Peter Cappelli has made a career of charting the often-tumultuous course of employment in America, from studies on downsizing and performance to a book on managing in a market-driven workforce. But Cappelli was taken aback by a theme that, in recent years, surfaced again and again in the mid-term exams of his MBA students.

The students are asked to write about their last job and how they were managed, and inevitably, people talk about why they left,” says Cappelli, the George W. Taylor Professor of Management and Director of the Center for Human Resources. “And many students began to report that they were in jobs that literally ended, and not just junior analyst positions in investment banking. In consulting and even some corporations, after three years or so, they are told to leave. Some people in partnerships were told that no one is eligible for partner promotion. In flat organizations, they are saying that beyond five years, they cannot see the next job up the hierarchy that could be open for them.”

“They don’t see a way to build a career, to advance,” says Cappelli. “And these are fast-track people.”

There’s little doubt that the traditional employment system of secure, lifetime jobs with predictable advancement and stable pay is, by most accounts, dead. How, then, do aspiring business leaders scratch their way up the corporate climbing wall? Cappelli hopes his latest project, a book he’s writing called The Path to the Top, will offer some answers.

In this first of two interviews, we talk to Cappelli about the past, looking at the early history of executive careers in the U.S. Our second interview, to be published in a forthcoming issue of the Wharton Alumni Magazine, will examine recent history and the future. “It wasn’t very satisfying, after my first book, to simply say what was no longer happening,” Cappelli says of the 1999 book The New Deal at Work: Managing the Market Driven Workforce. “It seemed to me that there is a story about how careers are evolving and that it makes sense to look at how we got here. The ultimate goal of the book, of course, is to come up with a new model — to show what careers are looking like now. There are a lot of things we know about the way careers worked in the past — about how pay was set, what determined advancement, etc. I’d like to be able to do the same thing for careers now. But initially, at least, I believe there’s something to be learned from the past.”

Let’s talk about how executive careers began. How did we get here?

The point to note at the very beginning of American industry is that there weren’t any executives. There were founders and worker bees. The founders outsourced most of what went on and didn’t really have managers or executives. Before William Durant went into the car business and created what became General Motors, he led a large carriage company that was strictly a marketing operation. The company contracted with a local builder to make the carriages that the Durant-Dort Carriage Company put its brand on and sold.

Early companies like Durant-Dort were organized very simply. The founders were at the top — they were the venture capitalists of their day and were paid sometimes huge amounts based on the value of their ownership. There were no managers — the founders handled everything themselves — the finances, distribution and sales. Henry Ford is a great example of this kind of owner-manager who ran everything himself, even resisting management systems like accounting. There’s a very telling story about Ford becoming irritated about some accounting issue one day, going into the head office at lunchtime, and throwing all the accounting books into the street. He didn’t care about organization, either, keeping track of his 446 salaried workers (in 1913) with no job titles, just by ranking them according to their pay level.

How was this possible?

It was possible to run these large companies without managers or executives because most of the tasks were outsourced, and this went beyond buying parts and supplies. Very few companies even sold their own products — they supplied merchants who sold the products for them. A company like DuPont did none of its sales or distribution at the turn of the last century. The company used hundreds of independent agents who were paid on commission and often represented more than one company.

But as companies grew it became more difficult for the founders to do everything themselves, and they then turned to people they knew and trusted — usually family members — to step in. Even then, there were few executives around the founder. Most of them came from the families — these were like today’s small businesses except that they had thousands of employees. There was the founder, his son, sometimes his wife.

DuPont is perhaps the best example of a company filling top jobs with members of the founding family, making college choices for young DuPont children based on job needs within the company — in essence picking their careers for them to prepare them for the executive track.

When did modern management structures begin to take shape?

It wasn’t until the railroads came along that we began to see organizations that were a little more sophisticated. The physical scale of railroad operations created enormous challenges. Running a train from one part of the country to another, picking up cargo and passengers along the way, changing crews and refueling, making connections with other trains, and doing all of this on a tight schedule meant a tremendous amount of coordination. The only way the railroads could be sure that everything was perfectly coordinated and on time was to do everything themselves — they had to develop internal systems of accounting and control and find people to run them. They had to hire middle managers to develop and administer these systems. The Pennsylvania Railroad was the largest and most complex of the railroad operations and is typically credited with creating middle management.

