Working under different rules.

Freeman, Richard B.

For the past four years, many members of the NBER's Program in Labor Studies have been examining how labor markets and income maintenance systems work in the major developed countries: the United States and its trading partners and competitors in the world economy. The "Working Under Different Rules" project, funded largely by a grant from the Ford Foundation, has focused on: the determination of wages and the inequality of wages in different countries (directed by Lawrence F. Katz and me); training of workers (directed by Lisa M. Lynch); income maintenance programs (directed by Rebecca M. Blank); works council modes of worker representation (directed by Joel Rogers and Wolfgang Streeck); and extreme poverty (directed by David E. Bloom). An additonal related project, culminating in the book described later in this NBER Reporter, contrasted labor markets in Canada and the United States (directed by David Card and me).

On May 7, 1993, the project's leaders presented summaries of the work of their research teams at a conference in Washington, DC. These summaries will be published in a volume titled Working Under Different Rules. Also, the research papers written for each of the projects will be published by the University of Chicago Press. My intention in this report is to provide just a brief overview of the entire project.

WHY LOOK AT FOREIGN LABOR MARKETS?

We conceived this project in response to the difficult time that many American workers have had in the past two decades. Real wages have fallen for the less-educated worker. Inequality in earnings and employment opportunities among workers with different characteristics has increased. Unionism in the private sector has declined. And, poverty has increased for large segments of the population, although not among the elderly.

Of course, the 1980s were difficult for workers in much of Europe and in Canada as well. In those countries, unemployment went from below to above U.S. levels. Perhaps even more important, unlike in the United States, unemployed people remained jobless (albeit with relatively generous benefits that partially induced the longer unemployment spells) for several years. For instance, 6 percent of the unemployed in the United States in 1991 were out of work for a year or more, compared to 37 percent in France, and 51 percent in Spain. The contrasting unemployment experiences of the United States and Europe in the 1980s generated widespread discussion of the American jobs miracle, and of the virtues of "flexibility American style." Some observers thought that the United States had all the "answers" to the economic problems of the 1980s, and that Europe had much to learn from us, while we had nothing to learn from them.

The basic premise of our project was more measured: that while the United States had some positive outcomes in the labor market in the 1980s, it also had some negative ones. Thus, perhaps there was something Americans could learn from the labor market experiences and the social programs of Europe, Japan, and Canada.

FINDINGS

Wage Inequality

Is the U.S. pattern of rising wage differentials and wage inequality among workers with different levels of education a universal development in advanced capitalist economies? Or, have some countries not experienced huge increases in wage inequality? The evidence collected for the 1980s shows that increases in wage inequality were most substantial in the United States and Great Britain. Because the increased inequality in Britain occurred while real wages were rising, low-paid British workers actually realized modest increases in their real wages over the decade. By contrast, low-paid American workers had sizable decreases in their real wages.

Wage inequality rose, but by much less, in Canada, Japan, and in continental Europe. Inequality barely changed in France and Italy, fell in the Netherlands, and rose much less in Sweden and Germany than in the United States. Nowhere did the ratio of the earnings of college graduates to less-educated workers rise as much as in the United States. Indeed, earnings differentials by education actually declined in the rapidly growing Korean economy during the same period.

What differentiates countries with wage differentials that are increasing slightly, or stable, from countries such as the United States, which experienced large increases in inequality? Basically, three things: (1) countries with fairly stable wage differentials place more emphasis on wagesetting institutions than the United States does; (2) they also either have maintained the strength of unions or have experienced smaller declines in unionization than the United States; and (3) they have a better training system for their less-educated workers.

Many European countries have wage-setting systems that are relatively centralized, either because of government policies or because of strong trade unions. In some countries, the government extends contracts from union to nonunion workers; this makes collective bargaining the key to wage determination, even with only moderate or modest union representation. In both Germany and France, for example, the ministers of labor extend contracts negotiated between employer federations and unions to all employers in that particular sector. The French rely heavily on the national minimum wage in setting wages for all workers. Wagesetting in Italy and Sweden is relatively centralized, too. Italy's Scala Mobile, a centrally negotiated wage for low-paid workers, pushed up the bottom of the earnings distribution in Italy in the 1980s. In Sweden, bargaining between the main employers' federation and the main blue collar union federation reduced wage inequality through 1983. But centralized wagesetting weakened in Sweden later in the 1980s, and earnings differentials began to widen.

