^ in large, complex organizations, can- j
not perform by themselves all tbe ,
A good managei is not one who
needs personal success or
who is people oriented - but
one who has a need for power.
Power Is the
Great Motivator
by David C. McClelland and David H. Burnham
What makes or motivates a good !
manager? The question is enormous |
in scope. Some people might say that ,
a good manager is one who is sue- ,
cessful - and by now most business ,
researchers and businesspeople ^
know what motivates people who ^
successfully run their own small
businesses. The key to their success
has turned out to be what psycholo-
gists call the need for achievement,
the desire to do something better or
more efficiently than it has been
done before. Any number of books
and articles summarize research
studies explaining how the achieve^
ment motive is necessary for a per-
son to attain success.
But what has achievement moti-
vation got to do with good manage-
ment? There is no reason on theoret-
ical grounds why a person who has a
strong need to be more efficient
should make a good manager. While
it sounds as if everyone ought to
have the need to achieve, in fact, as
psychologists define and measure
achievement motivation, the need
to achieve leads people to behave in
This aiticle was originally pub-
lished in March-April 1976, It won
the McKinsey Award for excellence
and has consistently been one of
the best'SeUingHBR reprints, Forits
republication as an HBR Classic.
David C, McClelland has written a
retrospective commentary.
126
tasks necessary for success. They |
must manage others to perform for |
the organization. And they must be
willing to do without immediate
and personal feedback since tasks
are spread among many people.
Tbe manager's joh seems to call
more for someone wbo can influ-
ence people than for someone who
does things better alone. In motiva-
tional terms, then, we might expect
the successful manager to have a
greater need for power than a need to
achieve. But there must he other
qualities besides tbe need for power
that go into tbe makeup of a good
manager. We will discuss bere just
what these qualities are and bow
they interrelate.
To measure the motivations of
ways that do not necessarily engen- managers, we studied a number of
them in different large U.S. corpo-
der good management. rations who were participating in
For one thing, because they fo- management workshops designed to
cus on personal improvement and improve their managerial effective-
doing things better by themselves, ness. (See the insert "Workshop
acbievement-motivated people Techniques.") We concluded that
want to do things themselves. For the top manager of a company must
another, they want concrete short- possess a high need for power, that
term feedback on their performance is, a concern for influencing people. ^
so that they can tell how well they However, this need must be disci-
are doing. Yet managers, particularly plined and controlled so that it is di-
rected toward tbe benefit of the in- I
stitution as a whole and not toward
the manager's personal aggrandize-
ment. Moreover, the top manager's
need for power ought to be greater
tban his or her need for being hked.
Measuring Managerial
Effectiveness
What does it mean when we say
that a good manager has a greater
need for power than for achieve-
ment? Consider tbe case of Ken
Briggs, a sales manager in a large
U.S. corporation who joined one of
our managerial workshops. About
I six years ago. Ken Briggs was pro-
moted to a managerial position at
headquarters, where he was respon-
sible for salespeople who service his
company's largest accounts.
In filling out bis questionnaire at
the workshop. Ken showed that he
correctly perceived what his job re-
quired of bim, namely, that be
should influence others' success
DRAWINGS BY PETER SIMPSON COOK
 more than achieve new goals him-
I shop leader, he became more and
self or socialize with his subordi- tive does? It is not enough to suspect
, more upset. He finally agreed, how-
nates. However, when asked with that power motivation may be im-
: ever, that tbe results confirmed feel-
other memhers of the workshop to I ings he had been afraid to admit to portant,- one needs hard evidence
I write a story depicting a managerial I himself or others. For years, he had that people who are better managers
^ situation. Ken unwittingly revealed I been miserable in his managerial than Ken Briggs is do in fact possess
I through his fiction that he did not ^ role. He now knew the reason: he stronger power motivation and per-
sbare those concerns. Indeed, he dis- I simply did not want, and he had not haps score higher in other character-
' covered that his need for achieve- been able, to influence or manage istics as well. But how does one de-
' ment was very high-in fact, over the ' others. As he thought back, he real- cide who is the better manager?
' ninetieth percentile - and his need ized he had failed every time he had Real-world performance measures
' for power was very low, in ahout the tried to influence his are hard to come by if one is trying to
' fifteenth percentile. Ken's high need staff, and he felt worse
' to achieve was no surprise-after all, than ever.
he had been a very successful sales-
man-but obviously his motivation ^ Ken had responded
to influence others was much less to failure by setting
than his job required. Ken was a very high standards -
little disturbed but thought that per- his office scored in '""'
haps the measuring instruments the ninety-eighth per-
were not accurate and that the gap centile on this scale - and by trying
to do most things himself, which rate managerial effectiveness in pro-
between the ideal and his score was duction, marketing, finance, or re-
not as great as it seemed. was close to impossible; his own ac-
search and development. In trying
tivity and lack of delegation conse-
Then came the real shocker. Ken's to determine who the better manag-
quently left his staff demoralized.
subordinates confirmed what his ers were in Ken Briggs's company,
Ken's experience is typical of those we did not want to rely only on their
stories revealed: he was a poor man-
who have a strong need to achieve >' superiors. For a variety of reasons
ager, having little positive impact on '
those who worked for him. They felt • but low power motivation. They iI superiors' judgments of their subor-
they had Httle responsibility dele- • may become very successful sales- '' dinates' real-world performance
gated to them, he never rewarded i people and, as a consequence, may may be inaccurate. In the absence of
but only criticized them, and the of- ' be promoted into managerial jobs for some standard measure of perfor-
fiee was not well organized but was ' which they, ironically, are unsuited. mance, we decided that the next best
confused and chaotic. On all those If achievement motivation does •
• mdex of a manager's effectiveness
scales, his office rated in the tenth to . not make a good manager, what mo- i
I would be the climate he or she cre-
fifteenth percentile relative to na- i • ates in the office, reflected in the
tional norms. ' morale of subordinates.
As Ken talked the results of the Almost by definition, a good man-
survey over privately with a work- • ager is one who, among other things,
helps subordinates feel strong and
David C, McClelland is a~four^der responsible, rewards them properly
and director of McBer and Com- for good performance, and sees that
pany; the director of research at things are organized so that subor-
Hay McBer, a human-resources dinates feel they know what they
management-consulting firm; Dis- should be doing. Above all, man-
tinguished Research Professor of agers should foster among subordi-
Psychology at Boston University: nates a strong sense of team spirit, of
and a professor emeritus of psychol- pride in working as part of a team. If '
ogy at Harvard University in Cam- a manager creates and encourages '
bridge. Massachusetts, He is the au^ this spirir, his or her subordinates '
thor of many books, including The certainly should perform better.
Achieving Society (Van Nostrand In the company Ken Briggs works
1961), Power: The Inner Experience for, we have direct evidence of a
(Irvmgton Publishers, 1975), and connection between morale and per-
Human Motivation (Cambridge formance in the one area where per-
University Press, 1988). David H formance measures are easy to find-
Burnham was president and chief namely, sales. In April 1973, at least
executive officer of McBer and three employees from this com-
Company when this article was pany's 16 sales districts filled out
originally published and now works questionnaires that rated their office
as a consultant in the Boston area for organizational clarity and team
spirit. Tbeir scores were averaged
HARVARD BUSINESS REVIEW January-February 1995
427

