You are on page 1of 20

Teaching Ideas

Bargaining Styles and Negotiation:


The Thomas-Kilmann Conflict Mode
Instrument in Negotiation Training
G. Richard Shell

Many negotiation courses and executive training programs cover the sub-
ject of bargaining styles. The Thomas-Kilmann Conflict Mode Instrument
(TKI) is a commonly used psychological assessment tool that helps students
and teachers probe this topic. The TKI measures the five conflict manage-
ment facets proposed by the Dual Concerns Model: competing,
collaborating, compromising, accommodating, and avoiding. The author
has used the TKI extensively in teaching executives about bargaining styles,
and discusses the strengths and weaknesses of it as a teaching aid. He also
presents research on the frequency with which various TKI scores are
reported in business programs. Finally, he provides thumbnail sketches of
typical bargaining behavior exhibited by people with very strong and very
weak predispositions for each of the five conflict modes. Some implications
of these behaviors for specific professional audiences are explored.

A person’s bargaining style can play a crucial role in negotiation. For


instance, the American real estate and casino entrepreneur Donald Trump is
well known to be (and takes pride in being) competitive to his core. At the
other end of the scale, Cable News Network personality Larry King is well
known to be (and takes pride in being) empathetic and easy to get along
with. If these two were to find themselves negotiating against one another,

G. Richard Shell is professor of legal studies and management at The Wharton School of the Uni-
versity of Pennsylvania, Philadelphia, Penn. 19104-6369. Email: shellric@Wharton.upenn.edu.
Among other publications, he is the author of Bargaining for Advantage: Negotiation Strategies for
Reasonable People (New York: Penguin, 1999).

0748-4526/01/0400-0155$19.50/0 © 2001 Plenum Publishing Corporation Negotiation Journal April 2001 155
both would be wise to think about their own and their counterpart’s bar-
gaining styles before making a move.
Like Trump and King, each of us has a unique combination of personal-
ity traits that impact the way we bargain. These behavioral instincts and
intuitions set boundaries on the range of situations within which we will be
at our peak in terms of effectiveness. In addition, they help define the types
of negotiations (and negotiation counterparts) we enjoy — and those we
find stressful, frustrating, or confusing.
What, then, is a bargaining style?
Bargaining styles, as I see them, are relatively stable, personality-driven
clusters of behaviors and reactions that arise in negotiating encounters. They
are, in the words of Gilkey and Greenhalgh (1986: 245), “patterns in individ-
uals’ behavior that reappear in various [bargaining] situations” through the
mechanism of “predispositions” toward particular courses of conduct. For
example, people who strongly dislike interpersonal conflict will likely carry
this dislike with them into many bargaining encounters. This trait will affect
their effectiveness when the negotiation shows signs of becoming confronta-
tional. Similarly, people who thrive on interpersonal conflict and explicit
disagreement in their personal relations will experience stress, and may
become less effective, when the bargaining situation calls for subtle tact and
diplomacy.
I believe knowledge of bargaining styles is critical to negotiation suc-
cess and ought to occupy a central place in negotiation training. Such
knowledge helps students gain perspective on their own actions, interpret
others’ behavior, and use feedback more constructively. A style-sensitive cur-
riculum leads students to assess their strengths and weaknesses in systematic
ways, helping them to devise individualized learning agendas. The timid stu-
dent can focus with greater clarity on why assertive people make him or her
uneasy. The competitively-oriented student can understand more deeply
why, to many others, winning isn’t everything.
But the topic of bargaining styles in teaching negotiation presents a
problem: how best to address this elusive subject. We negotiation teachers
are an eclectic bunch — lawyers, business and law school professors, dis-
pute resolution professionals, political scientists, psychologists, you name it.
Relatively few of us are professionally qualified to dig deep into the psyches
of our students. Nor, in some cases, do we wish to. Few things upset a nego-
tiation teacher or a class more than when a simulation accidentally exposes a
student’s deeply repressed compulsion, phobia, or fear. Yet dig we must (at
least a little) if we are to address the issue of bargaining styles. In this brief
essay, I shall discuss the way I gain entry into this topic.
My special focus here is on one of the more popular bargaining style
assessment tools, the Thomas-Kilmann Conflict Mode Instrument (TKI).
Written in the mid-1970s, the TKI is old news to many. Many teachers and
trainers have used the TKI for a variety of purposes, and an academic cottage
industry developed in the 1980s testing the reliability and validity of the TKI

156 G. Richard Shell Bargaining Styles and Negotiation


as well as other “conflict style” surveys. This research, taken as a whole,
revealed that there was no “holy grail” in the field of conflict assessment
tools (see, e.g., Womack 1988 and Knapp, Putnam, and Davis 1988). Indeed,
I used this assessment tool in the 1980s, but then temporarily abandoned it
as a “black box” that seemed to have little relationship to my students’
behavior. I subsequently returned to using the TKI, however, and found to
my surprise that, properly used, it is a rich source of insight for my MBA and
executive students.1
This essay presents some of the research and teaching notes I have
developed in using the TKI in my various programs. These include a “Whar-
ton-TKI Bargaining Styles Grid,” which records, in percentile form, the
frequency with which business executives in both executive and executive
MBA programs at Wharton have reported various TKI scores. This evaluation
tool summarizes the scores from a sample of more than 1600 such execu-
tives from all over the world and from a wide variety of professional fields. It
has proven to be a useful reference point for instruction on the style vari-
able. For example, using this grid as a point of reference, I have found that
clinical professionals in the health care field systematically report both less
competitive and more accommodating TKI scores than do executives in
more traditional businesses. Similarly, senior executives in law enforcement
(police chiefs) display significantly higher scores in avoiding than do my
business sample. Because professionals in health care and law enforcement
both must deal with business people in important matters, they find these
comparisons useful.
In addition, I present capsule summaries of what each of the five con-
flict modes described by the TKI translates into when viewed as aspects of
a bargaining style. I have observed many executives whose profiles exhibit
both very strong and, just as interestingly, very weak predispositions for
various TKI conflict modes. These extremes often translate into predictable
negotiation strengths and weaknesses. In my experience, there is a remark-
able stability to the predispositions people report as the foundation for
their styles — and an associated stability to the struggles people report
overcoming these predispositions to improve their practice. Occasionally,
someone’s style changes dramatically in the wake of negotiation training,
but I count these occasions as the rare exception rather than the rule. For
the rest, negotiation training is a source of information rather than transfor-
mation. People become better able to understand, explain and adapt their
behavior rather than change their predispositions. My notes on strengths
and weaknesses may serve as a point of departure for students discussing
bargaining styles in a classroom setting. While findings such as these do not
pass muster as rigorous, empirical research, they and the other observa-
tions I make in this essay are, I hope, suggestive of ways that teachers and
trainers can use the TKI to provoke student reflection. My focus, in other
words, is on using the TKI as a teaching tool — not on its construct validity
as a research instrument.

