Trends and Challenges for Work in the 21st Century

Opinions and views in these papers are those expressed by the author(s). They are not to be taken as expressions of support for particular positions by the Department of Labor. Please do not cite these papers without prior permission of the author(s).

AN OVERVIEW OF ECONOMIC, SOCIAL,

AND DEMOGRAPHIC TRENDS

AFFECTING THE US LABOR MARKET

Robert I. Lerman

Stefanie R. Schmidt

The Urban Institute

Washington, D.C.

The expansion of world trade, communication, immigration, capital flows, and multinational business activity has generated a great deal of political controversy in recent years. The campaigns of Ross Perot and of Patrick Buchanan provided the most-publicized outcries against globalization of trade and investment flows, and opposition to trade agreements. But similar concerns are voiced in writings by political commentators (Kuttner, 1997; Wolman and Colamosca, 1997; Greider, 1997) and by some economists (e.g. Freeman, 1995; Wood, 1995). The title of a 1995 article by Richard Freeman put the matter bluntly: “Are Your Wages Set in Beijing?”

During the early postwar period, at least through the late 1960s, a broad consensus in the U.S. favored liberal international trade policies. The U.S. took a leading role in promoting the opening of world markets, in sharp contrast to its protectionist periods in the 19th century and the 1930s (DeLong, 1998). In the financial arena, the U.S. decision to allow the value of the dollar to be determined in the free market played a role in the globalization of currency markets and ultimately other financial markets. By the 1990s, however, in the aftermath of post-1973 weak productivity growth, slow wage growth, and increasing inequality, the optimistic picture of globalization has been challenged. While economic theory continues to predict aggregate gains from trade, some see the net gains as too limited to offset large losses experienced by less-advantaged citizens, many of whom are said to lose from globalization.

The globalization debate raises several important questions about future workforce trends and public policy responses:

1. What is the impact of globalization on productivity and on the real wages and real incomes of workers as a whole?
2. How does globalization affect the demand for various groups of workers as well as the distribution of real wages and incomes?
3. To what extent does globalization limit the ability of governments to take constructive actions on behalf of the work force and disadvantaged groups?
4. What policies can permit the country to take advantage of the gains from globalization while minimizing its costs?

The literature on each of these questions is far too extensive to review in this paper. However, we provide some basic facts and draw on a range of analyses to clarify the various positions on these questions. The first step is to clarify the meaning of globalization. In addition to international trade, globalization embodies foreign direct investment, international financial flows, international migration, and cultural interactions.

Globalization: Trends and Patterns

Trade is clearly expanding throughout the world and strikingly so in the U.S. World merchandise exports increased over the postwar period, from 7 percent of world GDP in 1950 to 17.1 percent in 1993 (Krugman, 1995). Dramatic reductions in transportation and communication costs have certainly played a major role (Cooper, 1995; Council of Economic Advisers, 1997). Ocean freight per ton declined from $95 in 1920 to $29 in 1990 (1990 dollars); air transport per passenger mile fell from 68 cents in 1930 to 11 cents in 1990. By 1993, 29 percent of the value added of U.S. exports traveled by air. Policy developments leading to the opening of formerly closed markets have also played a major role in the expansion of postwar trade (Krugman, 1995).

For the U.S. we can see in Figure 4 the trend in the average of exports and imports as a proportion of the Gross Domestic Product. Note the two large jumps in the trade shares. Between the first oil shock in the early 1970s and the oil shock in 1979-1980, the trade share rose from under 6 percent in 1973 to over 10 percent in 1980. When oil prices fell in the mid-1980s, the trade share fell back somewhat to about 8.5 percent before returning to the 10 percent level in the early 1990s and reaching 12 percent by the late 1990s.

