It seems an unusual soruce for the most provocative information about small and medium-size companies to emerge in the 1980s: McKinsey & Co., one of the nation's leading management consulting firms, adviser to the giants of American business. But consider what the firm has produced. Allan A. Kennedy's Corporate Cultures (co-authored with Terrence E. Deal). Thomas J. Peters and Robert H. Waterman Jr.'s In Search of Excellence. Richard E. Cavanagh and Donald K. Clifford Jr.'s The Winning Performance of the Mid-Sized Growth Companies. Together these works brought a new vocabulary and new tools to the study of growth and the entrepreneur. Allan Kennedy's Corporate Cultures, published in June of 1982, would form the basis for many of the innovative ideas that would emerge from McKinsey. Kennedy brought an anthropological perspective to organizational theory; for better or worse, he argued, each company has its own unique culture, values, and standards communicated internally by style and dress, expectations and assumptions within the organization. Kennedy's study was theoretical, a search for a concrete framework to explain how companies actually work. His colleagues, on the other hand, would take an empirical approach, looking at excellent companies and trying to discover the roots of their success. At the core of both approaches, however, was the theory of an internalized culture.

Kennedy brought both the insights of the outsider and the approach of the scientist to his study. Born in 1943 in Montreal; reared in Cape Breton Island, Nova Scotia; and educated at Montreal's McGill University, Kennedy remains a Canadian citizen today, both bemused and bedazzled by the energy and spirit of the American entrepreneur. Trained as a nuclear physicist at Massachasetts Institute of Technology, he followed a convoluted career path -- a compater programmer for John Hancock Mutual Life Insurance Co., an organizer for the Student Nonviolent Coordinating Committee, a stint with IBM Corp. -- before coming to McKinsey in 1969.

At McKinsey, Kennedy found a home "It was a very rich smorgasbord for someone who is curious and wants to poke around, "he remembers. "The firm stays vibrant by letting people come in to pursue what ever interests them. "For the next 13 years, working for clients that included a half dozen of the Fortune 50, Kennedy would pursue what he now calls "a realworld business post-doctoral degree." But there were frustrations as well. Kennedy was fascinated by the booming computer companies, by the entrepreneurs of California's Silicon Valley and Boston's Route 128, few of which were on Mckinsey's client list. He chafed under the firm's proscription against junior staff actually going out and selling Mckinsey's services.

So just over a year ago, he took the entrepreneural plunge himself establishing Selkirk Associates Inc., an applications-software development house in Boston. INC.'s editors interviewed him there.

INC: Where did your work on Corporate Cultures begin?

KENNEDY: In 1975 I was put in charge of a McKinsey team that was sent to a company for cost reduction. It had a very straightforward problem. Its Japanese competitor was selling a product in the United States for less than it cost the U.S. company to manufacture it. We had about 250 middle managers working on the cost reduction team at its peak. We turned in something like 270 detailed, fully staffed recommendations for specific cost reductions. It was very detailed work, the best analytical work I ever saw at McKinsey.

INC: And?

KENNEDY: And after six months we'd increased costs by a couple million bucks. People within the company thought the analysis was so good that department after department was setting up a new department of cost effectiveness, then hiring staff. As consultants, we were damned worried about that -- you really do try to help your client as a consultant, contrary to popular wisdom. But for some stupid reason, they had all these recommendations, and they just weren't having an effect. And we started saying, "It's the culture in this damned place; they've grown so long and are so much into growth that people don't want to talk about the retrenching that cost reduction implies. People will attend a meeting and say, "Yeah, we ought to do that. But not me, I'm out launching a new product." It was sort of a cultural organizational problem that they just couldn't address.

INC: What was the root of the problem?

