Trust and Managerial Problem Solving
Trust and Managerial Problem Solving
Trust and Managerial Problem Solving
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/2393957?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
Dale E. Zand
This paper presents a model of trust and its interaction with information flow,
influence, and control, and reports on an experiment based on the model to test
several hypotheses about problem-solving effectiveness. The subjects were man-
agers and the independent variable was the individual manager's initial level of
trust. Groups of business executives were given identical factual information about
a difficult manufacturing-marketing policy problem; half the groups were briefed
to expect trusting behavior, the other half to expect untrusting behavior. There
were highly significant differences in effectiveness between the high-trust groups
and the low-trust groups in the clarification of goals, the reality of information
exchanged, the scope of search for solutions, and the commitment of managers to
implement solutions. The findings indicate that shared trust or lack of trust ap-
parently are a significant determinant of managerial problem-solving efectiveness.1
229
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
230 ADMINISTRATIVE SCIENCE QUARTERLY
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
Zand: PROBLEM SOLVING 231
relevant, comprehensive, accurate, and timely dance with agreements, and therefore will
information, and thereby contribute realistic have less need to impose controls on others,
data for problem-solving efforts. They will (see Figure 1). Consequently they will con-
have less fear that their exposure will be tribute to a decrease in social uncertainty,
abused, and will therefore be receptive to and be less likely to misinterpret the inten-
influence from others. They will also accept tions and the behavior of others. As a result,
interdependence because of confidence that underlying problems are more likely to be
others will control their behavior in accor- identified and examined, and solutions more
TRUST
Increase one's
vulnerability
to others whose
behavior one
cannot control
/ I
CONTROL INFORMATION
INFLUENCE
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
232 ADMINISTRATIVE SCIENCE QUARTERLY
Hypothesis
likely to be appropriate, 3. Search more
creative, and extensively
long- for
range. alternative courses of action,
Hypothesis 4. Have greater influence on
Hypotheses solutions.
It is not assumed here that trust alone will Finally, an increase in trust will increase
solve a technical problem; it is assumed that willingness to control one's own behavior,
group members collectively have adequate will increase confidence in the reliability of
knowledge, experience, and creativity to de- others, and will decrease efforts to control the
fine and solve a complex problem. It is also behavior of others, all of which will con-
assumed that it is possible to increase or de-tribute to increased satisfaction and motiva-
crease trust in members of a problem-solving tion. Hence, problem-solving groups with
group. high trust will:
On the basis of the model described, the Hypothesis 5. Be more satisfied with their
following differences can be predicted in the problem-solving efforts,
problem-solving behavior of groups with Hypothesis 6. Have greater motivation to
high and low trust. implement conclusions,
An increase in trust will increase the ex- Hypothesis 7. See themselves as closer and
change of accurate, comprehensive, and more of a team,
timely information. Problem-solving groups Hypothesis 8. Have less desire to leave their
with high trust will: group to join another.
P's intentions & P's behavior O's perception O's intentions &
expectations Restricts information P's behavior seeis as expectations
Not trusting Resists influence untrusting Not ti sting
0 untrustworthy Seeks to impose controls P untrustworthy
FIGURE 2. A MODEL OF THE INTERACTION OF TWO PERSONS WITH SIMILAR INTENTIONS AND EX-
PECTATIONS REGARDING TRUST
An increase in trust will increase the will- Let P denote one person and 0 the other.
ingness to influence others and the receptivity If (1) P lacks trust, (2) he will disclose
to the influence of others. Hence, problem- little relevant or accurate information, be
solving groups with high trust will; - unwilling to share influence, and will at-
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
Zand: PROBLEM SOLVING 233
tempt to control 0. (3) Assume 0 also lacks trust, but this study did not focus on a test
trust, (4) perceives P's initial behavior as of the spiral reinforcement model. The aim of
actually untrusting, and (5) concludes he this study was to examine the relation be-
was right to expect P to be untrustworthy; tween trust and problem-solving effectiveness
then (6) he will feel justified in his mistrust as formulated in the eight hypotheses.