Trade takes place mostly between industrial countries, but the U.S. share of imports from less-developed countries has increased. In 1972, U.S. exports of manufactured goods to less-developed, non-OPEC countries were 1.2 percent of GDP, while imports to the U.S. of manufactured goods from these countries stood at about 0.9 percent of GDP. By 1993, imports from these less-developed countries tripled to 2.7 percent of GDP while US exports to these countries doubled to 2.4 percent of GDP (Lawrence, 1996). By 1995, the percentage of manufacturing imports coming from developing countries had risen to 33.6 percent, up from

18 percent in 1973 (Collins, 1998). These imports were about 16 percent of value added in all of U.S. manufacturing. Exports to less-developed countries also increased substantially. The non-oil-developing countries took 39 percent of U.S. exports in 1995, well above the 24 percent figure for 1970. Often left out of these calculations is the fact that wages have been rising, even in countries considered less-developed. As Collins (1998) points out, the average wage paid to manufacturing workers in countries trading with the U.S. increased from 38 percent of U.S. wages in 1960 to 85 percent in 1992. This striking evidence demonstrates that the U.S. is not trading more with countries paying relatively low wages.

Expanding trade is only one dimension of globalization. Another major indicator is foreign investment, including direct investment by U.S. companies abroad and by foreign companies in the U.S. as well as financial investments by foreign citizens or companies in domestic stock, bond, or money markets. When foreign companies invest in the U.S. and U.S. companies invest abroad, economies generally become more integrated with the rest of the world.

The overwhelming amount of business investment and financial investment takes place between industrial countries. Nearly $80 billion of the $90 billion invested in the U.S. from abroad in 1997 came from Europe, Canada, and Japan. About $72 billion of the $114 billion 1997 foreign investment by U.S. companies went to Europe, Canada, and Japan. In recent years, direct investment in less-developed countries has expanded significantly, doubling between 1990 and 1995, while total U.S. direct investment abroad rose by 65 percent. The trend continued through 1997, as foreign direct investment in Latin America rose significantly.

Note that the difference in direct foreign investment amounted to about $24 billion, or less than 2 percent of gross investment in the U.S. Further, the payments from abroad to U.S. investors virtually match the payments to foreigners for their assets. In 1997, the U.S. paid about $257 billion to all foreign factors of production (mostly returns on capital), while U.S. residents and companies received almost the same amount, about $245 billion. On this measure, the rise of globalization has largely yielded outflows and inflows of similar magnitudes.

The similarity in returns masks much larger differences in the market value of assets. As of 1996, foreigners held about $5.1 trillion in assets, while the public and private sectors of the U.S. owned about $4.3 trillion. Since foreign-held assets in the U.S. rose more rapidly than U.S.-owned assets in other countries, foreigners were actually investing more in the U.S. than U.S. individuals and companies were investing abroad.

Another dimension of globalization is the growth of the international capital market. Financial flows across countries now dwarf the levels of only 10 years ago. Specialized markets, companies, and individuals have created the means by which organizations can manage risks (particularly foreign currency and interest rate risks) and make speculative investments. The international financial markets expanded in part with the growth in foreign trade and investment. But some are concerned that international financial markets take too much power away from national policymakers by constraining their ability to take various kinds of macroeconomic decisions and also to tax, to raise government spending, and to set labor standards.

Immigration to the U.S. has been expanding along with the growth of trade and financial flows. The share of net annual U.S. population growth accounted for by immigrants has reached about one-third, which equals its previous historical peak. About 1.1 million legal and illegal immigrants enter the U.S. each year and are concentrated in specific regions of the country. Moreover, immigrants are increasingly likely to come from less-developed countries. Between the 1950s and the early 1990s, the share of immigrants coming from Europe and Canada fell from about 67 percent to 21 percent, while the proportion from Asian and Latin American less-developed countries rose from 30 percent to 75 percent. Since today’s immigrants are somewhat less educated than native-born Americans, many worry about competition between immigrant workers and less-skilled native-born workers.

