This is where that question really matters.

Photographer: Simon Hofmann/Bongarts/

Who Is a Corporation Supposed to Serve?

Justin Fox is a Bloomberg View columnist. He was the editorial director of Harvard Business Review and wrote for Time, Fortune and American Banker. He is the author of “The Myth of the Rational Market.”
Read More.
a | A

The shareholders are the owners of a publicly traded corporation, and it is the job of the directors to look out for their interests. To many investors, financial journalists, corporate executives and even board members, these are self-evident truths.

As a matter of corporate law and history, though, they’re only partial truths, and recent ones at that. The idea that the corporation and board exist to serve shareholders, while it certainly had its adherents before then, really only became widely accepted in the 1980s and 1990s.

One 1976 academic paper in particular gets a lot of the credit or blame for this shift: “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure,” by Michael C. Jensen and William Meckling, two proud products of the University of Chicago economics tradition who at the time were at the University of Rochester. They described a conflict at the heart of corporate governance between the “principals” of a firm and the “agents” hired to run it.

The agents are clearly the managers of a corporation. It’s less certain who the principals are. Over the years corporations have been depicted as servants of the public, of their employees, of creditors, of shareholders and of other stakeholders. If you have to pick just one, shareholders do seem like the obvious choice -- and with the principal-agent model, you kind of do have to pick just one.

Over the past decade, though, there has been a flood of complaint that the principal-agent, shareholder-value-maximizing view of the corporation is wrong or at least incomplete. I'm sympathetic to these arguments, but taken together they can seem like the work of fuzzy-headed do-gooders. The principal-agent model may be wrong, but at least it's a simple, hard-nosed economic model. It will probably take another economic model to oust it.

It turns out there is a contender that’s been gaining adherents among corporate legal scholars for 15 years now, although it has yet to break through in the courts or public conversation. This model is called team production, and it has its roots in a 1972 paper by Armen Alchian and Harold Demsetz, two University of California-Los Angeles economists never mistaken for fuzzy-headed do-gooders. (They were both considered leading members of the Chicago school of economic thought, even though Demsetz spent only a few years there and Alchian never taught or studied there.) In "Production, Information Costs, and Economic Organization," they wrote:

Team production … is production in which 1) several types of resources are used and 2) the product is not a sum of separable outputs of each cooperating resource. 

The resources they had in mind were mainly people, and the puzzles they contemplated involved how to induce them all to work efficiently together. Corporations, they wrote, relied on managers to organize the team production, and then shareholders and competitive forces to police “managerial shirking.”

This isn’t wildly different from the principal-agent view, of course, and subsequent writings by other economists also emphasized the role of owners in efficiently organizing team production. In a paper published in 1998, though, University of Chicago finance professors Raghuram Rajan and Luigi Zingales pointed out a problem. A paraphrase:

[T]here are likely to be situations in which the owner of assets used in team production can extract a larger share of the gains from production by selling the assets rather than by making her own specialized human capital investments in the business. Knowing this, the other team members may be reluctant to make the specialized human capital investments they need to make.

That’s from "What must corporate directors do?" a new Brookings Institution paper by Margaret Blair of Vanderbilt Law School that nicely summarizes the intellectual history and practical implications of team production theory. In a principal-agent world, anything that clarifies the rights of principals and makes it easier to push out underperforming agents -- hostile takeovers, for example -- is a good thing. In the Raghuram-Zingales version of team production, it’s more complicated. In some cases, they wrote in “Power in a Theory of the Firm,” it can be most efficient for a “completely unrelated third party” to call the shots: “The third party holds the power so that the agents critical to production do not use the power of ownership against each other.”

When Blair read those words in 1998, she was working as a senior fellow in the Economic Studies Program at the Brookings Institution (she has an economics Ph.D., not a law degree). She recalls walking over to the office of Lynn Stout, a law professor then visiting Brookings, showing her the relevant passage, and saying, “That sounds like a board of directors, doesn’t it?”

Stout agreed, and the two of them wrote “A Team Production Theory of Corporate Law,” published in 1999. The paper argued that many aspects of the corporate legal tradition in the U.S. fit the team-production model much better than the principal-agent model. An example: the deference courts have long shown to directors’ business judgment in decisions on hostile takeovers and other contentious matters.

The Blair-Stout paper has since become the most cited corporate-law article of the past two decades. “It's been hugely successful in terms of the number of people who say we have to pay attention to this idea,” Blair says. Most of those people, she adds, are law professors under age 40 -- a bunch of whom gathered at the Seattle University School of Law last summer for a symposium on team production. Older scholars haven’t shifted from their focus on principal-agent problems, and team-production theory has yet to have much influence on how corporate law is taught. That just seems like a matter of time, though.

While the legal world plays catch-up, Blair sees hopeful signs in the technology and Internet industries:

Many of the firms that have sprung up in that space are led by entrepreneurs who say, “I have no intention of maximizing shareholder value. I'm investing for the long-term, and I've demonstrated that I have better ideas than you [the outside shareholders] do.”

That’s a team-production view, not a principal-agent view. It’s about bringing together resources in a corporation over a sustained period of time to produce cool things. It’s a big part of how technological and economic progress is achieved. It also has its limits. “That will end," Blair says. "There will come a time when Mark Zuckerberg runs out of ideas, and someone else will have to take control." When that time comes, principal-agent considerations will matter a lot. Until then, though, a team can get a lot more accomplished than an agent.

  1. Now head of India's central bank.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Justin Fox at

To contact the editor on this story:
James Greiff at