New Rules for a New Economy
Employment and Opportunity in Postindustrial America

Read BW's Review of This Book

CHAPTER ONE

Recreating the Prosperity of the Past
in the Economy of the Future

    The U.S. economy has grown steadily in the 1990s. By 1998, inflation and the federal budget deficit had melted away. Despite occasional jitters, the stock market continued to reach historic peaks. But regardless of what the business page says about consumer confidence at the time you read this book, Americans' anxiety about the future remains below the surface. Wages for all but a fortunate few have stagnated or fallen. Health care costs are up and insurance coverage down. Unemployment is low, but many people still fear the next round of layoffs. With factories moving offshore, big companies downsizing, and the public sector shrinking, workers are asking where the good new jobs will come from.

    There is good reason to worry. Automation and international competition continue to eat away at the manufacturing jobs that were once plentiful. Service industries now employ about three-quarters of the workforce, with many workers trapped in low-wage, dead-end jobs. This book is about the reasons and the remedies for stagnation and decline in American living standards, for growing wage inequality and employment instability. Both causes and cures are to be found in the service economy; that is where the bulk of the jobs are and where the jobs of the future will be created.

    Our starting point is the breakdown of the national economic system that generated prosperity for three decades after World War II, a period we call the Wonder Years. For a decade or more after the breakdown began, people clung to a past that could not be recovered. Now most Americans recognize that the past has gone for good. Yet no vision of the future has crystallized to take the place of people's past understandings of the economic world and their place in it: the old ideas of a job and career; how individuals and families achieve economic security; how employers, communities, and the nation prosper.

    In this book we outline a reconstruction of the contemporary economy that could generate a new sense of confidence and a new era of widely shared prosperity. We argue that many elements of a reconstructed U.S. economic system have already begun to appear. We try to put the pieces together, showing how they might yield better performance as well as greater economic security.

    The system we believe attainable would not be economic and social nirvana. Drudgery and hard physical labor would remain. Some Americans would still perform much more of this work, and earn much less, than others. Even if consumer price indices adequately accounted for new products and improved quality, we would not expect another period of quantitative growth--of more and more "stuff" every year--like the 1950s and 1960s. Still, the potential for qualitative improvements in services, for making work itself more rewarding, and for enjoying the fruits of growth in the form of shorter work time suggests an attainable future that Americans could look toward with optimism rather than unease.

Services Matter

Ever since Adam Smith wrote The Wealth of Nations in 1776, economic development has been virtually synonymous with manufacturing. Many people continue to regard production of services as subsidiary to production of goods, even a parasitic drag on economic performance. Although services have accounted for more than half of U.S. employment for more than fifty years (Figure 1), policymakers, economists, and business writers continue to focus on manufacturing. After all, for a quarter-century after World War II, productivity increases in manufacturing stimulated a virtuous circle of rising wages and growing consumer demand, corporate profits, more investment, job creation, and higher living standards. That era has ended.

    When public attention belatedly turned to service industries, powerful rhetorical images portrayed the service worker as a hamburger flipper or window washer, producing nothing lasting or solid. Pundits and politicians asked how Americans could expect to get along, much less prosper, taking in each other's laundry.

    Goods-producing industries still matter, of course. But the simple fact is that services now account for most of U.S. employment, while manufacturing has fallen to around 15 percent and continues drifting downward. More Americans now work in physicians' offices than in auto plants, in laundries and dry cleaners than in steel mills. Services, as a result, matter more than manufacturing when it comes to job quality. They also increasingly account for the rate at which living standards rise. But dependence on service jobs leaves many uneasy; say "dead-end job," and the image that comes to mind is McDonald's, not McDonnell Douglas.

    It is wrong to assume that the majority of service workers will necessarily be stuck in low-wage jobs with poor benefits and little security. The services offer millions of good jobs, as well as millions of bad ones. And though a large share of service jobs pay little more than the minimum wage, this is a consequence of particular features of the U.S. economy and its labor market, especially the spread of wage-based competition. With deregulation and new competition in sectors ranging from trucking to retailing to health care, employers have sought to cut labor costs, treating workers as interchangeable parts and accepting the high turnover and low commitment to the job that result.