KENNEDY: The average company middle manager spent 7.3 months in a job, then got promoted. That had been going on for years. When they were growing at 40% a year, they'd hire someone, then, 6 months later, they would have grown another 20%, and they'd need to move him up. That built up a mentality that said if you were a fast tracker at the company -- and the company is full of fast trackers -- you've got to get promoted in 6 months or you'll fall behind your peers. And that, in turn, promoted a mentality that said I just got my new job, now, what can I do in the next 3 months to make my mark so that I then have 3 months of shopping around my resume to get the next promotion. Now in 3 months he can't do a lay-off; if you just look at the mechanical steps, making up a roster of people, going through your due-diligence requirements with the union, and all those kinds of things, it takes longer than 3 months. So no manager in his right mind was ever going to take on something like cost reduction, because that meant he would be stuck in his job for 18 months. So one of our recommendations was that the chief executive officer send better signals to the organization about how cost reduction was more important than the traditional things that had been going on.

INC: How did that realization about culture lead to the book?

KENNEDY: Some of the ideas that were developed there were fed as fodder into a series of internal seminars at McKinsey. It got to be competitive in the best sense of the word; you'd go to a seminar, present your current set of ideas, listen to five other people's ideas from their client situations, then you'd have three months until the next seminar, thinking, "Christ I better go do something creative with this client so I'll have something to talk about." We built up a lot of organizational technology over the next year and a half, just through those informal meetings. We'd recognized that the informal side of the organization is very important, and we had a bunch of tools for diagnostic things we'd do to get a picture of what the informal organization is like.

INC: And those tools became the basis for both your book and Peters and Waterman's?

KENNEDY: My book was intended to be the beginning of a framework that people can use in a practical sense to begin to build measures of the informal organization. That is a predictable direction for a trained nuclear physicist. Peters came at the problem from the other end; he was trying to do empirical work, to say, Let's let the facts speak for themselves in terms of what really drives the best companies.

INC: The book came out in 1982, and, in less than a year, you'd left McKinsey and started your own business. Why the shift?

KENNEDY: When the book came out, I'd spent about three and a half years wrestling with cost problems, and I had a bad case of client fatigue. I'd actually thought about leaving McKinsey then. It's a great university laboratory, but at some point I wanted to go out and sell. But instead of leaving, I got talked into going back to Boston and opening the office up here. In Boston, I was spending more than half my time working with Route 128 companies. And those companies are full of entrepreneurs. You couldn't spend a day there without hearing, "Let's go call on a customer, let's go place an order." They're very real-world oriented. Most of them don't think very strategically, but they're terrific at action. And seeing all that action just heightened my frustration.

INC: Is that your definition of an entrepreneur, someone with an action orientation?

KENNEDY: It's action orientation, but it's more than that. Most entrepreneurs want to make something, to build something. There is real pride in taking a product and making it commercial, in having a new factory and seeing yourself progress. And when guys relax, like 10 o'clock at night or something, relaxation is talking about. Do you remember when we started out in that basement or in that garage? And now we've got corporate headquarters and we're building a factory next door. That's the kind of story they tell. It's a pride in building something. I absorbed that atmosphere, and I wanted to give it a try myself.

INC: Why applications software?

KENNEDY: We're all crazy about microcomputers, and I am, even from an organizational perspective. If you can put real computer horsepower in the hands of a single person for $5,000, it is going to revolutionize the way organizations work. People have been saying that for years, but the product is finally here. There is enormous opportunity for capital/labor substitution for productivity improvement.

INC: But how is that going to "revolutionize" the way organizations work?

KENNEDY: If you look at everything Tom [Peters] and I were writing about organizations, the leading organizational issue is how to kindle some kind of entrepreneurship. Nobody has a very good answer for that yet. But it is clear that the better places are better, simply because they are able to get more of their creative energy out. And I have a strong conviction that microcomputers are going to make an enormous difference. Take our system. We can set up a person as a full sales organization with all the functions of a sales support department, just because of the capacity of the computer. You can take a salesman and make him 10 times more productive. So the face of organizations is going to change dramatically over the next 10 to 20 years. If someone has his own microcomputer with the right application software, he doesn't need a big organization. He can go off on his own and do it. And statistics demonstrate that there is an overpowering trend toward people going out and taking entrepreneurial risks. In 1963 there were 186,000 new businesses formed in the United States. And in 1983 there were 580,000.

INC: What is behind that new trend toward risk taking?