Selected Evidence About Globalization’s Impact on the U.S. Labor Market Effects on Overall Incomes and Wages

Economists generally favor free trade as a method for raising real incomes. When individuals, firms, and/or countries specialize in production for which they have a comparative advantage, the benefits accrue to all of the trading partners. In one sense, free trade and investment flows at the international level are a broadening of competition at the individual country level. Extending competition to the international sphere broadens the scope for increasing allocative efficiency. It could even raise efficiency in a dynamic setting, since broader competition may stimulate innovation. Reduced barriers to trade can broaden export markets and thereby allow U.S. companies to reap larger returns on their innovations (Lawrence and Litan, 1998), can raise productivity by increasing competitive pressure on firms and by allowing U.S. firms to draw on the capital equipment and knowledge produced in other countries, can widen the variety of goods and services available to consumers, and can lower world prices. Lawrence and Litan cite World Bank estimates that consumers will gain between $100 and $200 billion per year simply as a result of one international trade agreement. Further substantial gains are expected to flow from the expanded international competition in telecommunications.

Nevertheless, some groups may lose from trade and may receive little or no compensation from the winners. The losses occur as some firms face lower prices and some workers must accept lower wages in the face of added competition. Were factors of production entirely mobile, firms and workers experiencing losses could shift away from their existing sectors toward more financially rewarding ones. But, given immobile capital and specific skills, changes in trade flows can impose capital losses. (Of course, other external forces—including changes in weather, technology, or government regulations—can create similar losses.) In addition, trade may add more competitive pressure to one factor of production (say, unskilled labor) than to another factor (say, skilled labor).

The theoretical impact of immigration is less clear. If immigrants simply expanded all segments of the nation’s labor force and capital stock in the same proportions, rising immigration need not lower the incomes of non-immigrants (Card, 1996). In fact, given economies of scale, immigration could actually raise incomes of non-immigrants. On the other hand, if the immigrant inflow was concentrated among low-skill workers and was not matched by added capital, then immigrants could gain at the expense of native-born workers. Immigration directly increases the supply of labor and thus could lower average wages as well as affect the distribution of wages. Given that immigration is concentrated in a small number of places within the U.S. any negative effects of immigration on wages should take place primarily in these locations. Yet, according to David Card (1996), effects on wages are difficult to observe. Most studies of the impact of immigration focus on its role in increasing earnings inequality, not on its effect on average incomes. We take up these issues in the next section.

The impact of foreign investment flows on workers is uncertain as well. In principle, the free flow of capital should raise the productivity of investments and certainly benefit owners of capital. One might think workers would lose if fellow citizens who own capital can invest their funds outside the country. If investments abroad substitute for investments in the home country, workers will work with less capital and find themselves producing less and earning less. But foreign investments might not substitute for local investments for two reasons: first, because a low expected return locally might cause capital owners not to make direct investments at all and second, because foreign investments may complement local investments, as when a company becomes more efficient by building low value-added parts in a less-developed country and high value-added components in its home country. Moreover, capital flows into as well as out of the country. If outflows ultimately lower the wages of workers, inflows may raise wages.

Several questions immediately arise from this textbook formulation. First, how large are the gains from globalization? Second, how are the benefits from globalization distributed? In particular, do some workers lose in absolute terms as well as relative to others in their societies?

Perhaps because of the complexity of the problem, few have attempted to estimate the overall income gains (or losses) from globalization. Even the effects of trade alone on national income are unclear. DeLong (1998) presents a figure of .3 to .7 percent of national income as the short-run costs to national income from protectionism over the 1800-1940 period in the U.S. However, he also points out that the long-run costs may have been much greater, since tariffs raised the prices of imported capital goods and led to a lower-investment, lower-wage economy. One respondent to the DeLong paper (Eichengreen, 1998) agreed that protectionism did not assist U.S. growth, while another (Temin, 1998) argued that early tariffs helped stimulate U.S. manufacturing and ultimately increased economic growth. Another ongoing debate is over the value of trade barriers in stimulating growth in Asian economies, particularly Japan and South Korea. While economists cannot agree on the impact of trade barriers on national incomes in the past, most favor open trade regimes, even those who strongly favor policies to compensate those negatively affected by trade.

