected that if you share, your payoff may be substantially
higher or lower than if you had not shared. In summary, since the report contains full
information, it is considered the riskiest option relative to the low risk option of not
sharing, or the moderate risk associated with sharing a report that only revealed partial
information.

 

 

 

Beta also has the option of giving you a similar report on their costs of using casings
and you may request their cost report if you wish.

During negotiations both you and Beta may verbally communicate any information you
wish, however, only information contained on official cost reports is considered
reliable and verifiable. Thus, any verbal statements or information may or may not be
“true”.

Are there any questions so far? I’d just like you to fill out this questionnaire
before we go on. Pass out pre—questionnaire (Make sure pair code on sheet matches
name-tag). Now, the last thing we're going to do is show you how to fill out the
proposal forms. (Hand out sample offer forms). Bring everyone into the same room.

95

 

Appendix C
Introduction Script - Casing Buyer

Good Afternoon and Welcome to The Beta Chip Company. I'm happy to inform you
that you've all been hired as Managers for Beta. I'm going to spend a few minutes
now to:

1. Explain what Beta Chips makes

2. Tell you about what your job is going to be at Beta as well as the current market

conditions Beta is facing.

3. Explain how you'll be paid and how it relates to your performance, and finally

4. I’ll show you a demonstration of how you might perform your task to familiarize
you with the decisions you’ll be making.

So to start, what does Beta make? (Show white computer chips). Beta Chips makes
computer chips which are installed in protective casings such as this (Show basic
casing). Once the chips are inserted in the casings the final product is shipped to our
customers to use in various electronic devices. Beta manufactures the white computer
chips while the casings are purchased from an outside supplier.

Your job today will be to purchase casings from a supplier. Beta has several plants
which operate independently, so you'll each be responsible for your own plant and
your own performance and each of you will deal with your own suppliers.

 

Market Condition - Threat

Currently, the industry that Beta operates in is facing very stiff competition from
overseas producers. As a result, the selling prices of both Beta’s product and our
suppliers casings have fallen to below profitable levels. Beta has decided to remain in
business for the near future and has sufficient resources to continue operating at the
present loss level for the next several periods.

Market Condition - No Threat

Currently, the industry that Beta operates in is facing no serious competition. As a
result, the selling prices of both Beta’s product and our suppliers casings are at
profitable levels. Beta has decided to remain in business for the near future and has
sufficient resources to continue operating at the present profit level for the next several
periods.

 

96

 

Now, let me explain how you'll be paid...

I’m going to start off by giving you each $9 (threat) or $7 (no threat). (Give each
participant $9 or $7). Now, depending on how successful you are at buying casings,
you may increase or decrease this amount. If you earn a profit, I’ll pay you more
money. If you earn a loss, you’ll have to pay back some or all of the $9 (or $7) to me.
Beta is a for profit company, and you should try to earn a profit and make as much
money as you can.

Your profits or losses are determined by the selling price of the final product, which is
fixed at $230 (threat) or $290 (no threat), minus the costs of purchasing the casings
and the costs of inserting the chips. You will meet face to face with a supplier and
negotiate the purchase price of the casings and conditions of the sales arrangement.

Based on current market conditions, I'm going to Show you how the profits/losses from
selling a final product using a Standard Casing would be calculated.

(Hand out cost sheet with covers — Show overhead)

Please turn to the first inside page. This first sheet is a sample profit calculation sheet.
The column marked proposals is where you will evaluate trial proposals that haven't
been agreed upon yet. The column marked agreement is where you'll fill in your
actual profit calculation based upon your final agreement.

 

Market Condition — Threat

The selling price of Beta’s final product is fixed at $230. Thus, the $230 would
represent your revenue. Suppose you meet with the supplier and agree to purchase a
standard casing at the current average price of $115. The $115 would represent your
"Casing costs."

Market Condition - No Threat

The selling price of Beta’s final product is fixed at $290. Thus, the $290 would
represent your revenue. Suppose you meet with the supplier and agree to purchase a
standard casing at the current average price of $145. The $145 would represent your
"Casing costs. "

 

 

 

Now, the cost of inserting the chips is based on six things: materials costs, two types
of labor costs, and three types of overhead costs.

Materials costs are based on the color of blocks used in the casing. If red blocks are
used, materials costs are $35, if yellow blocks are used the cost is $20 and if blue
blocks are used the cost is $5. Standard casings are made of yellow blocks, so
materials costs would be $20.

97

Assembly labor costs are based on how completely the casings are assembled when
they are shipped from the supplier. If they are completely unassembled, the cost is
$35, if they are partially assembled, meaning the caps are left off, the cost is $20. If
they are shipped fully assembled our labor costs are $5. Standard casings are shipped
partially assembled, so our assembly costs are $20.

Installation labor costs depend on the material used for the base of the casing. If the
base is made of tin, installation costs are $45. If it is made of plastic, the cost is $25
and if it made of steel, the cost is $5. The base on standard casings is made of plastic,
so installation costs are $25.

Overhead consists of storage, inspection and testing costs. Storage costs are based on
whether the casings are made in one, two, or three weeks from the time we order them.
If they’re made in three weeks, storage costs are $25. If they’re made in two weeks,
storage costs are $15 and if they’re made in one week the cost is $5. Standard casings
are made in two weeks, so storage costs are $15.

