Europeans have a lot of experience with governments
tinkering with technology policy. Bad experience.
From London, John Browning reports on why Europeans feel
that technology policy is an idea whose time is past.
Europe has an odd grip on Democratic economists. Some see Europe's interventionist economics as an injury to honest, free-trading America that cannot be ignored. Witness tough-talking Bill Clinton telling the American people that "if the Europeans want to continue to subsidize Airbus I'm not just going to roll over and play dead." Others see European policies as innovations to be imitated. Laura Tyson, head of Clinton's council of economic advisers, recently cited the Airbus project as an example of a successful defensive alliance between government and industry. But whether the question is to imitate or to retaliate, the conclusion often seems to be that the American government should take a more active role in promoting economic progress and innovation.
Europeans find this somewhat ironic. Many of their own economic interventions, like Airbus, were inspired by American intervention - Boeing received government subsidies for the development of military transports and used much of the basic technology in producing the 747. But most ironic - and perhaps dangerous for Clinton's America - is that most Europeans have been bitterly disappointed with the results of their economic interventions. They are busily reforming many of the same programs that some US policy makers would imitate.
For years, would-be American interventionists have urged their government to imitate Europe's ambitious program to develop next-generation, high-definition television (HDTV). Yet now that the interventionists finally have a sympathetic ear in government, the Europeans are spoiling the case by shamefacedly dismantling much of that program. Despite having spent nearly $1 billion developing a European HDTV, the European Commission's industry minister, Martin Bangemann, has said that Europe should instead fall into line with the new digital technologies being proposed in America.
Such embarrassments and disappointments leave Europeans somewhat puzzled by Americans' newfound enthusiasm for technological intervention. To quote Stephen Woolcock, a European economist at the Royal Institute of International Affairs: "The normal assumption in the United States, particularly among Democrats who want to promote the idea of an industrial policy, is that here in Europe we have coherent policies and that they are working. They are not. Technology and industrial policies are neither coherent nor working at the Community level, or in most European nations."
Statistics support Woolcock's claim. In 1978, the twelve nations that now make up the European Community more or less broke even on their trade in computers, telecommunications, and other information technology. By 1991, despite the best efforts of a variety of interventionist policies, Europe had sunk to an information technology trade deficit of $40 billion a year. In the case of semiconductors, Europe's share of production is only 10 percent of the world total - despite tariffs on foreign chips and generous programs to promote local production. Worse, though Europe accounts for roughly one-third of the world economy, its consumption of semiconductors has fallen to about one-sixth of the world total. With Europe's big electronics firms - Philips, Olivetti, Bull and Thomson - all mired in losses, no recovery is in sight.
Nor is there much optimism in Europe's laboratories. The region has only about 500,000 full-time researchers, compared to nearly 1 million in America. On average, Europe's research spending lags America's. In recent years, Europe's share of world patents has dropped, as has the number of European Nobel Prize winners.
Europe's disappointments in technology policy are not by themselves a conclusive argument that America should not try to create new policies of its own. If nothing else, Europe's failures should provide lessons from which the Americans can learn. The worry, however, is that America will not learn. The Clinton administration comes to its economic interventionism with very different arguments and expectations than those that have driven the Europeans. But it has reached - or will soon reach - very similar practical problems.
In Europe, technology policy has been made, time and again, not by the decisions of people with the biggest technical problems or the brightest ideas, but by those with the most powerful lobbyists. The results speak for themselves. This is the unhappy history that Clinton will either learn from or repeat.
Europe as Teacher
Britain and France represent the poles of technology policy in Europe. Britain's Conservative government dislikes intervention, and the British have spent much of the past decade trying to reduce government's role in the economy. France's Socialist government, by contrast, sees intervention as part of its duty in helping to achieve national greatness. The two different attitudes reflect each country's respective experiences with technology policies, experiences best summed by a comparison of telecommunication companies on both sides of the Channel.