The extent of union representation of workers also influences wage inequality. Both the United States and the United Kingdom had large declines in union penetration and, as noted, large increases in wage inequality during the 1980s. By contrast, Canada maintained a high level of union density and had smaller increases in inequality. Much of the greater rise in inequality in the United States than in Canada is attributable to the continued strength of Canadian unions.

Changes in the supply of workers with different levels of education also contributed to the differing trends in wage inequality. In the United States, growth of the college-educated work force decelerated greatly in the 1980s. By contrast, Canada and the Netherlands had sizable increases in the number of college graduates relative to high school graduates in the work force. As a consequence, Canada had only modest increases in the college--high school wage differential, and the Netherlands had modestly declining wage differentials.

In every country that we examined, the change in the relative supplies of workers with certain levels of education influenced relative wages. We would expect shifts in the industry or occupation mix of a country's demand for labor in favor of more-skilled workers to improve the relative earnings of those workers, too. But these shifts do not appear to have contributed substantially to the different trends in earnings inequality across countries. That is because all of the countries we study had similar shifts, favoring the more educated. Only in Japan and Korea do shifts in industry structure--notably the continued strength of manufacturing employment--help to account for their distinct change in relative wages.

Worker Training

Worker training differs across countries in many ways. Some countries base their training systems largely on the firm. Other countries rely more on government training programs, or on school-based or individual training decisions. Evidence on the returns to training for individuals in the form of wages, and for firms in the form of productivity, suggests that company-based training has the highest payoff. Presumably this is because it is linked more directly to the skills needed at the particular workplace. By contrast, informal "learning by doing" raises worker productivity in the short run but not in the long run. Government-led and school-based job training systems have only marginal effects on wages.

How do countries develop a successful firm-based training system? Germany and Japan are exemplary, and their experiences show that such a system requires considerable institutional support. In Germany, unions and works councils help to determine the content of training in apprenticeship programs. Upon completion of a program, trainees receive national certificates of skills. In addition to certification, the government provides schooling that complements the workplace training. Finally, government regulations make it costly to hire young people from outside of the apprenticeship system; local chambers of commerce also exert pressures against firms "poaching" trained workers.

In Japan, there is little mobility of labor in and out of large firms, but workers rotate jobs within firms. Thus, workers build skills in a variety of broadly defined jobs at a company. Furthermore, all high school graduates have mastered similar skills, and there are strong links between firms and schools. So a worker's performance in high school is the key to obtaining a job with a good firm that provides workplace training. Thus, the lesson from both Germany and Japan is that developing a good workplace-based training system requires a range of institutional support, from schools, government, and so on.

Do differences in training systems contribute to the different trends that we observe in inequality of earnings or in productivity growth? Our estimates show that formal firm-based training pays off in terms of earnings. Evidence on who gets training across countries shows that in the United States, most firm training programs are for white collar and educated workers. The limited training for blue collar workers is more remedial here than in other countries. Since training builds skills, and the more skills high school graduates have, the better they can compete with more-educated workers, it is likely that the better-trained, less-educated workers in, say Germany or Japan, are closer substitutes for college graduates than in the United States. Thus, when the job market favored more educated workers in the 1980s, there was less pressure for changes in earnings differentials in those countries than in the United States.

Worker Representation

In the United States, unions are the sole form of worker representation. But in recent years, union representation in the private sector has fallen to 11 percent of the workplace--comparable to its level before the New Deal. The National Labor Relations Board has frowned on company-sponsored committees of workers, viewing them as an anti-union device. So, many American workers are without a voice at their firms.

In contrast, all continental European countries have legally mandated works councils inside firms. How do these councils operate? The particulars of works councils differ across countries, but they have several common features.

In Europe, works councils are mandated legally in all firms above a given size. However, this does not mean that a government regulator forces councils onto firms whose workers and management do not want them; rather, either side can insist upon their formation. Works councils have legal rights to information about the performance of the firm as it relates to labor issues. They also have rights to "consultation" about changes in labor and personnel policies.

In Germany, but not elsewhere in Europe, the councils have the right to settle disputes with management by arbitration. In Canada, works councils deal solely with occupational health and safety issues. The mode of choosing workers to sit on a council is left to the enterprise, with the requirement that councilors be "representative." In all countries except Spain, councils cannot strike and do not negotiate wages.