and totaled to give an overall morale ^
score for each office. The percentage ^
gains or losses in sales for each dis- ^
trict in 1973 were compared with ^
those for 1972. The difference in ^
sales figures by district ranged from
a gain of nearly 30% to a loss of 8%,
with a median gain oi about 14%.
The graph "The Link Between
Morale and Sales" shows how, in
Ken Briggs's company, at least, high
morale at the beginning of the year
became a good index of how well the
sales division would actually per-
form throughout the year. Moreover,
it seems likely that the manager
who can create high morale among
salespeople can also do the same for
employees in other areas (produc-
tion, design, and so on), which leads
to better overall performance. What
characteristics, then, does a manager
need to create that kind of morale?
he Power Factor
In examining the motive scores of
more than 50 managers of both high-
and low-morale units in all sections
of the same large company, we found
Morale score [perceived organizational clarity pios team spiri
iThigher^thaii the fiftieth percentile of national norms.
128
that most of tbe managers - more
tban 70% -were high in power mo-
tivation compared with tbe aver-
age person. This finding confirms
that power motivation is important
for management. [Re-
member that, as we
use the term, power
motivation refers not
to dictatorial behavior
but to a desire to bave
impact, to be strong
and influential.) The
better managers, as
judged by the morale
of tbose working for tbem, tended to
score even higher in power motiva-
tion. But the most important deter-
mining factor of high morale turned
out to be not how tbeir power moti-
vation compared witb their need to
acbieve but whether it was higher
tban their need to be liked. Tbis rela-
tionship existed for 80% of the bet-
ter sales managers as compared with
only 10% of the poorer managers.
And tbe same held true for other
managers in nearly every part of tbe
organization.
n Morale and Sales
33 52
In the research, product develop-
ment, and operations divisions, 73%
of the' better managers had a stronger
need for power than a need to be
liked, as compared with only 22% of
ruinina the morale of
omce.
the poorer managers, who tended to
be what we term affiliative manag-
eis - whose strongest drive is to be
liked. Why should this be so? Sociol-
ogists have long argued that for a bu-
reaucracy to function effectively,
those who manage it must apply
rules universally; that is, if they
make exceptions for the particular
needs of individuals, the whole sys-
tem will break down.
The manager with a high need to
be liked is precisely the one who
wants to stay on good terms with
everybody and, therefore, is the one
most likely to make exceptions for
particular needs. If an employee asks
for time off to stay home and look |
after a sick spouse and the kids, the i
affiiiative manager agrees almost i
without thinking, out of compas- I
sion for the employee's situation. I
When former President Gerald
Ford remarked in pardoning Richard
, Nixon that Nixon had "suffered
enough," he was responding as an
, affiliative manager would because
, be was empathizing primarily with
' Nixon's needs and feelings. Socio-
• logical theory and our findings both
' argue, however, that the person
wbose need for affiliation is high
does not make a good manager. This
' kind of person creates low morale
' because he or she does not under-
' stand that otber people in tbe office
' will tend to regard exceptions to the
' rules as unfair to themselves, just as
' many U.S. citizens felt that it was
unfair to let Nixon off and punish
others who were less involved than
he was in the Watergate scandal.
So far, our findings are a little
', alarming. Do tbey suggest that the
. good manager is one who cares for
HARVARD BUSINESS REVIEW ianuary-February 1995
 , power and is not at all concerned
^ about the needs of other people? Not
I quite, for tbe good manager has
other characteristics that must still
I be taken into account. ile 'ontting of average sconas [national norms)
I Above all, the good manager's
20_ 30
I power motivation is not oriented to-
ward personal aggrandizement but Sense of responsibility
I toward the institution that he or she
I serves. In another major research
I study, we found that the signs of
controlled action or inhibition that
^ appear when a person exercises
imagination in writing stories tell a
' great deal about the kind of power
that person needs.' We discovered
• that if a high power-morive score is
I balanced by high inhibition, stories
I about power tend to be altruistic,
i That is, the heroes in the story exer-
. cise power on behalf of someone
•• else. This is tbe socialized face of
I power as distinguished from the
I concern for personal power, which is
eharacterisric of individuals whose
Scores for at least three subordinates of:
stories are loaded with power im-
agery but show no sign of inhibition a Affiliotive mar,agers iaffilia.ion greater »han power,, high mhib
^ or self-control. In our earlier study, • • Personal power managers (pawe, greater thoo offiliolion, ow
, we found ample evidence that the i • ln.t,tut,o,val managers (power greater fhan affilfatio.. hig
latter individuals exercise their I Averoge sco.es on sek.ted dimension b , s.bo.dina.e.
, power impulsively. They are more
, often rude to other people, they
drink too much, they try to exploit
others sexually, and they collect
tbeir employees to be more produc-
symbols of personal prestige such as In short, as we expected, affiliative
tive. Now let us compare them with
fancy cars or big offices. managers make so many ad homi-
affiliative managers (those in whom
' tbe need for affiliation is higher tban nem and ad hoc decisions that they
Individuals high in power and in I tbe need for power) and with tbe almost totally abandon orderly pro-
control, on tbe other hand, are more personal power managers (those in cedures. Tbeir disregard for proce^
institution minded; they tend to get whom tbe need for power is higher dure leaves employees feeling weak,
elected to more offices, to control tban for affiliation but wbose inbibi- irresponsible, and without a sense of
their drinking, and to have a desire tion score is low). what might happen next, of where
to serve others. Not surprisingly, we they stand in relation to tbeir man-
found in the workshops that the bet- In tbe sales division of the com- ager, or even of what they ought to
ter managers in the corporation also pany we cbose to use as an illus- be doing. In this company, the group
tend to score high on both power and tration, there were managers wbo of affiliative managers portrayed in
inhibition. matcbed tbe three
types fairly closely.
Three Kinds of Managers Tbe cbart"Wbicb Man-
ager Was Most Effec-
Let us recapitulate what we have tive?" shows bow tbeir
discussed so far and have illustrated subordinates rated tbe
with data from one company. The offices tbey worked in
better managers we studied - what on responsibility, or-
we call institutional managers-are ganizational elarity, and team spirit.
high in power motivation, low in af- Managers wbo are concerned about the chart were below the thirtieth
filiation motivation, and higb in in- being liked tend to have subordi- percentile in morale scores.
hibition. They care about institu- , nates who feel that they have little The managers who are motivated
tional power and use it to stimulate i personal responsibility, tbat organi- by a need for personal power are
zational procedures are not clear somewhat more effective. They are
1. David C McClelland, William N. Davis
Rudolf Kalin, and Eric Warner, The Drinking and that tbey have little pride iri able to create a greater sense of re-
Mai: INew York: The Free Press, 1972]. tbeir work group. sponsibility in their divisions and,
above all, a greater team spirit. They
HARVARD BUSINESS REVIEW January-February 1995
129