of P. Since (7) P sees O's behavior as un- To test the hypotheses derived from the
trusting, he (8) will be confirmed in his model, the research was designed so that half
initial expectation that 0 would not be trust- of the experimental groups started work on a
worthy and (2) P will behave with less trust business-management problem with a mental
than when he entered. set toward low trust and half with a mental
The interaction will continue around the set toward high trust. Mental set, as used
loop inducing 0 and P to behave with less here, includes intentions as to one's behavior,
and less trust until they arrive at an equilib- expectations as to the behavior of others, feel-
rium level of low trust, each attempting to ings such as anxiety or discomfort, and cogni-
minimize his vulnerability and to maximize tive frame used to interpret events and form
his control of the other. In the process the perceptions. In this research trust was not
effectiveness of problem solving will decrease. examined as a personality variable; that is,
After interaction has continued, each will an element of individual character, but as an
tend to hold more firmly to his entering induced attitude, one that the individual
beliefs. They will not have a reliable basis for could alter in a situation in which he was led
accepting or sharing influence, and the mu- to intend and to expect trust (or mistrust)
tual resistance to influence will arouse feel- from others as he attempted to solve a
ings of frustration in both. If they have a problem. Because trust as a personality trait
deadline, each will attempt to impose con- was not relevant, and to avoid alerting the
trols on the other. If P is O's organizational subjects to the issue of trust, no prior mea-
superior, he may command O's compliance, sures were taken of the subjects' attitude
which will reinforce O's mistrust. Usually, by toward trust.
the middle of the meeting the level of trust Subjects. Because of their high potential
will be lower than the initial level. for top management, upper-middle managers
Gibb (1964) offers support for the dyna- from all functions and product divisions of a
mics of this interaction. In observing small large, international electronics company were
group behavior he noted that the defensive periodically selected by their superiors, after
behavior of a listener generated cues which discussion with the corporate personnel staff,
subsequently increased the defensiveness of to attend an off-site, four week, in-residence
the communicator, resulting, if unchecked, inprogram in management development given
a circular pattern of escalating defensiveness. several times a year that accepted sixteen
The pattern of spiral reinforcement illus- managers at a time. Eight managers in each
trated in Figure 2 would operate construc- program, were randomly chosen to be sub-
tively if it is assumed that both P and 0 en- jects and distributed into two problem-
tered the relationship with trust in the other. solving groups, each with four members. The
Gibb (1964) observed that when defensive- remaining eight managers were observers;
ness was reduced, members were better able each was -randomly assigned to a group with
to concentrate on the content and meaning of four members and each group observed one
a message, became more problem oriented, problem-solving group. Data were gathered
and were less concerned about imposing con- in eight programs providing a total of sixty-
trols on each other's behavior. four subjects in sixteen problem-solving
groups and fifty-nine observers (five pro-
METHOD grams were short one manager) in sixteen
observing groups. There were no subordi-
The spiral reinforcement model of the nates, superiors, or peers from the same de-
dynamics of trust (Figure 2) has been pre- partment or division in any program. Inter-
sented to establish a theoretical rationale for views confirmed that the subjects and the
the methods used to induce different levels of observers did not know about the experi-
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
234 ADMINISTRATIVE SCIENCE QUARTERLY
ment, which was designed as a learning event to resign. Furthermore, he was told that
embedded in the program. expansion was not feasible because it would
Problem. The central problems involved reduce short-term profits, take more than a
(1) developing a strategy to increase short- year to build and start up a new plant, and
term profits without undermining long-termthe board was not likely to approve the fi-
growth of a medium-sized electronics com- nancing, so as a first step toward increasing
pany with very low return on investment, profits, he would have to announce his deci-
outdated manufacturing facilities, whose sion against expansion. The vice presidents
labor force had been cut 25 percent and had no knowledge of the president's dilemma
whose top management personnel had been when they started their thirty-minute prob-
changed and reorganized two years before, lem-solving meeting.
and (2) obtaining commitment to implement Induction of conditions of trust. The induc-
such a program despite strong managerial tion of the two levels of trust was accom-
disappointment because expectations of im- plished by operating on the following enter-
mediate investment for expansion and mod- ing beliefs of subjects: (1) the task
ernization would not be met. The situation, a competence of others, (2) norms on intro-
variation of one described by Maier et al. ducing information and new ideas, (3) norms
(1959), involved four executive roles: presi- on attempts to influence managers outside of
dent and vice presidents for marketing, man- one's primary responsibility, (4) likelihood
ufacturing, and personnel. Subjects were ran- that others would abuse trusting behavior,
domly assigned to the roles. and (5) competitiveness or collaborativeness
Procedure. All subjects and observers were for rewards.
given a written description of the production, In a high-trust group, a manager's entering
marketing, financial, and personnel difficul- mental set toward trust was shaped by the
ties of the company. following paragraph, which followed the
In the presence of the observers, subjects factual information in the role statement:
were told they were to conduct a meeting You have learned from your experiences during
lasting thirty minutes in the president's office
the past two years that you can trust the other
to make appropriate management decisions. members of top management. You and the other
Ostensibly, they were to demonstrate their top managers openly express your differences
decision-making competence to their fellow and your feelings of encouragement or of dis-
managers, the observers. appointment. You and the others share all rele-
Each subject was then given an additional vant information and freely explore ideas and
written statement with factual and attitudinal feelings that may be in or out of your defined
responsibility. The result has been a high level
information relevant to his function. He had
of give and take and mutual confidence in each
no knowledge of the role information given
other's support and ability.
to other subjects. The subjects privately read
and absorbed this problem information for Subjects in low-trust groups had a similar
twenty-five minutes so there would be mini- paragraph in their role information, but
mal need to refer to it during the meeting. worded to induce a decrease in trust.
