Entrepreneurs, Not Equations, Run the Economy

By Gene Epstein

Economist Amar V. Bhide is one of those lesser-order types who prefers to interpret the world with words, so his chances of making it to Stockholm remain rather dim. But in his recently published book, The Origin and Evolution of New Business (Oxford University Press, 2000), he does toss off some killer statements.

One of my favorites: "Differences in product attributes cannot easily explain why the chewing gum company Wrigley could build a global brand while lollipop companies did not."

In positing this "lollipop law" of business enterprise, Bhide (pronounced "BEHdee") is making a crucial contribution to the theory of industrial organization (IO) that is directly subversive of mainstream shibboleths. His chewing gum theorem says that firm size is mainly determined by the actions of entrepreneurs like Mr. Wrigley, and not so much by those "exogenous" factors so dear to the hearts of the IO theorists: for example, "receptiveness to advertising" or "barriers to entry" or even that old standby, "economies of scale."

Scale economies refer to the tendency for large plants to enjoy a cost advantage over small ones. Ergo, say the technological determinists, the facts of nature dictate the number of firms in a particular industry.

But in the first place, responds Bhide, firms are often a lot larger than scale economies would require, which surely suggests the influence of some entrepreneur. Witness General Motors, assembled by swashbuckling empire-builder William C. Durant and later refined by Alfred Sloan.

Moreover, says Bhide, the economies of mass production are mainly created by entrepreneurial initiative anyway, as the examples of Henry Ford and Andrew Carnegie can attest.

The flip side to that fundamental point truly drives it home: We can't assume, as mainstream theory would suggest, that all opportunities for scale economies have been exploited. Or as Bhide puts it: "If furniture could be mass-produced in the same way as automobiles, someone would have done so already." Indeed, citing the economist Joseph Schumpeter, he notes that "the cotton and wool industries provided opportunities to innovate in late 18th century England ... but only the cotton industry had the entrepreneurs who could take advantage of the possibilities."

What Bhide is really proposing is the need for a Copernican revolution in mainstream theory.

Copernicus, of course, pushed the then-subversive idea that the center of the universe was the sun and not the earth, as Ptolemy had propounded. Similarly, Bhide is part of a long line of thinkers who believe we must abandon the Ptolemaic view that central to our understanding of market process is this thing called "perfect competition," from which we derive the satellite concepts that go under the rubric of "imperfect competition."

Under perfect competition, "there are many small firms, each producing an identical product and each too small to affect the market price," to quote from the dean of textbooks, Economics (16th edition, 1998), by Paul Samuelson and his long-standing collaborator, William Nordhaus.

In other words, in this weird world, not much in the way of competition can exist. Since every firm is a price-taker, no one ever tries to increase market share by cutting the price. Since the products are identical, no one tries to introduce improvements. And while the authors never quite say it, production costs are also predetermined, which means creative cost-cutting is strictly forbidden.

This leaves Samuelson and Nordhaus free to enlighten us about the pointless fact that in the never-never world where everything is known, and where absolutely nothing need be left to chance, elegant cost curves can be drawn that will match with price.

But ironically, just to spice up their account of all those perfectly competitive curves, these two dismal scientists get creative. They invent the example of a presumably tiny bicycle manufacturer turning out "Fabiola's Fabulous Formfitters," never noticing that such a flamboyant attempt at branding probably means that the bicycles in question are not at all "identical" to those of the firm's "many" competitors, or at least are not perceived as such by consumers.

So instead of having to plow through sentences that begin with, "Starting at the maximum-profit output of 4,000 units, if Fabiola's sells one more unit ... ", the hapless undergraduate is provided with the opportunity to gain more insight by pausing to daydream about the market niche entrepreneur Fabiola was hoping for when he created his Formfitters -- and whether those easy-on-the-bottom bikes are just a passing fad or have a real future.

And that's just the point.

The entrepreneurial functions of riskbearing, innovation, and arbitrage should be made central to our understanding of market process. In The Origin and Evolution of New Business, the 44-year-old India-born Bhide shows that the entrepreneur is always seeking to be one among many, whether by establishing niche businesses like beauty salons and restaurants, or through the grander schemes of Frederick Wallace Smith, who created FedEx, or of Sam Walton, who through slow trial-and-error built Wal-Mart.

In other words, the search for some form of monopoly profit is what spurs their creative energy, and that simple insight blasts away the static models of perfect and imperfect competition that straitjacket our thinking.

As Bhide, currently Glaubinger Professor of Business at Columbia University, put it to me last week, "We know that most economic progress is not the result of the efficient allocation of existing resources. It's the result of entrepreneurial innovation."

And how that impulse toward innovation is best fostered is subject to a kind of Heisenberg Uncertainty Principle. Whether it's better for an industry to have many firms, a handful or just one acting alone mainly depends on the nature of the people who happen to be involved.

"If you don't have anybody with the will or capacity to take big initiatives," comments Bhide, "you'll have many companies competing. If there are several equally strong individuals in the industry, you end up with oligopoly [few sellers]. If you have one clearly superior entrepreneur, you'll end up with market dominance."

Now, static theory strictly informs us: The more firms, the better, since the whole point is to approach that perfectly competitive state in which all market power is abolished.

But if spurring innovation is what really matters, then there's no way to address that issue with any certainty. Indeed, it's often better for firms to complete less and cooperate more (that dread word: collusion), especially since the sharing of information can aid creative initiative.

Comments Bhide, "Take the extreme case of Microsoft's position as the dominant supplier of an operating system. The key question is not whether this form of monopoly will lead to price-gouging; it's whether breaking up the company will lead to faster rates of innovation. And the answer is, we don't know.

"But we do notice the company is putting out a stream of new products with more backward compatibility, since it has a commitment to its main products. And we know a company of that size and prestige has enormous potential to assemble a critical mass of creative talent."

On page 883 of the ninth edition of his textbook (1973), Samuelson displayed a graph showing that at certain projected rates of growth, the Soviet Union might out produce the U.S. as early as 1990. That killer theorem deservedly self-destructed. Let's hope for the same for the static theory on which it was