-1 I 11 HllllllllHlllllliH' IIIHIill' llji

 and high in inhibition. He exercised
can be thought of as managerial witb most businesses in tbe United his power in a controlled, organized
equivalents of successful tank com- States. Let us say frankly that we way. The stories he wrote reflected
manders such as General George think the bogeyman of authoritari- this fact. In one story, for instance,
Patton, whose own daring inspired anism has heen wrongly used to he wrote, "The men sitting around
admiration in his troops. But notice downplay the importance of power the table were feeling pretty good;
how in the chart these people are in managemem. After all, manage- tbey bad just finished plans for
still only in the fortieth percentile in ment is an influence game, Some reorganizing tbe company; the com-
the amount of organizational clarity proponents of democratic manage- pany bas been beset with a num-
that they create, as compared with ment seem to have forgotten tbis ber of organizational problems. This
the higb-power, low-affiliation, fact, urging managers to be more group, beaded hy a hard-driving,
concerned with people's brilliant young executive, has com- i
personal needs tban with pletely reorganized the company i
helping tbem to get structurally witb new jobs and re- |
things done. sponsibilities...." !
But much of tbe appar-
ent confUct between our This described how George him- I
wrongly used to findings and those of otb- self
and
was perceived by the company, I
shortly after tbe workshop, be I
er behavioral scientists in
the importance tbis area stems from the was promoted to vice president in i
charge of all sales. But George was |
in fact that we are talking
about motives, and be- also known to bis colleagues as a |
haviorists are often talking about ac- monster, a tough guy who would •
high-inhibition managers - institu- tions. What we are saying is tbat "walk over his grandmother" if she
tional managers. managers must be interested in play- I stood in the way of bis advance-
Managers motivated by personal ing the influence game in a con- I ment. He bad tbe rigbt motive com-
power are not disciplined enough trolled way. Tbat does not neces- bination and, in fact, was more lnter-
I
to be good institution builders, and sarily mean that they are or should I ested in institutional growth than
often tbeir subordinates are loyal to be autboritarian in action. On tbe , be was in personal power, but bis
them as individuals ratber than contrary, it appears tbat power-moti- i managerial style was all wrong. Tak-
to tbe institution they serve. When vated managers make tbeir subordi- I ing his cue from some of tbe top ex-
a personal power manager leaves, nates feel strong ratber tban weak. I ecutives in the corporation, he told
disorganization often follows. Tbe Tbe true authoritarian in action I people wbat tbey had to do, and he
strong group spirit that the manager would have the reverse effect, mak- 1 threatened them witb dire conse-
has personally inspired deflates. Tbe ing people feel weak and powerless. i quencesiftheydidnotdoit.
subordinates do not know wbat to Thus another important ingredi- i When George was confronted with
ent in the profile of a manager is his authoritarianism in a workshop,
do for themselves. he recognized tbat tbis style was
Of the managerial types, the in- ^ managerial style. In the illustrative counterproductive-in fact, in another
stitutlonal manager is the most company, 63% of tbe better manag-
ers (those whose subordinates had part of the study we found tbat it was
successful in creating an effective associated witb low morale - and he
work climate. Subordinates feel tbat higher morale) scored higher on the
democratic or coaching styles of subsequently changed to acting more
they have more responsibility. Also, like a coacb, wbicb was the scale on
those kinds of managers create high managemem as compared with only
22% of tbe poorer managers, a statis- which be scored the lowest initially.
morale because they produce tbe
greatest sense of organizational clar- tically significant difference. By con- George saw more clearly tbat his job
was not to force otber people to do
ity and team spirit. If such a manager ' trast, the latter scored higher on au- things but rather to help them to fig-
leaves, he or she can be more readily ' thoritarian or coercive management ure out ways of getting their johs
replaced by anotber because tbe em- I styles. Since the better managers done better for tbe company.
ployees have been encouraged to be were also higher in power moti-
loyal to the institution ratber tban ' vation, it seems that in action they
to a particular person. ' express their power motivation in Profile of the
' a democratic way, which is more Institutional Manager
Since it seems undeniable that a
manager witb a power orientation One reason it was easy for George
creates better morale in subordi- ' likely to he effective.
To see how motivation and style Prentice to cbange his managerial
nates tban one witb a people orienta- ', interact, style was that, as we saw in bis
tion, we must consider that a con- consider tbe case of George imaginative stories, he was already
cern for power is essential to good Prentice, a manager in the sales divi- having thoughts about helping oth-
management. Our findings seem to sion of anotber company. George ers characteristic of people witb the
fly in the face of a long and influen- bad exactly tbe rigbt combination of institution-building motivational
tial tradition of organizational psy- motives to be an institutional man-
ager. He was high in the need for pattern. In further examining insti-
chology, wbicb insists tbat authori- power, low in tbe need for affiUation, continued on page 133
tarian management is what is wrong
HARVARD BUSINESS REVIEW lanuary-February 1995
130
 • tution builders' thoughts and ac-
tbat it will make them less hard
• tions, we found they have four major Consider tbe results shown in the
driving, less expansion minded, and
I characteristics: chart "Managers Can Change Their
less committed to organizational ef-
! n Institution managers are more or- fectiveness. Our data do not support Style,'' where "before" and "after"
I ganization minded; that is, they tend their fears. workshop training scores are com-
^ to join more organizations and to pared. To judge by their subordi-
Those fears are exactly the ones nates' responses, the managers were
feel responsible for building up those
George Prentice had before he went clearly more effective after coming
organizations. Furthermore, they
to the workshop. Afterward, be was to terms with their style. The subor-
believe strongly in the importance of
a more effective manager, not de- dinates felt tbat tbey received more
' centralized authority.
spite bis loss of some of tbe sense of rewards, that tbe organizational pro-
' OThey report that tbey like to , his own importance but because of
• work, Tbis finding is particularly in- I it. The reason is simple: bis subordi- cedures were clearer, and that mo-
teresting because our research on , nates beheved afterward that be was rale was higher.
achievement motivation has led genuinely more concerned about the But what do those differences
• many commentators to argue that company than he was about himself. mean i:n human terms? How did the
achievement motivation promotes Whereas once they respected his managers cbange? Sometimes they
the "Protestant work ethic." Almost confidence but feared him, they now decided they should get into another
tbe precise opposite is true. People trust him. Once, he supported their line of work. This happened to Ken
wbo have a high need to achieve like image of him as a "big man" by , Bnggs, for example, who found that
to reduce their work by becoming talking about the new Porsche and tbe reason be was doing so poorly as
more efficient. Tbey would like to Honda be bad bought; when we saw a manager was because he had al-
see the same result obtained in less him recently, he said, almost as an most no interest in influencing oth-
time or with less effort. But manag- aside, "I don't buy things anymore." ! ers. He understood how he would
ers who have a need for institution- have to change in order to do well in
al power actually seem to like the his present job, but in the end de-
disciphne of work. It satisfies their ' Altering Managerial Style cided, with the help of management,
need for getting things done in an ' George Prentice was able to ' that he would prefer to work back
orderly way, i change his managerial style after ' into his first love, sales.
learning more about himself. But
• Tbey seem quite willing to sacri- I Ken Briggs moved into remainder-
does self-knowledge generally im-
fice some of tbeir own self-interest ' ing, helping retail outlets for his
prove managerial behavior-
for tbe welfare of the organization company's products get rid of last
they serve.
• They bave a keen sense of justice
^ It is almost as if they feel tbat people
I who work hard and sacrifice for the
I good of tbe organization should and
I will get a just reward for tbeir effort.
It is easy to see bow each of these
four concerns helps a person become
a good manager, concerned about
what the institution can achieve.
We discovered one more fact in
studying the better managers at
George Prentice's company. They
were more mature. Mature people
can be most simply described as less
egotistic. Somehow their positive
self-image is not at stake in tbeir
jobs, Tbey are less defensive, more
willing to seek advice from experts,
and have a longer-range view. They
accumulate fewer personal posses-
sions and seem older and wiser. It is
as if they have awakened to the fact
that they are not going to live forever
and have lost some of the feeling
that their own personal future is all U Before manoger training
that important. • After manager training