KENNEDY: The driving influence is personal economics. We have become a very wealthy middle-class country; individual incomes are a lot higher than they were 20 years ago. Statistics say the average incease was $23,000 in 1983, but the average income for the professional/managerial class is a lot higher. And now 60% of the households in the under-50 age bracket have two wage earners. So that creates an environment in which you can build a nest egg.

INC: Which permits people to take the risk of starting their own business?

KENNEDY: Exactly. It makes starting a business a lot less scary than if you were making $10,000 a year and could barely cover your mortgage. Now there is a margin for error. I think of the businessman right below me here. He runs a computer center -- a service bureau and training center. He used to work for a large company, but he always wanted to be on his own, not working in a big organization. So as soon as he could put aside enough money, by God, he took his idea for a business and did it. He realized he could live for six months. He didn't have to sell his house. His wife was producing some income to cover vars -- able costs. And six months seemed enough time to take a shot. In four months he was in the black, and he has himself a business now. In the 1950s, there were damn few people that had that possibility. In addition, a second factor is the shift of the economy into the service sector.

INC: Where there are low entry barriers to the growth businesses.

KENNEDY: Sure. If you were back in the 1900s trying to get into the car business, which was the growth field then, it took a lot of money to put up a factory. But where is the growth now? Service businesses generally. Probably software is the top. And while I'll tell you it ain't cheap -- it cost us at least a million dollars to get a decent buyable product off the ground -- actually that is a trivial investment. It is not $50 million or $100 million. In fact there are lots of little software houses out there that will do quite nicely on six months' work of a guy programming at home and then generating just enough revenue to keep his head above water. But it is his own show. He doesn't have to report to anyone. He is making something of value and getting a product out that he is proud of, then looking to see it grow.

INC: The classic American dream.

KENNEDY: Exactly. And that's the third reason for the explosion, something that has nothing to do with the past 20 years. this crazy country is inherently entrepreneurial. The great strength of the United States is in the independence of the people. Everybody and his brother inherently has the attitude of "I can get rich." Look at the pioneer movement. What in the world possessed the vast masses of people that populated this country to get in wagons and go rolling across strange territories to get their own piece of turf? There is something about the American mentality that is inherently independent. That is one of the great strengths of this country. I think it has been there forever; what we are seeing now are the economics to support it.

INC: What are the implications of this for the American economy?

KENNEDY: It is the most phenomenally good news in years. As a consultant, I've spent a lot of my time worrying ab out Japanese competition. And the Japanese are superb and fierce competitors. But Americans are going to beat the Japanese because of all the entrepreneurial activity. Look at the kind of spawning ground that Data General has been, for example. It develops one generation of a product, then four more companies spring up to develop the next generation. Six months later, four more respond to develop the next generation. Each one of those represent technology innovation in the real-world sense. It is a by-product of entrepreneurism. And I think that as the entrepreneurial trends continue, the United States is going to dominate the world markets because nobody will be able to keep up. In the large Japanese companies, they have no mechanisms for entrepreneurial activity.

INC: You use the word "trends." So you don't see this as a temporary phenomenon that will be replaced by another era of consolidation?

KENNEDY: If I had to go way out on a limb, I would say that large corporations as we have known them will not exist in 30 years' time. They won't be able to sustain their position. They won't be able to hold on to their people; everyone will know too many other people who have gone out on their own and made a fair amount of money. So all the good people are going to be spinning off.

INC: But isn't that model so far largely limited to technology markets?

KENNEDY: No. Only about 20,000 of the 580,000 businesses formed last year were technology businesses. The vast majority were companies like the one started by a friend of mine who has a telemarketing service for selling wine in Boston. He's got 25 telemarketers who phone high net worth individuals, and he delivers case lots of wine. He doesn't have a liquor store, but he is making money hand over fist. His business is growing like mad, with just a microcomputer to keep track of orders and with housewives who are part-time telephoners. There is no reason why he is not going to have a $100-million business in due course. If he's doing it in Boston right now with 25 telemarketers, there is no reason why he can't have it in most of the states in the union.

INC: So he in turn will become a larger business in his own right?