    In earlier decades, in service industries such as telecommunications and banking as well as in manufacturing, workers could start at the bottom of a firm's job ladder and expect to advance through learning, experience, and seniority. They could look forward to a career, a progression rather than a succession of jobs. Those who worked their way up in the 1970s and 1980s now confront an ongoing wave of mergers and corporate restructuring. These go by many names, including downsizing (or "rightsizing") and reengineering. Whatever the label, the consequences include layoffs, lateral transfers that may foreclose advancement opportunities, and, for those who keep their jobs this time, fear that the next wave will catch them too. At the same time, a shrunken social safety net and the decline of nuclear and extended families mean that those out of work have fewer resources to fall back on.

    In goods-producing industries, a combination of international and domestic competition drives restructuring. In services, the direct impact of international competition has been minor compared with deregulation and other new sources of domestic competition: unlike goods, only a few service products can be shipped or stored; most must be produced at the point of sale.

    When competitive forces lead to new and better service products, Americans benefit as consumers. But when competition leads to downward pressure on wages and benefits, Americans suffer as workers. And as wages below the top slice of the income distribution fall, workers can buy fewer goods and services. The loss of the opportunities that made the Wonder Years so promising for earlier generations also heightens social divisions.

    This book shows how to reverse recent labor market trends and the poor economic performance that has accompanied wage-based competition in many service industries. We argue, in our last two chapters, for a new set of institutions that will support higher wages for middle- and low-income workers and create more opportunities for those who seek to advance. These same institutions will also improve economic performance and service quality. The job for public policy, in short, is to catch up with the changes in the economy since the 1970s (indeed, since the last round of major U.S. labor policy innovation in the 1930s).

    The Wonder Years were wonderful not only because many workers could expect steady advances in pay and benefits but because the economy delivered in ways that people experienced every day--cars, boats, TVs, bigger homes, basic health care. (In some other respects, of course, the Wonder Years, a time of cold war and conformity, were anything but wonderful.) The reforms we propose should also contribute to tangible improvements in people's lives: child and elder care that enrich life rather than exacerbate guilt, better education, more responsive customer service, a revival of family and community life once wage increases for the many make the struggle to survive less consuming.

    This book is rooted in an analysis of work and business in the services. We use mostly service examples because we are trying to reorient debate away from the old economy and old concepts--away from manufacturing. Even so, the concepts we develop to make sense of the huge and diverse service sector also shed light on goods-producing industries. For this reason, as well as because the services now make up the bulk of employment, we see our last two chapters as a prescription for economywide renewal.

The Past: The Origins and Decay of the Wonder Years

The core features of the Wonder Years provide a basis of comparison for our analysis of the services today, a gauge of how the world has changed since white bread and bland beer gave way to microbakeries and microbreweries, and a historical and analytical starting point for asking how to construct a new period of prosperity.

    During the one hundred years beginning in 1870, real gross domestic product (GDP) per labor-hour in the United States rose more than tenfold, an astounding increase. On a per capita basis, real GDP rose almost seven times between 1870 and 1973.

    Mechanization and automation in manufacturing powered the American economic engine. The technological basis was laid after the Civil War, with the rise of the steel industry--initially to provide railroad rails--and the development of the "American system" of manufactures. The American system of assembling standardized interchangeable components was spawned in armories and factories that made axes, firearms, and clocks. It led to affordable mass-produced sewing machines and bicycles, later to automobiles and many other consumer goods. Railroads, telephone, and telegraph helped manufacturers coordinate their growing purchasing and distribution networks.

    Engineers and managers refined the American system in mass production industries, developing the methods still known as "scientific management." Organizations capable of administering high-volume standardized production also emerged. Alfred Chandler has documented with painstaking detail the creation of new administrative giants in a long list of industries.

    The big corporations that dominated major industries became vehicles for spreading mass production technologies through the economy. They had the capital and the control over markets necessary to invest in expensive, single-purpose equipment. Productive capacity rose at an unprecedented rate.