Many U.S. businesspeople fear ' Average scores on selected dimensions by more
tbis kind of maturity. Tbey suspect ' 50 solespBcple before and after their

HARVARD BUSINESS REVIEW January-February 1995

133

year's stock so that tbey can take on
each year's new styles. He is very
successful in this new role; he bas
cut costs, increased dollar volume,
and in time has worked himself into
an independent role selling some of
the old stock on his own in a way
that is quite satisfactory to the busi-
ness. And he does not have to man-
age anybody anymore.
In George Prentice's case, less
change was needed. He obviously
was a very competent manager witb
the right motive profile for a top
company position. Wben he was
promoted, he performed even more
successfully than he had previously
because be realized that he needed
to become more positive in bis ap-
proach and less coercive in his man-
agerial style.
But what about a person who does
not want to change johs and discov-
ers that he or she does not have the
right motive profile to be a manager?
The case of Charlie Blake is instruc-
tive. Cbarlie was as low in power
motivation as Ken Briggs, his need
to achieve was about average, and
his affiliation motivation was ahove
average. Thus he bad the affiliative-
manager profile, and, as expected,
the morale among his subordinates
was very low. When Charlie learned
that his subordinates' sense of
responsibility and perception of a re-
ward system were in the tenth per-
centile and that team spirit was in
the thirtieth, be was shocked. When
shown a film depicting three man-
agerial climates, Charlie said he
preferred what turned out to he the
autboritarian climate. He became
angry when the workshop trainer
and otber members in the group
pointed out the limitations of this
managerial style. He became ob^
' structive in the group process and
' objected strenuously to wbat was
being taught.
In an interview conducted much
later, Charlie said, "1 blew my cool, |
When I started yelling at you for be-
ing all wrong, I got even madder
when you pointed out that, accord-
ing to my style questionnaire, you
bet that tbat was just what I did to
my salespeople. Down underneath,
I knew something must be wrong.
Tbe sales performance for my divi-
sion wasn't so good. Most of it was
due to me anyway and not to my
salespeople. Obviously, their reports
that tbey felt I delegated very little
responsibility to tbem and didn't re-
ward them at all had to mean some-
thing. So I finally decided to sit
down and try to figure what I could
do about it. I knew I had to start he-
ing a manager instead of trying to do
everything myself and blowing my
cool at others because they didn't do
what I thought they should. In the
I end, after I calmed down on the way
I back from the workshop, I realized

