Size matters.

WHY ECONOMIES GROW: The Forces That Shape Prosperity and How We Can Get Them Working Again by Jeffrey Madrick Basic Books, $26.00

JEFFREY MADRICK'S Why Economies Grow is a short work with a big ambition: Debunk mainstream economics about the best way to promote faster economic growth. Most economists believe that the key to higher growth is more capital, better-trained workers and, most powerful of all, a steady stream of economic innovation. For Madrick, a country's economic development depends less on capital, labor, and new ideas than on how large its market is and how consolidated its industries are. Large and growing markets, he argues, produce economies of scale, and only when industries achieve such scale economies can more capital, labor, and new ideas produce faster growth. In the work's final analysis, the point of this novel thesis is not really to revise economic thinking, but to justify the return of big government spending that can pump up the market and promote those economies of scale.

"The American experience," he writes, "is the clearest example of the influence of large and growing markets on economic development" The book's strongest suit is a recounting of 19th and early 20th century economic history that reminds us that America's economic success is unimaginable without a vast and open domestic market that spurred business to develop new forms of production and distribution. The power of a large and growing market to drive economic progress is also useful to keep in mind when considering the rapid ongoing development of China and India.

Market size matters, but it hardly explains everything. Brazil and Russia with their vast domestic markets have repeatedly stumbled on the path of economic development, and nations with small internal markets flourished in Europe in the 18th and 19th centuries and in Southeast Asia in the late 20th century. So Madrick concedes that a strong economy needs capital, broad literacy and education, sound political and economic institutions, and much else. The point of focusing on the extent of a country's domestic market is to move his argument from an exposition on how economies develop to his own explanation for why economies grow.

The analysis of growth presented here is at once novel and nostalgic. As large markets drive economic development over decades, so strong market demand is said to drive growth year to year in a very particular way: by spurring firms to achieve vast economies of scale in the production of standardized products for those markets. For the author, the paradigm of sound growth is the 1950s, when everyone in the vast American market drove a Chevy or a Ford, bought the same cereals, wore the same grey flannel suits--and the economy grew faster than in any decade since.

To make this case and rehabilitate demand-driven economic policy, Madrick must not only refute conservative dogma on the primacy of capital and tax cuts, he must also discredit the neoliberal approach based on innovation and fiscal discipline. Here, the argument has to contend with the consensus of modern macroeconomics about what makes a modern economy grow. This consensus begins simply enough: The economy grows when the quantity or quality of its capital and labor increases, or the ways in which they are combined improve. When more people work or we invest more capital, GDP should rise; when workers' average skills improve or businesses use capital more efficiently, the economy should expand faster. When managers come up with' better ways to organize their workplaces and nm their firms, or innovators invent new ways of using labor and capital, growth should accelerate.

Following the work of Robert Solow (who received the Nobel Prize for his effort) and Edward Denison, most economists also agree on the relative importance of these factors. The media trumpeting the New Economy in the late 1990s went overboard, but economics supports their basic case: Solow, Denison, and others found that innovation in its various forms is the most important factor in U.S. productivity growth. At least 34 percent of U.S. growth from 1929 to 1982, for example, can be traced to the development of new products and processes; new ways of financing, marketing, and distributing goods and services; and new approaches to organizing and running a business. Innovation matters so much because it affords the economy a way of escaping the law of diminishing returns that otherwise would inexorably reduce the punch from simply adding more capital or labor.

Innovation plays the largest part, but only a part. The increasing numbers of workers and hours they work can account for 25 percent of growth over those years, and another 16 percent can be traced to workers' increased skills. After that comes the conservatives' favorite: Increases in the capital stock of equipment, factories, offices, and so on, explain 12 percent of the growth of those 53 years. A number of other factors also matter, including Madrick's favorite: Economies of scale realized by businesses with the emergence of regional and national markets can explain about 11 percent of growth. The economy's better use of labor has yielded a comparable effect, principally in women moving from household work to offices and Southern agricultural workers moving to industrial jobs in the North and Midwest.

Why Economies Grow never directly challenges this economic canon. Instead, it tries to recast the various factors as mainly consequences of growing demand. New technologies, new ways of marketing and distributing goods, and so on, all matter; but they are said to come about to meet the demand of expanding markets. The economic payoff for Madrick comes when the mass employment of these innovations produces new economies of scale.

This view requires some imaginative reinterpretations of recent economic developments. The productivity slowdown of the 1970s and 1980s is said to reflect not just the usual suspects like inflation and labor-force changes, but the rise of consumer demand for specialized products. This new era of "consumer choice" fragmented the mass market for standardized goods, producing countless niche markets that could never achieve real economies of scale. Similarly, we owe much of the productivity resurgence of the latter 1990s not to the efficient application of information technology (IT), but to a revival of a traditional mass market in standardized versions of IT products, with the attendant economies of scale for their producers.

In the end, the analysis lacks, well, any proof. But Madrick's point is not to provide data or evidence, but to shift our economic thinking away from factors like innovation and skilled labor, and back to the demand side of the economy. That provides the basis for his own agenda for prosperity: To keep domestic market demand growing, we need ongoing budget deficits to finance public investment, a higher minimum wage, expanded job programs, pensions and healthcare for all workers, more income support and training for the unemployed and so on, and on.

The question is: Without a lot of innovation, is the American market for traditional liberalism large enough to sustain its continued development?

ROBERT SHAPIRO, a fellow at the Brookings Institution and a columnist for Slate, directs the economic consulting firm Sonecon.