Free Trade Doesn't Work

by Ian Fletcher

  Silicon Valley is a famous success story of free enterprise, and to a large extent it deserves this reputation. Nevertheless, its rise was shot through with government support, without which it would probably never have existed. In fact, every place in the world where a semiconductor industry has developed, it has been the explicit target of state industrial policy.698

  The entire semiconductor industry is based upon the transistor, which was invented by Bell Laboratories in 1947. Bell Labs, however, was no product of free-market capitalism, but was the research wing of the old American Telephone and Telegraph (AT&T), a government-sanctioned monopoly. This company could only afford to support an expensive laboratory full of Nobel-caliber scientists precisely because it was a monopoly: protected from competitive pricing pressures, assured that no competitor would capture the commercial value of what it invented, and dedicated to the long term. It is Exhibit “A” against the canard that large, bureaucratic, government-subsidized companies protected from foreign competition can’t innovate. (This is not to say that these characteristics are positive goods in their own right, but it does rather suggest that the true determinants of industrial dynamism often lie outside laissez faire clichés.)699

  The semiconductor industry was a massive beneficiary of infant-industry subsidies from the start. As it hatched and grew in the late 1950s and early 1960s, close to 100 percent of its output was bought by the military,700 which needed expensive high-performance semiconductors for uses like missile guidance systems at a time when most consumer electronics still ran on vacuum tubes. Even as late as 1968, the Pentagon bought nearly 40 percent of the semiconductors produced in the U.S.701 Military demand enabled companies to stake their risky investments at a time when nobody else would buy their expensive cutting-edge technology.702 It enabled them to build the expertise that was later applied to civilian markets and achieve the scale economies needed to bring costs down into the range affordable for mass consumption.

  Aviation is another example of the dependence of America’s most successful industries on industrial policy. The entire 7x7 series of Boeing planes derives from the 707 launched in the late 1950s, which was the civilian twin of the KC-135 aerial-refueling plane built for the Air Force. Boeing actually lost money on its commercial aircraft operations for the first 20 years.703 To give further examples of the military lineage that made U.S. civil aviation possible:

  Lockheed sold commercial versions of its C-130, C-141, and C-5A. [The Lockheed L-1011, McDonnell-Douglas DC-10] and Boeing 747 were all spawned by technical advances on the engines used for the C5-A. In short, every generation of the new civilian air transport has relied heavily on technology developed for the military.704

  Other industries have been born from U.S. government industrial policy. The latest, of course, is the Internet, which derives from the military ARPANET built to enable communications between computers used for defense research. Even Google, the ultimate better-mousetrap free-market success story, was based on research done by founders Larry Page and Sergey Brin at Stanford while supported by National Science Foundation grant IRI-9411306-4 to research digital libraries.705 And the biotech industry has been incubated by, and has depended upon basic research funded by, the National Science Foundation and the National Institutes of Health.

  Unfortunately, if present trends continue, America’s harvest from federally funded industrial policy will inexorably diminish. Even the military itself is now lagging. According to John Young, former head of the Pentagon’s Advanced Technology and Logistics division:

  The [Defense] Department is coasting on the basic science investments of the last century and is losing the force multiplier advantage conferred by harvesting those investments. The last 15 years (since the demise of the Soviet Union) have seen the Department pull back substantially from many science areas. Yet, scientific knowledge is the underpinning of the current U.S. capability overmatch in most areas.706

  It seems Sputnik did us a bigger favor than we knew! The need to beat the Soviet Union appears to have been the decisive factor in disciplining the U.S. government to pursue an effective industrial policy, and when the Cold War ended, serious industrial policy seems to have ended with it.

  During the final push of the Cold War under President Reagan, the Defense Intelligence Agency and CIA created Project Socrates, whose purpose was to understand America’s declining economic and technological competitiveness and develop industrial policies in response. But just as this project was nearing fruition, the Berlin Wall came tumbling down. President Bush was ideologically hostile to industrial policy and systematically destroyed the project.707 He had Defense Advanced Research Projects Agency (DARPA) director Craig Fields reassigned and ordered all records of the project destroyed to frustrate Freedom of Information Act requests.708 Thus died what should have been the crowning achievement of Cold War industrial policy: a systematic codification of its economic insights. Ironically, some of the key staff of this project have since worked on economic strategy for foreign nations like Poland and Malaysia, which have diligently used this knowledge to compete with the U.S.

