The Myth of the Race to the Bottom

by Edward L. Hudgins
The author is director of regulatory studies at the Cato Institute and editor of Regulation magazine.

A frequently heard objection to in-creased freedom to trade is that Americans will lose jobs to other countries and see real wages decline as workers in Latin America, Asia, and elsewhere grow rich. The one objection actually raises three questions: First, is the United States experiencing a general economic decline? Second, are certain groups faring better while others fare worse? And third, if problems exist, is trade liberalization primarily responsible for them?

Theory explains and empirical evidence demonstrates that:

Americans on average still have the highest standard of living in the world. As America's economy has increasingly integrated with the global economy, average living standards have continued to rise. There has been no general decline in earnings.

Since the oil shocks of the early 1970s, economic growth has been less rapid than in the early post-war decades. But this situation does not seem to stem from imports or trade deficits.

Much of the real growth in earnings has gone to higher skilled and better-educated workers.

Lower skilled and less educated workers have not fared well in recent decades.

Although imports have affected certain less efficient American industries and enterprises, imports and trade deficits are not a significant cause of wage disparities.

Logical Problems

Many Americans fear their wages may decline because of trade. That fear of a "race to the bottom" is based on the erroneous premise that, with free trade, higher paid American workers will be forced to compete with lower paid workers in poorer countries, pulling down American wages in general. Jobs will always go to where costs are lower.

That negative view of free trade ignores the wealth-creating nature of free trade. With freedom to trade, all workers in all countries are able to use their labor to create the most valuable output possible in their situation. Less efficient workers might have to change jobs, possibly accepting lower earnings. And those individuals might have to upgrade their labor skills to raise their earnings. But the need to improve holds true for businesses as well as individuals facing competition. They must offer better products and services at prices acceptable to consumers if they are to continue in business. That is the natural course of market change whether the competition comes from domestic or foreign sources.

Best Use of Labor

Free trade and the division of labor might mean that in the United States more workers are employed writing computer software, creating new pharmaceuticals in research laboratories, running wheat farms with giant planting and harvesting equipment, or operating advanced machines in a highly automated textile mill. It might mean that in less developed countries more workers are employed assembling computer keyboards, plowing fields with small tractors, or using low-tech sewing machines to stitch collars onto shirts.

Low-wage workers in less developed countries generally do not have the educational background or practical experience to discover new medicines or write new software. They usually do not have access to the capital equipment necessary to farm thousands of acres in a season. And Americans would be foolish to replace a dozen individuals working a handful of huge combines with hundreds of individuals using hand plows to produce the same amount of crops. In either case of misallocated labor, the total amount of wealth created would be less than that created with division of labor and free trade. Hundreds of men hoeing a field would mean hundreds of men not producing other goods and services.

The Value of Labor

Another way to understand this situation is to recall that the value of any individual's labor -- that is, the price or wage it commands in the market -- depends on the value of the goods and services that the labor can produce. The labor of workers in industrialized countries produces more than that of workers in poorer, less developed countries. Thus American workers can trade their labor for higher wages.

As poorer countries develop, American market restrictions will not make American labor more productive or valuable. Restrictions will simply shield enterprises and workers from competition and thus from incentives to become more productive. The only way for workers to trade their labor for higher wages is to increase productivity.

Protectionists assume that wealth is a zero-sum game, that increased production in one part of the world necessarily means less wealth produced elsewhere. That is never the case when free trade is permitted.

A variation of the race-to-the-bottom argument maintains that the wages in low-wage and high-wage countries will converge if free trade is allowed. That argument will be considered in greater detail later. But in principle it, too, ignores the wealth-creating nature of free trade.

The Case of Puerto Rico

Some people maintain that industries flee the United States for lower wage countries with which America establishes free trade; they have a difficult time, however, explaining the situation in Puerto Rico. That Caribbean island is a commonwealth of the United States and thus can trade with the 50 American states with virtually no restrictions. In 1993 disposable per capita income was about $6,260 in Puerto Rico, compared with $18,552 in the 50 states. Given protectionist theories, one would expect full employment in Puerto Rico. In fact, unemployment is around 17 percent and has been in the double digits for decades. That compares with around 5.5 percent for the rest of the United States.