During the 1980s, when union membership fell in many countries and union influence declined in almost all countries, works councils flourished. In Canada, both management and labor agree that mandated health and safety councils have been superior to government regulation in overseeing workplace health and safety. In France, where unions are in disarray; in Italy, where many institutions are in disarray; and in Germany and Belgium, where unions remain strong, works councils provide a forum for cooperative labor-management relations and a place for workers to voice concerns and influence management decisions.

The general consensus for the countries studied in the NBER project is that councils help to create productive labor relations. This is true both where management has an extensive influence on councils, as in France, and where unions have an extensive influence on councils, as in Germany. Moreover, the growing role of councils in European countries with very different union and labor systems, and in a period when unionism is waning, suggests that they are a "robust" institutional form for labor-management relations. The councils seem more suited for dealing with the ongoing decentralization of collective bargaining and for building worker voice within enterprises than many traditional trade unions are.

Income Maintenance and Social Insurance Systems

Many economists and social observers, unfamiliar with the equivocal evidence for the widely studied American welfare system,(1) have blamed the more extensive income maintenance systems and employment regulations of Western Europe for continued high unemployment and related economic ills. There are some fairly clear examples of poorly constructed programs, including sick leave in Sweden, which have had substantial adverse effects on working; and of unemployment insurance programs that extend periods of joblessness. But our study of a host of different programs in different countries shows that, in general, these programs do not have major efficiency costs.

One reason is that many European programs require people to work in order to receive benefits. In France, for instance, the combination of time limits on welfare payments and the availability of public daycare leads many single women to leave welfare and go to work when their child reaches age three. In Sweden, employer benefits such as maternity leave are generous, so single mothers are more likely to have worked before they have a child. Other European programs, such as tenant protection or subsidies for homeownership, primarily affect residence decisions and have only a minor impact on labor mobility.

Another reason that European social programs do not necessarily have a negative effect is that firms or individuals find ways around them. For instance, employment security laws in Germany, France, and Belgium force firms to adjust their number of employees more slowly than they might like when market conditions change. But firms in these countries simply adjust the hours each employee works instead.

In Spain, there is considerable noncompliance with high payroll taxes mandated to pay for health care. In firms that evade those social security taxes, the workers are like American workers who do not have employer health coverage. However, most of the Spanish workers have family members in a firm that pays the social security taxes. Thus, those firms end up providing health care for the entire family.

In an attempt to increase labor market flexibility, several European countries sought to reform their income maintenance and employment regulations in the 1980s. Their success was marginal at best, and did not "cure" the European unemployment problem: unemployment in Great Britain was at least as high after Mrs. Thatcher's reforms as before. Both our studies of specific programs and the recent historical experience suggest that social protection programs should not be judged by their secondary effects on aggregate economic performance. Assessing these programs involves comparing how well they accomplish their goals as compared to their costs, which include the potential distortionary costs that result from their changing incentives in undesirable ways.

Extreme Poverty(2)

The failure of the United States to alleviate the problems of very poor people--those in the bottom 5 percent or so of the income distribution--stands in sharp contrast to its general economic success. Despite our high standard of living, the very poor are worse off in absolute terms here than in other countries. Those at the very bottom of the income distribution appear to be affected very little by overall income growth or the median level of income. Specific policies toward these people, not general economic growth, seem to be the only way to solve their problems.

But well-meaning policies do not always work. For example, Australia's generous funding for aborigines living in their native areas seemingly backfired by inducing low labor participation and extensive social problems. In contrast, European countries with welfare systems that require work to get benefits have managed to keep female-headed households out of extreme poverty. In part, this is because women in those countries who have out-of-wedlock births tend to do so later in their lives than American women do, after they have received an education or some work experience.

LESSONS FOR THE UNITED STATES

The "Working Under Different Rules" project had two main goals. First, we wanted to discover whether other advanced countries had labor market problems similar to those plaguing the United States. Then, we hoped to learn which policies or institutions had enabled those countries to overcome the labor market problems we share.

We found that some changes have been more substantial in the United States than elsewhere--the rise in income inequality and the loss of worker representation --and that extreme poverty is greater in the United States than in other countries. Since other countries have coped better than the United States with the shift in demand for labor away from less-skilled workers, with union weakness, and with the difficulties of the very poor, we learned at a minimum that these problems are not inexorable and irremediable. Even the modest differences in income maintenance policy between the United States and Canada--a stronger unemployment benefit system and family income maintenance system in Canada--greatly affect the poverty rate, for example. And the modest differences in labor law between the United States and Canada account for at least some of the differing pattern of unionization in the two countries, with consequences for income inequality.