In the early 1970s, both the (then-Labor) British and the French governments looked at telecoms and were inspired. Digital telecoms, they both decided, would become both a major market and a key element of the infrastructure needed by every developed country. So they both did the two obvious things: put together a coalition of local telecoms-equipment producers to build digital switches, and gear up the national telephone company for a program of investment in digital technology. But from similar beginnings, they reached very dissimilar ends.
France's telecom program reached its goals. Its national telecom network is now among the most modern in the world. Better, France's biggest telecoms-equipment supplier, Alcatel, is also the world's biggest telecoms-equipment supplier - leading both Siemens and AT&T with about one-sixth of the world market.
Britain's telecom program flopped. Until British Telecom was privatized in 1984, its budget was continually raided by ministers looking for cash to pay for their favorite social programs. Investment in digital switches lagged - though since privatization British Telecom has regained a lot of ground. Meanwhile, managerial and technical problems plagued the development of the British digital switch, called System X. So far, System X has won only one export order, to a Caribbean island.
So why the difference? In part, France's telecoms programs were better managed than Britain's. One of the reasons: The French were not in practice quite as dirigiste (interventionist) as they professed to be. Razeen Sally, a researcher who has studied the rise of Alcatel at INSEAD, Europe's leading business school, argues that a key to Alcatel's success was its ability to join with France Telecom to take charge of national policy. Having agreed-upon goals with the government, ministers effectively let the companies do what they thought needed to be done - a sort of governmentless government policy. In Britain, by contrast, Labor ministers took the idea of intervention seriously. And as each successive minister tried to add his own spin to telecoms policy, that policy became increasingly erratic and confused - until the Conservatives took government out of the loop by privatizing British Telecom and opening the industry to competition.
The Exception That Proves the Rule
With Alcatel as their example of what intervention can do, France has spent much of the past decade trying to rouse its European neighbors to technological activism - with decreasing success. There are two reasons why France's brand of dirigisme disenchants its neighbors. First, the French government has chosen wrong in technology more often than it has chosen right. Second, even France's successful interventions have imposed hidden costs and distortions on the economy - which may yet come back to haunt it.
It's no surprise that Alcatel has used its influence over French telecoms policy to help create the sort of environment in which it would like to compete. Predictably, that means prices for telecoms equipment and services have stayed higher than in more competitive Britain. But it also means that the range of services and equipment has generally remained narrower. France has been relatively slow to adopt new services like mobile telephony or
private voice and data networks. Should such services indeed start to grow rapidly at the expense of conventional cable-based services - as many in Britain and America predict they will - Alcatel's very strengths could become France's weaknesses.
Minitel, France's national data network, is another example of the distortions that an inspired government can bring to innovation. True, Minitel has had unprecedented success in bringing data networking to the masses. But it has also created a virtual monopoly. No sane businessperson is going to compete with a government that can afford to give away millions of terminals because it has a neat idea. So data-networking in France will henceforth develop at a pace dictated entirely by government - a pace that is looking very sluggish. Minitel's teletext technology and teeny-tiny keyboard would frustrate any American accustomed to data services like CompuServe, the Internet, or even Prodigy. And with French unemployment now well over 10 percent, improving Minitel is not high on the list of the government's priorities.
A glance beyond telecoms to the other technologies backed by the French government is even more discouraging for those who would imitate its dirigisme. Telecommunications mark France's only real interventionist success. The Concorde project will never recover the costs of making its supersonic aircraft. The nuclear power industry, which generates all of France's electricity, has been on dubious commercial ground since fossil-fuel prices failed to rise mercilessly, as had been predicted when France embarked on its nuclear program in the 1970s. The Ariane rocket launcher project owes more of its success to machismo than economics. Europe's HDTV program, which the French staunchly supported, flopped. And whatever Airbus's success in annoying Boeing and President Clinton, a variety of economic studies conclude that it has not returned more benefits to Europe's taxpayers than they have invested in subsidies - even on measures that include technological spin-offs.