  With the end of the Cold War, even the most basic elements of purely military industrial policy began to get short shrift. For one thing, the Pentagon ceased to care very much about buying American. In the words of then-Secretary of Defense Dick Cheney, a key figure in this shift, policies favoring American defense producers “raise questions about my spending money on things I could get cheaper elsewhere, and it raises the specter of having to rely upon less than first-rate technology in certain areas.”709 Thanks to nearly two decades of such policies, the U.S. is now unable to put a single military aircraft into the sky without using components made by potential adversaries. As a 2005 Defense Department report put it:

  The potential effects of this restructuring are so perverse and far reaching and have such opportunities for mischief that, had the United States not significantly contributed to this migration, it would have been considered a major triumph of an adversary nation’s strategy to undermine U.S. military capabilities.710

  The Pentagon is now facing a rash of counterfeit electronic components in military systems, which lays the U.S. open to the kind of deliberate sabotage we have ourselves employed against adversaries such as Saddam Hussein.711 We also now face politically motivated refusal of foreign suppliers to provide needed components. The best known case is a Swiss company, Micro Crystal AG, which refused to supply piezoelectric timing crystals for the guidance system of the Joint Direct Attack Munition (JDAM) smart bomb at the time of the Iraq war.712 (One surviving American company was found.) The military is not unaware of this problem, but is hamstrung by the political power of defense contractors, who find outsourcing parts very profitable.713

  INDUSTRIAL POLICY IN REVERSE: DEINDUSTRIALIZATION

  Deindustrialization thanks to bad trade policy is a more complex process than is usually realized. It is not just layoffs and crumbling buildings. It is, in fact, industrial policy in reverse. As a result, understanding industrial policy helps illuminate how industries die.

  When American producers are pushed out of foreign and domestic markets, it is not just immediate profits that are lost. Declining sales undermine their scale economies, driving up their costs and making them even less competitive. Less profit means less money to plow into future technology development. Less access to sophisticated foreign markets means less exposure to sophisticated foreign technology and diverse foreign buyer needs.714 When an industry shrinks, it ceases to support the complex web of skills, many of them outside the industry itself, upon which it depends. These skills often take years to master, so they only survive if the industry (and its supporting industries, several tiers deep into the supply chain) remain in continuous operation. The same goes for specialized suppliers. Thus, for example, in the words of the Financial Times’s James Kynge:

  The more Boeing outsourced, the quicker the machine-tool companies that supplied it went bust, providing opportunities for
Chinese competitors to buy the technology they needed, better to supply companies like Boeing.715

  Similarly, America starts being invisibly shut out of future industries which struggling or dying industries would have spawned. For example, in the words of Richard Elkus:

  Just as the loss of the VCR wiped out America’s ability to participate in the design and manufacture of broadcast video-recording equipment, the loss of the design and manufacturing of consumer electronic cameras in the United States virtually guaranteed the demise of its professional camera market....Thus, as the United States lost its position in consumer electronics, it began to lose its competitive base in commercial electronics as well. The losses in these related infrastructures would begin to negatively affect other downstream industries, not the least of which was the automobile....Like an ecosystem, a competitive economy is a holistic entity, far greater than the sum of its parts.716 (Emphasis added.)

  Free market economics systematically denies this greater-than-the-sum-of-its-parts aspect of the economy, as it assumes on principle that every part of the economy is always correctly priced by the market, rendering impossible any holistic effects in which the whole is worth more than their sum.

  The fruits of this reductionist way of thinking are visible all over the U.S. economy today. For example, the U.S. is today inexorably losing the position in semiconductors it built up with past industrial policy. This is visible in declining plant investment relative to the rest of the world. In 2009, the whole of North America received only 21 percent of the world’s investment in semiconductor capital equipment, compared to 64 percent going to China, Japan, South Korea, and Taiwan.717 The U.S. now has virtually no position in photolithographic steppers, the ultra-expensive machines, among the most sophisticated technological devices in existence, that “print” the microscopic circuits of computer chips on silicon wafers. America’s lack of a position in steppers means that close collaboration between the makers of these machines and the companies that use them is no longer easy in the U.S. This collaboration traditionally drove both the chip and the stepper industries to new heights of performance. American companies had 90 percent of the world market in 1980, but have less than 10 percent today.718

  The decay of the related printed circuit board (PCB) industry tells a similar tale. An extended 2008 excerpt from Manufacturing & Technology News is worth reading on this score:

  The state of this industry has gone further downhill from what seems to be eons ago in 2005. The bare printed circuit industry is extremely sick in North America. Many equipment manufacturers have disappeared or are a shallow shell of their former selves. Many have opted to follow their customers to Asia, building machines there. Many raw material vendors have also gone.

  What is basically left in the United States are very fragile manufacturers, weak in capital, struggling to supply [Original Equipment Manufacturers] at prices that do not contribute to profit. The majority of the remaining manufacturers should be called ‘shops.’ They are owner operated and employ themselves. They are small. They barely survive. They cannot invest. Most offer only small lot, quick-turn delivery. There is very little R&D, if any at all. They can’t afford equipment. They are stale. The larger companies simply get into deeper debt loads. The profits aren’t there to reinvest. Talent is no longer attracted to a dying industry and the remaining manufacturers have cut all incentives.