We also found that most social protection programs have only modest side effects on the efficacy of the labor market in Europe. Programs that require people to work before being eligible for benefits minimize the risk that people will choose to take government aid rather than working. The primary effects of many European programs are on the well-being of the people they were designed to help. Given this, it is not surprising that the diverse European efforts to solve unemployment by weakening labor regulations, enhancing labor market flexibility, and reducing the welfare state have not worked, at least not yet.

We observed that the effects of specific programs and institutions depend on the environment of other institutions and policies in which they operate. German and Japanese training institutions, in particular, require considerable institutional support to be successful. Thus, we learned that a set of interrelated programs has a greater chance of succeeding than a single program designed to resolve a given problem.

We also note that the experience of countries in the 1980s is consistent with two potential trade-offs. First, countries such as the United States that had a poor record in productivity and real wage growth had better experiences with employment. Except for Japan, no country managed to do well on both of these levels. Second, countries that maintained stable wage distributions had worse experiences with employment (again excluding Japan). This suggests a second possible trade-off--between income inequality and employment--in an era when market forces favored the better educated. The notion that any country has a "lock" on the right institutions or policies is simply not true, although the Japanese, as we all know, have an enviable growth record.

Specific papers in the "Working Under Different Rules" project reported on some successful and some unsuccessful labor market and income maintenance programs outside the United States. Of particular import for the United States are welfare programs that require or encourage, rather than discourage, work; and institutions that buttress training for less-educated workers. Of course, the United States has distinct features that make it infeasible to "import" foreign practices. What works in Japan, or Europe, or Canada, may not fit with American social and economic practices. And, it is logically possible (although unlikely) that what fails in some other advanced country might work here. Still, a solid body of information about foreign experiences does provide insights for the design of institutions or programs that might better our situation.

SOURCES

R. M. Blank, ed., Social Protection Versus Economic Flexibility: Is There a Trade-Off? Chicago: University of Chicago Press, 1994, forthcoming

Abraham, K. G. and S. N. Houseman, "Does Employment Protection Inhibit Labor Market Flexibility?

Lessons from Germany, France, and Belgium" Blank, R. M., "Public Sector Growth and Labor Market Flexibility: The United States Versus the United Kingdom"

Blank, R. M. and R. B. Freeman, "Evaluating the Connection Between Social Protection and Economic Flexibility"

SOURCES

R. M. Blank, ed., Social Protection Versus Economic Flexibility: Is There a Trade-Off? Chicago: University of Chicago Press, 1994, forthcoming Abraham, K. G. and S. N. Houseman, "Does Employment Protection Inhibit Labor Market Flexibility? Lessons from Germany, France, and Belgium" Blank, R. M., "Public Sector Growth and Labor Market Flexibility: The United States Versus the United Kingdom" Blank, R. M. and R. B. Freeman, "Evaluating the Connection Between Social Protection and Economic Flexibility" Borsch-Supan, A., "Housing Market Regulations and Housing Market Performance in the United States, Germany, and Japan" De La Rica, S. and T. Lemieux, "Does Public Health Insurance Reduce Labor Market Flexibility or Encourage the Underground Economy? Evidence from Spain and the United States" Hanratty, M. J., "Social Welfare Programs for Women and Children: The United States Versus France" Holtz-Eakin, D., "Health Insurance Provision and Labor Market Efficiency in the United States and Germany" Montgomery, E. B., "Patterns in Regional Labor Market Adjustment: The United States Versus Japan" Rebick, M. E., "Social Security and Older Workers' Labor Market Responsiveness: The United States, Japan, and Sweden" Scherer, P., "Trends in Social Protection Programs and Expenditures in the 1980s" Stafford, F. P. and S. Gustafsson, "Three Regimes of Childcare: The United States, the Netherlands, and Sweden"

D. Card and R. B. Freeman, eds., Small Differences That Matter: Labor Markets and Income Maintenance in Canada and the United States, Chicago: University of Chicago Press, 1993 Blackburn, M. L. and D. E. Bloom, "The Distribution of Family Income: Measuring and Explaning Changes in the 1980s for Canada and the United States" Blank, R. M. and M. J. Hanratty, "Responding to Need: A Comparison of Social Safety Nets in the United States and Canada" Borjas, G. J., "Immigration Policy, National Origin, and Immigrant Skills: A Comparison of Canada and the United States" Card, D. and W. C. Riddell, "A Comparative Analysis of Unemployment in Canada and the United States" Freeman, R. B. and K. Needels, "Skill Differentials in Canada in an Era of Rising Labor Market Inequality" Lemieux, T., "Unions and Wage Inequality in Canada and in the United States" Riddell, W. C., "Unionization in Canada and the United States: A Tale of Two Countries"