KENNEDY: Not if he's smart. He could take his business idea and expand it into the top 20 cities in the U.S. Then he'd have 20 guys managing 20 operations as hired help. But those guys wouldn't stick around. As soon as one realizes how simple it is, he's going to walk next door and set up the same operation so that he can get his own share. So if my friend is really smart, he'll try to franchise the operation, and it will wind up a $100-million concept with many owners participating.

INC: That may work fine for your friend, a small businessman staying small. But you forecast the end of the large corporation. Are you actually predicting that, say, IBM will dissolve?

KENNEDY: No large corporations will look the way they do now. If IBM wants to stay IBM in some fundamental way, it is going to have to change.

INC: Change to what?

KENNEDY: I don't think anybody has the right model for it right now. But remember what Jimmy Ling used to do with LTV -- spin off public stock as a subsidiary. That was pure financing, but it actually makes more sense from an organizational point of view. Take their best group of microcomputer designers. How can IBM hold on to them unless it is willing to spin off a subsidiary and let those guys take it public? Or else why would those guys design a machine for IBM? They've seen Don Massaro, say, leave Xerox [to start Metaphor Computer Systems Inc.] and get $15 million in venture money to design a new universal workstation. Don would have been crazy to stay at Xerox and do it for them. So if you are running IBM and you recognize that you aren't going to be able to keep a good designer, you can contract the work out. Or you can set up a subsidiary and give the key people equity, with IBM providing the seed capital, the building, the work, and the colleagues to talk to. You're going to end up with something like a holding company with a lot of these little individual subsidiaries.

INC: This would also be consistent with the belief, gaining in popularity, that smaller work units are the most productive.

KENNEDY: That's right. You get the highest productivity, morale, and performance from a small work force. A few corporations like Hewlett-Packard have recognized that. In fact the evidence seems to point to between 50 and 100 people as the most efficient work unit. If that is true, and if the economy is moving in that direction, then if you are running IBM, the biggest 10-year challenge you face is how to break IBM into the 10,000 little units of 100 people, each with an entrepreneurial stake. IBM will survive in some mode, but I think it will be very different from what we know right now.

INC: But what can IBM offer the people in the modules you're envisioning that will keep them from following the example you cited of Don Massaro, spinning off totally on their own?

KENNEDY Massaro got $15 million as venture capital money to start up his new computer company. I'm guessing that under that kind of a financing he ends up holding 20% of the equity, at best; venture people aren't going to fork over $50 million for 10% of the company no matter how good the guy is. So he's got 20% of that operation. So what is the difference for a similar spin-off from IBM if his venture financing comes from IBM?

INC: So your hypothetical model starts with venture capital from the larger parent. What are the other characteristics of the unit that make it different from the businesses we know today?

KENNEDY: There is a stronger sense of everyone participating in the success of the unit. I can see that merely by reflecting on my own experience here at Selkirk. There is nobody here who isn't well aware of how many orders are booked every day; you have that sense in most small companies. When I call our 22-year-old secretary and say, "Boy, we've got a hot prospect, and we have to get this piece of literature out to him," there is no mystery in her mind that if she works a little bit harder, gets that letter out, gets it in the mail, that that could produce an order. And an order could mean a rise in the value of her company stock, so there is a close identification toward the goals in the company.

INC: So the identification depends on sharing equity?

KENNEDY: I think the dynamic works independent of whether the equity is shared. The important thing is that people can identify with seeing the end result of their actions. And that is a phenomenon of scale. The strongest motivation is seeing something good happen as a result of your own input. So managers have to think differently. To get the top degree of effort from the individual and to make those individuals operate as a higher productivity entrepreneurial unit, you're going to have to operate in a different mode altogether.

INC: And you see this pattern happening eventually in all industries? Even to large-scale manufacturers? Could Boeing have built the 747 that way?

KENNEDY: Some industries are going to be slower, if the product requires a big, complex system. But they will move in that direction. We are only 20 or 30 years away from being able to use robots to automate an entire factory, so that part of the argument for large scale disappears. Boeing is a good worst-case example. I say to myself, suppose I was in the wing assembly. And the guys in the wing assembly and I say, 'Jesus Christ, if we took this outside, we could contract it back to Boeing at two-thirds the cost. We could make a pretty piece of change and we wouldn't have to put up with any of that bull about going to the monthly management meeting. " I'm not sure that a subcontracted model ultimately doesn't create more opportunity for higher productivity, for happier people on their own turf doing their own thing and creating their own product. Look at the success of the Japanese; much of it has to do with subcontractors as independent firms.