    No mechanism existed, however, to ensure that wages and purchasing power would rise in parallel with the economy's newfound capacity to produce. When the Great Depression brought a prolonged period of stagnant demand and corporate losses, there was nothing to stop massive layoffs and real wage cuts from building to a vicious circle in which falling demand pushed the entire economy into deep decline.

    Widely shared perceptions that inadequate consumer demand had helped trigger the Depression added to support for New Deal labor and social welfare reforms. Congress passed legislation that helped create the institutional pillars of the Wonder Years, buttressing the economy against a future downward spiral fed by declining wages.

(1) The National Labor Relations Act of 1935 (also known as the Wagner Act) protected workers' rights to organize and bargain collectively. It contributed to the rise of labor unions in the steel, auto, rubber, electrical equipment, and other manufacturing industries. After World War II unions and collective bargaining established industrywide wage and benefit standards in core manufacturing industries. Postwar union contracts linked annual increases in real wages to the roughly 3 percent annual rate of national productivity growth.

(2) The Fair Labor Standards Act of 1938 established national standards for the minimum wage, overtime pay, and restrictions on child labor. In the next thirty years periodic increases kept the national minimum wage above 40 percent of the average manufacturing wage. The federal minimum wage ensured that workers in many low-wage, nonunion industries also shared in the benefits of productivity growth.

(3) The Social Security Act of 1935, the foundation of the current social security system, established national contributory old-age pensions. The same legislation established a federal unemployment insurance tax and incentives that induced states to create unemployment insurance systems. When older or unemployed workers dropped out of the labor force, their consumption would no longer fall precipitously.

    After World War II the United States entered the Wonder Years. The nation emerged from the war with unmatched industrial power. Not only were U.S. industries undamaged, but they had in many ways been strengthened by government investments in technology and production capacity. Conversion from military production proved less disruptive than many had feared. As other economies rebuilt, American industry in 1953 accounted for 45 percent of world manufacturing output. With foreign competition a minor concern, major firms settled into generally comfortable domestic accommodations. Union contracts, minimum wage increases, and unemployment insurance payments joined Keynesian fiscal and monetary policy to dampen the business cycle, complementing mass production by sustaining consumer demand. Jobs were plentiful and so were profits.

    This pattern could not last. It was easier for Europe and Japan to catch up than for the United States to stay ahead. The big American steelmakers were complaining of "unfair" foreign competition by the end of the 1950s. Japan improved its manufacturing capabilities fast enough to become the world's low-cost supplier of goods ranging from ships to television sets in the 1960s. By the late 1970s imports accounted for almost 30 percent of U.S. automobile sales. No longer could U.S. Steel and General Motors function as price leaders, ensuring handsome margins for themselves and adequate, if usually smaller, profits for their domestic rivals and suppliers.

    During the 1960s productivity rose at an average annual rate of 2.8 percent. The rate of increase fell below 2 percent in the 1970s and has since been even lower, around 1 percent a year. Low productivity growth put a damper on wage growth; without rising wages, demand growth slowed.

    While international competition buffeted and broke the old oligopolies in manufacturing, deregulation, beginning under President Jimmy Carter in the late 1970s, contributed to steadily growing competitive pressures in a wide range of domestically oriented industries, including financial services, transportation, and telecommunications. At the same time, new entrants and changing consumer preferences brought new competitive pressures to industries such as retailing. In the 1990s attempts to contain rising costs brought competition even to the health care sector.

    Manufacturing and service firms alike sought to reduce costs and improve quality and flexibility by automating and reorganizing their workplaces, closing off many internal job ladders, and cutting into well-paying jobs that had earlier been open to less-educated workers. Some manufacturers moved low-skilled, labor-intensive production to Asia and Mexico. In the 1980s Fortune 500 firms began to downsize, computers showed up on desktops, and employers merged, consolidated, and began to rely more heavily on part-time and temporary workers.

    Meanwhile, the institutional framework of U.S. labor relations had been fraying. Male-dominated industrial unions adapted poorly to the rising percentage of women in the labor force and an expanding service sector. Advising corporations on how to remain "union-free" became a thriving cottage industry. Union coverage declined steadily to today's pre-Wagner Act levels--11 percent of the private nonagricultural labor force.