In truth, many factors in addition to low wages attract enterprises. These include levels of worker skills; the cost and reliability of water, electricity, and other utilities; transportation infrastructure; access to input resources; and proximity to markets. More important is the freedom of individuals and enterprises to earn and retain profits, and freedom from confiscatory levels of taxation and from heavy-handed government regulations that hamper business creation. Further, if a state or country maintains a welfare system that removes incentives to work, the economy will suffer.

Puerto Rico is not a free market paradise; and this is the best explanation for its continuing poor economic performance. For many American industries, it would make little economic sense to locate in Puerto Rico or in other countries.

America's Economic Place in the World

There are those who believe that other countries are overtaking the United States economically or that America is a country of shrinking wealth. They need only compare this country with others to be disabused of those notions.

Still the Largest Economy

America still has the world's largest economy. Calculations by the Organization for Economic Cooperation and Development (OECD) place America's gross domestic product (GDP) in 1994 at $6,649.8 billion. The next largest economy, Japan, has a GDP of $2,593 billion, not quite half the size of America's. The reunited Germany comes in third at $1,601.7 billion. (See Figure 1.)

America's Purchasing Power

As mentioned earlier, Americans on average have the highest standard of living. Such comparisons are never completely accurate or exact, but some measures are better than others.

World Bank's misleading numbers. There are several ways to compare average earnings in different countries. The World Bank comparisons sometimes quoted by doomsayers are misleading. When comparing per capita GDP in various countries, the World Bank takes a country's average per capita income in local currency and multiplies it by the exchange rate to show the equivalent in dollars. By that measure Americans in 1995 appeared have lower wages than the Japanese, Norwegians, and Danes and to tie with the Swedes. But ex-change rates do not accurately reflect differences in living standards. Rather, they reflect factors such as the international demand for a currency to be used for trade or investments.

Workers are paid in local currencies and make purchases in local currencies. A Japanese worker is paid in yen and must purchase food, housing, and all other goods and services in yen. Thus, a closer look finds that the average Japanese spends 25 percent of income on food compared with 15 percent or less for the average American, and as much as double the portion of income for housing than does an American. Two joint U.S.-Japanese government studies showed, respectively, consumer prices in Japan to be on average 41 percent higher than in the United States in October 1989 and 37 percent higher in April 1991.

OECD figures. A better overall measure for comparing living standards is purchasing power parity (PPP) as calculated by the OECD. PPP adjusts to a certain extent for exchange rate and trade flow effects, although imperfections remain. For example, some countries keep their national accounts in ways different from the majority of developed countries, creating anomalies in the statistics. Still, PPP better reflects real living standards than does the World Bank measure.

Using PPP, Americans are found to have higher average purchasing power than citizens of other industrialized countries. (See Figure 2.) In 1994 the average American earned $25,515, the Japanese $20,756, the German $19,675, and the Swede $17,442.

These numbers also show that there has been no decline in America's per capita GDP over the past decades as America's economy has become more integrated into the world economy.

Jobs in America

Strong job creation. The American employment situation is also better than in other industrial countries. Over the past decade more jobs have been created in the United States than in all the other major industrial countries combined. Between 1982 and 1990 America added 17.8 million workers to payrolls, an 18 percent increase. That rapid job creation occurred even though America's trade deficit was reaching record levels. (See Figure 3.)

The rate of job creation in the United States compares favorably with the 6.1 million jobs or 10 percent growth for Japan during the same period and 1.8 million jobs or 6.6 percent growth for West Germany. The average job growth rate in the European Community countries during that period was 8 percent, and in the industrialized OECD countries it was 8.8 percent. Furthermore, most of the jobs created in the United States were in the private sector, whereas a high proportion of the net employment gains in Europe was in the public sector.

Lower unemployment. Unemployment in Western Europe now averages over 10 percent, or twice as high as America's rate of around 5 percent. The rate for western Germany rose from 6.3 percent in 1991 to around 10 percent in 1995. Unified Germany's unemployment rate has hovered between 10 percent and 12 percent, with a post-war record of more than 4 million out of work.

Sweden, once considered the socialist country that could guarantee full employment, saw unemployment jump from 1.6 percent in 1990 to around 12 percent in 1994. Swedish employment figures have always been deceptive because that country's method of keeping national accounts differs from those in other countries. The Swedish government regularly places the unemployed in government-sponsored public works or job training programs and lists these individuals as "employed." Now, even by their own measure, the Swedes have had to abandon any pretense of government-created full employment.

