Paul Krugman: an old piece on national competitiveness


By Paul Krugman

Almost nobody — in business or government — would disagree with this statement: “Today America is part of a truly global economy. To maintain our standard of living, we must learn to compete in an ever tougher world marketplace. That’s why high productivity and product quality have become essential. We need to move the economy into high-value sectors that will generate jobs for the future. And the only way we can be competitive is if we forge a new partnership between government and business.”

The problem is: It’s baloney. In reality, there is almost nothing to our fixation with national competitiveness, or its central idea — that every country is like a giant corporation slugging it out against rivals in global markets. The U.S. and Japan are simply not competitors in the same way that, say, Ford competes with Toyota. Any country’s standard of living depends almost entirely on its own domestic economic performance, and not on how it performs relative to other countries. That’s not just my view; it’s what most economists think.

Why should you care? One important reason: Countries that wrongly think they are in a competitive struggle over who gets the spoils are more likely to fall into a major trade war. Such conflicts don’t destroy economies the way real wars do, but they harm everyone — and it takes a long time to recover from them. When protectionist measures like the infamous Smoot-Hawley tariff shattered the global trading system between the two world wars, trade among industrial countries didn’t regain its 1914 peak until 1970.

The other risk is that true believers in competitiveness — I call them strategic traders, to point up their obsession with winning export battles — often advocate pouring huge sums of taxpayer money into projects they hope will create jobs or build prestige but that make almost no economic sense. Example: the billions of dollars France has spent propping up its computer industry. Leading American strategic traders — among them, Labor Secretary Robert Reich, Clinton health policy adviser Ira Magaziner, and MIT economist Lester Thurow — have argued for similarly extravagant (and misguided) investments to enhance national competitiveness.

To understand why the doctrine of strategic trade drives economists into a fury, go back to that imaginary quotation. It was carefully constructed to illustrate six fundamental misconceptions. Let’s do the numbers.

(1) “Part of a truly global economy.” Strategic trade rhetoric implies both that the typical U.S. business or worker is now producing for global markets and that the extent of globalization is historically unprecedented. Neither is true.

In 1992, exports were 10.6% of U.S. GDP, imports were 11.1%. This was substantially more than the 4% on both sides of the ledger in 1960. Surprisingly, however, the numbers haven’t changed much since 1980: The big increase in the importance of international trade to the U.S. economy took place in the 1970s, not the 1980s. And even so, the numbers remain modest.

Now remember the strategic traders’ favorite analogy: that a country is merely a corporation writ large. As we’ve just seen, the theoretical corporation called America Inc., even after globalization, sells almost 90% of its output to its own workers and shareholders. How many companies do anything close to that? None. Clearly countries are nothing at all like corporations.

One might argue that while most U.S. output is still sold to ourselves, a much larger proportion is sold in domestic markets that are newly subject to international competition, such as automobiles or computers. This is, however, only a little bit true. More than 70% of U.S. output consists of services rather than goods, and most services are effectively insulated from international competition because they are hard to transport. Collectively, services account for only about 20% of U.S. trade.

Nor is the degree of U.S. dependence on world trade without precedent — or even unusual. Most historians of the international economy date the emergence of a truly global economy to the Forties — the 1840s, when railroads and steamships reduced transport costs to the point where large-scale shipments of bulk commodities became possible. International trade quickly surged. By the mid-19th century, the leading economy of the day, Great Britain, was exporting more than a third of its GDP — three times as much as the U.S. exports today. Britain eventually invested about 40% of its savings overseas every year. And an era of mostly open borders was marked by international migration that dwarfs anything recent. (Where was your great-grandmother born?)

Why does this matter? Strategic traders reject conventional economic wisdom on the grounds that it is no longer relevant in a global economy — but even classical economic theory, developed mostly by English economists in the 19th century, applied to an economy that was far more dependent on international trade and investment than the U.S. is today. So what looks like wisdom to the unwary looks like ignorance and shoddy thinking to anyone who knows some economic history.

(2) “Competing in the world marketplace.” Again, strategic traders hold that countries compete with each other in the same way that Nike competes with Reebok; they attribute the long stagnation of middle-class living standards in the U.S. to a failure to do this effectively.

What’s wrong with their thinking? At a conceptual level, the most basic point about international trade is that it is not a zero-sum game. Companies like Nike and Reebok are almost purely rivals: Only a tiny fraction of Nike’s sales are to Reebok workers, and vice versa. So one’s success tends to be at the other’s expense. But the major industrial countries, while they sell products that compete with each other, are also each other’s main export markets and main suppliers of useful imports. If anything, a successful European or Japanese economy helps the U.S. by providing us with larger markets.