Workshop Techniques or is likely to he able to adjust to

We derived the case studies and
data used in this article from a
number of workshops we eon-
ducted, during which executives
learned about their managerial
styles and abilities as well as how
to change them. The workshops
also provided an opportunity for
us to study which motivation
patterns in people make for the
best managers.
At the workshops and in this
article, we use the technical
terms need for achievement,
need for affiliation, and need for
power. The terms refer to mea-
surable factors indicating motiva-
tion in groups and individuals.
Briefly, those characteristics are
measured by coding managers'
spontaneous thoughts according
to how often they think about
doing something better or more
efficiently than before (need for
achievement), ahout establishing
or maintaining friendly relations
with others (need for affiUation),
or about having an impact on oth-
ers (need for power). When we
talk about power, we are not talk-
ing about dictatorial power but
about tbe need to be strong and
influential.
When the managers first ar-
rived at the workshops, they were
asked to fill out a questionnaire
about their jobs. Each participant
analyzed his or her job, explain-
ing what he or she thought it re-
quired. The managers were asked
to write a number of stories about
pictures of various work situa-
tions we showed them. The sto-
ries were coded according to
how concerned an individual was
with achievement, affiliation, or
power, as well as for the amount
of inhibition or self-control they
revealed. We then compared the
results against national norms.
The differences between a per-
son's job requirements and his or
her motivational patterns can of-
ten help assess whether the per-
son is in the right job, is a candi-
date for promotion to another job.
fit the present position.
To find out what kind of mana-
gerial style the participants had,
we then gave them a question-
naire in which tbey had to choose
how they would handle various
realistic work situations in office
settings. We divided their answers
into six management styles or
ways of dealing with work situ-
ations. The styles were demo-
cratic, affiliative, pacesetting,
coaching, coercive, and authori-
tarian. The managers were asked
to comment on the effectiveness
of each style and to name the
style they preferred.
One way to determine how ef-
fective managers are is to ask the
people who work for them. Thus,
to isolate the characteristics that
good managers have, we asked at
least three subordinates of each
manager at the workshop ques-
tions about their work situations
tbat revealed characteristics of
their supervisors according to six