  PCB manufacturers need raw materials with which to produce their wares. There is hardly a copper clad lamination industry. Drill bits are coming from offshore. Imaging materials, specialty chemicals, metal finishing chemistry, film and capital equipment have disappeared from the United States. Saving a PCB shop isn’t saving anything if its raw materials must come from offshore. As the mass exodus of PCB manufacturers heads east, so is their supply chain.

  It’s the big picture that needs to be looked into. There isn’t one single vertically integrated North American shop that could independently supply a circuit board. Almost every shop stays in business supported solely by revenues from ‘brokering’ Asian boards.719 (Emphasis in original.)

  All over America, other industries are quietly falling apart in similar ways.

  Losing positions in key technologies means that whatever brilliant innovations Americans may dream up in small start-up companies in future, large-scale commercialization of those innovations will increasingly take place abroad. A similar fate befell Great Britain, which invented such staples of the postwar era as radar, the jet passenger plane, and the CAT scanner, only to see huge industries based on each end up in the U.S.

  America’s increasingly patchy technological base also renders it vulnerable to foreign suppliers of “key” or “chokepoint” technologies. In the words of Laura D’Andrea Tyson:

  Because semiconductors are an indispensable input throughout the electronics complex, strategic control over their supply by a concentrated Japanese oligopoly poses a threat to downstream producers throughout the world.720

  One form this takes is the refusal of oligopoly suppliers to sell their best technology to American companies as quickly as they make it available to their own corporate partners.721 It doesn’t take much imagination to see how foreign industrial policy could turn this into a potent competitive weapon against American industry. For example, Japan now supplies over 70 percent of the world’s nickel-metal hydride batteries722 and 60-70 percent of the world’s lithium-ion batteries.723 This will give Japan a key advantage in electric cars.

  IMPOSSIBLE NOT TO HAVE AN INDUSTRIAL POLICY

  Because of the myriad impacts that government decisions have upon industries, there is no option of “not having” an industrial policy. There is only good and bad industrial policy. In the words of James C. Miller III, chairman of the Federal Trade Commission under Reagan, “Any discussion of industrial policy should begin with recognition that we have one. The issue is what type.”724 A nation that refuses to have a conscious industrial policy will still have a de facto industrial policy because the sum of its short-term tactical choices will amount to a long-term strategic choice whether intended or not.

  If nothing else, the brute fact of foreign mercantilism means that the option of genuinely free trade has long since been taken away from us. Again, D’Andrea Tyson:

  We must not be hoodwinked by the soothing notion that, in the absence of U.S. intervention, the fate of America’s high-technology industries will be determined by market forces. Instead, they will be manipulated by the trade, regulatory, and industrial policies of our trading partners.725

  Free trade and the absence of deliberate industrial policy are not neutral choices, free of government interference; they are positive strategic bets in their own right, which will only pay off if their key underlying economic assumption is true: pure free markets, at home and abroad, are always best. Taking an ideological stand against “central planning” misses the point, because the central planning that has rightly disgraced itself is socialist central planning, something entirely different. Similarly, ideological fulmination against “government picking winners” misunderstands the role that federal support plays. As Michael Borrus, founding general partner of the Silicon Valley venture capital firm X/Seed Capital, explains, referring to the National Institute of Science and Technology’s Advanced Technology Program:

  ATP is sometimes labeled with the profoundly misleading and profoundly misinformed characterization of ‘picking winners and losers.’ That is, frankly, flat wrong. No investor, private or public, picks winners and losers in technology innovation. Rather, it is the market (customers) that does the picking. By contrast, with ATP and other federal technology programs, the government is really helping to plant long-term technology seeds in areas of private market failure or acute public need. Some of those technology seeds will sprout, others will not. But the planting, the activity as a whole, must go forward if long-term economic gains are to be effectively harvested.726

  Opponents of industrial policy claim to oppose all industrial
policy, but actually only oppose varieties they disapprove of. Despite the laissez faire myth that industrial policy was discredited with the end of the Cold War, worldwide, as Dani Rodrik explains:

  The reality is that industrial policies have run rampant during the last two decades—and nowhere more than in those economies that have steadfastly adopted the agenda of orthodox [free market] reform. If this fact has escaped attention, it is only because the preferential policies in question have privileged exports and foreign investment—the two fetishes of the Washington Consensus era—and because their advocates have called them strategies of ‘outward orientation’ and other similar sounding names instead of industrial policies.727

  Export processing zones are one example of this industrial-policy-by-an-other-name. These receive duty-free access to raw material and component inputs, tax holidays on corporate, personal, and property taxes, exemption from usual regulations (including labor laws), and subsidized infrastructure. Another example is the wide array of subsidies, ranging from tax advantages to one-stop-shop help navigating local bureaucracy, given to encourage foreign direct investment (FDI).