Inspection costs are based on whether casings are shipped unbagged, single-bagged, or
double-bagged. If they’re shipped unbagged, inspection costs are $45. If they’re
shipped single-bagged, the cost is $25 and if they’re shipped double-bagged, the cost is
$5. Standard casings are single—bagged, so inspection costs are $25.

Finally, testing costs are based on the size blocks used in the casings. Testing costs are
$45 if large blocks are used, $25 if medium blocks are used, and $5 if small blocks are
used. Standard casings are made of medium blocks, so testing costs are $25.

 

Market Condition — Threat

Thus, the total costs of inserting the chips would be $130. So, your overall profit
would be $230 minus $115 to purchase the casing minus $130 to insert the chip, which
equals a loss of $15. Now as far as your pay goes, I’ll give you 7% of any profits you
make, but you’ll have to give me 7% of any losses. So on this deal, you’d owe me 7%
of $15, or $1 .05. Are there any questions about how profits or losses are calculated?

Market Condition - No Threat

Thus, the total costs of inserting the chips would be $130. So, your overall profit
would be $290 minus $145 to purchase the casing minus $130 to insert the chip, which
equals a profit of $15. Now as far as your pay goes, I’ll give you 7% of any profits you
make, but you’ll have to give me 7% of any losses. So on this deal, I’d pay you 7% of
$15, or $1.05. Are there any questions about how profits or losses are calculated?

 

98

N ow underneath this page are more blank profit calculation sheets you can use in each
of the meetings today. Does everyone have a pencil with an eraser and a calculator?
Again, the column marked proposals is for you to evaluate proposals during a given
meeting, sort of like scratch paper. The final column marked agreement is where you
should calculate your actual profit from the final agreed upon arrangement.

The next thing I’m going to do is explain a little about our suppliers.

Alpha Casing is Beta’s primary supplier. Alpha’s profits are determined by subtracting
their costs of producing the casing, from the price you agree to pay them.

 

Market Condition - Threat

If you can’t come to an agreement with Alpha in the allotted time, you will both earn
an automatic loss of $120, and you’ll each have to pay 7% of that or $8.40 (Show
supplemental information on bottom of profit/ loss calculation sheet).

Market Condition - No Threat
If you can’t come to an agreement with Alpha in the allotted time, you will both earn

an automatic loss of $90, and you’ll each have to pay 7% of that or $6.30 (Show
supplemental information on bottom of profit/ loss calculation sheet).

 

 

When you go to negotiate with Alpha, you have the option of giving them Beta’s
official cost report. (Show on overhead).

 

Cost Report Format - STC

It shows Beta’s materials and labor costs and what drives them. Overhead is shown as
just a lump sum from using a standard casing. So, if you give Alpha this information
they will know exactly what your labor and materials costs are, but they won’t know
what drives your overhead costs.

Cost Report Format - ABC

It shows Beta’s materials, labor and overhead costs and what drives them. So, if you
give Alpha this information they will know exactly what all of your costs are.

 

There are both advantages and disadvantages from giving the cost report to Alpha.
One advantage is that it may help you and the Alpha manager figure out cheaper ways
to produce casings, because our costs may be linked to theirs. Thus, you may be able
to increase overall profits and your own earnings. In addition, you only have 30
minutes in which to reach an agreement, and sharing the cost report may help you
reach agreement quicker.

99

The disadvantage is that it will partially (STC) fully (ABC) reveal to Alpha what you’ll
earn on any potential deal because they’ll know the revenue you get from selling the
final product, the amount you’ll pay for the casing and some (STC) all (ABC) of your
costs of using the casing. The Alpha manager could use this information to obtain a
better deal for himself, thus, reducing your profits.

In past negotiations, the cost report has been kept confidential and the standard casing
agreement has been the result. It is predicted that if you don’t share the cost report,
your agreements and payoffs will be very close to the standard arrangement.

 

Cost Report Format - STC

Choosing to share the cost report is considered riskier than keeping it confidential
because it’s expected that if you share, your payoff may be higher or lower than if you
had not Shared. However, since the report only contains partial cost information,
sharing it is less risky than if full cost information had been revealed. In summary, it is
riskier to share the report than keep it confidential, but it’s not as risky as if all the cost
information were revealed.

Cost Report Format - ABC

Choosing to share the cost report is considered much riskier than keeping it
confidential because it’s expected that if you share, your payoff may be substantially
higher or lower than if you had not Shared. In summary, since the report contains full
information, it is considered the riskiest option relative to the low risk option of not
sharing, or the moderate risk associated with sharing a report that only revealed partial
information.

 

Alpha also has the option of giving you a similar report on their cost of producing
casings and you may request their cost report if you wish.

During negotiations both you and Alpha may verbally communicate any information
you wish, however, only information contained on official cost reports is considered
reliable and verifiable. Thus, any verbal statements or information may or may not be

’9

“true .

Are there any questions so far? I’d just like you to fill out this questionnaire before we
go on. Pass out pre-questionnaire (Make sure pair code on sheet matches name-
tag). Now, the last thing we're going to do is Show you how to fill out the proposal
forms. (Hand out sample offer forms). Bring everyone into the same room.