Andrew Carnegie also deserves a great deal of credit for creating the modern executive career in corporations by transferring many of the operating procedures of railroads into manufacturing. He worked his way up from a telegraph operator into a job at the Pennsylvania Railroad and eventually became superintendent of its Western Division, the most important division in that railroad. Along the way he absorbed many of the railroad’s elaborate operating principles, especially the idea that performance standards could be created for every job and that individual managers should be responsible for meeting those standards.

Carnegie came up with the idea of an operating or line executive based on the superintendent’s job he once held on the railroad, and it’s not a coincidence that the top job in a Carnegie steel mill and later in all steel plants had the superintendent title.

Carnegie was a guy who gave people opportunity and moved them up quickly if they did the right things. His most famous executive was Charles Schwab, who started as grocer boy to Carnegie executive Captain Bill Jones and went on to head Bethlehem Steel. Carnegie wanted to promote executives from within and to get to know the men who stood out.

But other than the Carnegie Steel Works, it’s hard to see many examples of corporations elevating employees from their own ranks to these new executive positions during this time period. Carnegie gave managers a great deal of discretion and responsibility and promoted people based on merit, which was not common.

The DuPont Powder Company was a good example of a typical company at the turn of the century. Dupont had many different explosives products targeted to different customers that were all managed under one central administrative structure. When Pierre DuPont took over the company through a leveraged buy-out of family interests, he reorganized the company and created the first multi-divisional organizational structure, separate divisions organized around individual product lines. This model created a wave of executive jobs because each of these divisions acted as its own entity and needed its own leaders.

As they grew, DuPont and other companies brought in outside talent through acquisitions. A company like General Motors acquired 25 companies in the space of a couple of years, while DuPont acquired 60, and when they did that they took in the entrepreneurs that started those companies. These entrepreneurs took the new executive positions in the multi-divisional companies.

Like DuPont and other corporate giants of its day, General Motors was a product of its acquisitions. Williams Durant created GM be­tween 1908 and 1910 by acquiring Buick Motors and then adding 25 companies, 11 of which were independent car manufacturers. Not surprisingly, the founders of these companies became the first executives of GM. Henry Leland Cadillac, for example, ran the new Cadillac division under the GM label.

So the creation of new executive jobs did not really create career paths. For the generation in business before World War II, the way to get ahead in a career was uncertain and complicated. Yes, it was possible to rise within a company, but internal advancement was a slow process. Promotions took a long time because there weren’t many senior management positions to fill, and vacancies came only with death and retirement. Openings were usually filled by family members and friends of the owners; few managers worked in jobs that fit their education or training. The best way to get to the executive suite was first to create your own company, second, invest enough in it to have controlling interest, or third, be a family member or intimate associate of a founder.

How did World War I begin to change things?

World War I meant that companies had to become more serious and sophisticated in their efforts to develop and retain managers. Before the war companies largely relied on something called the “drive” system to motivate employees. This basically was motivation by fear – companies used the threat of being fired as a way fuel performance. But as labor markets tightened, this obviously stopped working, and companies had to begin to adopt new methods of management. The most important of these was the idea of promotion from within.

Corporations also looked to the military for ideas, adopting selection tests to hire candidates and place them in specific jobs, a technique pioneered by the Army at the beginning of the war. Most of these practices were aimed at production workers, but they also affected the managerial ranks in the 1920s, where there were also major shortages. At that time, manager candidates were hired into specific functional areas and “training” was really just a very brief orientation. For most young managers, it was sink or swim.

Some large companies began training-based development programs aimed at focused development rather than haphazardly hoping a new hire would grow into the company. Others, like the Public Service Company of Northern Illinois, pioneered “high-potential” programs to select out employees who seemed to have an aptitude for senior management positions.

But most of these programs were small and ad hoc. A real breakthrough came at General Electric, where the scale of operations and need for talent warranted a major employee development effort. GE began to grow very quickly in the 1920s, and to meet the growing demand for managers, the company began to hire new college graduates with no real work experience not only into technical positions, but also into managers spots. The new hires were sent to GE plants for a one-year program either on a business training track or a “test” engineering program for the technical jobs. After this one-year period, the trainees essentially applied for permanent jobs within GE.

But the movement to develop employees was seriously set back by the Depression. The huge decline in business meant that companies had more than enough managers to meet their needs. Larger firms like GE maintained some of their programs for de­veloping managers, but most companies dropped theirs altogether.

What about World War II?