Not only is Europe suffering from high unemployment levels, European unemployment tends to be longer term than in the United States. In 1989, 90 percent of unemployed Americans were out of work for less than six months. Only 6 percent were out of work for more than a year, compared with 44 percent of the French and 49 percent of Germans.

Unemployment in Japan rose from around 2.2 percent in 1990 to 3.3 percent in late 1996. Although 3.3 percent is low compared with the U.S. rate, it is high by Japanese standards. The wages of many Japanese workers, unlike those of most Americans, are flexible. As much as one-third of the average worker's income is paid in the form of a bonus based on how well a company performed that year. As a result, during short-term economic slowdowns, Japanese firms effectively cut workers' pay, thus avoiding the need to lay off workers. Nonetheless, most large Japanese firms are reevaluating their "employment for life" policies in the face of current economic problems.

Offsetting layoffs. In the United States in recent years, well-publicized layoffs by large companies have fueled calls for government action to protect American jobs. What generally gets little attention is the job creation that might not have occurred had some industries been protected.

For example, much consternation accompanied AT&T's 1996 announcement of planned layoffs of 40,000 workers. But job creation in other telecommunications enterprises and sectors has been strong. Since 1985 MCI employment has grown from 12,000 to 48,000. Jobs with Sprint rose from 27,000 in 1985 to 52,000 today. The number of cable operators and programmers grew during that period from 24,000 to 112,000. No one announces 40,000 new hires at once, so job growth rarely makes the headlines.

Unmeasured Progress

The wage situation in the United States bears closer examination. First, however, consider that such measures of economic well-being do not fully reflect a form of progress that is important to most Americans: the availability of new goods and services. Although the dollar value of such goods and services is generally reflected in economic statistics, their contribution to quality of life usually is not.

Deregulation or freer trade generates some benefits. For example, beginning in the late 1970s, airline deregulation cut the real, inflation-adjusted cost per passenger-mile flown--that is, the cost of tickets--by perhaps one-third. Families that might not have been able to afford a vacation can now visit friends, relatives, or places of interest in faraway parts of the country or world. Other families that might have had to drive to their destinations, wasting vacation days on the road and substantially increasing the risks of an accident, now gain more leisure time and safety by flying.

Other benefits come from technological advances. Medical techniques for heart surgery have advanced over the past decades, saving thousands of lives. More than 2,000 heart transplants now take place annually in the United States. The consumer electronics revolution has brought video cassette recorders and microwaves into most homes. The personal computer revolution can now link anyone to vast stores of information and a worldwide web of communications and knowledge.

Furthermore, in many cases, goods and services that were costly and thus luxuries for the well-to-do have become affordable, as prices dropped, for average Americans.

A Closer Look at Income

Although one can say, simply, that Americans on average are doing well compared with citizens of other countries, a more refined picture of U.S. economic conditions is useful.

Income Per Capita

America fares well compared with other countries in average per capita GDP. But what has been the trend over recent decades? Per capita GDP shows growth in recent decades except during the 1982 and 1990 recessions. Between the end of the Korean War in 1954 and the first oil shock in 1973, per capita GDP grew just over 2 percent annually. From 1974 through 1989 the average annual growth rate was only 1.6 percent. Was this slowdown due to the effects of trade?

W. Michael Cox and Beverly J. Fox of the Federal Reserve Bank in Dallas have observed that in the 84-year period between 1869 and 1953, the country's average per capita GDP grew at a rate of around 1.6 percent, the same as during the post-1973 period. Cox and Fox believe the country could have done better during the post-1973 period, but they do not suggest that more open world markets hampered growth. Nor are they alone in suggesting that the 1950s and 1960s were decades of extraordinary growth resulting from the huge amounts of resources and manpower shifted from the war effort into domestic production.

Furthermore, during the period between the end of the 1982 recession and 1989, when the U.S. trade deficit reached record levels, annual growth in per capita income averaged around 3 percent, well over 1.6 percent annually.

Consumption Per Capita

Average per capita GDP is not synonymous with earnings; GDP includes more than just income -- for example, investments. And GDP does not include other kinds of economic activity -- for example, part of the value of sales of used cars and other used goods.