Negotiation Journal April 2001 157


Background: Instruction on Personal Differences
The study of personal bargaining styles in negotiation and the use of teach-
ing materials to probe this factor date back almost as far as do organized
courses on bargaining. Teachers have used four main models to explore this
topic: personal styles as reflected in gender and cultural identities; coopera-
tive versus competitive “types”; interpersonal orientation (“IO”); and the
Dual Concerns Model, which measures relative concern for one’s own out-
come versus relative concern for the other party’s needs. As we shall see, the
five distinct conflict modes measured by the TKI derive from the Dual Con-
cerns tradition.
Perhaps the most visible way in which the topic of personal differences
enters the negotiation curriculum is through the lens of gender and culture.
Every simulation-based negotiating seminar with a mixed-gender audience
addresses gender as an issue, if only implicitly, in the way various men and
women experience themselves and each other in bargaining exercises.
There is a rich experimental and normative literature on gender, with The
Shadow Negotiation, a new book by Deborah Kolb and Judith Williams
(2000), being but the latest (and best) summary.
Cultural differences are also common topics in negotiation training.
Here, as with gender, there is abundant research on which to draw, but the
classroom experience is considerably richer when students and teachers
actually exhibit different cultural biases in their bargaining behavior. For
example, different cultures specify different expectations for how negotia-
tions should proceed — with profound effects on the styles people learn as
participants in these cultures. A recent report on teaching negotiation dis-
cussed an instance of the difficulty one American instructor had teaching
students from Thailand about brainstorming techniques. The difficulty?
There is no word in the Thai language or concept in Thai culture for brain-
storming.2 Without knowledge of Thai society, negotiation instruction on
collaboration could not begin.
Beyond personal differences based on sex and culture, instructors often
draw out the simplest distinction of personal bargaining styles, that between
“cooperative” and “competitive” people. The former is oriented toward
helping others achieve their goals. The latter is oriented toward “winning”
and achieving the greatest outcome for themselves. Professor Gerald
Williams’s empirical studies of lawyer-negotiators in the late 1970s and early
1980s (1983: 20-40) lent depth to the basic “two-style” model of bargaining
personality in finding that the great majority of his sample fell generally into
either “cooperative” or “competitive” camps. Williams then went on to list
the concrete behaviors that appeared to be associated with “effectiveness”
and “ineffectiveness,” both within and across these two categories.
Professors Rubin and Brown added nuance to the cooperative/competi-
tive model by introducing the notion of “interpersonal orientation” as an
organizing principle for the personality aspect of bargaining behavior. In The
Social Psychology of Bargaining and Negotiation, Rubin and Brown (1975)

158 G. Richard Shell Bargaining Styles and Negotiation


taught that people varied in their degree of responsiveness toward others’
behavior and emotional states. People are either high “IOs” or low “IOs.”
High IOs — people who pay close attention to others’ behavior and emo-
tions — can be either cooperatively or competitively oriented, depending on
whether they are focused on helping others (cooperative) or defeating them
(competitive). Low IOs, however, are relatively inattentive to others and are
“individualistically” motivated. They seek to achieve their own goals, period.
In effect, the IO model introduced a third “type” to add to the competitive
and cooperative ones — individually-oriented people seeking to achieve
their own goals without reference to the other party.
The fourth and most popular avenue into the subject of bargaining
styles derives from the Dual Concerns Model. The two “concerns” of this
model are the concern for the “self” and the concern for the “other.” Intro-
duced by Blake and Mouton in the mid-1960s (see Blake and Mouton 1964),
the Dual Concerns Model traces five different behavioral classifications: com-
peting, collaborating, compromising, accommodating, and avoiding. The
classifications come from placing the two concerns into relationship with
one another in five dynamic pairs. A dominant, high concern for the self’s
goals and associated desire to limit the other’s results is a competing orienta-
tion. A balanced, high level of concern for both the self’s and the other’s
goals and results is a collaborating mode. A balanced, moderate level of con-
cern for both self and other brings out the compromising orientation. A
dominant, high concern for the other’s goals without much concern for the
self translates into an accommodating approach. Finally, a disinclination
toward placing the self in conflict with others to pursue any goals at all is
avoiding.3
The Role of Personality Assessment Tools
In treating the subject of styles as part of a negotiation curriculum, many
teachers have found that some sort objective personality assessment is a use-
ful way to help students gain perspective on the topic. The TKI is one such
instrument and has been in use for some 25 years. Professors Thomas and
Kilmann developed the TKI expressly to elicit and test the five conflict
modes posited by Blake and Mouton’s model (see Kilmann and Thomas
1977). The TKI thus tracks the Dual Concerns framework exactly.
The TKI recommends itself as an especially useful tool for probing bar-
gaining styles in a classroom setting. I have found among its virtues:
• ease of administration (it takes only about ten minutes to take and score);
• relative freedom from social desirability biases in the way statements in
the instrument are presented;
• conflict styles that match up with strategy concepts widely used in the
negotiation literature; and
• significant congruence between the TKI styles students report and their
perceptions of their own behavior across a set of simulations.