Treatments. Subjects were randomly as- The reward system was operated on by
signed to one of two group conditions: an information placed only in the president's
entering mental set toward high or low trust. statement. In the high-trust condition, the
The factual data about production, market- president was led to see his relation to his
ing, finance, and so on was identical in both vice presidents as collaborative. His role
conditions, and all vice presidents were led statement said that "although the Board's
to expect that the president would announce decision considered you specifically, since
approval of a long-studied plant expansion. you appointed the current top management
In both conditions the president's state- team it is likely that the Board will go out-
side for a successor and possibly other vice
ment told him that on the preceding day, he
had received an ultimatum from the board ofpresidents."
directors demanding an increase in profits In the low-trust condition the president
within one year or else he would be forced was led to see his relation to his vice presi-
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
Zand: PROBLEM SOLVING 235
dents as potentially competitive. His role in Table 1, with the chi-square value for each
statement said that since the board's ultima- item.
tum pertained to him, it was possible that The subjects' rating of level of trust con-
they might appoint one of the vice presidents firms that the induction of high and low trust
as his successor. The vice presidents in both was successful (p < .001) after one-half hour
conditions were given no information about of problem discussion. This result, although
whether their relation to the president was not a direct test of the spiral reinforcement
potentially competitive or collaborative. model, does offer support for it.
All subjects were told that "whenever in- Since the observers used only their per-
formation is incomplete, introduce whatever sonal standards for their ratings, it is note-
facts and experiences seem reasonable under worthy that they had little difficulty recog-
the circumstances." nizing the behavior indicative of low or high
trust (p < .001).
Observers. In addition to reading the writ-
The hypotheses about differences between
ten general description of the company's
groups with high or low trust were confirmed
problems, before observing the problem-
by the responses of the subjects (items 2-9,
solving meeting, the observers were told of
p < .001) and observers (items 3-9, p < .001;
the vice presidents' factual basis for seeking
item 2, p < .05).
and expecting to get final approval for plant
Qualitative Differences. There were also
expansion and that the president had re-
observable qualitative differences in the com-
ceived a one-year ultimatum from the board
prehensiveness and creativity of the problem
the preceding day, but they were given no
solving of the two groups.
information about the attitudinal parts of the
High-Trust Groups. In the high-trust
statements.
groups the president consistently disclosed
Measures. After thirty minutes, group dis-
voluntarily the board's demand for better
cussion was stopped and each subject and short-term performance. These teams, after
observer completed a questionnaire with initial frustration with the disapproval of
eight or nine items. The respondent was to immediate expansion, dealt with the short-
indicate whether in his group, or the group range plans to increase profitability and then
he observed, there was "much" or "little" of began to design long-range plans for moderni-
the property described in each item. zation and expansion that they would present
The items were: (1) trust, (2) openness to the board.
about feelings, (3) clarification of the group's Short-range plans emerging from the dis-
basic problems and goals, (4) search for cussion among the vice presidents included
alternative courses of action, (5) mutual in- straightforward proposals to review the prod-
fluence on outcomes, (6) satisfaction with uct line, to identify and promote sales of
the meeting, (7) motivation to implement high-profit items, and to cut back output of
decisions, (8) closeness as a management low-profit items. Their more creative pro-
team as a result of the meeting. The subjects' posals, flowing from substantial changes in
questionnaire had a ninth item: "As a result their perceptions, included, for example, leas-
of this meeting would you give little or much ing space in a nearby vacant plant, rearrang-
serious consideration to a position with ing work flow, selectively modernizing equip-
another company?" The written statement ment that would provide greatest cost benefits
could only suggest to each subject an enter- and require minimal capital, subcontracting
ing mental set toward high or low trust. By standard components, and rapidly converting
the end of the meeting each subject's level oftwo new products from research to produc-
trust would depend on the extent to which tion. In one group the managers agreed to
his entering beliefs were confirmed by the invest their personal savings to help finance
behavior of the other managers. modernization, to show the board their strong
commitment to the company's future.