Another measure of economic progress is consumption, which is the ultimate goal of economic production. The trend in the growth of consumption, Cox and Fox find, runs slightly below but parallel to the growth in per capita GDP. These consumption figures confirm the upward economic trend found in per capita GDP.

Average Earnings or Wages

One measure of economic well-being is hourly wages in the form of cash or money income. That measure shows actual deterioration since 1973, with a loss of about one-half percent per year through 1993. Critics of free trade use that statistic to justify their claim that Americans on average are growing poorer. But wages alone do not tell the whole economic story.

Average Total Compensation

Total compensation for work includes more than cash wages. Of great importance over past decades has been the rise in the value of nonwage and nontaxed income, principally health care and retirement contributions from employers as well as increased paid vacation.

Cox and Fox found that total per capita compensation rose by about one-half percent per year between 1973 and 1993. They observed that from 1953 to 1993 the portion of payrolls devoted to health benefits rose from 3 percent to 14 percent, and the portion going to retirement jumped from 5 percent to 13 percent.

Gary Burtless, a scholar at the Brookings Institution, found that employers' contributions to workers' health and welfare plans grew from 4.5 percent of personal income in 1973 to 6.6 percent in 1993. The government contribution in the form of Medicaid payments grew from 1.8 percent to 5.2 percent on average during that period.

Real Personal Income

A more inclusive gauge of economic well-being is real personal income per capita, which includes wages; benefits, such as employer contributions to health care; and other factors such as rents, interest, dividends, and government transfers. Cox and Fox found an average annual growth rate in real personal income of 1.65 percent between 1973 and 1989. That rate virtually matches their finding of 1.64 percent average annual per capita GDP growth during the same period. They further point out that paid vacation and holiday time have increased by about seven days over the past two decades. Such income gains are impressive.

Median Family Income

Another indicator of economic well-being is median family income. That measure differs from average per capita income in two ways. First, the unit of analysis is a group of individuals. Second, the median (as opposed to average) family income means the same number of families make more than the median as make less. Cox and Fox find that median family income has virtually stagnated, increasing only one-tenth of one percent per year between 1974 and 1993.

Critics of free trade also point to that statistic as evidence of the adverse effects of open markets. But international trade does not explain the situation. Burtless cites Census Bureau statistics showing that average family size decreased over the past two decades, from 3.44 in 1973 to 3.2 in 1993. Thus, a seemingly stagnant median family income is divided among fewer individuals, leaving more income per individual family member. Looking at income per person within the average now-smaller family, income is seen to go up. According to Burtless, per capita income per median family member increased 7.4 percent over the 20-year period.

Burtless also notes that average rather than median family income increased 13.2 percent between 1973 and 1993. Taking into account smaller family size, that means average real income per family member increased by 21.6 percent during that period.

Consumer Price Index Adjustments

A recent finding with regard to the consumer price index (CPI) may lead to calculations of economic progress. The special Advisory Commission to Study the Consumer Price Index, headed by former Bush administration Council of Economic Advisers chairman Michael J. Boskin, reported to the Senate Finance Committee in December 1996 that the rate of price increases in government statistics has been overstated by about 1.1 percent per annum. Such a CPI overstatement means that real purchasing power has been understated in the past. Thus, estimates by Burtless, Cox, Fox, and others probably understate the level of economic progress for poorer Americans.

Implications of the CPI revision must still be worked out for measurements of living standards. But they initially indicate that consumers, and thus workers in lower paying positions, are better off than previously thought.

Problems for Unskilled Workers

So far the evidence shows that there has been no general decline or even stagnation in the average American's purchasing power since 1973, and certainly not during the 1980s when trade deficits hit record highs. That does not mean there are no problems in some sectors of the economy or segments of the workforce. To round out the earnings picture, it is necessary to examine these problems and determine how trade contributed to the situation.

Critics of the Reagan administration in the 1980s argued that many of the 18 million net new jobs created during that decade were low-paying service positions, often labeled "burger flipper" jobs. Now it is clear that many of those jobs were higher paying and required higher levels of skill and education. So what is the actual situation for low-skilled positions?