    With the decline of unions came the erosion of the social policies supported by labor. The minimum wage roughly tracked productivity for the two decades after 1947, then lost half its value relative to productivity over the next three decades (see Figure 5, Chapter 8). The share of unemployed workers receiving unemployment insurance benefits fell. By the end of President Bush's term most people understood that the old days were gone, even if they weren't quite sure why.

Making Sense of the New (and Old) Economy

The narrative above is familiar. This book goes beyond it to develop an analytical framework that describes the economy of the Wonder Years and of today. We then use that framework to define policies designed to point the nation toward a new era of prosperity.

    We describe the economy using three concepts: work systems, business organization, and career paths. Each type of work system relies on a different basic approach to organize production and motivate or control how hard and well people work. Examples range from assembly lines, to janitorial labor with mop and broom, to the bureaucratic incentives of promotion and pay found in banks and insurance companies, to professions such as medicine and the law. Work systems emerge from a complex interaction among firms, employees, and technological and other contextual influences.

    By business organization we mean the ways that firms structure their operations. To a majority of Americans, the most familiar forms of business organization are still the giant corporations--GM, Sears, AT&T (and newer firms such as Microsoft on which the media shower equal attention)--and the small establishments that consumers pass by and deal with daily. Within these organizations, managers decide which goods or services to offer and how to produce or distribute them. Relationships between firms, in turn, are coordinated through the market, that is, through "arm's-length" buying and selling between firms that have no enduring ties and will buy from someone else if they can get a better deal. Patterns of business organization in the services have always differed from those in manufacturing because of the smaller average size of service establishments and the prevalence of franchise agreements. In both services and manufacturing, recent trends in business organization include the creation of more enduring and cooperative links between separate firms (as, for example, in health care alliances).

    Career paths structure the movement over time of workers through the labor market. These paths are shaped by the decisions of individual workers, groups of workers (informal communities, professional associations, labor unions), and owners and managers. The job ladders within large firms illustrate one kind of career path (e.g., from file clerk to secretary to office administrator). Another example is the movement of skilled craft workers in the construction trades from one worksite to another. As we define them, career paths do not necessarily imply that workers have "good" jobs. Nor does movement always lead to advancement in pay, skill, or responsibility.

    Public policy affects economic outcomes not only directly but indirectly through its influence on work systems, business organization, and career paths (Figure 2). An obvious direct effect comes from the legally mandated minimum wage. Examples of indirect impacts include antitrust laws that affect business organization and antidiscrimination laws that affect work systems and career paths.

The Wonder Years Revisited

Table 1 provides a brief comparison of the four work systems analyzed, in detail in Chapters 3 and 4. The core of the middle class in the Wonder Years enjoyed one-company careers in vertically integrated firms. One major group consisted of blue-collar workers, many with "tightly constrained" jobs (including assembly-line jobs, although there were never as many as in the popular imagination). During the Wonder Years large numbers of middle-class Americans also found employment (as secretaries, middle managers, factory quality inspectors, and supervisors) in a work system we call semiautonomous. In this system, good wages, seniority-based pensions and promotion opportunities, and other bureaucratic incentives tied workers to firms and elicited effort and commitment. Since they generally stayed with one employer a long time, workers could acquire and apply firm-specific know-how. They learned the company's products, its "standard operating procedures," how to work effectively with others in the organization whose cooperation they needed. Most of those in tightly constrained blue-collar and semiautonomous white-collar jobs did not work for GM or AT&T. They worked for the GM supplier or dealer, the community hospital, the local bank or supermarket. But rapid economywide growth translated into promotion opportunities and security even in smaller firms.

    The top of the labor market, in the Wonder Years as today, consisted of employees in the high-skill autonomous work system. This category includes professionals, upper-level managers, craft workers, and many technicians. In the high-skill autonomous system, employees, not their bosses, take primary responsibility for performance. Professional and craft pride, reinforced by career incentives, motivates effort and performance. Some high-skill autonomous workers during the Wonder Years spent their entire careers in one or a few organizations. Others, such as skilled construction workers, had careers that spanned many employers. Formal institutions (e.g., professional associations and hiring halls, portable credentials and benefits) supported informal networks of professional and craft workers, reducing the social and economic cost to the worker of moving to a new employer.