Negotiation Journal April 2001 159


In semester-long MBA classes, I administer the TKI early in the course
and give a general lecture on bargaining styles sometime in the first two
weeks. I hand out the Wharton-TKI Bargaining Styles Grid (see Figure One)
in class and walk students through the process of translating their raw scores
into percentile scores, discussing what various “high” or “low” scores may
mean. I also encourage students to use their journals to reflect explicitly on
their TKI scores, and make possible connections between their TKI scores
and their results in early simulations and out-of-class bargaining encounters.
This emphasis on self-evaluation continues throughout the course. Midway
through the course, in connection with a team-based negotiation exercise, I
present the class with a list displaying everyone’s TKI scores. I do this to
emphasize the importance of taking bargaining styles into account both in
team planning and in preparation for facing a counterpart.
In executive seminars, I follow much the same pattern, although the
time devoted to the TKI is necessarily shorter. For a one-day seminar, to save
time, I try to have the TKI administered prior to the workshop. The owner
of the TKI, Consulting Psychologists Press, Inc., has a website that enables
participants to fill out the instrument online.4 If it is not possible to get the
results in prior to the workshop, I have participants fill out the TKI first
thing, before the lecture part of the program begins. I then introduce the
TKI and lecture on bargaining styles immediately after the participants have
completed their first, short bargaining simulation. This timing has the virtue
of presenting participants with a recent, salient data point of their own bar-
gaining behavior on which to reflect as they study their TKI scores and get
feedback from their first workshop counterpart. In the week-long executive
negotiation workshop at Wharton, we typically discuss TKI scores by the
end of the second full day of training.
Before going into more detail on the strengths and weaknesses of the
TKI, I want to emphasize that it is by no means the only psychological
assessment tool used in negotiation classes. A recent informal survey on
negotiation pedagogy (which is available on the website of the Program on
Negotiation at Harvard Law School, www.pon.harvard.edu/events/hewlett)
reveals that instructors use a variety of other assessment tools to probe the
issue of personality and bargaining styles. In addition to the TKI, instructors
report using a multi-rater system developed at Harvard’s Kennedy School
(Allred 2000), the Machiavellian Scale (Christie and Gies 1970), the Myers
Briggs Type Indicator, the Inventory of Personal Conflict Management Styles,
and the FIRO-B questionnaire.5
Some courses use multiple assessment tools and go the extra step of
having clinical psychologists as part of the negotiation teaching team. Profes-
sors Greenhalgh and Gilkey (1986) reported that their course at Dartmouth
included the use of eight different psychological tests, including the TKI, as
part of the instruction. The Harvard Law School, meanwhile, features an
intensive behavioral segment (see Bordone 2000) designed with the help of
psychological clinicians, in which students practice coping with conflict sit-

160 G. Richard Shell Bargaining Styles and Negotiation


uations that are especially stressful for them. One assumes that different stu-
dents identify which situations cause them stress based in part on their prior
experiences and personal predispositions.
The Thomas-Kilmann Conflict Mode Instrument: A Closer Look
With these background materials on style in mind, I now want to turn to a
closer, more substantive examination of the Thomas-Kilmann Conflict Mode
Instrument. The TKI is popular beyond the confines of negotiation classes.
Indeed, it is one of the most widely used instruments of its kind in the
world. I interviewed Professor Kilmann in the late 1990s in connection with
writing a chapter on bargaining styles for a recent book (see Shell 1999). Kil-
mann reported that several million copies of the instrument had been used
in classrooms and training over the twenty-plus years since he and Professor
Thomas conceived it in the 1970s. This record is all the more remarkable, he
told me, in light of the fact that the instrument was commercially copy-
righted by its authors almost as an afterthought, on the chance that there
might be a market for an assessment tool dealing with conflict behavior.
Professors Thomas and Kilmann did not develop the TKI with bargain-
ing or negotiation in mind. Rather, they were interested in finding a
measurement device for probing the validity and independence of the five
conflict modes hypothesized by Blake and Mouton. Prior to the develop-
ment of the TKI, research on the Dual Concerns Model had been plagued by
problems. In particular, variance in results appeared to be strongly linked to
subjects’ desire to exhibit socially desirable traits rather than to their actual
preferences for one conflict mode over another. For example, people did not
wish to be seen as avoiding conflict and so they would select the more
socially praiseworthy trait of collaborating.
Thomas and Kilmann addressed this problem head-on by pairing sim-
ple, equally desirable (or undesirable) phrases representing each conflict
attitude (such as “I am usually firm in pursuing my goals” for competing with
“I try to find a compromise solution” for compromising) and forcing subjects
to choose between the statements in each pair. The TKI displays several
phrases for each of the five conflict modes and requires subjects to select
between statements in 30 different pairs, forcing choices even when both
statements in a pair appear equally preferable or not preferable to the sub-
ject. Sentences representing each of the five conflict styles appear 12 times
in the instrument, so the maximum score for any style is 12 while the mini-
mum score for each style is 0. The simplicity of the statements,6 their
repetition, and the need for subjects to select them as against other, equally
compelling or repelling statements minimizes social desirability variance.
What does the TKI, in fact, measure? And how reliable is the TKI within
its psychological domain? The question of what it measures is relatively
straightforward: the TKI measures self-reported preferences or predisposi-
tions for using the five conflict modes posited by the Dual Concerns Model
when subjects are forced to choose between statements exemplifying these