The Education Factor

Earnings for those in the lowest income brackets have stagnated or declined, a fact apparently linked to levels of education. Cox and Fox found that the premium for better educated workers rose significantly between 1972 and 1992. In 1972 a high school dropout made only about 62 percent of the income of a high school graduate, whereas a worker with an advanced college degree made 172 percent of the income of the high school graduate. By 1992 the dropout was making only 58 percent of the high school graduate's income, whereas the worker with an advanced degree was making 254 percent. Income for all groups with some college education rose after 1972. Thus, there has been a growing premium for better educated and higher skilled workers over the past two decades.

Anecdotal evidence also suggests that the educational factor is responsible for poor earning power. A CEO reports that his factories now hire more college graduates to work the shop floor, whereas 20 years ago high school graduates were adequate. The change stemmed in part from a need for higher skilled workers. But it also reflected a problem with the quality of workers coming out of high school. That quality has declined in recent decades, with such basic characteristics as punctuality and the maturity to hold down a job lacking.

American Enterprise Institute scholar Jagdish Bhagwati and Vivek H. Dehejia seem to confirm the pattern. They observe that the adjustment difficulties of lower skilled workers and the resulting lower earnings could have been caused by the rise of lucrative alternatives such as drug-dealing, the fall in the quality of schools, and the collapse of the family and hence the fall in the motivation and aptitude for getting educated among those affected.

The Technology Factor

The role that education plays in explaining stagnant earnings for part of the workforce suggests that advances in technology also play a role. Bhagwati and Dehejia note that "the happy experience of the 1950s and 1960s may have been due to technical change that was substantial, was more uniformly spread among exportables and importables, and was more neutral than biased whereas, in the 1980s, it has probably been slower . . . has been more focused on skills-intensive exportables, and has been more skills-biased."

That is, when technological advances can be easily applied in manufacturing facilities employing semi-skilled workers, the differential effects on wages will be minimal. But the information revolution associated with personal computers, advanced software, and the Internet gives a clear advantage to workers with "knowledge." Indeed, knowledge itself is often the product being produced and sold. Furthermore, technological advances in recent decades have helped certain goods and services compete well with imports and indeed become major exports.

Bhagwati and Dehejia also suggest that the earnings gap is due in part to "the greater transferability of workplace-acquired skills by the skilled. An accountant handling IBM, for example, can shift his acquired know-how readily to a new job at Caterpillar or Chrysler, but working better on the assembly line for autos at Ford may not transfer to working at a blast furnace in Pittsburgh, or for that matter to flipping hamburgers at MacDonalds [sic]."

Transferability of certain high-tech skills would seem to be high. For example, someone who develops one kind of computer software would probably be able to retool easily to develop another kind.

The Trade Factor

Explanations other than trade seem best to explain wage differences. Does trade have any effect?

No doubt imports and international competition help determine which sectors and workers realize higher profits and incomes and which ones find their goods, services, and skills of less value. If the American market were flooded with a cheap, foreign-made product of reasonable quality, American firms would have to respond with changes. That is the point of competition, to allow for specialization.

Competing with technology. In a competitive situation such as that an industry could face several options. It might shut down its operation; it might move all or part of its operation overseas; it might specialize in the lines of products and services in which it remained competitive; or it might become more efficient, for example, through the introduction of new, less labor intensive technologies.

For example, during the recession in the early 1980s, the textile and apparel industries in America laid off hundreds of thousands of workers. That downsizing was no doubt due in part to foreign competition. Since the late 1950s, however, U.S. trade barriers had restricted textile and apparel imports. Those barriers had delayed the industry's adjustment to competition and the introduction of new technology.

By the mid- and late 1980s, those industries, especially the textile segment, recovered and became very profitable. For example, in 1987 textile mills ran at about 94 percent of capacity, compared with 81 percent for American manufacturing as a whole. Total textile and apparel production rose by 6.5 percent between 1985 and 1986, with record consumption of inputs such as cotton, wool, and other fabrics. Profits rose by 46 percent in 1986. Yet in 1985 the U.S. merchandise trade deficit was $122.2 billion and in 1986 it was $145 billion.

One secret to the success of American textile manufacturers was their use of better technology, much of it imported. Anecdotal evidence also suggests another surprising factor. While touring several textile factories in North Carolina during the period of expansion, this author was told that the mills were experiencing problems with a lack of manpower. Apparently the demand for workers in newer, high-tech industries in the region forced some of the textile mills to increase wages to attract workers.