    At the bottom of the labor market were "unskilled" labor-intensive jobs: janitors and gardeners, waiters and waitresses in all but elite restaurants, nurses' aides and hospital orderlies. In these jobs, the nature of the tasks prevents continuous, close supervision. Neither employers nor employees systematically analyze the work process to improve efficiency or quality. Low wages and the modest cost of capital equipment discourage employers from attempting the type of rationalization typical of capital-intensive production processes. For this reason, we term this category of work system "unrationalized labor-intensive." In the Wonder Years some people in unrationalized labor-intensive jobs moved up by moving out--into other work systems with other employers. For workers in both unrationalized and tightly constrained work systems, collective bargaining and the minimum wage kept pay rising roughly in step with salaries in the other two work systems.

    Many Americans see the post-World War II period as wonderful in hindsight because, unlike today, workers' aspirations and opportunities were more or less in balance. White men, at least, could take for granted that they would make it into a well-paid tightly constrained or semiautonomous job by some time in their twenties. Others were not so fortunate. African Americans, emboldened by the civil rights movement in the 1960s, challenged their relegation to the most arduous and mind-numbing work. The women's movement took issue with female segregation into unrationalized jobs and bureaucratic, semiautonomous occupations that paid less than male-dominated job categories. Since the 1960s, aspirations have risen. Since the 1970s, opportunities, especially for those with modest educational credentials, have narrowed. Widening gaps between aspirations and opportunities have left many Americans anxious, uncertain, confused.

The New Economy: The Origins of Economic Anxiety

Table 2 summarizes wage and employment changes by work system between 1979 and 1996. (Changes in occupational definitions make comparisons for years before 1979 treacherous.) Employment in the unrationalized labor-intensive work system remains at about a quarter of the total, as in 1979, while growth in professional and technical jobs has increased the share of the high-skill autonomous work system. Semiautonomous jobs have declined relative to other categories. Some semiautonomous jobs have migrated to the high-skill category. In other cases, employers competing on the basis of low wages have replaced semiautonomous jobs with unrationalized labor-intensive work. Retailing illustrates the substitution. In department stores facing competition from both discounters and specialty outlets, many nonprofessional positions that once paid decent wages and benefits and offered advancement opportunities--classic semiautonomous jobs--have been transformed into low-paying, high-turnover positions.

    The wage picture is even grimmer. Since 1979 pay in the unrationalized labor-intensive and tightly constrained work systems has fallen relative to the high-skill system. The decline of unions and the real value of the minimum wage contributed to the widening gap. Part of the drop in union coverage reflects the difficulty of organizing service workers, many of whom work in small establishments spread across large industries such as retailing.

    Within firms where the semiautonomous work system still predominates, career trajectories have become less predictable, in part because fewer of these jobs are in large firms with well-marked career paths. With deregulation, companies such as AT&T have reduced payrolls and dismantled their internal hierarchies. Wage inequality has grown as firms have contracted out work once performed by their own employees. Janitorial, food-service, and security jobs that once paid big-company wages and benefits are more often isolated in low-cost suppliers that rely on low-paid contingent workers.

    The analysis summarized above and explored in detail in later chapters points to three underlying roots of economic anxiety. The first is low pay and lack of upward mobility for those stuck in unrationalized labor-intensive jobs, as well as in a few expanding, tightly constrained settings such as telephone call centers. These kinds of jobs rarely lead to a step up on a big-firm job ladder. A second cause is the insecurity that troubles those with good jobs. Office workers, managers, and sales employees, lacking specialized skills in high demand or portable credentials, fear the loss of income and status that may follow from a forced change of employer.

