Until the 1990s, retraining ruled at companies like IBM. Big Blue, which promised lifetime employment to its workforce, moved its employees every few years and when it did, taught them new jobs.


But when IBM’s traditional dark suits and white shirts gave way to knit shirts and khakis, the company’s commitment to lifetime employment — and thus retraining — waned. Under former Chairman Lou Gerstner, the formerly paternalistic employer laid off tens of thousands of employees.


Why did the computer maker scotch retraining for workforce “churning” — that is, laying off employees with obsolete skills and replacing them with workers offering newer skills? It was simply bowing to the temper of the times, according to Wharton management professor Peter Cappelli, who says such an approach is increasingly common in today’s workplace.


“In the economy now, change is faster, and the odds that your skills will need to be updated have increased,” Cappelli says. “The question becomes, is your employer going to reinvest in you or move on to someone else?” 


As director of the school’s Center for Human Resources, Cappelli wanted to know why a few companies have in fact remained committed to retraining, even in the new ramped-up business climate. “The question seems central to understanding why some employers and some jobs are ‘good’ while others are not,” Cappelli notes in a recent paper entitled, “Social Capital and Retraining,” forthcoming in the journal Industrial Relations.


In the contemporary context, Cappelli explains in his paper, “corporate restructuring has become the main driver of job insecurity.” He cites an American Management Association survey in which 66% of the employers responded that “downsizing in their companies during the 1990s was driven by internal restructuring and reengineering, in contrast to more traditional explanations that relate job loss primarily to business cycles. And roughly a third of all companies reported that they were hiring new workers during layoffs in order to get the new skills they need to accommodate their restructuring plans.”


This process of restructuring by laying off and hiring, or churning, “externalizes the costs of restructuring to the laid-off employees and increases the demands on other providers of skills in society,” Cappelli writes. Retraining, however, suggests the opposite approach to restructuring “in that it internalizes restructuring costs, stabilizing employment and expanding overall skill levels in the process.”


Consequently, the decision to churn or retrain “is increasingly central to discussions about the responsibilities that employers have to workers and society,” Cappelli says. He defines retraining as “the decision to invest in the skills of workers who would otherwise be at risk of losing their jobs unless they acquire new skills.”


Having placed his research in the context of today’s workforce, Cappelli then asks the question: What distinguishes companies that emphasize retraining from those that do not? His answer: Companies that employ a large amount of “social capital” are more likely to retrain workers.


What is social capital? It’s a tight network of relationships within a workplace, Cappelli says. “It’s where you and I work together in ways that are idiosyncratic enough that we need to know a lot about each other. Imagine dancers. There, the social capital is significant. In order to perform well, dancers have to understand their partners’ personalities and movements.”


Companies that employ social capital often emphasize formal teamwork. “A well-functioning team is a profound notion,” Cappelli says. “It doesn’t just mean being in the same office.” Many people, for example, say they work with colleagues when, in fact, their day-to-day tasks rarely overlap. The extent of their joint work is fetching coffee for each other, kibitzing in the hallway or sharing tired jokes via e-mail. Occasionally, they might offer each other advice. Well-functioning teams, however, “don’t just share what they know. They know what to ask each other. They know each other’s strengths.”  


In his paper, Cappelli discusses the idea of synergies in the context of social capital in the workplace. “Because it is an asset that exists between individuals rather than within each individual, social capital may suggest why it could make sense to reinvest in and retain individuals even if their job-specific skills are obsolete: The relationships they maintain with others may create value that extends beyond their ability to perform their current job.”


Cappelli goes on to suggest a direct connection between social capital and retraining “that turns on the make-or-buy choice that underlies the retraining decision. If a firm chooses not to retrain, it replaces existing employees with new ones. In the process social networks in the workplace are disrupted, and social capital is destroyed. If a firm does retrain, it preserves social networks and retains social capital.”


In an economic sense, Cappelli says, “social capital can be thought of as a particular type of fixed investment that can be preserved through retraining … It may take less of an investment to retain redundant employees to make a contribution than to hire new ones because the former always have important firm-specific investments in social capital. But the investment is in relationships, not skills.”


White-collar endeavors that demand a high degree of social capital range from advertising firms to surgeons’ operating-room teams. Among blue-collar workers, social capital exists among groups of tradesmen who work together on construction project after construction project. “These relationships rely on trust. You can’t easily keep track of who’s done what. There’s a high degree of mutual obligation that’s difficult to cover in a formal contract,” says Cappelli. He suggests a test: “If you can spell everything out in a formal contract, then it’s a business with low social capital.”