HARyARD BUSINESS REVIEW January-February 1995

134

that It is not so bad to make a mis-
take; it's bad not to learn from it "
After the course, Charlie put his
plans into effect. Six months later,
his subordinates were asked to rate
him again. He attended a second
workshop to study the results and
reported, "On the way home, r was
nervous. I knew I had been working
With those guys and not selling so
much myself, but I was afraid of
what they would say about how
things were going in the office
When I found out that the team
spirit and some of those other low
scores had jumped from around the
thirtieth to the fifty-fifth percentile
criteria: (1) the amount of confor-
mity to rules the supervisor re-
quires, |2) the amount of respon-
sibility they feel they are given,
(3) the emphasis the department
places on standards of perfor-
mance, (4) the degree to which
rewards are given for good work
eompared with punishment
when something goes wrong, {5) Those in Stage II are interested
the degree of organizational clar-
ity in the office, and (6) its team
spirit.' The managers who re-
ceived the highest morale scores
(organizational clarity plus team
spirit) from their subordinates
were determined to be the best
managers, possessing the most
desirable motive patterns.
We also surveyed the subordi-
nates six months later to see if companies.
morale scores rose after managers
completed the workshop.
We measured participants on
one other eharacteristic deemed
important for good management-
maturity. By coding the stories
that the managers wrote, which
HARVARD BUSINESS REVIEW fanuary-February 1995
BE BETTER
I was so delighted and relieved that I
couldn't say anything all day long."
THAN THE BEST
When he was asked how he acted "An essential guide for corporate leaders
differently from before, Charlie said, rt can significantly raise the level of man-
''In previous years when eorporate agement thinking."
-MICHAEL MACOBY, principal
headquarters said we had to make consultant to AT&T's Workplace of the Future
] 10% of our original goal, I had
called the salespeople in and said "Evaryone in business should read this
in effect, 'This is ridiculous; we are book." —NORMAN R. AUGUSTINE,
CEO, Martin Marietta
not going to make it, but
you know perfectly well
a what will happen if we
don't. So get out there and
f to work your tails off.' The
if." 20 hours
result was that I worked
a day and they
did nothing.
"This time I approached the sales-
people differently. I told them three
things. First, they were going to have .
to do some sacrificing for the com- i
pany. Seeond, working harder is not !
going to do much good because we '
are already working about as hard as THE DEMOCRATIC
CORPORATION
continued on page 138 ' A Radical Prescription for Recreating Corporate
Amer[ca and Rediscovering Success
RUSSELL LACKOFF $23.00, 272 pp
revealed their attitudes toward Will hft the spirits of even the most
authority and the kinds of emo- pessimistic... Use this book as a mirror
for your company. Its lessons are mind-
tions displayed over specific is- opening."
sues, we were able to pinpoint Soundview Executive Book Summaries
managers at one of four stages in
the progress toward maturity "Highiy pro vocative...shows the path to
People in Stage I are dependent on creating brilliant products while simul-
taneously renewing the whole firm "
others for guidance and strength. - J I M WOMACK, Principal Research Scientist.
MIT Japan Program
primarily in autonomy. In Stage
III, people want to manipulate
others. In Stage IV, they lose their
egotistic desires and wish to serve
others selflessly.^
The conclusions we present in
this article are based on work-
shops attended by more than 500
managers from some 25 U.S. cor-
porations. We drew the examples
in the charts from one of those
L Based on Cf 1, Litwm and R.A. Strmge"r's
Monvat,oncmd Organization^} CuLte
(Boston: Division of Research, Harvard THE PERPETUAL
Business School, 1966|.
ENTERPRISE MACHINE
2. Based on work by Abigail Stewart, as re- Seven Keys to Corporate Renewal Through
ported m David C, McClelland's Successful Product and Process Development
H. KENT BOWEN, KIM B. CLARK, CHARLES A
HOLLOWAV, and STEVEN C. WHEELWRIGHT '
$27.50, 436 pp.
FORD To charge,
1-800-451-7556
(M-F, 9-5 EST)
U S I N E S S
UNIVERSITY PRESS
 1972 sales, an increase that played a i
^ salespeople - and lower m his ten- large part in turning the overall com- ;
we can. What will be required are , dency to try to do everything him^ ; pany performance around from a $ 15 ,
special deals and promotions. You , self He asks people to sacrifice tor ^ million loss in 1972 to a $3 million i
are going to have to figure out some | the company. He does not defensive- , profit in 1973. The company con- |
new angles if we are to make it. , ly chew them out when they chal- i tinued to improve its performance ,
