The Computer Chip vs. Potato Chip Debate

Economists long ago put to rest the error that British economist Adam Smith made when he argued that manufacturing should be given primacy in a country’s economy. Indeed, in Book II of “The Wealth of Nations,” Smith condemned as unproductive the labors of “churchmen, lawyers, physicians, men of letters of all kinds, players, buffoons, musicians, opera singers, opera dancers, etc.” We may agree with Smith — and Shakespeare — about the uselessness of lawyers perhaps, but surely not about British actor and director Laurence Olivier or Italian opera singer Luciano Pavarotti. But the larger issue is the role of manufacturing in a nation’s economy. This issue has been discussed for decades but has become particularly relevant in the United States and other leading nations in the wake of the global crisis.

In Britain in the mid-1960s, Nicholas Kaldor, a world-class Cambridge economist and an influential adviser to the Labour Party, raised an alarm over “deindustrialization.” His argument was that an ongoing shift of value added from manufacturing to services was harmful because manufacturing was technologically progressive, whereas services were not. He even got a Labour Chancellor of the Exchequer, James Callaghan, to introduce in 1966 a Selective Employment Tax, which taxed employment in services more heavily than employment in manufacturing — a measure that was reversed in 1973, once it was realized that it would hit the tourist industry, which generated badly needed foreign exchange.

Kaldor’s argument was based on the erroneous premise that services were technologically stagnant. This view no doubt reflected a casual empiricism based on the mom-and-pop shops and small post offices that English dons saw when going outside their Oxbridge colleges. But it was clearly at odds with the massive technical changes sweeping across the retail sector and eventually the communications industry, which soon produced faxes, mobile phones and the Internet.

In fact, the dubious notion that we should select economic activities based on their presumed technical innovation has been carried even further, in support of the argument that we should favor semiconductor chips over potato chips. While rejection of this presumption landed Michael Boskin, chairman of President George H.W. Bush’s Council of Economic Advisers, in rough political waters, the presumption prompted a reporter to go and check the matter for himself. It turned out that semiconductors were being fitted onto circuit boards in a mindless, primitive fashion, whereas potato chips were being produced through a highly automated process — which, by the way, is how Pringles chips rest on each other so perfectly.

The “semiconductor chips versus potato chips” debate also underlined a different point. Many proponents of semiconductor chips also presumed that what you worked at determined whether, in your outlook, you would be a dunce (producing potato chips) or a modernist (producing semiconductor chips).

I have called this presumption a quasi-Marxist fallacy. Marx emphasized the critical role of the means of production. I have argued, on the other hand, that you could produce semiconductor chips, trade them for potato chips, and then munch them while watching television and becoming a moron. On the other hand, you could produce potato chips, trade them for semiconductor chips that you put into your personal computer and become a computer wizard. In short, it is what you “consume,” not what you produce, that influences what sort of person you will be and how that affects your economy and your society.

Ignorant of the extensive deindustrialization debate in 1960s Britain, two University of California Berkeley academics, Stephen Cohen and John Zysman, started a similar debate in the United States in 1987 with their book “Manufacturing Matters,” which claimed that without manufacturing, a viable service sector is untenable. But this argument is specious. One can have a vigorous transportation industry with trucks, rail and air cargo moving agricultural produce within and across nations, as countries such as pre-Peronist Argentina, Australia, New Zealand and modern Chile have done very successfully.

Cohen and Zysman argued that manufacturing was related to services like “the crop duster to the cotton fields, the ketchup maker to the tomato patch,” and that if you “offshore the tomato farm … you close or offshore the ketchup plant. … No two ways about it.”

My reaction was: “As I read the profound assertion about the tomato farm and the ketchup plant, I was eating my favorite Crabtree & Evelyn vintage marmalade. It surely had not occurred to me that England grew its own oranges.”

While these episodes reflected academic obsession with manufacturing and therefore died early deaths, the same cannot be said for the latest revival of the “ manufacturing fetish” in the United States and Britain. The latest flirtation with supporting manufacturing has come from the current crisis, especially in the financial sector, and is therefore likely to have greater prospects for survival. The fetish is particularly rampant in the United States, where the Democrats in Congress have gone so far as to ally themselves with lobbyists for manufacturing to pass legislation that would provide protection and subsidies to increase the share of manufacturing in the nation’s gross domestic product.

Because of the financial crisis, many politicians have accepted this argument in a virtual throwback to Adam Smith. Financial services, the argument goes, are unproductive — even counterproductive — and need to be scaled back by governmental intervention. It is then inferred that this means that manufacturing must be expanded. But this does not follow. Even if you wanted to curtail financial services, you could still focus on the multitude of nonfinancial services.

Diesel engines and turbines are not the only alternatives. Many services — such as professional therapy, nursing and teaching — are available. The case for a shift to manufacturing remains unproven for the simple reason that it cannot be proved.