Moreover, the purpose of international trade — the reason it is useful — is to import, not to export. That is, what a country really gains from trade is the ability to import things it wants. Exports are not an objective in and of themselves; the need to export is a burden that a country must bear because its import suppliers are crass enough to demand payment.

But isn’t it a fact that the stagnation of U.S. living standards has been in large part due to a failure to compete effectively on world markets? No, it’s not a fact. From 1979 to 1989 the real compensation of all U.S. workers rose 5.8%, while productivity rose 5.1%. These are purely domestic variables — that is, productivity is not measured relative to other countries, and no data about global market shares or anything that involves the global economy are taken into account. Yet the two series rose by almost exactly the same (disappointing) amount. So we got almost exactly the growth in living standards we would have gotten if the U.S. were alone in the world and we had no international trade at all.

(3) “High productivity.” Just about everyone now agrees that the U.S. economy needs higher productivity. Most people, however, are confused about why. The most popular explanation is that we need to be productive in order to compete in the global economy. That was the explanation President Clinton gave in February 1993, when he tried to justify an economic package that included painful tax increases. But it’s wrong. We need to be more productive in order to produce more, and this would be true even if the U.S. were completely without foreign competitors or customers.

To illustrate, let’s consider three questions. First, what happens to a country whose productivity is inferior to that of the countries it trades with? The common view is that it will suffer. After all, if you aren’t better than your rivals in something, how can you sell anything on world markets? The right answer is that being less productive than your trading partners poses no special problems. Of course, a country whose productivity is low across the board is not going to have a high standard of living; but that has nothing to do with the fact that it must coexist with more productive nations. In fact, the option of exporting to those superior “competitors” the things you don’t make too badly, and importing from them the things you do, delivers a somewhat higher living standard than a country with very low domestic productivity might otherwise enjoy. This is the basic point David Ricardo made in 1817 when he first expounded the principle of comparative advantage — a principle that every student learns in Economics 101 and that most politicians persist in ignoring.

Second question: What happens to a country whose productivity growth lags behind that of its rivals? The common view is that it is in big trouble — after all, a corporation that systematically fails to match competitors’ productivity gains is not going to stay in business. The right answer, however, is that how fast productivity is growing abroad, and whether we are ahead of or behind the pack, is irrelevant.

From World War II until 1973, productivity in the U.S. rose 2.8% annually. After 1973, it rose only 0.9% annually, a rate generally slower than in other advanced nations. If the rest of the world did not exist, this slowdown would have reduced growth in U.S. incomes by 1.9% per year. Any effects from lagging behind foreign competitors contributed at most another 0.1 percentage point. The moral: What matters for the trend in U.S. living standards is our domestic rate of productivity growth — period.

Third question: Which is more important, productivity growth in industries that must compete with foreigners, or productivity growth in sectors that produce for sheltered domestic markets? Most people would answer that productivity growth among companies that compete internationally is more important. The right answer, again, is that what matters for the U.S. standard of living is the overall productivity of American workers. It doesn’t matter whether they are competing with foreigners or producing only for the domestic market. This has one major implication that runs counter to much conventional wisdom: Service productivity matters more than manufacturing productivity. To be exact, since about 70% of the value added in the U.S. economy is in services, vs. 20% in manufacturing (with the remainder in agriculture, construction, and mining), a percentage point gain in service productivity is worth 3 1/2 times as much as an equal gain in manufacturing.

(4) “High-value-added sectors.” Strategic traders claim the best way to improve living standards is to encourage investment in sophisticated industries like computers and aerospace, which provide high value added per worker. Wrong again. Why is value added per worker in some businesses higher than in others? It isn’t enough to assume they are just better businesses. If they were, capital and labor would flood into them, competing those high returns away. (Markets may be imperfect, but they aren’t stupid or sluggish.) In fact, the usual reason value added per worker is high in some industries is that other inputs, such as capital or skill, are high there as well. Since the economy has limited supplies of capital and skill, encouraging industries that use those scarce resources intensively may well lower instead of raise per capita income.

What is most striking, however, is that advocates of “high value” industries, like Robert Reich and Lester Thurow, apparently haven’t bothered to check which industries actually do have high value per worker. As the table on this page reveals, it turns out that the real high-value industries in the U.S. are extremely capital-intensive sectors like cigarettes and oil refining. High-tech sectors that everyone imagines are the keys to the future, like aircraft and electronics, are only average in their value added per worker.