Commission of Compromise
Given the huge political differences between the member states of the European Community, it is not surprising that mutual suspicion and jealousy play an important role in shaping European policy. The European Commission has the unenviable task of reconciling the dirigiste French with the laissez-faire British while repairing other shades of policy in between. In addition, it must also try to ensure that resources are fairly distributed between rich nations and those like Greece and Portugal, which have not yet been able to afford technology policy, let alone technology for that matter.
Disagreements and mutual suspicion have helped shape two key elements of the Commission's policies towards new technologies. One is a competition policy that has vigorously discouraged member states from trying to promote their own favorite technology or industry. With the possible exception of former French Prime Minister Edith Cresson - whose vision of France leading a united Europe into economic warfare against the Japanese electronics industry was liberally doused with cold water in 1991 - politicians seem to desire intervention less than they fear having to explain to the voters back home why their neighbors' special interest groups can have subsidies they can not.
The second calming compromise built into the Commission's technology policy is a focus on pre-competitive research. Much of the $7 billion (4 percent of the region's civilian research spending) goes into set-piece programs that try to promote international cooperation in research far enough away from the marketplace that the Commission cannot be accused of playing commercial favorites. The ESPRIT program focuses on information technology; JESSI supports research into semiconductors; and RACE promotes broadband telecoms. Such programs typically pay about half the costs of research, but require that research projects involve a consortium of corporate and university researchers from several European countries.
The problem with keeping research a safe distance from the marketplace, however, is that it stays a safe distance from the market. The Eurocrats in charge of ESPRIT politely describe industry's take-up of the program's research results as "disappointing." Less politely, a survey of the program by leading European industrialists - led by Wisse Dekker, ex-chair of Holland's Philips - recommended that the program be scrapped, together with JESSI and RACE. None of the programs, they argued, had actually done what they were set up to do. Nor are they likely to.
Because the Commission understandably insists that European taxpayer money goes only to European industries, researchers cannot take advantage of technological advances made overseas. Not only does this hamper efforts to create international standards, it also too often encourages European researchers to squander resources re-inventing technologies already created elsewhere.
By promoting European research consortia, the Commission has fostered cooperation between European scientists. But the Dekker report argues that it has done so at the expense of links between researchers and the industrialists who will take their ideas to market. To overcome this problem, Dekker proposes a new policy based on "vertical" alliances between researchers and industry. This is presumably an idea shared by Martin Bangemann, who also assumed responsibility for information-technology research when he took over as EC industry minister in January 1993.
In effect, this policy begins to create a sort of techno-industrial complex, run by bureaucrats and big companies largely for bureaucrats and big companies. Given that Europe's big companies have a record of technology decision-making every bit as bad as government's, there is no reason to believe that this group will be any more efficient than the military-industrial complex that preceded it - except that they will have a far harder time selling $10,000 coffee-makers and $4,000 toilet seats to customers with an option to buy elsewhere.
Dekker himself was at the forefront of a classic example of the sort of trouble that governments and big companies can get into when they decide to work together: HDTV. In haste to develop a cinema-quality TV that could compete with the Japanese and Americans, Dekker, as head of Philips, and his counterpart at France's Thomson disastrously underestimated the progress of digital technology. They chose an analog technology, called MAC, as the centerpiece of Europe's competitive weaponry. The HDTV program was led by Philips and Thomson, but also used subsidies to try to bring together an alliance focused on commercializing the technology. Unfortunately, American researchers, not blessed by government attention, created digital HDTV - which means that Europe's MAC would be outdated by the time it could be deployed. So instead of building new vertical alliances, Bangemann's first challenge will be to see what he can salvage from the last one.
Uber Alles
Disasters elsewhere highlight the marvel of Germany's technology policies. Germany manages to promote technological innovation and diversity of decision-making at the same time. Instead of concentrating resources on a few favored technologies, German policies concentrate on diffusing technology through the economy. In the process of diffusion the government does have opportunities to give innovation a nudge or two in the directions ministers would like
to see it go, but most of the choices about which technologies to back - and how to back them - are made by individual firms in close touch with their markets.
