Managing the Increasing Complexity of Large Organizations
If you look at the history of political theory—and basically
when we talk organizational structure we are talking governance
and political theory—there are two strands in the Western
tradition. One is the constitutionalist, which basically says,
“What are the right laws so that even mediocre people can function
and so that the evildoers are at least confined?” The other
one, which went by the name of the “education of the Christian
prince” and also goes back to the Greeks and even earlier, says,
“How do we form the rulers so that even if there are turbulent
times and the rules aren’t clear, we get the best there is, and we
get virtue and leadership into the system?”
In the last 30 years, we have emphasized the constitutional
approach. Let me say that this is normal in this country, where
business basically started out taking its structural concepts from
the American Constitution. If you look at our organizational
theory, it is tremendously influenced by the Constitution, for better
or worse. I’d say largely for better because at least it preaches
the need to think through the limitations of power. At the same
time, we have had some approaches that are opposite—that start
out with, “How do we form people?”
Oh, some of you may be old enough to remember that I once
was considered a pioneer of human relations, which was an attempt
to offset somewhat the constitutionalist approach with an
approach on the formation of people. And there is organization
development. But let’s face it. Those two things haven’t worked.
They have been minor corrections. And now we will have to
look at the formation of people very seriously, simply because for
some of the problems we have to solve there is no other way.
Let me give you an example. If you look at multinationals
today, they are mostly nineteenth century in their structure,
with a parent company and with subsidiaries that manufacture
the same products for their own home market. But increasingly,
you get incestuous relationships in which the specs come out
of Detroit, the design comes out of Germany, the body comes
out of Brazil, and the transmission out of Mexico—like the
Ford Fiesta. Or if you look at IBM office products, resources
are organized not in terms of products but in terms of stages of
production with labor-intensive work done in one place, with
design done in another, with a lot of the technology done where
the technologists are.
And let me say that technologists are proving remarkably resistant
to migration. Those French specialists prefer to work in
a research lab near the Louvre to working in Hoboken. I can’t
figure it out, but they do. And you can’t get them to move or
even to go down to Connecticut, and so you have to do the research
where the researchers are. It’s no longer true as it was in
the ’50s that researchers are so badly paid in Germany or Austria
or Japan that they’ll only too gladly take that job in Peoria. They
don’t anymore. And so you will have to think through how you
organize and integrate not components but stages. The product
is then sold where the customers are, which is likely to be in developed
countries. And so that subsidiary you have in France is
not a traditional one.
But even more difficult is the subsidiary in Colombia, where,
if you are a pharmaceutical company, Colombia is a developed
market. In fact, the only industry for which developing countries
are fully developed markets is the pharmaceutical for the simple
reason that pharmacy is the only part of modern health care a
very poor country can really afford. It’s the cheapest part, and it
does 60 percent of the job. And so Colombia spends proportionally
more per capita on prescription drugs than most developed
countries, and yet it’s a small market because of the small urban
population—roughly similar to one good sales district in a developed
country, let’s say Manchester or Kansas City.
And yet that head of your subsidiary there has to be a distinguished
man because he will matter a great deal. So he probably
is the most distinguished medical administrator, former dean
of the medical school, and a minister of affairs. How does he
relate to the top? He has to be an equal because he’ll negotiate
with the government there now, with the Catholic nun who
buys drugs for Mercy Hospital. You’ll need his input. He’s a very
distinguished man; very few of that caliber could you get into
your own organization. How do you structure him? And so you
have all kinds of new complexities, which make the simple, traditional
structure very hard. Then one has to say, “First, we need
new structural principles.” The answer to it is, “We ain’t got
none. We have patchwork.”
We had in the history of organization two very simple principles.
One is from [French mining engineer and early twentiethcentury
management theorist] Henri Fayol, and the other one
was what [General Motors Chairman] Alfred Sloan and then
I codified as “federal decentralization.” And they worked where
they worked like a charm, but they have strict requirements and
severe limitations.
You know, when I first heard of the Bell telephone system,
it was beautifully organized and very simple, a good operating
company, and 98 percent of their business was within their territory
because in those days 98 percent of all calls and 98 percent
of all revenue were local calls. Nobody had heard of computer
transmission or what have you. And so, when you look at it today—
and you all know the Bell telephone system is desperately
trying to reorganize itself into regulated and unregulated and
quasi-regulated businesses—you no longer have a local system
that’s interconnected, a long-distance system with local peripherals.