Negotiation Journal April 2001 161


modes in 30 paired sequences. The TKI gives subjects no cues as to the
social context for their choices (i.e., the TKI does not prompt subjects to
make their choices based on what they would do in any given conflict, rela-
tionship, or social setting). Thus, the TKI tends to prompt subjects to select
modes based on “overall” predispositions rather than analyses of “best prac-
tices” for resolving any particular conflict-of-interest situation.
Whether these self-reports are “true” measures of traits (i.e., whether
my own impression that I prefer to accommodate is matched by my behav-
ior and by others’ impressions that I do, in fact, accommodate) is not so
clear. Part of the validity question hinges on use of vocabulary — what I
mean by accommodating may not be the same as what you mean. Another
issue is bias — people tend to rate others based in part on their own predis-
positions. A highly competitive person may see someone who is only
“moderately” competitive as lacking competitive zeal. A noncompetitive per-
son, by contrast, may see that same “moderately” competitive person as
quite competitive indeed.7
Thankfully, negotiation teachers using the TKI can provide their own
check on validity through the feedback students give each other. Peer ratings
are a common feature of many negotiation courses, and students may use
these ratings to see just how others’ perceptions of their behavior matches
with their self-reported predispositions for conflict modes. Even if the corre-
lation is imperfect, the self-reflection involved in this sort of analysis
advances students’ understanding of the strengths and weaknesses inherent
in their bargaining styles. The utility of the TKI for teaching purposes has sel-
dom been challenged, even by researchers who have questioned its validity
as a research tool. For example, in one survey on the TKI undertaken by Pro-
fessor Womack in 1988, the author noted that the TKI got “high marks” from
her sample of 14 trainers and teachers for both ease of use and effectiveness
in uncovering individual differences. Womack (1988a) concludes that “used
appropriately, [the TKI] can be beneficial to both trainers and researchers.”
The Wharton-TKI Bargaining Styles Grid
In order to use the TKI effectively in negotiation training, one needs to adapt
the raw scores derived from taking the TKI to the bargaining context. The
first step in this process is to plot a student’s TKI scores against some rele-
vant sample of peers that reflects the counterparts he or she will encounter
in bargaining practice. Consulting Psychologists Press distributes a grid with
the TKI instrument to serve this purpose; but it is based on a relatively small
sample of several hundred U.S. managers that some teachers (myself
included) have found unsatisfactory. To address this issue, I have collected
TKI scores from over 1600 global executives participating in negotiation
training sessions I have led.
In Figure One, I present the grid. This Figure displays the various TKI
scores provided by my sample in percentile form. For example, a score of
“6” in Figure One under the “avoiding” mode places a person reporting that

162 G. Richard Shell Bargaining Styles and Negotiation


Figure One
Wharton-TKI Bargaining Styles Grid

Competing Collaborating Compromising Avoiding Accommodating


100%
12 12, 11 12 12, 11 12, 11, 10
11 10 9
11
10 10 9
8 8
90% 9 9
10
High 25%

7
8
7
80% 8

75% 7 9 6

6
70%
7

6
60% 8 5
5
Mid 25%

50% 5 6

7
4
40% 4

4
5

30%
6
3
25% 3
3
20% 4
Low 25%

5
2 2
2
10%
3 4
1 2 3 1 1
0 1 2, 1
0% 0 0 0 0
Sample: 1682 participants in Wharton School executive programs led by Professor G. Richard Shell.

Negotiation Journal April 2001 163


“6” just below the 75th percentile of my sample for this predisposition. Such
a score indicates that the person has a relatively strong predisposition
toward avoiding in conflict situations. Recall that statements representing
each conflict mode appear a total of 12 times in the TKI. Thus, a score of “6”
means that the subject selected the avoiding statement one-half the times
such a statement appeared. Although this may seem like a “moderate,” half-
way score, most people within my sample of global business executives
select the avoiding statement far fewer than half the times it appears — mak-
ing a “6” score quite high for the avoiding mode.
A score of “6” under the compromising mode, by contrast, places the
subject just below the 30th percentile — indicating a relative lack of predis-
position for this way of handling conflict. Thus, each mode has a slightly
different set of percentile scores associated with each raw score between 0
and 12.
The way to use this grid in a classroom is simple: students fill out the
TKI, then locate and circle their raw scores for each mode on the grid. I then
ask them to connect the five circles with lines to form a simple graph. For
example, someone with a TKI score of 7 for competing, 8 for collaborating,
10 for compromising, 2 for avoiding, and 3 for accommodating would cre-
ate a graph that looks like the one depicted in Figure Two. This profile
suggests that the person has strong predispositions (75th percentile or
higher) for three modes: competing, collaborating, and compromising. At
the same time, this person’s instincts for avoiding conflict and accommodat-
ing other people’s needs are weak (below the 25th percentile). Overall, a
person with this profile might be expected to be quite assertive and to view
interpersonal confrontation as a functional and necessary aspect of human
communication. If I were to reverse this profile and observe someone with
scores in the 75th percentile for avoiding and accommodating — and the
25th percentile for the other three modes — I would expect to encounter a
relationship-sensitive person who dislikes confrontation (and therefore feels
awkward in negotiation) and who is strongly inclined to solve the other per-
son’s problem when conflict arises.
In using this grid, it is worth remembering what the TKI is measuring:
predispositions. People may score below the 25th percentile on competing
and still be capable of behaving in a competitive manner, should the situa-
tion or other party force them to do so. Their low TKI competitive score
simply suggests that they have little predisposition for this conflict mode and
probably do not enjoy hard-nosed, winner-take-all approaches to negotiation.
Depending on the extent of their professional experience, this low score
may also mean that their instincts regarding competitive tactics may be less
well developed than those of counterparts who have a stronger predisposi-
tion for the competitive mode.
The Wharton-TKI Grid may help students assess their aptitudes for
negotiating in particular situations or with particular counterparts. Someone
with a strong predisposition toward collaborating will, all else being equal,