Third, I'm going to back you up. I m , lenge him but tries to figure out , in 1974 with an 11 % further gain m i
eoing to set a realistic goal with each i what their needs are so that he can ,
influence them. He realizes that his . sales and a 38% increase in profits, i
of you. If you make that goa but • job is more one of strengthening and : Of course, not everyone can be ^
don't make the company goal, 1 U , reached by a workshop. Henry Car- ,
see to it that you are not punished. supporting his subordinates than ot ter managed a sales office for a com-
But if you do make the company criticizing them. And he is keenly pany that had very low morale
goal, I'll see to it that you will get interested in giving them lust re- (around the twentieth percentile) be-
some kind of special rewards.'' wards for their efforts.
I The changes in his approach to his fore he went for training. When
When the salespeople challenged • iob have certainly paid off. The sales I morale was checked some six
Charlie, saying he did not have I figures for his office in 1973 were up ' months later, it had not improved.
enough influence to give them re- I, more than 16% over 1972 sales and ', Overall sales gain subsequently re-
wards, rather than becoming angry, i' up still further in 1974 over 1973 I fleeted this fact - only 2% above the
Charlie promised rewards that were I sales. In 1973, his gain over the pre- ' previous year's figures.
in his power to give ^ such as longer I vious year ranked seventh in the na- ' Oddly enough, Henry's problem
vacations. ', tion, in 1974, it ranked third. And he '
, was that he was so well liked by
Note that Charlie has now begun , wasn't the only one in his company I everybody that he felt little pressure
to behave in a number of ways that ' to change. Always the life of the
we found to be characteristic of the ! to change managerial styles. Overall ', party, he is particularly popular he-
good institutional manager. He is, I sales at his company were up sub- I cause he supplies other managers
above all, higher in power moti- I, stantially in 1973 compared with
vation - the desire to influence his
Retrospective Commentary
AT&T follow-up research did
to a higher level much more often
1

provide an opportunity to observe

Two important changes have
occurred in the workplace since
David H. Burnham and I wrote
"Power Is the Great Motivator"
in 1976. The hig, old-fashioned
hierarchical organizations we
studied have flattened out. And
female managers have entered
the workplace in full force.
Yet our findings ahout manage-
ment style still hold true, regard-
less of a manager's gender, for the
type of hierarchical organization
we observed in the article. Suc-
cessful managers- what we called
institutional managers - have a
strong need for power (that is, for
influencing others) that is greater
than their need to be Uked, and
they exhibit self-control.
That finding was confirmed
through subsequent research, in-
cluding a study of people who
were promoted up the managerial
ranks at AT&T. Of people who
joined AT&T between 1956 and
I960, the institutional managers
had been promoted 16 years later
than other types of managers, we
found. We also discovered that
the same managerial characteris-
tics predicted future success tor
both men and women who en-
tered the AT&T system between
1976 and 1980.
However, in a recent study ot
PepsiCo, a large, decentralized
company, we found that having
a high need for achievement con-
tributes more to success than
does a high interest in influenc-
ing other people. In fact, the need
for power was often a handicap
in that company, That confirms
earlier convincing evidence that
a constant concern for improve-
ment, for growing the husiness in
a cost-efficient way, characterizes
successful managers of small
companies, which is what many
subsidiaries of PepsiCo essen-
tially function as.
To show conclusively how
women manage, we still need
much more information. But the
the subtleties m style differences
between the best female and male
managers. Using an objective
coding system to analyze the sto-
ries managers wrote ahout hypo-
thetical situations, Ruth Jacobs, a
principal at McBer and Company^
and I found the women much
more consensus oriented than the
men. In particular, the women
seemed to think ahout power as a
resource that can be used to influ-
ence outcomes on the job and to
focus the competencies of the
people who work for them. Men
in the study, on the other hand,
tended to think of power more as
an end in itself, as something
they can use to react against or
take power away from others in
authority. Men saw power as a
way to supersede others in power;
women rarely did.
Since 1976, my work focus has
returned to a lifelong interest m
competencies in management.