100

 

 

Appendix D
Sample Negotiation Script - Threat Condition

Now, the last thing we're going to do is show you how a typical meeting might go.
(Hand out sample offer forms).

All proposals and agreements must be made out in writing on a contract proposal such
as the one you have there. In the negotiation, the buyer from Beta Chips is
responsible for making the first offer and will fill out proposal number 1. So, as a
buyer, suppose my first offer is to purchase a standard casing for $105. I'd fill out the
form like so. . .and I'd give the proposal to the casing seller. Now the numbers in this
example are strictly hypothetical and may or may not have any resemblance to actual
proposals. This just a demonstration. If the seller agrees, we'll both sign each others
copy by placing a check next to our titles at the bottom. If the seller wishes to propose
a different offer, they would fill out the next proposal form. There will be a box on
your meeting table labeled unaccepted proposals, just put the old proposals in there.
So, as a seller, suppose I'm not happy with that offer. I'd place it in the box and
suppose I counter with the offer to sell a standard casing for $125 (fill out proposal).
I'd hand the proposal to the buyer. Again, the buyer can accept it or propose
something else. So, as a buyer suppose I decide to purchase a nonstandard casing such
that the blocks are blue instead of yellow for $125. I'd fill out the form like this (fill
out proposal). Anything that differs from the standard casing has to be written in the
space marked “casing specifications”. Everything besides the block color would be the
same as in the standard casing. Now, suppose the seller agrees to this - we both place a
check by our titles on each others copy and we each keep a copy. At this point we'd
fill out the "agreement" column on our profit calculation sheet. As soon as you’re
finished, please notify one of us (we’ll be outside in the hall).

At any time, you both have the option of stopping the negotiations and taking the losses
from no agreement which were explained to you before.

Are there any questions on how to fill out the contract proposals? You’ll have
approximately 30 minutes in which to come to agreement and again, you have the

Option of sharing your official cost report with the other negotiator.

Now, due to changing conditions, you may or may nOt deal with each other in future
negotiations.

Take each pair to separate room to negotiate.

101

 

 

Appendix E
Sample Negotiation Script - No Threat Condition

Now, the last thing we're going to do is Show you how a typical meeting might go.
(Hand out sample offer forms).

All proposals and agreements must be made out in writing on a contract proposal such
as the one you have there. In the negotiation, the buyer from Beta Chips is
responsible for making the first offer and will fill out proposal number 1. So, as a
buyer, suppose my first offer is to purchase a standard casing for $135. I'd fill out the
form like so. . .and I'd give the proposal to the casing seller. Now the numbers in this
example are strictly hypothetical and may or may not have any resemblance to actual
proposals. This just a demonstration. If the seller agrees, we'll both sign each others
copy by placing a check next to our titles at the bottom. If the seller wishes to propose
a different offer, they would fill out the next proposal form. There will be a box on
your meeting table labeled unaccepted proposals, just put the old proposals in there.
So, as a seller, suppose I'm not happy with that offer. I'd place it in the box and
suppose I counter with the offer to sell a standard casing for $155 (fill out proposal).
I'd hand the proposal to the buyer. Again, the buyer can accept or propose something
else. So, as a buyer, suppose I decide to purchase a nonstandard casing such that the
blocks are blue instead of yellow for $155. I'd fill out the form like this (fill out
proposal). Anything that differs from the standard casing has to be written in the space
marked “casing specifications.” Everything besides the block color would be the same
as in the standard casing. Now, suppose the seller agrees to this - we both place a
check by our titles on each Others copy and we each keep a copy. At this point we'd
fill out the "agreement" column on our profit calculation sheet. As soon as you’re
finished, please notify one of us (we’ll be outside in the hall).

At any time, you both have the option of Stopping the negotiations and taking the
profits from no agreement which were explained to you before.

Are there any questions on how to fill out the contract proposals? You’ll have
approximately 30 minutes in which to come to agreement and again, you have the

option of sharing your official cost report with the other negotiator.

Now, due to changing conditions, you may or may not deal with each other in future
negotiations.

Take each pair to a separate room to negotiate.

102

 

 

Appendix F

Profit/Loss Calculation Sheet - Seller

 

Proposals

Final Agreement

 

Sales Revenue (Selling Price)

 

Materials Costs - Base

$5 if base is made of Tin or

$15 if base is made of Plastic or

$25 if base is made of Steel
(Standard Casing base is made of Plastic)

 

Materials Costs - Blocks
$5 if casing blocks are Red or
$25 if casing blocks are Yellow or
$45 if casing blocks are Blue
(Standard Casing blocks are Yellow)

 

Labor Costs
$5 if casings are Unassembled or
$25 if casings are Partially assembled or
$45 if casings are Fully assembled
(Standard Casings are Partially Assembled)

 

Overhead Costs - Setup
$5 if made of Large blocks or
$20 if made of Medium blocks or
$35 if made of Small blocks
(Standard Casings use Medium blocks)

 

Overhead Costs - Supervision
$5 if made in Three weeks or
$25 if made in Two weeks or
$45 if made in One week
(Standard Casings are made in Two weeks)