164 G. Richard Shell Bargaining Styles and Negotiation


Figure Two
Example of An Individual’s TKI Profile

Competing Collaborating Compromising Avoiding Accommodating


100%
12 12, 11 12 12, 11 12, 11, 10
11 10 9
11
10 10 9
8 8
90% 9 9
10
High 25%

7
8
7
80% 8

75% 7 9 6

6
70%
7

6
60% 8 5
5
Mid 25%

50% 5 6

7
4
40% 4

4
5

30%
6
3
25% 3
3
20% 4
Low 25%

5
2 2
2
10%
3 4
1 2 3 1 1
0 1 2, 1
0% 0 0 0 0
Sample: 1682 participants in Wharton School executive programs led by Professor G. Richard Shell.
Negotiation Journal April 2001 165
have a talent for interest-based bargaining techniques. Such techniques as
brainstorming will come naturally to them and will even seem “obvious.”
Highly competitive people will feel more comfortable than others when
faced with hard-nosed bargaining moves. Those with high accommodating
scores may find their predispositions quite useful in relationship-sensitive sit-
uations — such as consulting — in which attention to others’ needs and
feelings is critical.
People often display preferences for several of the TKI conflict modes.
And people sometimes have several modes that they distinctly do not prefer.
When a TKI trait falls within the middle band of percentile rankings (roughly
between the 25th and 75th percentiles), it usually means the mode is a
more-or-less available resource, which may be called upon as the situation
and personality of the other party dictates.
I have also found that different professional subgroups exhibit distinc-
tive style preferences. For example, I have taught many nurses and nurse
executives in the U.S. health care industry in various programs. These indi-
viduals’ TKI scores show much less of a predisposition for competing and
considerably more of a predisposition for accommodating than is the case
for business executives. Considering the overall professional focus of nurses
on caring for others, these scores make sense. People may select a care-giv-
ing profession in part because serving others’ needs gives them personal
satisfaction. Executives in the nonclinical side of health care exhibit no such
trends — a fact that may help explain some of the reoccurring communica-
tion problems between “suits” (the business people) and “coats” (the clinical
staff) that plague some health care institutions. I have encountered a number
of such professional subgroup styles, including police chiefs and diplomats
(both groups are high “avoiders”); Wall Street investment bankers (high
“competitors”); and financial services and pharmaceutical sales teams (high
“compromisers” and “accommodators”).
Some Characteristics of Negotiators
Exhibiting the Five Conflict Modes
Over the years, I have discussed personal TKI profiles with many hundreds
of executives and other professionals. In these conversations, I have tested
various style-based hypotheses with them for confirmation or disconfima-
tion. In the subsections that follow, I summarize this experience by
commenting on the bargaining strengths and weaknesses that may be exhib-
ited by negotiators in a relatively high (75th or higher) percentile or
relatively low (25th or lower) percentile for each of the five conflict modes.
For shorthand, I refer to people with each trait by the name of the conflict
mode itself (e.g., “high accommodator” or “low compromiser”). I am thus
assuming in these comments that the person is “high” or “low” in the sub-
ject conflict mode only — and “in the middle” for all other modes.
Instructors should be aware that this is a convenient rather than a realistic
assumption, but it does permit me to discuss some important implications of

166 G. Richard Shell Bargaining Styles and Negotiation


TKI traits. Interaction of various combinations of preferences will, of course,
affect the way negotiators and their counterparts experience and manage
bargaining behavior in any given situation.

Accommodating
Negotiators Strongly Predisposed Toward Accommodating. Negotiators
with a strong predisposition toward accommodation derive significant satisfac-
tion from solving other peoples’ problems. They often have good
relationship-building skills and are relatively sensitive to others’ emotional
states, body language and verbal signals. This is a great trait to call on when
working on negotiating problems within teams, bargaining in sales-based “rela-
tionship management” roles, or providing many types of customer services.
In terms of weaknesses, high accommodators sometimes place more
weight on the relationship aspect of negotiations than the situation may war-
rant. In such cases, they are vulnerable to more competitively-oriented
people. High accommodators who are taken advantage of in such situations
may then experience resentment, further impeding their effectiveness.
Weakly Predisposed Toward Accommodating. Negotiators with low
accommodation scores are relatively uninterested in the other party’s emo-
tional states, needs, and frames of reference. They also have a tendency to
hold out stubbornly for their view of the “right” answer to a negotiating
problem. Where the low accommodator is an expert who understands the
negotiation problem better than others at the table, this trait will assure that
a group spends time considering the objectively “best” outcome. However,
others may perceive the low accommodator as stubborn to the point of
being unreasonable.

Compromising
Negotiators Strongly Predisposed Toward Compromising. People with
a strong predisposition toward compromising are eager to “close the gap” in
negotiations based on fair standards or formulae. When time is short, or
when the stakes are small, a marked predisposition toward compromise can
be a virtue. Others will see the high compromiser as a relationship-friendly,
“reasonable person.” However, high compromisers often rush the negotia-
tion process unnecessarily to reach the closing stage of the process and, in
their haste, may make concessions too readily. This can be costly when a lot
is riding on a negotiation outcome. They also do not discriminate carefully
among various fair criteria, some of which may be more advantageous to
their position than others. High compromisers tend to be satisfied with an
outcome as long as it is supported by any face-saving reason.
Weakly Predisposed Toward Compromising. People with a weak pre-
disposition for compromise are, almost by definition, men and women of
principle. Their great strength is their ability to summon passion and com-
mitment when serious matters of principle and precedent are at stake in a
negotiation. Their great weakness is their tendency to “make an issue” of

Negotiation Journal April 2001 167


everything — finding issues of principle where others see only issues relat-
ing to money or relative convenience. By arguing at length about things
others see as secondary, the low compromiser risks being seen by others as
stubborn — a person who elevates consistency over substance. Low com-
promisers therefore sometimes have trouble closing deals when issues of
principle arise and time is short.
A comparison between low accommodators and low compromisers is
instructive. Low accommodators can (more quickly than most) become
attached to their preferred solutions. Low compromisers, by comparison,
become attached on their preferred principles and fairness arguments. In
both cases, they may irritate other people, acquiring reputations for being
stubborn.