HARVARD BUSINESS REVIEW Tan uary-February 199.S

138
 with special hard-to-get brands of
cigars and wines at a discount. He I was often a real help to others and gamzation. The top managers shown
uses his close ties with everyone to I the company. As a result, Henry did here have a high need for power and
bolster his position in the company I not have to cope with such ques- an interest in influencing others
even though it is known that his of- tions at all. He had so successfully
both greater than their interest in be-
fice does not perform well compared , developed his role as a likable, help- ing liked by people. The manager's
with others. I ful friend to everyone in manage-
I ment that, even though his sales- concern for power should be social-
people performed badly, ized ^ controlled so that the institu-
he did not feel under any tion as a whole, not only the individ-
pressure to change the ual benefits. People and nations
way he managed people. with this motive profile are empire
What have we learned builders; tbey tend to create high
from Ken Briggs, George morale and to expand the organiza-
Prentice, Charlie Blake, tions they head. But there is also
and Henry Carter? We danger in this motive profile,- as in
His great interpersonal skills be- have discovered what motives make countries, empire building can lead
came evident at tbe workshop when an effective manager - and that I to imperialism arid authoritarianism
he did very poorly at one of the busi- change is possible if a person has the I in companies.
ness games. When the discussion right combination of qualities. I The same motive pattern that pro-
turned to why he had done so badly i duces good power management can
and whether he acted tbat way on Oddly enough, the good manager i
I also lead a company to try to domi-
the job, two prestigious participants m a large company does not have a I^ nate others, ostensibly in the inter-
immediately sprang to his defense high need for achievement, as we de- 'I ests of organizational expansion
explaining away Henry's failure by hnc and measure that motive al- 'I Thus It IS not surprising that big
argumg tbat the way be did things though there must be plenty of that 'I business has had to be regulated pe-
motive somewhere in his or her or I riodically by federal agencies.

Similarly, the best managers pos-

That has meant taking a close
look at people who are outstand-
ing managers and trying to break
down exactly how they go about
work vis-a-vis the way less effec-
tive managers do. The work de-
veloped as a reaction to studies
that seem to resurface every 25
years or so (as they have done
again recently) asserting a link
between intelligence and overall
competence. Intelligence, those
studies claim, is a hereditary fac-
tor that cannot be altered and is
therefore the only relevant mea^
sure of a person's "competence "
My work has shown the oppo-
site to be true. In fact, resuhs
of the Scholastic Aptitude Test
given to college-bound studejits
among others, relate little to how
competently those people man-
age in the workplace later in life
People who scored exceptionally
well on SATs often later func-
tioned poorly as managers, and
people with only average scores
often made the best managers It
IS not intelligence that separates
the best people from the worst
when it comes to job perfor-
mance. To measure competency
levels, I coded interviews with
managers about everyday work
situations for specific behaviors
including self-control, self-confi-'
dence, an ability to get a consen-
sus from people, and strong moti-
vations for achievement, power
or both. ' Summarized in this way, what we
Indeed, sucb motivational
characteristics (as illustrated by
the mstitutional managers in our
article) continually emerge as
wbat separates world-class man-
agers from mediocre ones. That
IS as true today as wben we
wrote the article, and it applies to
women as well as to men. What
has happened in the meantime is
that we bave a better idea of what
combinations of motives and
other competencies that we can
measure in interviews create
managerial success in the new
decentralized organizations
' sess two characteristics that act as
I regulators ^ a greater emotional ma-
I ^""ty'^liere there is little egotism,
, and a democratic, coaching manage-
rial style. If an institutional power
motivation is checked by maturity
It does not lead to an aggressive'
egotistic expansiveness. That means
managers can control their subordi-
nates and influence others around
them without having to resort to co-
ercion or to an authoritarian man-
agement style.
have found out through empirical
and statistical investigations may
sound Uke good common sense. But
It IS more than common sense; now
we can say objectively what the
characteristics of the good manager
are. Managers of corporations can se-
lect tbose who are likely to be good
managers and train those already in
managerial positions to be more ef
more confidence.
-4u(iiorsnote.^thename7 and case
materiahnthisarticle are disguised.
Reprint 95108 ~ ~~

HARVARD BUSINESS REVIEW January-February 1995

139