 

 

Overhead Costs - Handling
$5 if casings are Unbagged or
$20 if casings are Single-bagged or
$35 if casings are Double—bagged
(Standard Casings are Single-bagged)

 

Overall Profit/Loss

 

 

 

Additional Information;

Beta’s final product sells for a fixed price of $230 (threat)/$290 (no threat)

Standard Casing selling price (based on past sales):

If you can’t come to agreement;

You will earn a loss of $120 (threat)/$90 (no threat)

$115 (threat)/$145 (no threat)

The manager of Beta will earn a loss of $120 (threat)/$90 (no threat)

103

 

 

Appendix G

Profit/Loss Calculation Sheet - Casing Buyer

 

 

Proposals Final Agreement
$230 (threat) $230 (threat)
Sales Revenue $290 (no thrt) $290 (no thrt)

 

Casing costs (Purchase Price)
(Standard Casing price: $145)

 

Materials Costs:
$35 if casing blocks are Red or
$20 if casing blocks are Yellow or
$5 if casing blocks are Blue
(Standard Casings are Yellow)

 

Labor Costs - Assembly
$35 if casing are Unassembled or
$20 if casings are Partially assembled or
$5 if casings are Fully assembled
(Standard Casings are Partially assembled)

 

 

Labor Costs - Chip Installation
$45 if casing base is Tin or
$25 if casing base is Plastic or
$5 if casing base is Steel

(Standard Casing bases are Plastic)

 

Overhead - Storage Costs
$25 if casings are made in Three weeks or
$15 if casings are made in Two weeks or
$5 if casings are made in One week
(Standard Casings are made in Two weeks)

 

Overhead - Inspection Costs
$45 if casings are Unbagged or
$25 if casings are Single-bagged or
$5 if casings are Double-bagged
(Standard Casings are Single-bagged)

 

Overhead - Testing Costs
$45 if casings are made of Large blocks or
$25 if casings are made of Medium blocks
$5 if casings are made of Small blocks
(Standard casings are made of Medium blks)

 

 

Overall Profit/Loss

 

 

 

 

Additional Information: Standard Casing selling price (based on past purchases):
$115 (threat)/$l45 (no threat). If you can’t come to agreement: You will earn a
loss of $120 (threat)/$90 (no threat) and the manager of Alpha will earn a loss of
$120 (threat)/$90 (no threat).

 

104

Appendix H
Standard Casing Production Cost Report - ABC Format
The cost to produce standard casings is based on the following items:

1. Materials Costs - Base
$5 if base is made of Tin
$15 if base is made of Plastic
$25 if base is made of Steel
Standard Casing base is made of Plastic $15

2. Materials Costs - Blocks
$5 if casing blocks are Red
$25 if casing blocks are Yellow
$45 if casing blocks are Blue
Standard casing blocks are Yellow $25

 

3. Labor Costs
$5 if casings are Unassembled
$25 if casings are Partially assembled
$45 if casings are Fully assembled
Standard casings are Partially assembled $25

4. Setup Costs
$5 if casings are made of Large blocks
$20 if casings are made of Medium blocks
$35 if casings are made of Small blocks
Standard casings are made of Medium blocks $20

5. Supervision Costs
$5 if casings are made in Three weeks
$25 if casings are made in Two weeks
$45 if casings are made in One week
Standard casings are made in Two weeks $25

 

6. Handling Costs
$5 if casings are Unbagged
$20 if casings are Single-bagged
$35 if casings are Double—bagged

Standard casings are Single-bagged SA)
Standard Casing: Total Cost $130

105

Appendix I

Standard Casing Production Cost Report — STC Format

The cost to produce standard casings is based on the following items:

1. Materials Costs - Base
$5 if base is made of Tin
$15 if base is made of Plastic
$25 if base is made of Steel
Standard Casing base is made of Plastic

2. Materials Costs -

Blocks

$5 if casing blocks are Red

$25 if casing blocks are Yellow

$45 if casing blocks are Blue
Standard casing blocks are Yellow

3. Labor Costs

$5 if casings are Unassembled

$25 if casings are Partially assembled

$45 if casings are Fully assembled
Standard casings are Partially assembled

4. Overhead Costs

Standard Casing: Total Cost

106

$15

$25

$25

SQ

$130

 

Appendix J
Standard Casing Usage Cost Report - ABC Format
The cost to insert chips into a standard casing is based on the following items:

1. Materials Costs
$35 if casing blocks are Red
$20 if casing blocks are Yellow
$5 if casing blocks are Blue
Standard Casing blocks are Yellow $20

2. Labor Costs - Assembly
$35 if casings are Unassembled
$20 if casings are Partially assembled
$5 if casings are Fully assembled
Standard casings are Partially assembled $20

3. Labor Costs - Chip Installation
$45 if casing base is made of Tin
$25 if casing base is made of Plastic
$5 if casing base is made of Steel
Standard casing bases are Plastic $25

4. Storage Costs
$25 if casings are made in Three weeks
$15 if casings are made in Two weeks
$5 if casings are made in One week
Standard casings are made in two weeks $15