The environment for Germany's technological success is as peculiar as the policies that fostered it. Its economy is spectacularly innovative and successful. But it is not particularly strong in either of the two industries which most countries would regard as the measure of their techno-muscle: computers and biotechnology. Germany's powerful green lobby has effectively banned most biotechnology, which it regards as just another way of messing up Mother Nature. And the combined bureaucracy of Siemens and the Bundespost sits like a lead weight - albeit a very solidly engineered one - on Germany's computer business. Instead, the country's success comes from the vast collection of small- and medium-sized states called the Mittelstand.
Many of the Mittelstand companies are part of machinery, metalworking, and other industries struggling against competition from low-wage-labor countries in Asia. They cannot be cheaper; they can only be better. One way that the German government helps keep them better is through a network of 50 or so subsidized research establishments called the Fraunhofer Institutes.
Each Fraunhofer is set up as a partnership between government and a university whose primary goal is to conduct contract research for local companies. The research work is done by students - many of whom will get a Ph.D. in the process - on subjects chosen by industry. But the work is heavily subsidized by government. This unique balance between university, business, and government is the key to the Fraunhofers' success.
In 1992, the total budget for the Fraunhofers was about $600 million. Only one-third of that money comes from the private companies in whose name the research is ostensibly being done. But the government's two-thirds share is managed to keep the private sector firmly in charge. One-half of the government money (one-third of the total) is matching funds for individual research contracts. Most of the other half is expected to be spent on equipment - so that each Fraunhofer can act as a demonstration area for the latest technology in its field. If the flow of new research contracts dries up, so too does the government money.
Repeat After Me
Surveying the history of Europe's technology policies, the obvious question for Clinton's team is: How could so many bright Europeans work so hard for so long and create so many screw-ups? The answer is that making decisions about technology is just not something that governments are naturally good at. To see why that might be so, indulge in a brief fantasy.
Imagine for a moment that you are a civil servant: intelligent, hard-working, selfless (don't chuckle, this is a fantasy). Imagine also that we can ignore special interests, greed, and power-mongering (suppress that smirk). Now ask yourself three questions:
* Are you comfortable with the prospect of taking money from, say, a warehouse supervisor raising four kids in a tract house and giving it, however indirectly, to a 29-year-old hot-shot computer programmer living in a bayside condo and driving a Porsche?
* Could you justify, in the name of national competitiveness, research projects that give taxpayer money to foreigners?
* Which of these two requests for research money would you pick: a heavily-researched, plodding proposal from a reputable firm, or a visionary tome, in places only comprehensible to those with both a Ph.D. and an active imagination, from a small firm started by researchers who, although widely considered brilliant in their fields, have never managed to hold down a job for more than eighteen months at a time?
Herein lies the essential conundrum. In theory, governments love to support innovation. But in practice, most innovation is done by privileged people taking big risks in global markets. That is just not the sort of game which fair, responsible governments can win.
The necessity of constructing policies within these limitations is the lesson from Europe. And it is far from clear whether Clinton and his team will learn that lesson by study alone. Some of Clinton's policies threaten a re-run of Europe's mistakes. Al Gore's insistence that only the government knows best how to build data networks seems at odds with his well-intentioned desire to spread this crucial technology as widely as possible through the economy. More worrying still are Clinton's own hints that in the name of "fair trade," he will grant favors to companies who feel themselves injured by foreign competition - the resulting trade war of tit-for-tat subsidies for "strategic" industries spells mutually assured destruction far more surely than any arms race.
But there are also hopeful signs that Clinton may actually be the new-model Democrat that his campaign literature claimed he was. Ron Brown's Commerce Department has been given a hefty budget to expand the number of manufacturing technology centers to over 100 - which, like Germany's Fraunhofers, will bring new technology to small- and medium-sized companies. Better, Robert Reich's education budget has been swelled with billions of dollars to help Americans learn to use tomorrow's technology.
In one of the more stirring moments of his first State of the Union address, Clinton asked Americans "to accept responsibility for the future of this country," and reminded them that "every one of us can be an engine of change in our own lives." In technology, as in many other fields, Clinton's success will be measured by the extent to which he keeps his side of the bargain he proposes. Responsibility is about making choices and paying for them, not just paying.