The old organization cannot work and doesn’t. And so
how do you organize? We don’t know the answers; we know
patchwork. And we will all have to live with things that quite
clearly are full of friction and present problems.
We’ve tried to build our organizations as close to mechanical
models as possible because it is simple. [Scientific management
pioneers] Frederick Taylor and Henri Fayol both assumed that
you know what you are doing. You know, a coal mine mines coal.
It’s obvious, isn’t it? Well, you are now in a period in which the
real challenge is to decide what you are doing in the context of
technological change or market change.
The one axis of organization that you’ll say you need is a skeleton.
And all of you know a land animal that is more than six
inches tall needs a skeleton; it can’t be held together by heart and
skin anymore. The organization chart, with its lines of authority
and its reporting, is a skeleton. Now, we always have some
problems with that. If you have a divisional structure, the relationship
between your corporate comptroller and your divisional
comptroller is not a simple one and can’t be decided one way or
another. You all know that when you get to technology, it doesn’t
work if you have coordinating groups, and it doesn’t work any
other way.
Yet how do we organize the new within the old? In fact, can
we organize it? If you look at the last 30 years, the obstacles to
entrepreneurship were exceedingly high in terms of the tax laws,
in terms of our credit structure. Now, we are not as bad as the
Japanese where the better you do in school, the bigger your employer
has to be. We don’t believe that bigger is better and very
big is best, but we’ve gotten pretty close to it. We are told that the
bright ones want to go into small business. That’s what they say.
But when you look at where they are three years later, they are all
with Citibank, and for good reasons. Citibank can afford people
who don’t earn their keep yet. Citibank can afford that overhead,
and that little entrepreneur with $2 million in sales can’t. But he
also has no time to train anybody. He has to throw them in, and
also they learn the wrong things. One has to learn system.
Before you can say what we violate in the system, you have to
master it. You can’t write free verse until after you have learned
to write a sonnet. I once studied composition with one of the
most advanced of modern composers [the Austrian] Anton Webern.
(I almost became a musician.) And I thought he would
let me write the kind of stuff he wrote. And he said, “My dear,
Peter, you are taking liberty with the variation form. [Joseph]
Haydn gets it after 30 years. First of all, you will never be a
Haydn, and you have been at it 30 days.” And I had to learn to
write the orthodox variation. And then, after I had done that for
a year badly, he said, “Now, maybe you can take one liberty, but
be careful.” And, I brought it in, and he said, “I was wrong. You
aren’t ready yet.” And he was right.
And, that’s why students go to the GEs and the IBMs: not
only because they pay better, but also because they have a training
program. We have been amazingly effective at having small
entrepreneurs, and yet, when you look at the things that need to
be done, an enormous amount must come out of existing, large
organizations simply because the capital needs, the people needs,
and the planning needs are so great.
So, how do you organize your entrepreneurial within the
managerial? Again, the answer is largely not structure but people and compensation systems and managing individuals and placing them.
The human relations people made one dreadful mistake,
which we now realize. They talked of human nature. That was
the reason why we in the human relations movement were totally
ineffectual. We talked of a collective called the human being
in society. We preached individuals, but we did not really look
at them. We didn’t manage them. We managed workers, supervisors.
One has to do that, but still it explains why we were
ineffectual and why the very human beings whom we thought
we befriended rejected us. The organization development people
believe that structure is an obstacle and an impediment and has
to be, if not eliminated, at least bent to the human being. They
are equally ineffectual.
The skeleton has to be rigid. One cannot adapt the skeleton
to the individual, to people, but one has to fit people as individuals
so that they can learn what the individual can learn. We will
have to challenge people: “It’s your job to think through who has
to know and understand what you are trying to do and to make
yourself understood. Don’t wait for the information specialist;
that’s futile. It’s your job to say who needs to know what you are
going to concentrate on and also who depends on you for what.
And then to go to that person and say, ‘This is what I think you
look to me for,’ which is the only way to develop relationships.”
Every client of mine tells me of the terrible personality problems
that we have. Nonsense. Personality problems are very rare
in organizations. Organizations have to be very tolerant. Misunderstandings
are common. Personality problems are almost
unheard of. But what is common is that you don’t know what the
next fellow is doing because he hasn’t told you, and you haven’t
asked him. And so you assume he must do what to you is obvious,
and when he does something else you think he is either
stupid or malicious. No, he only marches by his own drummer
and hasn’t told you what it is and you don’t hear it. So, we will
have to demand a great deal of responsibility from individuals
for making themselves effective in a system.