5. Inspection Costs
$45 if casings are Unbagged
$25 if casings are Single~bagged
$5 if casings are Double—bagged
Standard casings are Single-bagged $25

6. Testing Costs
$45 if casings are made of Large blocks
$25 if casings are made of Medium blocks
$5 if casings are made of Small blocks
Standard casings are made of Medium blocks $2;

Total cost of using a Standard Casing $130

107

 

Appendix K
Standard Casing Usage Cost Report - STC Format
The cost to insert chips into a standard casing is based on the following items:

1. Materials Costs
$35 if casing blocks are Red
$20 if casing blocks are Yellow
$5 if casing blocks are Blue
Standard Casing blocks are Yellow $20

2. Labor Costs - Assembly
$35 if casings are Unassembled
$20 if casings are Partially assembled
$5 if casings are Fully assembled
Standard casings are Partially assembled $20

3. Labor Costs - Chip Installation
$45 if casing base is made of Tin
$25 if casing base is made of Plastic
$5 if casing base is made of Steel

Standard casing bases are Plastic $25
4. Overhead Costs SL5

Total cost of using a Standard Casing $130

108

 

 

Appendix L

Contract Proposal Form

Date Time Pr

 

Instructions - Proposer: Fill out the form as you wish and then give it to the other
manager.

Instructions - Receiver: Take and examine this proposal. If you AGREE to it, tell
the other manager and place a check next to your title at the bottom of the form. Have
the other manager put a check by his title. If you do NOT AGREE, fill out the next
contract proposal.

Meeting # Proposal #

Check one:

_I propose to buy/sell the standard casing at a price of $

_I propose to buy/sell a non-standard casing with the specifications
described below at a price of $

Casing Specifications: Items not specified will remain as in the standard casing.

If you both agree to the terms above, each manager places a "check“ next to his
title AFTER you come to agreement. If you do not both agree to this offer, DO
NOT MARK.

I accept this offer: Office Use Only:

Manager, Alpha Casings Verified by:

Manager, Beta Chips

109

 

Appendix M
Pre-negotiation Questionnaire
Date/Time Pair

1. Please check one:

I am going to sell casings to a customer

I am going to buy casings from a supplier
Please answer the following questions by circling the extent to which you agree or
disagree with each statement. Circling 1 means you strongly disagree with the

statement; 4 means you neither agree nor disagree with the statement; 7 means you
strongly agree with the statement.

2. During the negotiation I can calculate all of my own costs.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

3. If the other negotiator and I agree to exchange a Standard Casing at the Standard
price, I will lose money.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

4. If I give my official cost report to the other negotiator, they will know what my
labor and materials costs are (and what drives them) but NOT what drives my overhead
costs.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

5. During the negotiation, l have the option of giving my official cost report to the
other negotiator or keeping it to myself.

1 2 3 4 5 6 7

Strongly Neutral Strongly
Disagree Agree

110

 

6. If I give my Official Cost Report to the other negotiator, they may be able to use
the information to increase their own profits at my expense.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

7. The industry in which the other negotiator and I operate in is currently facing
strong competition which has driven prices below profitable levels.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

8. If I give my Official Cost Report to the other negotiator, they will know what
determines ALL of my costs and what those cost amounts are.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

9. Please rate how risky you feel it would be to share your Official Cost Report with
the other negotiator.

1 2 3 4 5 6 7

Not risky Somewhat Very
at all risky Risky

111

 

Appendix N
Exit Questionnaire

Date/ Time Pair

 

 

1. Please check one:
I sold casings to a customer
I bought casings from a supplier
Please answer the following statistical questions. All information will be kept
anonymous and confidential.
2. Sex (circle one): M F
3. What degree are you currently pursuing (circle one):

Undergraduate Masters MBA Ph.D. _

4. Grade point:

 

5. National/Cultural Background: United States or Canada _ European
Asian Latin America Other (please specify)
6. Have you ever had a job where negotiation with others was required? Yes No

If yes, please describe
Please answer the following regarding the negotiation you just completed.
7. I gave my official cost report to the other negotiator: YES NO

8. The other negotiator gave me his/her official cost report: YES NO
9. We came to an agreement: YES NO

10. Prior to the experiment you just completed, how well did you know the other
negotiator?

1 2 3 4 5 6 7
Had never Had met Know him/her
met before him/her but very well
did not know
very well

112

11. Prior to the experiment you just completed, how would you characterize the
nature of your relationship with the other negotiator?

1 2 3 4 5 6 7
Personal Stranger Close personal
rival friend

Please answer the following questions by circling the extent to which you agree or
disagree with each statement. Circling 1 means you strongly disagree with the
statement; 4 means you neither agree nor disagree with the statement; 7 means you
strongly agree with the statement.

12. I think the other negotiator and I cooperated as a “team” to find the most cost
effective way of making casings.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

13. Overall, I didn’t really trust the other negotiator.
l 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree
14. I am satisfied with the amount I earned on the negotiation.
1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

15. IfI have to negotiate again, under the same market conditions I will NOT share my
official cost report with the other negotiator.