If effects on overall incomes are difficult to estimate, what about the impact on average real wages? Lawrence (1996) is content to argue that the expansion of trade could not have done much to reduce average wages in the U.S. His focus is on manufacturing, since it is this sector where the growth in imports from less-developed countries has taken place. After accounting for the wage premium paid in manufacturing, for the increase in manufacturing imports, and for the possibility that the threat of imports forced unionized manufacturing workers to accept lower rates of wage growth, Lawrence concludes that expanded trade could only have lowered wages by far less than one percent (0.1 to 0.3 percent) between 1978 and 1989.

A recent report by Scott, Lee, and Schmitt (1997) projects larger negative impacts on wages of U.S. workers. They attribute U.S. job losses of about 2.4 million in 1994 to the nearly $100 billion increase in the trade deficit in goods and services that took place between 1979 and 1994. As the authors themselves state, it may be inappropriate to attribute the job loss figure to trade instead of to broader macroeconomic factors. They still contend that trade has a negative impact on U.S. workers because of the induced shift in the composition of jobs. Their empirical results are far from conclusive in indicating a negative trade effect on average wages, since they suggest that imports tend to reduce jobs requiring less education than the economy-wide average while exports tend to increase jobs among highly educated workers.

An alternative perspective emphasizes the potential positive impact of openness on productivity and, ultimately, wage growth. A set of case studies conducted by McKinsey Global Institute (Lewis et al. 1992; Bailey and Gersbach, 1995) estimates how productivity differs between the U.S. Germany, and Japan in a range of specific manufacturing and service industries and explores the reasons for cross-country differences. The findings for manufacturing suggest that the more exposed a country’s industry is to global competition, the higher the productivity. Companies forced to compete with the best internal and external producers must achieve high productivity and quality or risk losing sales, profits, and even staying in business. The McKinsey studies of selected service industries yielded similarly positive effects on productivity from the competitiveness of the environment. The authors concluded that managers in industries facing an intensely competitive environment are forced to concentrate on their company’s economic well-being, an emphasis that ultimately leads to higher productivity. The McKinsey case studies cast doubt on the idea (Piore, 1998) that companies succeed internationally by pursuing a low road of low wages, while accepting low productivity. However, since the McKinsey analyses deal only with selected industries, they may miss the experience of industries following the low-road model.

One topic that has attracted little or no attention in the literature is the impact of trade on the economy’s ability to reach low unemployment rates with minimal inflation. Certainly, pushing unemployment rates to 4.5 percent and sustaining the current expansion are important for U.S. workers, especially the less advantaged groups. Trade flows and the threat of foreign competition may well serve as a brake on price increases that otherwise might emerge in a more closed economy. On the other hand, open economies may be more unstable, both because of the larger potential impact of exchange rate shocks, supply shocks, and demand shocks. At this point, macroeconomists are still trying to understand why the U.S. economy has managed to maintain low unemployment rates without generating inflation. If a more open economy is one of the reasons, then policymakers will rightly be wary of actions that restrict trade.

Foreign investment is another element of globalization that could affect average U.S. wages. Some writers cite foreign investments or the threat of foreign investments by U.S. companies as weakening the power of workers and ultimately lowering their wages. The problem extends to the well-educated, according to Wolman and Colamosca (1997), who point to foreign investments by Citibank and Hewlett-Packard in Asia as “only the beginning of the trend toward corporate ‘outsourcing’…of highly skilled labor.” In 1995-1996, affiliates of non-bank U.S. companies operating abroad (BEA, 1998) employed about seven million workers, or about 5 percent of the U.S. workforce. However, foreign affiliates operating in the U.S. employed about 5 million workers in the U.S. Thus, taking account of all direct foreign investment, there are about 2 million jobs fleeing abroad not offset by jobs coming into the country. This figure constitutes about 1.5 percent of the U.S. workforce. Moreover, these data do not take account of the higher investment in the U.S. resulting from the fact that the U.S. attracts more inflows of capital than U.S. companies and individuals spend on foreign investments. One study (Wade, 1996) suggests that even multinational enterprises generally move only their most routinized operations abroad and that such changes as “just-in-time” inventory management and increased specialization weaken the incentives to disperse production globally and encourage locations near final markets.