If you look at the history of political theory—and basically
when we talk organizational structure we are talking governance
and political theory—there are two strands in the Western
tradition. One is the constitutionalist, which basically says,
“What are the right laws so that even mediocre people can function
and so that the evildoers are at least confined?” The other
one, which went by the name of the “education of the Christian
prince” and also goes back to the Greeks and even earlier, says,
“How do we form the rulers so that even if there are turbulent
times and the rules aren’t clear, we get the best there is, and we
get virtue and leadership into the system?”
In the last 30 years, we have emphasized the constitutional
approach. Let me say that this is normal in this country, where
business basically started out taking its structural concepts from
the American Constitution. If you look at our organizational
theory, it is tremendously influenced by the Constitution, for better
or worse. I’d say largely for better because at least it preaches
the need to think through the limitations of power. At the same
time, we have had some approaches that are opposite—that start
out with, “How do we form people?”
Oh, some of you may be old enough to remember that I once
was considered a pioneer of human relations, which was an attempt
to offset somewhat the constitutionalist approach with an
approach on the formation of people. And there is organization
development. But let’s face it. Those two things haven’t worked.
They have been minor corrections. And now we will have to
look at the formation of people very seriously, simply because for
some of the problems we have to solve there is no other way.
Let me give you an example. If you look at multinationals
today, they are mostly nineteenth century in their structure,
with a parent company and with subsidiaries that manufacture
the same products for their own home market. But increasingly,
you get incestuous relationships in which the specs come out
of Detroit, the design comes out of Germany, the body comes
out of Brazil, and the transmission out of Mexico—like the
Ford Fiesta. Or if you look at IBM office products, resources
are organized not in terms of products but in terms of stages of
production with labor-intensive work done in one place, with
design done in another, with a lot of the technology done where
the technologists are.
And let me say that technologists are proving remarkably resistant
to migration. Those French specialists prefer to work in
a research lab near the Louvre to working in Hoboken. I can’t
figure it out, but they do. And you can’t get them to move or
even to go down to Connecticut, and so you have to do the research
where the researchers are. It’s no longer true as it was in
the ’50s that researchers are so badly paid in Germany or Austria
or Japan that they’ll only too gladly take that job in Peoria. They
don’t anymore. And so you will have to think through how you
organize and integrate not components but stages. The product
is then sold where the customers are, which is likely to be in developed
countries. And so that subsidiary you have in France is
not a traditional one.
But even more difficult is the subsidiary in Colombia, where,
if you are a pharmaceutical company, Colombia is a developed
market. In fact, the only industry for which developing countries
are fully developed markets is the pharmaceutical for the simple
reason that pharmacy is the only part of modern health care a
very poor country can really afford. It’s the cheapest part, and it
does 60 percent of the job. And so Colombia spends proportionally
more per capita on prescription drugs than most developed
countries, and yet it’s a small market because of the small urban
population—roughly similar to one good sales district in a developed
country, let’s say Manchester or Kansas City.
And yet that head of your subsidiary there has to be a distinguished
man because he will matter a great deal. So he probably
is the most distinguished medical administrator, former dean
of the medical school, and a minister of affairs. How does he
relate to the top? He has to be an equal because he’ll negotiate
with the government there now, with the Catholic nun who
buys drugs for Mercy Hospital. You’ll need his input. He’s a very
distinguished man; very few of that caliber could you get into
your own organization. How do you structure him? And so you
have all kinds of new complexities, which make the simple, traditional
structure very hard. Then one has to say, “First, we need
new structural principles.” The answer to it is, “We ain’t got
none. We have patchwork.”
We had in the history of organization two very simple principles.
One is from [French mining engineer and early twentiethcentury
management theorist] Henri Fayol, and the other one
was what [General Motors Chairman] Alfred Sloan and then
I codified as “federal decentralization.” And they worked where
they worked like a charm, but they have strict requirements and
severe limitations.
You know, when I first heard of the Bell telephone system,
it was beautifully organized and very simple, a good operating
company, and 98 percent of their business was within their territory
because in those days 98 percent of all calls and 98 percent
of all revenue were local calls. Nobody had heard of computer
transmission or what have you. And so, when you look at it today—
and you all know the Bell telephone system is desperately
trying to reorganize itself into regulated and unregulated and
quasi-regulated businesses—you no longer have a local system
that’s interconnected, a long-distance system with local peripherals.
