• 08 Jan 2001
  • What Do You Think?

Have We Extended the Boundaries of the Firm Too Far?

by James Heskett

Summing Up

"What we are looking at is a fundamental challenge to our assumptions about which corporate structures work," commented Daniel Hayes in response to the recent piece on the future bounds of the organization. Raman Julka was just as dramatic: "The generic value chain described by Michael Porter is being fragmented by organizations; value addition ... is taking place in numerous forms ..." Their comments reflected the general assumptions among respondents that organization boundaries will continue to expand, with better or worse (ranging up to "ruinous") results for those engaged in it.

Those organizations thought to be most able to extend their boundaries are those able to work in a "componentized" manner, with "certain pieces of technology or innovation that can be outsourced" (Sarang Kulkarni), those able to employ the open standards of the Internet and willing to transfer databases to a "rival supplier" in the event of an unsatisfactory relationship (Joshua Doherty), and those maintaining a highly focused in-house R&D capability while outsourcing small portions of strategic needs associated with this capability in order to preserve proprietary information (Kulkarni).

Doherty was concerned that organizations might extend their boundaries without a clear strategy. He offered a "2x2" rule-of-thumb strategy for doing this based on the levels of strategic risk and in-house ability associated with the action. Where both are high, he suggests "insourcing." Where both are low, outsourcing is the answer. Where ability is high and strategic risk low, an effort to establish a separate business opportunity (by "spin off" or other means) might be in order.

Of course, the toughest decisions often are associated with situations in which strategic risk is high but in-house capability is low. Here Doherty suggests a strategic alliance with the purpose of upgrading in-house capability as at least one alternative. Those who have spent a great deal of time examining strategic alliances and their somewhat checkered record might offer a word of caution here. They would ask: Do the justifiably selfish purposes that bring an organization to a strategic alliance contain the seeds of the alliance's demise?

The technological, social, and legal environment appears to favor further extension of organization boundaries. In the future, will this phenomenon be limited primarily by the environment, or by the ability of managers to take advantage of it? What do you think?

Original Article

A byword of the new economy has been "partnering." Small start-ups have pursued, in the words of my colleague Tom Eisenmann, the "get big fast" strategy by focusing on core activities while doing the rest by partnering with others, including competitors.

A large, successful firm such as Cisco Systems has used partnering, often involving minority investments in many partners, to enable its employees to work only on high value-added activities that yield more than $600,000 in revenues for each of its more than 25,000 employees. (The phenomenon is described in a case, "Cisco Systems: Are You Ready?", that I coauthored recently.)

Even GE places high value on what it calls "boundaryless" behaviors among its executives, encouraging them to exchange ideas with each other and, when beneficial to the company, with managers outside GE as well. In short, it is becoming more and more difficult to define the boundaries of an organization.

It is the latest manifestation of a lecture given 68 years ago by Ronald Coase, now professor emeritus at the University of Chicago Law School. Prof. Coase set forth a theory designed to help set limits on organizational boundaries. He proposed that so-called "transaction costs" (which he regarded as quite high at the time, thus providing a justification for the vertical integration of large organizations) set limits on behaviors that we refer to today as partnering, the forming of alliances, and outsourcing.

Transaction costs resulted from the difficulties of communicating between organizations, obtaining full disclosure, and in general exercising oversight in such relationships. Coase's ideas were thought to be so valuable that they earned him the Nobel Prize in Economics in 1991.

One of the most important phenomena of the Internet-based economy is the drastic reduction in transaction costs for those wishing to engage in partnering. Thus, Cisco relies on contract manufacturers for most of the final assembly of its products and nearly all of its basic production, employing the Internet in managing relationships designed to respond to customer needs.

It obtains a large proportion of innovative ideas from its partners, some of whom it eventually acquires. And it is engaged in building a network of more than 4,000 academies in which its partners train tens of thousands of the technicians needed each year to keep the Internet functioning. Often its partners are potential competitors.

Are there limits to boundaryless behavior? If carried too far, can it result in loss of control over the quality of goods and services delivered by the supply chain? Can reliance on others for innovation lead to the neglect of in-house R&D capability? Can it result in the "leakage" of strategic information to partners who are also competitors? When partners fail, can it in fact create a "house of cards" effect, as firms with interlinked investments have to write them down on their books? Given the potential for these behaviors, are transaction costs, regardless of the Internet, as low as they seem?