Effects on the Distribution of Wages and Incomes

The primary concern about globalization is its impact on the distribution of wages and incomes in developed countries, including the U.S. One indicator of this concern is the vast literature on trade’s distributional effects that emerged in the 1990s. Moreover, political movements opposing expanded trade emphasize trade’s supposedly devastating effects on less-skilled or middle-skilled workers.

The underlying reason globalization is said to harm low-skilled U.S. workers is that the rest of the world has abundant low-skilled workers who are paid a small fraction of the wages paid to comparable U.S. workers. Once foreign less-skilled workers are allowed to compete with less-skilled U.S. workers, the wages of the two groups will converge, lowering U.S. wages toward those paid in other countries. The added competition could come via trade (where less-skilled work becomes embodied in goods) or immigration (where less-skilled workers directly raise U.S. labor supplies).

Other explanations based on economic theory account for all changes in price, production, and consumption, but use a variety of simplifying assumptions about technologies and mobility (Cline, 1997). According to a commonly cited theory (Stolper-Samuelson), increased trade in goods produced with relatively high proportions of a particular factor (say, unskilled labor) will cause a reduction in the price of that factor and an increase in its use. Applied to the debate over trade to the U.S. this theory offers a mechanism by which the expanded trade from less-developed countries could lower the wages of the less-skilled and increase wage inequality.

Before discussing some of the empirical work aimed at isolating the impact of globalization on wage inequality, it is useful to distinguish among measures of wage inequality. Trade could affect the price (wage rate) and/or quantity (number of workers) of each category of labor. Wage rates alone may not capture a reduced demand for less-skilled U.S. workers, since minimum wages or other wage rigidities could prevent wages from falling and lead to large employment reductions. Thus, for example, European trade with less-developed countries could be affecting European labor markets via increased unemployment of the less-skilled but still not generate increased wage rate differences between skilled and less-skilled workers.

Most studies examine the effect of trade on the wage rate differentials between two categories of workers. Some authors use the wage differential by education, such as the percentage wage gap between workers with a BA degree and workers with only a high school diploma or between the college-educated and non-college workers (Baldwin and Cain, 1994). Others focus on the wage differential between production and non-production workers (Bernard and Jensen, 1994; Leamer, 1998; and Sachs and Shatz, 1998). The choice matters because the wage gap by educational category rose much more dramatically than the wage gap between production and non-production workers. More importantly, analyses of fixed categories of labor may yield misleading results even about group wage differentials because they ignore changes in the composition of the categories.

A good example is the wage gap by education. The median earnings of college graduates grew from 1.3 to 1.7 times the earnings of high school graduates between 1979 and 1995. But, because of substantial increases in educational attainment, the groups did not represent the same proportions of the labor force. In 1979, the average high school graduate with no college ranked at the 40th percentile of the educational distribution of the U.S. labor force; by 1993, the ranking of the average high school graduate had dropped to the 28th percentile. The decline in average ranking of college graduates was much smaller, declining from the 90th percentile in 1979 to the 87th in 1995. If the quality of the worker depends partly on native ability that is correlated with educational attainment, then some of the widening gap in wage differentials by educational category may simply be due to the changing average abilities of the two groups. A higher wage gap between 87th and 28th percentiles in the ability distribution than between the 90th and 40th percentiles is hardly surprising nor an indicator of rising inequality. Similar compositional changes may well explain some, if not all, of the increased differentials between non-production and production worker wages.

The compositional problem takes on added force since studies of trade impacts on age differentials between categories generally ignore distributional changes within categories. Another problem is that, when examining the effects of trade on the prices of particular industries, analysts measure the skill intensity of industries by fixed categories, by education or production-nonproduction workers.