RESULTS
Low-Trust Groups. In low-trust groups,
Measures of Trust. The responses of sub- the vice presidents had difficulty understand-
ing the basis for the president's decision
jects and of observers are reported separately
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
236 ADMINISTRATIVE SCIENCE QUARTERLY
Subjects Observers
Condition Condition
against expansion and his desire for short- dents. In several groups the president threat-
range profits. In several groups they asked ened to dismiss a vice president.
him if there were reasons behind his decision Conversation among subjects of the low-
other than those he had disclosed, but he trust groups after they had answered the
steadfastly refused to reveal information questionnaire, showed the high defensiveness
about the board's demands. As a result the and antagonism they had induced in each
vice presidents in low-trust groups could not other. For example, half the vice presidents
sense how close the company might be to said that they were so discouraged they had
reorganization and possibly dissolution. They started to think of looking for another job in
spent most of the meeting disagreeing with the middle of the meeting, and several said
the president by repeating their basic argu- they hoped the president's plane would be
ments for immediate expansion. Finally, after hijacked or crash. The president usually re-
prolonged frustration, the president would torted that he had decided to dismiss them
impose directives on the group. Usually he before the next meeting.
would demand review of the product line to Discussion. One might contend that the
eliminate low-profit items. If there was any managers were attempting to follow rigidly
creativity it came from the president, who the attitude toward trust suggested in their
was desperately seeking a solution in spite of briefing, but in the debriefing interviews, the
the resistance of his vice presidents. Occa- managers said that after their meeting had
sionally, the president would propose that it started, their level of trust varied in response
might be possible to lease space in a nearby to the behavior of the other managers. In
vacant plant, but his idea would be discarded low-trust groups, for example, about half of
as unworkable by the belligerent vice presi- the vice presidents said that by the end of the
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
Zand: PROBLEM SOLVING 237
meeting they found themselves trusting one but the creativity and comprehensiveness of
or another vice president more than they its solutions had suffered in comparison with
expected to and trusting the president much other high trust groups.
less than when they had started. Because of the many limitations of the
That the pattern of spiral reinforcement experiment-that is, the small number of
requires all members of a group to hold subjects, data gathered over several years,
similar intentions to trust (or not trust) may the study conducted within the context of
be too stringent a condition. The following a management development program-the
anecdotal evidence suggests that several study was restricted to conditions in which
members with similar intentions may be all managers in a group had the same initial
sufficient. An unanticipated incident illus- level of trust. The condition of mixed trust,
trates how difficult it may be for one in which some members would tend to trust
person acting alone to break the reinforce- and others would tend to mistrust, was not
ment pattern even though he has formal included; but one could predict that the
power. In one low-trust group, in an effort effects on creativity and comprehensiveness
to behave with trust toward his vice presi- of solutions, and on motivations to implement
dents, the president early in the meeting dis-solutions might be intermediate between
closed that the board wanted better profit those of the high-trust and low-trust groups.
performance in one year or else might ask for The two incidents described above are con-
his resignation, but this attempt to show trust sistent with such a prediction.
did not alter the emergence of low-trust Furthermore the problem used in this
behavior among the vice presidents. Indeed, study was quite complex, required that the
in interviews after the meeting, the vice participants generate the alternatives, and
presidents said they interpreted the presi- had no unique, optimal solution. There might
dent's statement as a means of shifting blame be less of a difference in the output of high-
to the board for his decision not to approve trust and low-trust groups working on highly
expansion, so that instead of increasing their structured problems; that is, problems with
trust, his behavior confirmed their mistrust. clear, tangible goals, with well-defined in-
Also, they interpreted the president's com- formation, with alternatives provided, and
ment that he might be forced to resign as with a unique solution. Theoretically, the
evidence that the board did not trust him, so structure inherent in the problem might re-
they should not either. Two vice presidents duce a group's susceptibility to the social un-
in this group said that by the middle of the certainty generated by low-trust behavior.
meeting they were thinking about how they On the basis of the data in Table 1, however,
might hasten the president's resignation. It it would seem that, given similar member
seems that behaving with high trust towards competence, groups that develop high trust
others who are not trusting will not neces- would solve problems more effectively than
sarily induce trust, and if one does so it is low trust groups, that is, they would do better
wise to limit one's increase in vulnerability. in locating relevant information, in using
Another illustration of the difficulty of their members' skills to generate alternatives,
interrupting the spiral reinforcement pattern and in eliciting commitment.
occurred in a high-trust group, in which the The data also indicate that patterns of low-
president did not reveal the board's demand trust and high-trust group behavior are recog-
for short-term profits. The vice presidentsnizable by untrained observers. Possibly the
consistency between the responses of subjects
said that the president seemed troubled, and
and observers was increased by the fact that
asked him if he was explaining all the reasons
behind his decision not to expand. In the they were all managers in one company,
debriefing interviews, after they learned presumably exposed to a common organiza-
about the president's predicament, one vicetional culture, but any such effect was prob-
president turned to the president and said,ably offset by the fact that they came from
"Why didn't you tell us? We could have donewidely separated divisions, and some were
so much more to help you and ourselves." foreign nationals who had worked in overseas
subsidiaries.