Competing with low-wage foreigners. Some critics of free trade might concede that higher skilled American workers have little to fear from foreign competition. But they would still maintain that lower skilled American workers will always find their wages falling as low-skilled jobs move overseas, raising wages for foreigners.

Burtless notes several objections to the assumptions behind that scenario. First, it is not clear whether technologies that make use of low-skilled labor are identical in developed and less developed countries. Thus, lower-skilled labor might be more useful in certain industries or aspects of production in developed countries.

Second, returns to scales of production in developed and less developed countries will not necessarily be constant. And finally, the wage-convergence scenario often assumes that trading partners will not specialize in separate export items. In fact, complete specialization does occur.

Low wages with no import competition. Imports may indeed be the principal cause of stagnant or declining wages in import-sensitive industries, so these industries might understandably turn to less labor intensive, higher technologies to remain competitive. But, as Burtless points out, one would not expect that reaction in industries not subject to competition from low-priced imports. In fact, one might expect the opposite. With abundant unemployed workers, industries not subject to import competition might well become more labor intensive.

Burtless examined industries that were highly affected, moderately affected, and not affected at all by trade. Within the highly affected industries he found relatively small wage disparities between higher and lower earners and thus, presumably, between higher and lower skilled workers. In industries not highly affected by trade he found greater disparities.

The trend toward divergence between high and low earners in industries least affected by trade and in those most affected by trade has been the same since 1969. Low-wage, and presumably lower skilled, workers in the heavily trade-affected industries saw their wages slip in comparison with higher skilled workers in those industries. But low-wage workers in the least trade-affected industries saw their wages slip as well in comparison with their higher skilled colleagues in the same industries. In other words, import pressures are not the primary cause of wage stagnation in certain segments of the workforce.

The Government Factor

Finally, the greatest change in the American political and economic regime over the past three decades has been the explosive growth in the size of the federal government, with massive redistribution of wealth, regulation of the economy, and welfare state policies. These, more than anything else, have robbed businesses of capital and flexibility, robbed workers and families of income, and created incentives that punish the productive.

The United States has not gone as far as Western Europe in socializing its economy and undermining private property and contract rights. Thus America has a better employment situation than does Europe. But the kind of government policies that have so harmed Europe have harmed America as well, with lower skilled workers as civilian casualties.

A recent example of the effect of the government factor is found in America's business productivity. Between 1982 and 1990 productivity rose by 17.4 percent. But since 1992 it has grown by only 0.3 percent. The effects of the record tax hikes in 1990 (which took effect in 1991) and in 1993 (which took effect in part retroactively) must be prime suspects for productivity's battered condition.

Productivity is crucial to increasing real purchasing power. Put simply, if Americans do not continue to increase the value of goods and services, they cannot increase their earnings. The problem lies not in imports. It lies in federal tax and regulatory policies.

---

References

Bhagwati, Jagdish, and Dehejia, Vivek H. "Free Trade and Wages of the Unskilled -- Is Marx Striking Again?" In Bhagwati and Marvin H. Kosters, eds. Trade and Wages: Leveling Wages Down? Washington, D.C.: AEI Press, 1994.

Burtless, Gary. "Trends in the Level and Distri-bution of U.S. Living Standards, 1973-1993." Conference paper delivered at the ASSA meeting. San Francisco, January 1996.

Burtless, Gary. "Widening U.S. Income Inequality and the Growth in World Trade." Prepared for the Tokyo Club meeting. Dresden, Germany, 15 October 1995.

Cox, W. Michael, and Fox, Beverly J. "What's Happening to America's Income?" Policy Backgrounder, no. 138. Dallas: National Center for Policy Analysis, 1996.

Hudgins, Edward L., "The West at the Crossroads," in The 1994 Joint Economic Report, Joint Economic Committee, U.S. Congress, May 9, 1994.

"Service Sector Productivity." Washington: McKinsey Global Institute, 1992.

Yager, Loren. "Price Comparisons Between the Japanese and the U.S. Markets." Santa Monica: RAND Corporation Center for U.S.-Japanese Relations, 1991.

Freedom to Trade: Refuting the New Protectionism is Copyright 1997 by the Cato Institute. All rights reserved.

---

| Home | Table of Contents | Publications | Cato Institute |