(5) “Jobs.” Many strategic traders blame America’s failures in international competition for the loss of “good jobs” in manufacturing, with the unfortunate workers forced either into unemployment or into much lower-paying service jobs. The image of the former steelworker now earning minimum wage flipping hamburgers is deeply embedded in popular perceptions.

In fact, the U.S. economy has been the great job engine of the advanced world, with a 38% increase in employment from 1973 to 1990, compared with 19% in Japan and only 8% in Europe. Now it is true that real wages have stagnated. But is this because workers have been forced out of good manufacturing jobs into low-paying service jobs? No, for two reasons. First, manufacturing jobs are not all that well paid (the hourly wage in manufacturing is only 10% higher than in other jobs). The stagnation in real wages was not because good jobs in manufacturing were lost, but because real wages for all jobs that didn’t require a college education stagnated or fell, as technological change reduced demand for less skilled workers. And second, the widespread belief that the U.S. has lost its manufacturing base in the face of foreign competition is simply wrong.

Deindustrialization and the “hollowing out” of the economy never happened. True, the share of manufacturing in U.S. value added and employment has been falling for many years. But this trend is common to all industrial countries. In fact, manufacturing’s share of the economies of Germany and Japan has declined as fast as or faster than that in the U.S. Nor is there any mystery about the trend. Essentially, it is driven by the combination of relatively fast productivity growth in manufacturing and limited demand for manufactured goods. The general public prefers to spend most of the annual increase in its income on services. The result: Demand can be satisfied by a static or even falling number of factory workers.

The story should sound familiar: It’s exactly what happened to agriculture 50 years earlier. Very few Americans now live on the farm, not because our farmers are uncompetitive, but because they are so productive we don’t need many of them. And America’s “deagriculturalization” has proceeded in spite of consistent trade surpluses in farm products.

What appears to make the manufacturing story different is that since 1980 the U.S. has consistently run trade deficits in manufacturing, and there is no question that some industries, such as shoes and apparel, have shrunk under the impact of import competition. But is this mainly because of a failure to be competitive in manufacturing?

The answer is an overwhelming no. In 1992 the U.S. trade deficit in manufactured goods was $62 billion. Now remember that much of a dollar of “manufactured” exports indirectly represents services such as health care purchased by the manufacturer. (GM’s largest supplier is not a steel company but Blue Cross/Blue Shield.) Input-output studies of the U.S. economy give us a pretty good estimate of the hidden service component of manufactures trade: Only about 60% of a dollar of manufactures sales represents manufacturing value added. Thus, we should scale down the merchandise trade deficit by 40% to estimate the overall impact of trade on the industrial base. This suggests that, at most, competitive problems in U.S. manufacturing, as measured by our trade balance, reduced value added in these industries by 60% of $62 billion, or $37 billion, in 1992.

That may sound like a big number, but keep in mind that even a supposedly “deindustrialized” America in 1992 still had a value added in manufacturing of more than $1.1 trillion. We have just suggested that in the absence of a trade deficit that number might have been about $37 billion larger; that’s a difference of only 3.3%.

(6) “A new partnership.” This is, of course, the bottom line. Unfortunately, as we’ve just seen, a government-business partnership guided by the tenets of strategic trade would almost certainly lead to bad policies, since it would be founded on excruciatingly bad economics — as bad, in its own way, as the extreme supply-side doctrine popularized by a different set of policy entrepreneurs during Ronald Reagan’s presidency.

My advice is to consider a proper understanding of the real relationship between productivity and competitiveness as a kind of test of the reliability of supposed experts, in and out of government. The issues involved are not hard to sort out — we’re not talking quantum mechanics here. So if you hear someone say something along the lines of “America needs higher productivity so that it can compete in today’s global economy,” never mind who he is, or how plausible he sounds. He might as well be wearing a flashing neon sign that reads I DON’T KNOW WHAT I’M TALKING ABOUT.

Where would productivity gains bring the biggest payoff for the U.S.?

A. Computers

B. Aerospace

C. Autos

D. Fast food

Answer: D. Fast Food

Value added per worker, 1991 figures

Strategic traders say the U.S. should pour money into industries with high value added per worker because they will generate the jobs of the future. Will they? Look at this list and decide for yourself.


Cigarettes $823

Petroleum refining $270

Autos $112

Tires & inner tubes $101

Aerospace $86

Electronics $74

All manufacturing $73

Originally published, 3.7.94

Leave a Reply

Your email address will not be published. Required fields are marked *