The group's level of trust had remained high,
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
238 ADMINISTRATIVE SCIENCE QUARTERLY
Finally, this study revealed that theory and confirmation is needed and how much re-
research on group forces have had only a quires further investigation.
minor impact on the thinking of managers. Finally, this research offers evidence that a
The managers in this study were among the social phenomenon, trust, can significantly
best educated and the most sophisticated to alter managerial problem-solving effective-
be found in corporate organizations. After ness.
completing the questionnaires, but without
any information about the trust model, they
were brought together and asked for their Dale E. Zand is a professor of management
explanation of what had happened in the two in the Graduate School of Business Adminis-
groups. They consistently responded that the tration at Netw York University.
outcomes were the result of the personalities
RE.ERENCES
of the men (who had been randomly as-
signed to the different roles) or the presi- Baldwin, Alfred L., Joan Kalhorn, and Fay
dent's style (which they interpreted as Hoffman Breese
autocratic or democratic) or the time he 1945 "Patterns of parent behavior." Psy-
chological Monograph, 58, 268, pp. 1-
stated his decision not to expand (early or
75.
late in the meeting). The possibility that a
Blake, Robert R., and Jane S. Mouton
shared level of trust, that is, a group force or
1964 The Managerial Grid. Houston: Gulf
a belief held by several or all members of a Publishing.
group, could constitute a social reality which Bradford, Leland P., Jack R. Gibb, and Kenneth
could significantly affect problem-solving D. Benne
effectiveness was not mentioned. 1964 T-Group Theory and Laboratory
Method. New York: John Wiley.
CONCLUSIONS Deutsch, Morton
1962 "Cooperation and trust: some theoreti-
The findings of this study confirm the cal notes." In Marshall R. Jones (ed.),
hypotheses derived from the model. The Nebraska Symposium on Motivation.
results indicate that it is useful to concep- Lincoln, Nebraska: University of Ne-
tualize trust as behavior that conveys ap- braska Press, pp. 275-319.
Dunnette, Marvin D., and John P. Campbell
propriate information, permits mutuality of
1968 "Laboratory education: impact on
influence, encourages self-control, and avoids
people and organizations." Industrial
abuse of the vulnerability of others.
Relations, 8, 1, pp. 1-27.
It appears that when a group works on a Fiedler, Fred E.
problem, there are two concerns: one is the 1953 "Quantitative studies on the role of
problem itself, the second is how the mem- therapists' feelings toward their pa-
bers relate to each other to work on the tients," in Orval H. Mowrer (ed.),
problem. Apparently in low-trust groups, Psychotherapy: Theory and Research.
New York: Ronald Press, Ch. 12.
interpersonal relationships interfere with and
Gibb, Jack R.
distort perceptions of the problem. Energy
1961 "Defense level and influence potential
and creativity are diverted from finding com- in small groups." In Luigi Petrillo and
prehensive, realistic solutions, and members Bernard M. Bass (eds.), Leadership
use the problem as an instrument to minimize and Interpersonal Behavior. New York:
their vulnerability. In contrast, in high-trust Holt, Rinehart Winston, pp. 66-81.
groups there is less socially generated un- 1964 "Climate for trust formation." In
certainty and problems are solved more effec- Leland P. Bradford, Jack R. Gibb and
tively. Kenneth D. Benne (eds.), T-Group
Theory and Laboratory Method. New
This study also offers qualitative support
York: John Wiley, pp. 279-301.
for the spiral-reinforcement model. It sug-
House, R. J.
gests that mutual trust or mistrust, among 1967 "T-group education and leadership
members of a group, are likely to be re- effectiveness: a review of the empiric
inforced, unless there is marked or prolonged literature and a critical evaluation."
disconfirming behavior. Exactly what dis- Personnel Psychology, 20, 1, pp. 1-32.
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms
Zand: PROBLEM SOLVING 239
This content downloaded from 210.57.222.69 on Sat, 03 Mar 2018 12:31:20 UTC
All use subject to http://about.jstor.org/terms