    A third root of economic anxiety is the apparently stagnant economic performance of the services, which limits the fruits available to distribute. We trace low measured productivity growth in part to the nature of service sector work processes. In unrationalized labor-intensive jobs, exemplified by nursing homes and other low-wage social services, employers don't care about improving performance because workers are so cheap. Workers do not have the training, knowledge, or power to improve performance on their own. Even in some tightly constrained and semiautonomous service work, low or stagnant wages and the high turnover they generate discourage firms from investing in workers or more energetically seeking performance improvement. Indeed, managerial approaches to performance improvement in the services to a surprising extent reflect preoccupation with the "deskilled" jobs of the mass manufacturing era, even though the tightly constrained work system has never been very prominent in the service sector.

Imagining Some Solutions: A Policy Preview

The policies proposed at the end of this book directly attack the low wages, rising job insecurity, and lackluster economic performance that we see as the roots of economic anxiety.

    Since wages are too low in the poorest-paying work systems, they should be raised. This will not happen by relying on the market as it now operates. Public policy can support wage levels in two basic ways: by raising the hourly minimum and by encouraging collective bargaining. The federal minimum wage (now $5.15) remains about 25 percent below its real value in 1968 (which would be about $7 today) and less than half what it was in 1968 relative to labor productivity. The minimum wage should be raised substantially. This would also push up wages for workers paid somewhat above the minimum.

    The one-employer career may not have disappeared, but even the largest firms make it plain today that there are no guarantees. With declining employment attachments, firms are less likely to support the skill development needed both for improved economic performance and for career advancement. (In Chapter 5 we explain why workers' skills and experience are the key to achieving large improvements in performance in many services.)

    There is no point in wishing for an end to downsizing and reengineering (which in some cases make their own contributions to performance improvement). The only practical way to address the mismatch between worker aspirations and actual prospects for security and mobility is to construct career paths that cut across firms. That way workers can look forward to "staircase" careers in which they can move up by moving to another employer. Public policies to encourage staircase careers include support for developing skill standards, training institutions, and job referral systems coordinated by multifirm labor and management groups.

    The third leg of our policy stool is an expansion of "reinvented," more craftlike unions. These are critical for raising wages, building new career ladders, and improving workers' performance in many service occupations. Current policies, however, effectively deny representation to many workers who move among small firms. Framed in the 1930s, U.S. laws were designed to encourage unionization at big factories. These laws are poorly suited to an era of small service establishments and transient employment attachments. Labor law reform should support union formation and bargaining across multiple employers (e.g., in retailing or office work). Employers would benefit through access to better prepared, more capable workers.

    Taken together, these and other policies discussed in Chapters 7 and 8 would create conditions that foster better economic performance within firms, clusters of firms, and occupational groups. Our policies would limit "destructive" competition, based on working harder and for less, in favor of "constructive" competition. They would allow public policy and institutions to catch up with the changes that have accumulated in work systems, business organization, and career padis since the 1930s.

    We anticipate three sets of responses to the policy proposals summarized above. Some oppose wage-setting policies, maintaining that they raise unemployment and undermine economic efficiency. But as we argue in Chapter 8, allowing the minimum wage to fall erodes efficiency gains over time because firms need not improve performance to stay in ' business. There is, in addition, little evidence that recent hikes in the minimum wage have cut into job creation. And even with a big increase in the minimum wage, we suggest in Chapter 8 that the reinvention of social policy could keep unemployment low.

    Second, some may object that low-wage foreign competition in a global economy must eventually prove overwhelming. Regardless of one's view of the consequences of wage-based foreign competition for U.S. manufacturing, only about 10 percent of the service workforce holds jobs directly exposed to international competition. Adding manufacturing and agricultural jobs would roughly triple the number of jobs potentially exposed to foreign competition, but even this total represents less than 30 percent of the labor force.

    The great bulk of service products are nontradable and will remain so. It is true that information technologies, low-cost transportation, and reductions in trade barriers may increase cross-border services trade (e.g., back offices moved to Ireland or the Caribbean, software upgrades produced in India). But it is domestic competition, in which American workers compete only with one another, that holds down wages and benefits for those without highly specialized skills and knowledge. In the less mobile parts of the services, states and localities, as well as the national government, can set rules for competition without worrying that they will be undercut by others outside U.S. borders.