1 2 3 4 5 6 7

Strongly Neutral Strongly

Disagree Agree

16. If I have to negotiate again, I would prefer to negotiate with the same person.
1 2 3 4 5 6 7

Strongly Neutral Strongly
Disagree Agree

113

17. During the negotiation, I felt like I was competing against the other negotiator.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

18. I believe the arrangement we agreed upon was the most cost effective way to
produce casings.

1 2 3 4 5 6 7
Strongly Neutral Strongly
Disagree Agree

19. I think the other negotiator earned (check one)

Much more than I did
A little more than I did
About the same as I did
A little less than I did
Much less than I did

114

 

 

 

For each of the following pairs of choices, please place a check next to the option

you prefer.

1. __ winning $10 for sure or
2. __ winning $10 for sure or
3. __ winning $10 for sure or
4. __ winning $10 for sure or
5. __ winning $10 for sure or
6. __ winning $10 for sure or
7. _, winning $10 for sure or
8. __ winning $10 for sure or
9. __ winning $10 for sure or

115

a 10% chance of winning $20 and a
90% chance of winning $0

a 20% chance of winning $20 and a
80% chance of winning $0

a 30% chance of winning $20 and a
70% chance of winning $0

a 40% chance of winning $20 and a
60% chance of winning $0

a 50% chance of winning $20 and a
50% chance of winning $0

a 60% chance of winning $20 and a
40% chance of winning $0

a 70% chance of winning $20 and a
30% chance of winning $0

a 80% chance of winning $20 and a
20% chance of winning $0

a 90% chance of winning $20 and a
10% chance of winning $0

 

Appendix 0
Agreement Verification Form

Date/Time Pair Start Time End Time

 

The buyer (Beta) did/did not share their report with the seller (Alpha).
The seller (Alpha) did/did not share their report with the buyer (Beta).

 

Alpha Casing Finafigreement

 

Sales Revenue (Selling Price)

 

Materials Costs - Base
$5 if base is made of Tin or
$15 if base is made of Plastic or
$25 if base is made of Steel
(Standard Casing base is made of plastic)

 

 

Materials Costs - Blocks
$5 if casing blocks are Red or
$25 if casing blocks are Yellow or
$45 if casing blocks are Blue
(Standard Casing blocks are Yellow)

 

Labor Costs
$5 if casings are unassembled or
$25 if casings are partially assembled or
$45 if casings are fully assembled
(Standard Casings are Partially Assembled)

 

Overhead Costs - Setup
$5 if made of large blocks or
$20 if made of medium blocks or
$35 if made of small blocks
(Standard Casings use medium blocks)

 

Overhead Costs - Supervision
$5 if made in three weeks or
$25 if made in two weeks or
$45 if made in one week
(Standard Casings are made in two weeks)

 

Overhead Costs - Handling
$5 if casings are unbagged or
$20 if casings are single-bagged or
$35 if casings are double-bagged
(Standard Casings are single-bagged)

 

 

 

 

Overall Profit/ Loss x .07 =

 

 

116

 

Beta Chips

Final Agreement

 

Sales Revenue

$290(threat)/$230 (no threat)

 

Casing costs (Purchase Price)

 

Materials Costs:
$35 if casing blocks are red or
$20 if casing blocks are yellow or
$5 if casing blocks are blue
(Standard Casings are Yellow)

 

Labor Costs - Assembly
$35 if casing are unassembled or
$20 if casings are partially assembled or
$5 if casings are fully assembled
(Standard Casings are partially assembled)

 

Labor Costs - Chip Installation
$45 if casing base is tin or
$25 if casing base is plastic or
$5 if casing base is steel
(Standard Casing bases are plastic)

 

Overhead - Storage Costs
$25 if casings are made in three weeks or
$15 if casings are made in two weeks or
$5 if casings are made in one week
(Standard Casings are made in two weeks)

 

Overhead - Inspection Costs
$45 if casings are unbagged or
$25 if casings are single-bagged or
$5 if casings are double-bagged
(Standard Casings are single—bagged)

 

Overhead - Testing Costs
$45 if casings are made of large blocks or
$25 if casings are made of medium blocks
$5 if casings are made of small blocks
(Standard casings are made of medium
blocks)

 

 

Overall Profit/Loss

 

 

117

 

References

Anderson, S. 1995. A framework for assessing cost management system changes: The
case of activity-based costing implementation at General Motors, 1986-1993.
Journal of Management Accounting Research 7: 1-51.

Argyris, C., and R. Kaplan. 1994. Implementing new knowledge: The case of activity-
based costing. Accounting Horizons 8 (September): 83-105.

Ashkenas, R. 1990. A new paradigm for customer and supplier relationships. Human
Resource Management 29 (4): 385-396.

Bazerman, M. 1983. Negotiator judgment: a critical look at the rationality assumption.
American Behavioral Scientist 27 (2): 211—228.

Bazerman, M., T. Magliozzi, and M. Neale. 1985. Integrative bargaining in a
competitive market. Organizational Behavior and Human Decision Processes
35: 294-313.

Bensaou, M. 1997. Interorganizational cooperation: The role of information technology
- An empirical comparison of US. and Japanese supplier relations. Information
Systems Research 8 (2): 107-124.

Bensaou, M. and N. Venkatraman. 1995. Configurations of interorganizational
relationships: A comparison between US. and Japanese automakers.
Management Science (September): 1471 -1492.