While measuring impacts on wage differentials between groups simplifies the analysis, the approach does not directly address trends in wage inequality. a broad concept that incorporates all workers. Nevertheless, authors of some trade studies label their findings as examining impacts on wage inequality when they are actually measuring trends in wage differentials between fixed categories of workers. The distinction is important because it is clearly possible for wage differentials by education to widen at the same time that overall wage inequality does not change at all (Lerman, 1997a). Further, one recent analysis (Lerman, 1997b) found that the inequality of wage rates across all hours worked in the U.S. economy was essentially constant between the mid-1980s and the mid-1990s, the period when expanded trade flows was said to have increased inequality significantly.

Despite these and other limitations, a large number of studies have yielded estimates of trade’s impact on wage differentials. The studies are far too numerous to review in this paper. Yet even a cursory review of the findings reveals a wide divergence across studies in the proportion of the increased differential attributed to the growth in trade. Examples of studies concluding that trade exerted at most a minor (zero to 10-20 percent) impact on wage differentials include Lawrence (1996), Baldwin and Cain (1997), and Krugman (1995), while Wood (1994, 1995), Leamer (1998), and Sachs and Shatz (1998) argue for substantial (over 20 percent) impacts from trade alone. This strikingly wide variation comes from differences in methodology, in time periods, and in measures of wage differentials. One methodology involves measuring the factor content of imports and exports to determine the impact of trade on the implicit supply of less-skilled workers. An alternative is to examine the impact of trade on prices in trade-sensitive industries, especially those trade-sensitive industries that employ large numbers of less-skilled workers. Price reductions in these sectors would be indicative of trade’s impact on wage differentials.

The work by Cline (1997) takes a distinctive tack by examining trade’s impact on both the net changes in wage differentials and the gross changes in wage differentials. He points out that the net impact on wage differentials was the sum of several gross disequalizing components minus such equalizing components as the rising supply of skilled workers throughout the world. Thus, the factors contributing to rising differentials together add up to more than 100 percent of the net changes. On the basis of Cline’s (1997) elaborate simulation analysis, trade effects represent a high share (33 percent) of the net changes in wage differentials, but less than 10 percent of the gross changes. He clearly regards the smaller figure as the most appropriate.

Cline’s simulation yields projections of the impact of trade expansions on the absolute as well as the relative position of the unskilled. Freezing existing trade protection levels at their 1993 levels and ruling out additional free trade agreements would do little to boost wages of the unskilled. Relative wages of the unskilled would fall substantially if trade encompassed the entire U.S. economy, including the sectors now producing nontraded goods and services. However, unskilled workers would still gain in absolute terms because of the efficiency gains linked to trade.

Only a few studies provide a combined estimate of the effects of trade and migration on wage differentials. The effects of migration have proved surprisingly difficult to identify. As Card (1996) notes, while one would expect migration to exert its largest impact on areas in which migrants concentrate, studies of even sudden inflows of foreign immigrants reveal little or no impact. Nevertheless, in a recent study, Borjas, Freeman, and Katz (1997) find that the high inflow of immigrants with no high school education accounted for up to half of the increase in wage differentials between high school graduates and dropouts. At the same time, these authors find that the combined impact of immigration and trade from less-developed countries explains no more than 10 percent of the widening wage gap between college graduates and high school graduates.

Overall, the evidence for large impacts of globalization on wage differentials and wage inequality is limited. Susan Collins, editor of the most recent collection of articles on the subject, concludes that the “available evidence suggests that globalization (including trade and immigration) may explain 1 to 2 percentage points of the 18 percentage point overall change in the wages for high school-educated workers relative to those who are college educated.” However, Collins points out that analyses to date have not taken account of several indirect effects of globalization and that future work may reveal a considerably larger role for the integration of the world economy.

Whatever the effect of trade and immigration on skilled-unskilled wage differentials, these two forces exerted a smaller impact on family income inequality. One reason is that since the mid-1980s overall inequality increased much less than wage differentials by education and skill. In fact, according to some comprehensive measures of inequality of wage rates across all hours worked, wage rate inequality has not risen since the mid-1980s (Lerman, 1997b). Although the role of changes within the top 3 to 5 percent of the wage distribution is in dispute (Lerman, 1997b; Bernstein and Mishel, 1997), wage inequality across the bottom 95 percent of hours worked was flat. Moreover, inequality of annual earned income among 25- to 54-year-olds actually declined between 1986 and 1995.