Through his policies, Clinton can push technology and innovation out into the
economy through education, manufacturing technology centers, universities, and the encouragement of dialogue between researchers and businessmen. Or he can pull technological decision-making into the hands of government and lobbyists - by promoting special treatment for favorite industries or politician's pet technologies.
Just how easy it is to slip over the line from a policy that promotes choice to one that diminishes it was demonstrated by Clinton's trip to Silicon Valley in late February. The Silicon Valley business leaders who flocked to support Clinton have no trouble deciding what is a strategic industry: It's them. (We're rich, we're successful - of course we're strategic.) As soon as they finished applauding Clinton's promises to decentralize government decision-making, they immediately began to ask which of their pet technologies Clinton would back. Visualization? Networking? Semiconductors? That is hardly decentralizing.
Now only Clinton and his advisers can decide who will make the choices about technology. But the lesson from Europe is that if they decide to make all the choices themselves, they will fail.
(c) 1993 Wired magazine
SIDEBAR-
The Wonks Debate
Europe's interventionism comes from tradition and instinct; America's by high-flying intellectual debate. While the roots of many of Europe's interventions lie in the traditions of socialism and the necessities of economies blasted by the Second World War, the new-style enthusiasm that fires many Democratic policy-makers comes from mathematical theories constructed by a new generation of economists - notably Paul Krugman at MIT, but also Stanford's Brian Arthur, Berkeley's Paul Romer, and Laura Tyson, who was at Berkeley until Clinton appointed her chair of his Council of Economic Advisers. Their achievement was to devise elegant theoretical justifications for what European politicians have always believed instinctively: Government intervention in specific industries can, in theory, be good for the whole economy.
This is a minor revolution in economic theory. Prevailing wisdom among economists has been that intervention aimed at promoting specific industries is at best futile and at worst a perverse form of charity towards competitors. This wisdom rests on two planks: first that competition is robust, and second that diminishing returns to scale ensure that companies find it harder and harder to make money from a product as its market share grows.
If these two assumptions are true, then dumping or other "predatory" trade practices are charities in disguise. If a company wants to sell goods at a price below the cost of making them - either because it's government subsidized or because it's just plain stupid - the smart thing to do is buy as much as possible at that price. Even if a subsidized company does get a big market share by selling its wares at ridiculously low prices, it will never be able to use that position to recoup its losses because, the traditional theory goes, new competitors will spring back into the market as soon as it raises prices.
The new economic activists find several faults with this theory. First, competitors might not be able to spring back quite as resiliently as the theory contends. Disused skills, closed factories, and dented confidence can, in reality, be hard to recover. Second, some "strategic" industries, they argue, may be worth more than the direct profits to be made from them because they provide indirect benefits to the rest of the economy - for example, by training workers or providing a sort of technological fertilizer. Equally important, many information- and innovation-based industries do not seem to suffer from the same diminishing returns to scale as do traditional ones. On the contrary, where other industries complain of monopolies, technology industries boast of "standards."
All of this means that with a nudge here and a subsidy there, governments can in theory promote vast prosperity. That vision has set political hearts going pitter-pat all over Washington. Unfortunately, the practical problems with this vision are at least as large as the theoretical opportunities. In theory, there is also a sure-fire way to make money on the stock market: Sell when stocks are going down and buy when they are going up. But this theory begs the very same questions as does the new theory of strategic industries. How, in practice, do you decide what is going to be successful, important, and strategic?
Europe's experiences demonstrate two points. First, without an answer to this practical question, technology policy does more harm than good because it sinks vast resources into dead-end technologies like Europe's HDTV or the Concorde. Second, governments do not have such answers. At various times and in various places, Europeans have tried all sorts of ways to make these choices. Their experiences show that politicians are at best no wiser than ordinary citizens but, because they are more powerful, their mistakes have larger and more damaging consequences.
John Browning (browning@well.sf.ca.us) is a London-based technology writer and consultant. After twelve years with The Economist, he is juggling various writing projects and research into artificial intelligence.