Bettis, R., and V. Mahajan. 1985. Risk/return performance of diversified firms.
Management Science 31 (7): 785-799.

Bowman, E. 1980. A risk/return paradox for strategic management. Sloan Management
Review 21: 17-31

. 1982. Risk seeking by troubled firms. Sloan Management Review 23: 33-42.

Brodt, S. 1994. “Inside information” and negotiator decision behavior. Organizational
Behavior and Human Decision Processes 58: 172-202.

Buckless, F. A., and S. P. Ravenscrofi. 1990. Contrast coding: A refinement of
ANOVA in behavioral analysis. The Accounting Review 65 (4): 933-945.

Camevale, P., and D. Pruitt. 1992. Negotiation and mediation. Annual Review of
Psychology 43: 531-582.

Chapman, T., J. Dempsey, G. Ramsdell, and M. Reopel. 1998. Purchasing and supply
management: no time for “lone rangers." Supply Chain Management Review
(Winter): 64-71.

Cole, R., and T. Yakushiji, eds. 1984. The American and Japanese Auto Industries in
Transition. Ann Arbor, MI: University of Michigan.

Cooper, R., and R. Slagmulder. 1998. Strategic cost management: Cost management
beyond the boundaries of the firm. Management Accounting (March): 18-20.

Cusumano, M., and A. Takeishi. 1991. Supplier relations and management: A survey of
Japanese, Japanese-transplant, and US. auto plants. Strategic Management
Journal 12 (8): 563-588.

Dawes, R. 1988. Rational Choice in an Uncertain World. Fort Worth, TX: Harcourt,
Brace, Jovanovich, Inc.

Dedera, C. 1996. Harris Semiconductor ABC: Worldwide implementation and total
integration. Journal of Cost Management 10(1): 44-58.

118

 

 

Drake, A., S. Haka, and S. Ravenscrofi. 1999. Cost system and incentive structure
effects on innovation, efficiency and profitability in teams. The Accounting
Review 74 (3): 323-345.

Drumheller, H.K. 1993. Making activity-based costing practical. Journal of Cost
Management 7 (2): 21-27.

Dyer, J. 1996a. Specialized supplier networks as a source of competitive advantage:
Evidence from the auto industry. Strategic Management Journal 17 (4): 271-
291.

. 1996b. Does governance matter? Keiretsu alliances and asset specificity as
sources of Japanese competitive advantage. Organization Science 7 (6): 649-
666.

Edelman, K. 1997. Suppliers are from Saturn, customers are from Jupiter. Across the

Board 34 (9): 24-29.

Ellram, L. 1995. Partnering pitfalls and success factors. International Journal of
Purchasing and Materials Management 31 (2): 36-44.

Ellram, L., and O. R. V. Edis. 1996. A case study of successful partnering
implementation. International Journal of Purchasing and Materials Management
32 (4): 20-28.

Fiegenbaum, A., S. Hart, and D. Schendel. 1996. Strategic reference point theory.
Strategic Management Journal 17: 219-235.

Fiegenbaum, A., and H. Thomas. 1988. Attitudes toward risk and the risk-retum
paradox: Prospect theory explanations. Academy of Management Journal 31 (1):
85-106.

Foster, G. and D. Swenson. 1997. Measuring the success of activity-based management
and its determinants. Journal of Management Accounting Research 9: 109-142.

Gupta, M. and R. King. 1997. An experimental investigation ofthe effect ofcost
information and feedback on product cost decisions. Contemporary Accounting
Research 14(1): 99—127.

Helper, S. 1991. How much has really changed between US. automakers and their
suppliers? Sloan Management Review 32 (4): 15-28.

Johnson, HT, and RS. Kaplan. 1987. Relevance Lost: The Rise and Fall of
Management Accounting. Boston, Massachusetts: Harvard Business School
Press.

Johnston, R. and P. Lawrence. 1988. Beyond vertical integration - the rise ofthe value-
adding partnership. Harvard Business Review 66 (4): 94-101.

Kahneman, D., and A. Tversky. 1979. Prospect theory: An analysis of decision under
risk. Econometrica 47: 263—291.

Kahneman, D., J. Knetsch, and R. Thaler. 1991. The endowment effect, loss aversion
and the status quo bias. Journal of Economic Perspectives 5 (1): 193-206.

Kaplan, R.S., D. Weiss, and E. Desheh. 1997. Transfer pricing with ABC. Management
Accounting (May): 20-28.

Laughhunn, D., J. Payne, and R. Crum. 1980. Managerial risk preferences for below-
target returns. Management Science (26): 1238-1249.

Lax, D., and J. Sebenius. 1986. The Manager as Negotiator. New York, NY: The Free
Press.

119

 

Lee, D. 1997. The impact of poor performance on risk-taking attitudes: A longitudinal
study with a PLS causal modeling approach. Decision Sciences 28 (1): 59-80.

Luce, R. D., and D. von Winterfeldt. 1994. What common ground exists for descriptive,
prescriptive and normative utility theories? Management Science 40 (2): 263-
279.