Avoiding
Negotiators Strongly Predisposed Toward Avoiding. High avoiders are
adept at deferring and dodging the confrontational aspects of negotiation. As
a positive attribute, avoidance can be experienced by others as tact and
diplomacy, permitting groups to function in the face of dysfunctional, hard-
to-resolve interpersonal differences. Professional diplomats are avoiders, as
are skillful “office politicians.” I have also noticed that people who enjoy
working in hierarchies have higher than normal avoiding scores — profes-
sionals such as police officials, military officers, and even surgeons. When
interpersonal conflict is a functional aspect of organizational or group life,
however, high avoiders can be a bottleneck in the flow of important infor-
mation about the intensity of people’s preferences. And when interpersonal
conflicts fester, they sometimes get worse, leading to all manner of prob-
lems. Finally, high avoiders often pass up many perfectly legitimate
opportunities to negotiate.
Weakly Predisposed Toward Avoiding. Low avoiders have little fear of
interpersonal conflict. Indeed, they may in some cases enjoy it. As negotia-
tors, they have a high tolerance for assertive, hard-nosed bargaining. They
can fight hard against their bargaining counterpart all day and share drinks
and stories with the same person in the evening. Low-avoid scores are help-
ful in such professions as labor-management relations, litigation, and mergers
and acquisitions work. But beware: People with low-avoid scores sometimes
lack tact, and are sometimes viewed by others as confrontational. In bureau-
cratic settings, low avoiders may be seen as troublemakers who refuse to
leave well enough alone.

Collaborating
Negotiators Strongly Predisposed Toward Collaborating. High collabo-
rators enjoy negotiations because they enjoy solving tough problems in
engaged, interactive ways. They are instinctively good at using negotiations
to probe beneath the surface of conflicts to discover basic interests and per-

168 G. Richard Shell Bargaining Styles and Negotiation


ceptions. They enjoy the continuous flow of the negotiation process and
encourage everyone to be involved. Business development executives, entre-
preneurs, corporate lawyers, real estate developers, and the best brokers in
any field often display high collaboration scores. By the same token, people
with a strong predisposition for collaborating sometimes needlessly trans-
form relatively simple situations into more complex (and interesting)
occasions to practice their skills. This can irritate other people who want
closure, who lack time to invest in a matter, or who do not wish to risk trig-
gering interpersonal conflict. A high collaborator with a very low competing
score can also be at risk against a highly competitive counterpart, who will
happily let the collaborator “work the problem” and devise an elegant
answer before moving to capture the lion’s share of the gains.
Weakly Predisposed Toward Collaborating. Low collaborators dislike
using the bargaining process as a forum for their creativity. These negotiators
prefer having problems clearly specified before the negotiation starts and
like to stick to their bargaining plan once a meeting is underway. When the
matters being negotiated are so inherently complex that real-time brain-
storming is the best way to proceed, low collaborators may become a
bottleneck, slowing the process down. One way for the low collaborator to
compensate for this weakness is to make liberal use of breaks in the bargain-
ing process to gather their thoughts and reset their strategy.

Competing
Negotiators Strongly Predisposed Toward Competing. Like high collab-
orators, high competitors also like to negotiate. But they enjoy it for a
different reason: negotiating presents an opportunity to win what they see
as a game or sport based on a set of practiced skills. Competitors can be
found in any professional field, but they are most in their element when the
stakes are high, time is limited, and bluffing is possible. Litigation attorneys
who enjoy their work tend to be highly competitive people, as do invest-
ment bankers and hedge fund managers. Such negotiators have excellent
instincts about such matters as leverage, deadlines, openings, final offers,
ultimata, and similar aspects of tough, traditional bargaining. The competi-
tive style can dominate the bargaining process, so competitive people can
sometimes be hard on relationships. In addition, competitive negotiators
instinctively focus on the issues that are easiest to count in terms of winning
and losing — like money. They may overlook nonquantitative issues that can
yield substantial value.
Weakly Predisposed Toward Competing. People with a weak predispo-
sition for competing think negotiations are about much more than winning
and losing. Rather, the goal in negotiations is for the parties to treat each
other fairly, avoid needless conflict, solve problems or create trusting rela-
tionships. Others often view people with low competing scores on the TKI
as especially nonthreatening. This can be a strength in many professional set-