A second reason is that family income inequality depends not only on wage inequality but also on the number of earners per family, the composition of families, and the correlation between earnings of various family members (Burtless, 1998b). According to Burtless’s calculations, only about one-third of the rise in family income inequality arises from increases in wage inequality. Thus, even if trade and immigration accounted for as much as one-third of the growth in wage inequality (a high estimate in the literature), trade would have contributed less than 10 percent of the rise in family income inequality.

One apparent omission from this literature is the effect of trade on the relative prices paid by skilled and unskilled workers and by low- and high-income households. Much of the concern about wage inequality effects arises from surging imports from less-developed countries. Since the goods produced by these countries are generally at the low end of the quality spectrum and include essentials (especially clothing), it may well be that the benefit of lower prices for these commodities due to trade are disproportionately large among low-income individuals. If so, analyses of trade would have to take account of this relative consumption price effect as well as the relative wage effect. Of course, a full analysis of this issue might not yield any differences across groups in the direct gains from price changes.

Globalization’s Impact on Progressive Government Policies

By limiting the autonomy of national governments to conduct economic policy, globalization could either harm or help workers. According to Robert Kuttner (1997), worries about the reaction by global capital markets force governments to adopt contractionary economic policies. Competition to retain and expand business investment might discourage governments from imposing corporate taxes. Spending and tax limitations might force a cutback in social welfare spending. And global market pressures could pressure governments to deregulate labor and product markets.

Judging the impact of globalization on macroeconomic policy and macroeconomic outcomes is difficult. Integration into the world economy or closer integration into regional alliances (as in the European Union) appears to discourage countries from running large fiscal deficits and permitting high rates of inflation. But in some cases, as in Japan today, it is pressures from abroad that are encouraging policymakers to pursue expansionary policies. Even when the short-run effect of globalization causes governments to emphasize contractionary macroeconomic policies, the long-run impacts are certainly as likely to yield positive as negative outcomes. Tight fiscal policies, by drawing investments, can lower interest rates, thereby promoting economic activity.

In the case of the U.S. the key question is whether price competition resulting from open trade improves the economy’s ability to sustain very low unemployment rates without inducing additional inflation. To the extent that trade exerts this impact, the gains are likely to be largest among the less-skilled, since their employment is most responsive to reductions in the aggregate unemployment rate. In the recent expansion, from 1992 to the first half of 1998, for example, the employment-population ratio of high school dropouts rose by over 10 percent while the employment-population ratio of college graduates remained constant.

Economic integration need not weaken national authorities’ ability to set high labor standards for two important reasons. First, such policies as mandated benefits may not raise the costs of production to the extent that the payroll taxes are borne by the workers in the form of lower pay. Second, the gains in higher productivity from labor standards may outweigh any impact on costs.

So far, evidence is scanty that globalization is generating irresistible pressures toward the convergence of policy regimes, according to Eddy Lee (1997) of the International Labor Organization. Lee cites the wide and continuing differences in labor market institutions in Japan, Germany, and the U.S. as indicative of the ability of countries to pursue different national approaches while remaining integrated in the world economy.

Financial destabilization resulting from globalization is another worry. There is little doubt that some financial crises have been exacerbated by the scale of foreign liquid investments. Certainly, the current run from Asian currencies has complicated Asia’s ability to recover from the recent shocks. However, when countries experience a financial crisis, a non-market dimension of globalization—institutions such as the International Monetary Fund and the World Bank—can provide liquidity and credibility that prevent even greater outflows of currency. On the other hand, the measures proposed by the IMF for some of the Asian countries may be overly restrictive and contractionary.

Ironically, globalization effects on financial stability might well lead to even larger inflows of foreign capital into the U.S. Thus, while the U.S. might lose sales because of the weaknesses of foreign economies, the capital inflow might raise investment in the U.S.