McGowen, A. S., and T. P. Klammer. 1997. Satisfaction with activity-based cost
management implementation. Journal of Accounting Research 9: 217-238.

McKinnon, SM. and W.J. Bruns, Jr. 1992. The Information Mosaic. Boston, MA:
Harvard Business School Press.

McMillan, J. 1990. Managing suppliers: Incentive systems in Japanese and US.
industry. California Management Review 32 (4): 38-55.

Miller, K., and P. Bromiley. 1990. Strategic risk and corporate performance: An analysis
of alternative risk measures. Academy of Management Journal 33 (4): 756-779.

Neale, M., and M. Bazerman. 1985. The effects of framing and negotiator
overconfidence on bargaining behaviors and outcomes. Academy of Management
Journal 28 (1): 34-49.

Ness, J. A., and T. G. Cucuzza. 1995. Tapping the full potential of ABC. Harvard
Business Review 73 (J uly-August): 130-138.

Newman, D. 1980. Prospect theory: Implications for information evaluation.
Accounting, Organizations, and Society 5 (2): 217-230.

Ochs, J ., and A. Roth. 1989. An experimental study of sequential bargaining. The
American Economic Review (June): 355-384.

Olsen, R. 1997. Prospect theory as an explanation of risky choice by professional
investors: some evidence. Review of Financial Economics 6 (2): 225-232.

Payne, J ., Laughhunn, D., and Crum, R. 1980. Translation of gambles and aspiration
level effects in risky choice behavior. Management Science 26 (10): 1039-1060.

Pruitt, D., and P. Camevale. 1993. Negotiation in Social Conflict. Pacific Grove, CA:
Brooks/Cole.

Rapoport, A., I. Erev, and R. Zwick. 1995. An experimental study of buyer-seller
negotiation with one-sided incomplete information and time discounting.
Management Science 41 (3): 377-394.

Richardson, J. 1993. Restructuring supplier relationships in US. manufacturing for
improved quality. Management International Review 33 (Special Issue): 53-67.

Roberts, M.W., and K.J. Silvester. 1996. Why ABC failed and how it may yet succeed.
Journal of Cost Management 9 (4): 23-35.

Rosenthal, R., and R. L. Rosnow. 1985. Contrast Analysis: Focused Comparisons in
the Analysis of Variance. Cambridge, U. K.: Cambridge University Press.

Roth, A. 1995. Bargaining Experiments. In The Handbook of Experimental Economics,
edited by J .H. Kagel and A. Roth. Princeton, NJ: Princeton University Press.

Roth, A., and M. Malouf. 1979. Game-theoretic models and the role of information in
bargaining. Psychological Review 86: 574-594.

Roth, A., and J .K. Mumighan. 1982. The role of information in bargaining: An
experimental study. Econometrica 50 (5): 1123-1142.

Rubinstein, A. 1982. Perfect equilibrium in a bargaining model. Econometrica 50: 97-
109.

 

 

120

Shank, J .K., and V. Govindarajan. 1992. Strategic cost management and the value chain.
Journal of Cost Management (Winter): 5-21.

Shields, M.D., and MA. McEwen. 1996. Implementing activity-based costing systems
successfully. Journal of Cost Management 9 (4): 15-22.

Simonson, I., and A. Tversky. 1992. Choice in context: Tradeoff contrast and
extremeness aversion. Journal of Marketing Research 29 (3): 281-295.

Singh, J. 1986. Performance, slack, and risk taking in organizational decision making.
Academy of Management Journal 29 (3): 562-585.

Stewart, T. 1995. The information wars: What you don’t know will hurt you. Fortune
131(11): 119-121.

Swenson, D. 1995. The benefits of activity-based cost management to the manufacturing
industry. Journal of Management Accounting Research 7: 167-180.

Thompson, L. 1991. Information exchange in negotiation. Journal of Experimental
Social Psychology 27 (2): 161-179.

Thompson, L., and R. Hastie. 1990. Social perception in negotiation. Organizational
Behavior and Human Decision Processes 47: 98-123.

Tversky, A., and D. Kahneman. 1986. Rational choice and the framing of decisions.
Journal of Business 59 (4): 8251-8278.

. 1991. Loss aversion in riskless choice: A reference dependent model. The
Quarterly Journal of Economics 56 (4): 1039-1061.
. 1992. Advances in prospect theory: Cumulative representation of uncertainty.

Journal of Risk and Uncertainty 5: 297-323.

Tversky, A., Slovic, P., and D. Kahneman. 1990. The causes of preference reversal. The
American Economic Review 80 (1): 204-217.

Tversky, A., and I. Simonson. 1993. Context-dependent preferences. Management
Science 39 (10): 1179-1189.

Varian, H. 1992. Microeconomie Analysis, 3'd edition. New York, NY: W.W. Norton
& Company, Inc.

Von Neumann, J ., and O. Morgenstem. 1944. Theory of Games and Economic
Behavior. Princeton, NJ: Princeton University Press.

Wiseman, R., and L. Gomez-Mejia. 1998. A behavioral agency model of managerial
risk taking. Academy ofManagement Review 23 (1): 133-153.

121

 

 

IIIIIIIIIIIIIIIIIIIIIIIII

ill/11111!lllllHill1111111111111111111