Negotiation Journal April 2001 169


tings in which the ability to gain trust is a critical skill. However, when large
stakes are on the table and other, highly competitive negotiatiors are pre-
sent, the low competitor may be at a disadvantage.
Some Questions People Frequently Ask Regarding the TKI
In using the TKI as an instructional tool, I have encountered a number of
reoccurring questions from students and executives. Below, I share some of
the more common questions — along with some suggested answers.
1. Is there an “optimal’ TKI score for negotiators?
No. There is no single “right” style for negotiation effectiveness. Rather,
people with preferences or aversions for each of the five styles tend to dis-
play certain systematic strengths and weaknesses. These strengths and
weaknesses, in turn, either help or hurt in given circumstances and with
given people on the other side of the table. For people who engage in trans-
actional negotiations as a profession, higher competitive and collaborative
scores will be an indication that they enjoy their work. For people who do a
lot of relationship-based sales or consulting, higher accommodating and
compromising scores may be a sign that they feel comfortable in their jobs.
Professional diplomats, by contrast, sometimes report higher-than-usual
scores for avoiding conflict.
In other words, your TKI score may be one indication of how naturally
your style “fits” the professional setting in which you negotiate. But there is
no single profile that works best for all negotiators.
2. Which style traits are exhibited by people who enjoy the bargaining
process?
In general, people with higher scores in both competing and collabo-
rating enjoy the negotiation process somewhat more than people who have
low scores in both categories. High competitors like to win and high collab-
orators like to develop elegant, stable solutions to the problems underlying
the negotiation. Both can be very assertive. By contrast, people with lower
scores in these two categories tend to prefer the other three style modes, all
of which focus (in different ways) on reducing conflict, bringing closure to
negotiations, and emphasizing the other party’s needs in relationships.
3. What does it mean if a person prefers several styles?
Each person has his or her own unique combination of preferences.
Many have strong predispositions toward more than one style. People tend
to assess the situation and people they face, check to see if their most pre-
ferred (and therefore familiar) style is appropriate, and then proceed. If their
most preferred style is inappropriate (e.g., they are a high avoider and they
are selling their used car), they tend to shift to their next most preferred
style. This shifting of approaches can also occur in the middle of a negotia-
tion if the preferred style is not working to advance the process.
Some style combinations yield characteristic results. For example, as
noted earlier, someone high in both competing and collaborating tends to

170 G. Richard Shell Bargaining Styles and Negotiation


be comfortable in many negotiating situations in which the stakes are large.
Someone high in both competing and avoiding, meanwhile, is somewhat
one-dimensional as a negotiator and tends to project a “we’re-doing-this-my-
way-or-I’m-hitting-the-road” attitude. If this person cannot use their
competitive tactics, they tend to exit the situation or delegate the bargaining
to someone else. Someone with high scores in both accommodating and
compromising, by contrast, is very relationship-oriented and will tend to be
seen as relatively easy to get along with as they move fluidly between an ori-
entation toward solving the other person’s problem and an orientation
toward simple, fast, fair allocations.
One of the many useful aspects of using the TKI is to notice these and
other, different combinations. An experienced teacher will find that their
hypotheses are confirmed in a surprising number of cases. But the more
interesting work is in prompting students who have various combinations to
reflect on and report on their experience of their own predispositions as a
way for them to advance their own self-understanding.
4. What if a person’s scores are all in the middle percentiles — i.e., he or
she has no strong preferences?
TKI scores in the midrange percentiles often indicate that the conflict
style in question is relatively accessible and can be called out as the occasion
demands. Some people score in this middle range for all five attributes, indi-
cating that they have a very adaptable style that can serve them well in most
situations. Such people may still be at a relative disadvantage, however,
when facing off against equally experienced people who have much more
definite preferences. For example, a “moderately” competitive negotiator
facing a highly competitive counterpart in a situation that rewards a compet-
itive approach may need to summon all their energies and instincts to stay
even with their counterpart’s moves. Their counterpart will experience less
stress and “stretch” in this situation.
5. Do TKI scores change over time?
TKI scores are a function of the time and place one fills out the instru-
ment. If one has had a recent, very bitter conflict that has ruined a
relationship, there may be a tendency to regret this event and select more
accommodating TKI choices than would otherwise be the case. Similarly, if
one has recently been taken advantage of in a tough negotiation, then one’s
TKI scores might reflect a desire to correct this by selecting more competi-
tive statements than would ordinarily be true.
If the user can place their mind in “neutral” and simply do their best to
select which of the statements in each pair better expresses their overall pre-
ferred attitude, however, the TKI score will be relatively stable over time. In
my experience, the direction of these scores remains relatively steady over a
semester-long course, with the same modes being preferred to others — if
by somewhat different margins.

Negotiation Journal April 2001 171


Conclusion
No single model of personality exhausts the variety of ways in which people
experience themselves and others in negotiation. Neither the Dual Concerns
Model nor its most popular measurement tool — the TKI — comes close to
a full catalogue of the conflict modes in the human psyche. But the TKI is a
very good place to start such a catalogue.
With the foundations of the five conflict modes hypothesized by the
Dual Concerns Model firmly in mind, negotiators are better able to look criti-
cally and intelligently at the unique combinations and nuances of behavior
that they and their counterparts bring to the bargaining table. And as a
teaching tool, the TKI is one of the best ways I have found to introduce the
bargaining style factor to students as a negotiation variable. Knowledge of
personal bargaining styles brings with it perspective, confidence, a greater
sense of control over the bargaining process, and skill at interpreting what
the other party intends by its words and actions. These are worthwhile
insights for anyone seeking greater personal effectiveness at the bargaining
table.