After the Second World War, things really began to change. During and after the war, companies suddenly started growing like crazy. And during the Depression, they’d really not hired anybody. So with the wartime economic boom, they had a big demand for managerial talent, and they hadn’t hired anybody or developed anybody for a decade. So they faced a huge shortage of talent. During the war, a lot of senior executives stayed on past retirement for patriotic reasons. As soon as the war was over, many of them retired. So there was this huge wave of retirements and nobody in the pipeline. When the war ended, many of the women who filled jobs during the war, including administrative and supervisory positions, left those jobs and went home. American companies had big-time, serious shortage of management talent.

The other thing that most of us forget about is that a lot of executives in that era tended to die in office. One third of the managers who were age 45 in the 1950s were expected to die before they were 65. There’s a great quote from Ben Moreell, the chairman of the Jones and Laughlin Steel Company, who described the unstable succession process that resulted as “When a great president dies a vacuum is created, and into that vacuum is swept the nearest guy who hasn’t had a coronary.”

So this is when everything about modern careers really began. They initially began by trying to see if they could meet their talent needs by hiring from each other. But they just couldn’t do it – the shortage was too great. So they plunged into this model of developing people internally. Companies borrowed an enormous amount about development from the military and began taking it incredibly seriously.

What did the model look like?

Pretty much everything about managing employees and developing them systematically that companies do today was more sophisticated in the 1950s. They did fancy personality tests and IQ tests, elaborate rotation programs and high-potential programs, and sent people for executive education programs.

Innovative programs like job rotation, “junior” boards of directors, Harvard’s Advanced Management program, executive development committees headed by division presidents or vice presidents were also put into place. By 1954, 63 percent of large companies were using standardized personality tests to assess loyalty and potential in their hiring decisions. The amazing thing is that these assessments had nothing to do with a person’s present job performance. They were all about potential and having the traits associated with executive roles.

At GE, a huge manufacturing leadership program was based on the model of increasing job responsibility, classroom work, and training and individual counseling. Those that did well progressed immediately into a job rotation program that took them to jobs at two or three plants. The most promising of these then went on to an advanced management program, a high-potential program based on special project and case study work.

How successful were the programs?

They were hugely successful in the sense that they got a massive pool of talent developed and coming through the system fast. In 1956, GE hired a record 1,800 new college graduates and 85 percent of them went into one of 10 training programs. That same year, the company created its Management Research and Development Institute, an off-site training center for top executive programs.

These high-potential programs were basically designed to quickly replace the guys at the top. They wanted college-educated trained people in those jobs and didn’t want them to progress slowly through the ranks. In some ways, the story is that they were so successful that they caused the problems later on – the bloated middle ranks of American companies. They had so much talent coming through and they kept these programs in place and kept producing all these people. Then by the end of the 1960s, the economy starts to slow down, the companies stop growing, and they’ve got this huge pool of talent that is stuck without opportunities for advancement, and the programs are still churning out new managers.

The programs became an end in themselves. By the 1970s, the system was clogged and people were in dead-end jobs and getting frustrated. A huge restructuring wave began about 1981, and by that point, companies implicitly realized that they had way too much talent, which is why we saw the downsizing wave begin with a particular focus on white-collar jobs. They moved from a model in the 1970s where the prime directive was to develop more talent to a model of the 1980s where the directive was to get rid of talent. And all these development programs pretty much fell by the wayside.

Today, what’s interesting is companies are talking about more or less exactly reinventing what existed in the 1950s, and in some cases they don’t even know what they are reinventing. The foundation of the old models was very bureaucratic looking organizations that had lots of clearly defined jobs, job titles, and hierarchies. You could just look at an organization and know where your next job was and where the promotion path was and people could see how long it would take to advance. None of that is true now. Organizations are more fluid and flattened. The old model was all about how that happened and how people advanced and got ahead, and there was a fair amount of criticism about how stifling that all was. My long-term view on this is that the internal development model was more or less an anomaly — a World War II driven anomaly.

In some organizations, leaving to go to business school is now literally the next step up a career ladder – everybody is expected to quit, get an MBA, and then think about coming back. The companies have changed their job ladder so that they have entry-level jobs that literally go no where. In other organizations, it’s just unclear where they go. My plan, as my book evolves, is to clarify the mystery of careers today, to answer questions about what it takes to move beyond the dead ends.

Where will the book go from here?

From here the story moves to the present and to patterns that are appearing in modern careers. The idea is to understand what factors determine who gets to the top of modern companies and why. The processes appear to be quite different, especially the apparent necessity to move across companies in order to get ahead. But whether there is a simple description for how all that happens is not yet clear.