The old organization cannot work and doesn’t. And so
how do you organize? We don’t know the answers; we know
patchwork. And we will all have to live with things that quite
clearly are full of friction and present problems.
We’ve tried to build our organizations as close to mechanical
models as possible because it is simple. [Scientific management
pioneers] Frederick Taylor and Henri Fayol both assumed that
you know what you are doing. You know, a coal mine mines coal.
It’s obvious, isn’t it? Well, you are now in a period in which the
real challenge is to decide what you are doing in the context of
technological change or market change.
The one axis of organization that you’ll say you need is a skeleton.
And all of you know a land animal that is more than six
inches tall needs a skeleton; it can’t be held together by heart and
skin anymore. The organization chart, with its lines of authority
and its reporting, is a skeleton. Now, we always have some
problems with that. If you have a divisional structure, the relationship
between your corporate comptroller and your divisional
comptroller is not a simple one and can’t be decided one way or
another. You all know that when you get to technology, it doesn’t
work if you have coordinating groups, and it doesn’t work any
other way.
Yet how do we organize the new within the old? In fact, can
we organize it? If you look at the last 30 years, the obstacles to
entrepreneurship were exceedingly high in terms of the tax laws,
in terms of our credit structure. Now, we are not as bad as the
Japanese where the better you do in school, the bigger your employer
has to be. We don’t believe that bigger is better and very
big is best, but we’ve gotten pretty close to it. We are told that the
bright ones want to go into small business. That’s what they say.
But when you look at where they are three years later, they are all
with Citibank, and for good reasons. Citibank can afford people
who don’t earn their keep yet. Citibank can afford that overhead,
and that little entrepreneur with $2 million in sales can’t. But he
also has no time to train anybody. He has to throw them in, and
also they learn the wrong things. One has to learn system.
Before you can say what we violate in the system, you have to
master it. You can’t write free verse until after you have learned
to write a sonnet. I once studied composition with one of the
most advanced of modern composers [the Austrian] Anton Webern.
(I almost became a musician.) And I thought he would
let me write the kind of stuff he wrote. And he said, “My dear,
Peter, you are taking liberty with the variation form. [Joseph]
Haydn gets it after 30 years. First of all, you will never be a
Haydn, and you have been at it 30 days.” And I had to learn to
write the orthodox variation. And then, after I had done that for
a year badly, he said, “Now, maybe you can take one liberty, but
be careful.” And, I brought it in, and he said, “I was wrong. You
aren’t ready yet.” And he was right.
And, that’s why students go to the GEs and the IBMs: not
only because they pay better, but also because they have a training
program. We have been amazingly effective at having small
entrepreneurs, and yet, when you look at the things that need to
be done, an enormous amount must come out of existing, large
organizations simply because the capital needs, the people needs,
and the planning needs are so great.
So, how do you organize your entrepreneurial within the
managerial? Again, the answer is largely not structure but people and compensation systems and managing individuals and placing them.
The human relations people made one dreadful mistake,
which we now realize. They talked of human nature. That was
the reason why we in the human relations movement were totally
ineffectual. We talked of a collective called the human being
in society. We preached individuals, but we did not really look
at them. We didn’t manage them. We managed workers, supervisors.
One has to do that, but still it explains why we were
ineffectual and why the very human beings whom we thought
we befriended rejected us. The organization development people
believe that structure is an obstacle and an impediment and has
to be, if not eliminated, at least bent to the human being. They
are equally ineffectual.
The skeleton has to be rigid. One cannot adapt the skeleton
to the individual, to people, but one has to fit people as individuals
so that they can learn what the individual can learn. We will
have to challenge people: “It’s your job to think through who has
to know and understand what you are trying to do and to make
yourself understood. Don’t wait for the information specialist;
that’s futile. It’s your job to say who needs to know what you are
going to concentrate on and also who depends on you for what.
And then to go to that person and say, ‘This is what I think you
look to me for,’ which is the only way to develop relationships.”
Every client of mine tells me of the terrible personality problems
that we have. Nonsense. Personality problems are very rare
in organizations. Organizations have to be very tolerant. Misunderstandings
are common. Personality problems are almost
unheard of. But what is common is that you don’t know what the
next fellow is doing because he hasn’t told you, and you haven’t
asked him. And so you assume he must do what to you is obvious,
and when he does something else you think he is either
stupid or malicious. No, he only marches by his own drummer
and hasn’t told you what it is and you don’t hear it. So, we will
have to demand a great deal of responsibility from individuals
for making themselves effective in a system.