172 G. Richard Shell Bargaining Styles and Negotiation


NOTES

1. The occasion for my revisiting the Thomas-Kilmann Conflict Mode Instrument was the
arrival on our faculty of a practice professor, Stuart Diamond, who is a graduate of Harvard Law
School and a former affiliate of Conflict Management Inc. of Cambridge, Mass. Stuart believed the
TKI was a particularly useful teaching tool, and his enthusiasm convinced me to try using it again.
2. Negotiation pedagogy: A research survey of four disciplines (at 61). This report was pre-
pared for the spring, 2000 meeting at Harvard University of scholars affiliated with Hewlett
Foundation-supported negotiation centers. It is available at the Program on Negotiation’s website,
www.pon.harvard.edu/events/hewlett. See also Ron S. Fortgang’s (2000) description and analysis
of this project.
3. Rounding out the psychological aspects of the standard negotiation course curriculum,
experimentally-grounded ways of discovering various psychological biases have become a com-
mon topic. This set of materials has little to do with bargaining styles as such. Rather, it
emphasizes cognitive errors to which people of all races, genders, and interpersonal orientations
are prone.
Led by Max Bazerman and Maggie Neale (1992) — building on foundations laid by Daniel
Kahneman and Amos Tversky (1979) and Kahneman, Slovic, and Tversky (1982) — negotiation
teachers have learned to instruct their students about systematic biases in human cognition that
can cloud negotiator judgment. The experimental evidence for such phenomena as “fixed pie”
biases, gain-loss framing, availability, escalation of commitment, and overconfidence is strong, and
classroom demonstrations of many of these biases are relatively easy to conduct. The pedagogical
goal is to help people learn to recognize cognitive biases, overcome them in appropriate situa-
tions, and become more effective negotiators as a result.
Like materials on gender, race, or culture, materials on cognitive psychology complement
rather than substitute for content related to bargaining styles. The biases are not predispositions
toward handling interpersonal conflict situations so much as they are hard-wired quirks in the
human information-processing system. Negotiations are merely one form of decision making in
which such biases are troublesome.
4. The TKI may be licensed for use by its current owner, Consulting Psychologists Press, Inc.
The telephone contact for this firm is 800-624-1765. The world wide web address is: www.cpp-
db.com/products/tki
5. Faculty at Harvard’s Kennedy School of Government have more recently developed and
used in their negotiation courses a web-based “multi-rater feedback tool” based on a variation of
the Dual Concerns model (see Allred 2000). This multi-rater system probes a student’s effective-
ness at using five “facets” of conflict management: integrating, asserting, accommodating,
avoiding, and a combination facet called “accommodating/avoiding.” Students rate themselves
along these dimensions and, in turn, are rated by other people with whom they have interacted in
conflict situations (including course simulations). The results of these surveys are collated and
compared.
The multi-rater system drops “compromising” as a conflict mode from the Dual Concerns
model in favor of the combination of “accommodating/avoiding.” The big difference between the
multi-rater and TKI instrument is not the conflict modes, however. Rather, it is the feedback,
which students get in the multi-rater profile on their self-reported styles from people who have
experienced these styles in action. This sort of “reality check” gives students helpful perspective
and allows them to evaluate their effectiveness. It is, of course, possible to use the TKI or any
other assessment tool in exactly the same way by asking students to rate each other during the
course on their relative skill at using the relative personality-based skills.
Another psychological tool used for exploring the spectrum of personality types is the
“Machiavellianism Scale” developed by Christie and Geis (1970). People who score high on these
tests (“high Machs”) tend to win in loosely structured coalition games, manipulate others for self-
gain more often, and lie more frequently than do low Machs. High Machs are usually seen as a
special type of competitively oriented person.
6. One TKI study has suggested that the TKI categories may be a bit too simple and that the
avoiding and accommodating styles really embody a single style preference for sidestepping “emo-
tional unpleasantness” (Konovsky, Jaster, and McDonald 1989). This study did not involve the
context of negotiation, however, when a tendency to accommodate by addressing others’ needs
may be more easily distinguished from a tendency to avoid the negotiation situation altogether.
7. For other studies summarizing research on the TKI, see Womack (1988) and Knapp, Put-
nam, and Davis (1988).

Negotiation Journal April 2001 173


REFERENCES

Allred, K.G. 2000. Distinguishing best and strategic practices: A framework for managing the
dilemma between creating and claiming value. Negotiation Journal 16(4): 387-397.
Bazerman, M.H. and M.A. Neale. 1992. Negotiating rationally. New York: The Free Press.
Blake, R.R. and J.S. Mouton. 1964. The managerial grid. Houston: Gulf Publications.
Bordone, R.C. 2000. Teaching interpersonal skills for negotiation and for life. Negotiation Journal
16(4): 377-385.
Christie, R. and F.L. Geis, eds. 1970. Studies in Machiavellianism. New York: Academic Press.
Fortgang, R.S. 2000. Taking stock: An analysis of negotiation pedagogy across four professional
fields. Negotiation Journal 16(4): 325-338.
Gilkey, R.W. and L.Greenhalgh. 1986. The role of personality in successful negotiating. Negotia-
tion Journal 2(3): 245-256.
Kabanoff, B. 1987. Predictive validity of the MODE Conflict Instrument. Journal of Applied Psy-
chology 72(1): 160-163.
Kahneman, D. and A. Tversky. 1979. Prospect theory: An analysis of decision under risk. Econo-
metrica 47: 136-291.
Kahneman, D. D. P. Slovic, and A. Tversky. 1982. Judgment under uncertainty: Heuristics and
biases. Cambridge: Cambridge University Press.
Kilmann, R.H. and K.W. Thomas. 1977. Developing a forced-choice measure of conflict-handling
behavior: The ‘Mode’ Instrument. Educational and Psychological Measurement 37: 309-325.
Knapp, L., L.L. Putnam, and L.J. Davis. 1988. Measuring interpersonal conflict in organizations:
Where do we go from here? Management Communication Quarterly 1(3): 414-429.
Kolb, D.M. and J. Williams. 2000. The shadow negotiation: How women can master the hidden
agendas that determine bargaining success. New York: Simon and Schuster.
Konovsky, M.A., F. Jaster, and M.A. McDonald. 1989. Using parametric statistics to explore the
construct validity of the Thomas-Kilmann Conflict Mode Instrument. Management Commu-
nication Quarterly 3(2): 268-290.
Rubin, J.Z. and R.B. Brown. 1975. The social psychology of bargaining and negotiation. New
York: Academic Press.
Shell, G.R. 1999. Bargaining for advantage: Negotiation strategies for reasonable people. New
York: Penguin.
Williams, G.R. 1983. Legal negotiation and settlement. St. Paul: West.
Womack, D. F. 1988. A review of conflict instruments in organizational settings. Management
Communication Quarterly 1(3): 437-445.
———. 1988a. Assessing the Thomas-Kilmann Conflict Mode Survey. Management Communica-
tion Quarterly 1(3): 321-349.

174 G. Richard Shell Bargaining Styles and Negotiation

You might also like