Overall, globalization’s effects on economic policy have been more a matter for speculation than for rigorous research. Only a modest amount of research is available on the questions of 1) the impact of globalization on policy; and 2) the effects of any globalization-induced policies on country outcomes.

Policy Implications of Globalization Trends

Surprisingly, disagreements over the impact of globalization are more common among economists than are differences over policy. Even those who estimate negative effects of trade on unskilled U.S. workers continue to favor free trade and open markets. Economists who believe trades negative impact is minimal nevertheless favor training and other policies to help low-skill workers.

Policy responses generally fall into three main categories: 1) support for additional training; 2) wage subsidies for low-wage workers; and 3) special assistance for workers displaced because of increased imports. Other proposals include raising the minimum wage, encouraging unionization, and requiring labor standards in other countries.

If the impact of trade is to reduce the demand for less-skilled workers relative to high-skilled workers, the natural response is to train less-skilled workers to attain higher skills. By reducing the supply of less-skilled workers and increasing the supply of skilled workers, improved education and training increases average wages while narrowing the gap between low-skilled and skilled workers. Trade-induced declines in the relative demand for the less-skilled could be fully offset by training-induced reductions in the relative supply of the less-skilled. If so, we would not expect any change in relative wages.

The nation’s capacity for training and retraining in response to trade would be enhanced with improved systems for identifying sectoral skill demands, successful training models, and certification for occupational skills. In many fields, limited information about training and certification may slow the adjustment to changing demands. Incentives for continuing and even lifelong learning could also smooth the adjustments to trade-induced shifts in demand. Were U.S. workers engaging in training on an ongoing basis, they could react more easily to all sources of restructuring, including not only trade but also technological change. Finally, improving the nation’s school-to-career systems, especially with youth apprenticeship, would provide another vehicle for matching skills demanded by employers with the mix of training obtained by students and workers. Employers are likely to offer places only in apprenticeship fields for which they anticipate strong demand. Instead of having the mix of training places determined by external programs, apprenticeship and similar approaches directly transmit employer demands to potential trainees.

Since expanded education and training may not spur enough of a shift out of the unskilled categories, many favor helping low-wage workers by providing wage subsidies. One such policy is the 1993 expansion of the Earned Income Tax Credit, implemented in 1996, which provides a 40 percent subsidy to earnings to families with two or more children on earnings up to about $9,000.

Some policies to compensate the workers displaced by trade have been enacted, but their effect has been small. Trade Adjustment Assistance was criticized for serving primarily to extend unemployment benefits and not retraining workers. The special NAFTA program requires recipients to take up training, but some are wary of accepting benefits because workers obtaining benefits cannot take an alternative job.

One approach, suggested by Lawrence and Litan (1997), is to complement training programs with some explicit compensation for the loss of a trade-related job. To compensate workers for their actual losses—as indicated by the difference between what workers earn in their new jobs and what they earned previously—they propose wage insurance. Under their scheme, dislocated workers who were in their job for some minimum period (say, two years) would qualify for half the loss of earnings they may experience after gaining a new job. Compensation would last for a limited period of two to three years.

This plan raises a number of questions. Would the subsidy extend to workers with high levels of earnings, say $60,000 before displacement and $40,000 after displacement? If so and if the percentage of earnings reductions in postdisplacement jobs does not vary with the initial wage level, then the size of the subsidy would increase, the higher was the worker’s former earnings. Perhaps more troubling is the recognition that low-wage workers indirectly affected by trade might obtain little or no benefits while middle- and upper-wage workers gain substantial amounts.

Immigration policy is another tool for affecting the relative scarcity of unskilled workers. Currently, migrants average only slightly less education than native-born workers. However, they are concentrated at the bottom and the top of the educational distribution. Estimates of their impact on wage differentials suggest that immigrant inflows are much more important than trade in lowering the relative wages of high school dropouts. If these estimates are accurate, then lowering inflows of low-skilled immigrants might improve the relative market position of residents who did not complete high school.

Globalization is only one factor influencing the low-skilled labor market. In order to examine policies for assisting low-skill workers, we must take a close look at the market for their services.