Compare high ‘social capital’ sorts of jobs to, say, an accounting firm, where a set of formal rules — the Generally Accepted Accounting Principles — govern what employees do. The rules are the same no matter where an accountant works, which makes it relatively easy for an accountant to switch jobs. As Cappelli describes it in his paper, “Some organizations rely on bureaucratic management and work organization practice based on rigid rules and procedures for decision making that are designed in part to be relatively impervious to social relations and resilient to employee turnover. The classic example of assembly line operations based on the principles of scientific management seem to fit that model in that they reduce opportunities for social relationships to affect the work process … Work systems based on teamwork and empowered groups, in contrast, rely much more heavily on the social relationships between employees and therefore on social capital to operate effectively.”


For his research, Cappelli looked at more than 3,000 employers that had responded to a 1994 U.S. Census Bureau survey on employment practices. The survey asked specifically whether the employer retrained workers who would otherwise be laid off due to economic changes at their establishment. He also looked to see whether skill requirements for their jobs had risen, which suggested a need for retraining, and whether they had excess operating capacity, which suggested the need for a possible layoff.


Since Cappelli also had to find companies that relied on social capital, he looked for those that employed self-managed teams and Total Quality Management programs, which depend on problem-solving teamwork. In addition, he sought out companies that allowed flex-time, which requires that workers cooperate in setting schedules and making sure tasks gets completed.


He found a strong statistical relationship between companies that retrained workers with outdated skills and those that employed social capital. “The story here is that workers are bringing something other than their skills; remember, these people had obsolete skills. But what they bring is they know everybody else, know how to get along with them, and know everybody’s strengths and weaknesses. As a result, when they are retrained, they will make progress faster.”


In his analysis, Cappelli ruled out three conventional explanations of why companies retrain. They don’t retrain because 1) they face high hiring costs, which make it costly to replace workers 2) offer lots of training of other kinds as a rule or 3) have other employee-friendly policies. None of those held up to statistical analysis; Cappelli couldn’t find a relationship between them and the likelihood that a company would retrain workers. For example, just because companies strive to be “good employers” by offering benefits such as health insurance, family leave and profit-sharing doesn’t mean they will offer retraining. 


Cappelli did find a relationship between companies offering employee stock options and retraining. “Stock options are back-loaded compensation. You have to stay around to get them, and retraining helps people stay around.” Then again, companies might not want people with lots of options to stay. After all, if the workers were forced to leave, the company wouldn’t have to hand them shares of its stock. But “companies that have opportunities to cheat like that don’t do it very often,” Cappelli points out. “They would develop a bad reputation, which is worse than not offering the options in the first place.”


Cappelli’s paper grew out of his service on a committee empanelled by the Russell Sage Foundation in New York. The foundation, which supports social-science research, asked a group of scholars to examine how more companies could be induced to retrain workers with obsolete skills. The committee members disagreed over whether it is possible to make companies retrain workers without fundamentally changing the way those companies operate and whether the problem was merely a technical one of finding the right incentives and penalties.


But Cappelli, a labor economist by training, believes employers might have fundamental reasons for taking different paths. “I’m skeptical of the carrot-and-stick approach. It would take massive carrots and massive sticks to get all companies to retrain, and it would ultimately change the way they operate.”


Different firms have different strategies because they operate in different markets, Cappelli says. Sometimes, employing teams and keeping those teams together makes sense. Other times, it doesn’t. Consider the difference between SAS Institute, a North Carolina-based maker of statistical-analysis software, and other software firms. SAS, the nation’s largest privately held software company, has turnover that is a fraction of the industry average, typically about 4% a year. As a result, it retrains its programmers as it enters new markets.


“SAS isn’t trying to develop new browsers and move into markets where nobody has been before. It is trying to adapt and extend a product it has had for 30 years. As a result, they need to keep the employees with the specific knowledge of its product. Companies that are doing new and different things, on the other hand, benefit from having new people with different ideas,” Cappelli says. “Do you want all software companies to look like SAS? It would be a peculiar business world if you didn’t have variation. The variation is there for a reason.”


Cappelli concludes in his paper: “Employers who retrain workers do so at least in part to preserve the social capital that exists in worker relationships. Specifically, the use of work practices like self-managed teams and TQM rely on that social capital to operate effectively … The TQM result may also reflect social capital beyond coworkers, including relationships with customers and suppliers. These results point to the importance of ‘strong-tie’ social capital of the kind that develops in close working relationships.”


If anything, the message of Cappelli’s paper goes to workers more than their employers. “My advice is caveat emptor,” he says. “If you walk into a company with a lot of contractual relationships, a lot of low social capital-type work, don’t expect that it’s going to make a big investment in you if business turns down.”