INC: Or look at Control Data.

KENNEDY: That is a perfect example. They have taken some portion of their human-resource function and held it responsible for generating outside income to support its budget, in fact encouraged it to be out in the marketplace. And Control Data is going to find that it is getting better service, more professional service, more economical service from a group that is functioning in the independent marketplace.

INC: But suppose that group really takes off. Suddenly they don't have time to do the work for the parent company. How does a manager keep enough control to guarantee that his needs are met?

KENNEDY: To go back to the IBM example -- how does John Opel run the 10,000-module IBM? He does Control Data kinds of things with as many staff groups as he can, setting them up as independents. But I think there is something to the culture of the company, making people feel part of a bigger family. So he is going to work pretty hard to keep "think" signs on people's desks and to keep a big blue label on the sports jackets of the people in these independent units. So they'll be independent and they'll have their own name and they'll be up logging their own services -- but they'll have a strong sense of loyalty to the parent, loyalty that is nurtured consciously by management.

INC: What about smaller companies? Are they, too, going to be forced to break themselves up into these independent, equity-sharing modules?

KENNEDY: I make the same argument for the small, overly bureaucratized companies: They'll get killed if they don't adapt. They may even have to adapt faster because they probably don't have the resources to survive otherwise. On the other hand, they probably have the means to adapt faster.

INC: But do they have the will? It's hard to imagine an entrepreneur who has built a business over, say, 20 years, gleefully agreeing to give up the fruit of his efforts.

KENNEDY: It is a question of role models. Those small bureaucratic companies undoubtedly started as entrepreneurial ventures. If you went back 20 years they were saying, "Gee, I got the company off the ground; I'm now a capitalist. I can afford to hire managers. I can get out of the day-today datails. Idon't have to make sales calls." A lot of consulting for small companies came from businessmen saying, "I'm big enough now that I can afford a consultant." It is the entrepreneur grabbing the fruits of victory. Now why did he choose those particular fruits? What the hell role models did he have around? He'd look at GE or Procter & Gamble and say, "Boy, I'm going to be like that some day." But that's not the ethic any more.

INC: What is the new ethic?

KENNEDY: Today it's, Look who is making a bundle this week. Who opened a garage operation and managed to go public and cash in for $20 million? Even guys who really have it made in that conventional sense are green with envy. On Route 128 the ethic is not to build IBM, the ethic is to make yourself a pot of money and get your own piece of turf.

INC: But doesn't that assume that every small business person will have equal access to capital, not just the high-technology companies?

KENNEDY: The weak point in my argument is how the capital markets work. If you look at Xerox, for example, look at Shugart and Versatec -- they've got dozens of those divisions. And they got those because at some point the owner wanted to cash in and figured that he'd get a better deal from Xerox than from taking the company public. The capital markets are more geared toward large companies, and I'm not good enough at financial stuff to sit here and paint the model of how the capital-market regulations ought to be changed to facilitate what I think is the important economic trend.

INC: And that, according to your theory, is the trend toward decentralization that is going to reverse the movement toward industrial and business consolidation?

KENNEDY: I think it is all one big wave. We used to live in a highly decentralized society, before the Industrial Revolution, with countless shops and craftsmen. The steam engine and the Industrial Revolution came, and guys figured out that you can put 1,000 people out there making textiles and make a hell of a lot of money. So they built some big corporate models. That was the new game in town, getting the economies of scale. But they got to a scale that was unmanageable -- General Motors was the first to recognize that. So GM had this wonderful idea to set up divisions. It took the world a while to catch on, but in fact what's happened is that company after company has decentralized. And things have worked better. So they decentralize a little more. So decentralizations have been steady. And I think all the evidence says that if you want to get down to the optimal size, you'll get down to the 100-person unit we have been talking about. And I would argue that the companies that push the farthest in that direction are going to be the next wave of successful companies.