    Finally, some might suggest that the best solutions to wage dispersion and dead-end jobs are found in education and training. Education is a good in itself. But education, standing alone, cannot make low-wage jobs disappear. As we argue in Chapter 4, low-wage jobs exist because some work systems are not organized to raise performance in ways that might lead to higher wages. Unless those work systems are reorganized, jobs for those in the lower part of the wage distribution will not pay better or offer greater career opportunities. More education will not automatically prompt reorganization, because low-wage work systems have little place for higher skills. Indeed, more education and training absent other policies could simply make worker frustration more widespread. Rising numbers of Americans would face opportunities that fall short of their aspirations.

Building a Postindustrial System

In the next chapter we compare jobs and productivity performance (to the extent it can be measured) in the services with those in manufacturing. The organization of the rest of the book follows our three main analytical constructs. Chapters 3 through 5 deal with work systems. Chapter 6 turns to business organization. Chapter 7 examines career paths. Chapter 8 considers how work systems, business organization, and career paths might be reshaped by public policy to generate more good jobs and better economic performance.

    Our book is also organized to help readers envision a "system-building" process parallel to the one that created the manufacturing-centered prosperity of the Wonder Years. As we saw, three distinct but intertwined developments combined to create that prosperity. In stage 1, technological and organizational innovations--new products, rationalization, mechanization, and scientific management--led to dramatic rises in productivity. Stage 2 saw the creation of institutions that helped spread mass production methods and related innovations through the economy. The key institutions were corporations with enough market power to be confident that, after making large-scale investments, they would be able to profit from a correspondingly large volume of sales. Stage 3, completed in the 1930s and 1940s, ensured the purchasing power necessary to keep the national economy expanding. It involved the creation of demand-sustaining labor market institutions (industrial unions, pattern bargaining, the minimum wage) and social policies (unemployment insurance and social security).

    Over the next generation or so the United States can choose to emulate the three-stage evolution that led to the Wonder Years.

    Stage 1: Recent innovations in service sector work processes have begun to suggest new ways of improving performance. In Chapter 5 we argue that learning by workers, individually and collectively, can generate performance gains sufficient to support better jobs and better pay.

    Stage 2: As the technological and organizational basis for performance improvement in services develops, the nation will need mechanisms that can propagate the new approaches, much as big corporations and labor market policy spread mass production methods. In Chapter 6 we suggest that recent changes in corporate strategy and structure are not enough to diffuse more productive approaches throughout the service economy. Although fluid and fragmented forms of business organization make corporations more nimble, they undercut the capacity of individual firms to support the training, learning, and employment security essential to high-quality, high-productivity strategies. In Chapter 7, therefore, we examine multifirm structures that could foster adoption of the new approaches to performance improvement and simultaneously give workers a wider range of employment and learning opportunities.

    Stage 3: In the creation of the old industrial system, the final step was the resolution of the problem of aggregate demand and unemployment. We discuss the manifestation of this issue in the new economy briefly in Chapter 8. Today, there must be enough paid jobs and social support that vulnerable workers are not forced to take subsistence jobs in the low-wage unrationalized labor-intensive work system. We speculate that creative social policy could make it easier to move among paid work, education, family responsibilities, and community service. Harking back to the preindustrial era, people would always have useful activity to perform in one of these realms. And the ability to move among them could keep the need for paid work in balance with the available jobs.

The analysis of particular firms, industries, and work systems in the chapters that follow shows that competition can operate in constructive or destructive ways. That argument would likely be taken for granted if we were talking about the world of sports, where the game is a more obviously artificial construct. Basketball's competition committee, for example, modifies the rules periodically to maintain audience appeal. Why? Because coaches and players keep trying to beat the old rules. Several years ago the committee instituted severe penalties for flagrant fouls and prohibited hand checking. The new rules help ensure that fluidity and athletic skill, rather than brute force and barely contained violence, remain the keys to success on the court. The competition committee, in other words, acted to discourage destructive competition.

    The economic world is also a human construct, not a state of nature. It is based on competition guided by rules. The issue is the type of economic competition and organization that the rules, from property rights to labor laws, encourage. Today they encourage widespread low-wage, low-skill competition and fail to encourage widespread improvement of service sector economic performance. Better economic and social outcomes require better rules.