R. B. Freeman, ed., Working Under Different Rules, New York: Russell Sage Foundation, 1993, forthcoming

Blank, R. M., "A Trade-Off Between Social Protection and Economic Flexibility?" Card, D. and R. B. Freeman, "Small Differences That Matter: Canada Versus the United States" Freeman, R. B. and L. F. Katz, "Rising Wage Inequality: The United States Versus Other Advanced Countries" Freeman, R. B., "Learning from Foreign Experiences" Lynch, L., "Payoffs to Alternative Training Strategies at Work" Rogers, J. and W. Streeck, "Workplace Representation Overseas: The Works Council Story"

R. B. Freeman and L. F. Katz, eds., Differences and Changes in Wage Structures, Chicago: University of Chicago Press, 1994, forthcoming Abowd, J. and M. Bognanno, "International Differences in Executive and Managerial Compensation" Abraham, K. G. and S. Houseman, "Earnings Inequality in Germany" Blanchflower, D. G. and A. J. Oswald, "International Wage Curves" Blau, F. D. and L. Kahn, "The Gender Earnings Gap: Some International Evidence" Edin, P. and B. Holmlund, "The Swedish Wage Structure: The Rise and Fall of Solidarity Wage Policy" Erickson, C. and A. Ichino, "Wage Differentials in Italy: Market Forces, Institutions, and Inflation" Freeman, R. B. and R. Gibbons, "Getting Together and Breaking Apart: The Decline of Centralized Collective Bargaining" Gregory, R. and F. Vella, "Aspects of Real Wage and Employment Changes in the Australian Male Labor Market" Katz, L. F., G. Loveman, and D. G. Blanchflower, "A Comparison of Changes in the Structure of Wages in Four OECD Countries" Kim, D. and R. Topel, "Labor Markets and Economic Growth: Lessons from Korea's Industrialization, 1970-90" Krueger, A. B. and J. Pischke, "A Comparative Analysis of East and West German Labor Markets Before and After Unification" Schmitt, J., "The Changing Structure of Male Earnings in Britain, 1974-88"

L. M. Lynch, ed., Training and the Private Sector: International Comparisons, Chicago: University of Chicago Press, 1994, forthcoming Berg, P. B., "Strategic Adjustments in Training: A Comparative Analysis of the United States and German Automobile Industries" Bishop, J. H., "The Impact of Previous Training in Schools and on Productivity, Required OJT, and Turnover of New Hires" Blanchflower, D. G. and L. M. Lynch, "Training at Work: A Comparison of U.S. and British Youths" Cameron, S. V. and J. J. Heckman, "Determinants of Young Male Schooling and Training Choices" Dolton, P. J., G. H. Makepeace, and J. G. Treble, "Public and Private Sector Training of Young People in Britain" Elias, P., E. Hernaes, and M. Baker, "Vocational Education and Training in Britain and Norway" Groot, W., J. Hartog, and H. Oostebeek, "Returns to Within-Company Schooling of Employees: The Case of the Netherlands" Hashimoto, M., "Employment-Based Training in Japanese Firms in Japan and the United States: Experiences of Automobile Manufacturing" Oulton, N. and H. Steedman, "The British System of Youth Training: A Comparison with Germany" Soskice, D., "Reconciling Markets and Institutions:

The German Apprenticeship System" Weiss, A., "Productivity Changes Without Formal Training"

J. Rogers and W. Streeck, eds., Employee Participation and Works Councils, Chicago: University of Chicago Press, forthcoming Bernard, E., "Works Councils in North America" Brulin, G., "Sweden" Escobar, M., "Spain" Freeman, R. B. and E. Lazear, "Economic Theory of Works Councils" Levitas, A. and M. Federowicz, "Poland" Lukacs, J., "Hungary" Muller-Jentsch, W., "Germany" Regalia, I., "Italy" Rogers, J. and B. Wooten, "United States" Streeck, W. and S. Vitols, "Multinational European Companies" Visser, J., "The Netherlands"