But Where Are the Jobs?

It is not at all clear that President Clinton's economic policies add up to a coherent program capable of restoring robust economic growth and good jobs. In particular, the administration has placed exaggerated hopes on the imagined connections among education and training, productivity, and earnings.

The administration faces two fundamental economic imperatives. It must create a lot of jobs, and it needs to create more good jobs. To do this, it must temporarily lift the U.S. economy far above its recent groth rate of about 2 percent a year. At the pre-Inaugural economic summit in Little Rock, economists debated whether the U.S. economy needed any fiscal stimulus. Those who thought it did called for amounts ranging from $15 to $60 billion, less than 1 percent of U.S. GDP. In April, a very modest $15 billion stimulus package was filibustered to death. To put that figure in perspective, in Japan where recent trend growth has been closer to three percent per year, the debate was between stimulus proposals ranging from $90 to $200 billion. The $117 billion package just approved comes to fully three percent of Japan's GDP--12 times the rejected Clinton stimulus.

It is doubtful that a macroeconomic policy of gradual deficit reduction, absent even minimal short-term stimulus, will take us beyond slow growth and a jobless recovery in 1993 and 1994. The challenge of creating good jobs is even more daunting. The Republican view is that there's no way to know where good jobs come from or to improve the jobs that industry produces; private investment goes where it gets the best return, and the jobs naturally result. Clintonomics, in contrast, claims that policy interventions can improve the jobs that the economy would otherwise generate. In particular, higher productivity is said to be the means to well-paid jobs, and education and training the keys to productivity.

While policy changes can undoubtedly enrich jobs, there are some problems with the Clinton view. For one thing, the evidence for tight links among skills, productivity, and wages is circumstantial at best--especially in the current context of fierce global competition, high unemployment, downward pressure on wages, and weak demand. Further, even were the links more solid, it is not evident that Clinton's policies are the right ones to produce the skills that will produce the high-wage jobs.

To date, Clintonomics has gotten both the demand and supply sides of the equation wrong. Despite candidate Clinton's refusal to make deficit reduction central, President Clinton has become something of a born-again deficit-hawk, pushing a policy that will only retard growth. And on the supply side, Clinton has basically echoed the views of his predecessor, George Bush, in emphasizing the supposedly low quality of the work force as the explanation for the proliferation of low-wage jobs.

At first glance, the evidence for a productivity-wage link seems strong. Over time, Japan and much of Northern Europe have closed the productivity gap between themselves and the U.S. in many traded-goods industries, and their relative wages have indeed risen. The U.S. productivity growth slowdown that began sometime between 1967 and 1974 coincides with falling average real wages since 1973.

But it is not at all clear from this history what is the cause and what is the effect. During the postwar boom, increased education and training did not drive growth. Rather, a macroeconomic climate of high demand and rising real wages made possible the steady improvements in productivity. In the 1940s, 1950s, and 1960s, these gains co-existed with relatively strong unions and regulated economies, so that productivity gains could translate into wage gains, and continue the virtuous circle of high demand, rising productivity, rising wages, and high demand.

The productivity slowdown after 1973 was a worldwide phenomenon, but the relationship between wages and productivity began to diverge significantly, both across countries and across industries. Among major market economies, real wages fell only in the U.S. and U.K.--in the latter even though U.K. productivity growth dwarfed France and Germany's throughout the 1980s. Many third world nations that experienced rapid productivity increases--Mexico, Brazil, and indeed most non-east Asian poor countries--suffered huge drops in real wages.

Subscribe to The American Prospect

Furthermore, within the U.S., wages fell most in sectors in which productivity growth slowed least. Yet it was in precisely these sectors (auto parts, electronics assembly, apparel) that manufacturing was most likely to move offshore. Conversely, wages rose most in activities--financial management, corporate and tax law, hospital administration, and physician services--in which productivity is hardest to measure and in which most observers point not to U.S. competitive success--indeed, none are in traded goods sectors--but to the need for cost containment.

Supposedly, higher skills translate into higher productivity automatically. As evidence, proponents of this view point to the sharply higher return that accrues to years of schooling (and especially to advanced degrees) since the early 1970s and infer that education is key to higher earnings. The rising advantage that an MBA has over a high-school dropout, for example, is offered as proof that skill requirements are growing. The Clinton administration, like the Bush administration before it, also credits business complaints about shortages of skilled labor and about the difficulty of finding literate, numerate, problem-solving employees.

A look at productivity data by industry, however, actually finds faster productivity growth in sectors with a higher proportion of less-educated workers. Moreover, employers that complain about skill shortages usually turn out to be those unwilling to pay for the better workers.


Though it would shock most economists, major industries such as autos, computers, and appliances exhibit huge productivity differentials--of as much as 40 percent--across companies and even across nearly identical plants within and among companies. Fully half of the roughly 25 percent productivity gap between Ford and General Motors plants is explained by differences in capacity utilization. In short, demand explains productivity and productivity growth at least as much as the quality of factor inputs such as labor.

This is true of national industries as well as countries. As Notre Dame economist Candace Howes has argued, in nearly every case that Clintonomics cites as examples of the success of the skill-productivity-wages model--auto and consumer electronics in Japan, machine tools in Germany, ceramics in Northern Italy--one also finds rapid demand growth, and that domestic demand growth benefitted from (explicit or tacit) trade policy favoring domestic producers.

Why did wages and productivity converge by country, and to a lesser extent by industry, during the postwar boom? To some extent, this was a fortuitous convergence of historic forces. The enormous jolt to demand caused by World War II (in the U.S.) and the stimulus of postwar reconstruction (in Western Europe) coincided with the stability afforded by U.S. economic hegemony and the Bretton Woods system, coupled with the institutional power of unions and the prestige of national economic planning in most of the West.

In a marketized global economy, however, there's no reason to think that wages and productivity will continue to move in lockstep. Indeed, recent events suggest just the opposite. Consider the remarkably small productivity gap between Mexican and U.S. auto and electronics assembly plants, contrasted with a wage gap of at least seven-to-one. Even without NAFTA, investment in both industries is tilting toward low-wage Mexico. That puts downward pressure on wages and hence demand here, causing U.S. plants to run below full capacity. That further reduces their productivity advantage, which in turn speeds the southward shift of investment.

With weak demand at home, smart companies continue to skimp on capital investment--the real source of high-wage jobs. Indeed, for all the talk of "knowledgeware" and human capital, the statistical link between physical capital spending per employee and wages is so strong that there's not much left for productivity, education, or anything else to explain. For all the talk of Japan's "lean production," its manufacturing sector invests 60 percent more per worker than ours. The trick is that they do it in the smaller supplier shops, which are much more modern than in the U.S., where large firms relentlessly squeeze their smaller suppliers, which in turn squeeze their workers rather than invest in modern equipment and methods.

As for training and education, while no one should be against people knowing more, consider that Taiwan, Mexico, and South Korea all offer better educated entry-level manufacturing workers today than the U.S. We probably could, and should, catch up by the year 2000, but it's highly unlikely that we could make the skills of our work force an absolute competitive advantage, especially relative to wages.

Improved labor quality is a worthy goal, but only large and effective national investment could measurably improve it in four, or even ten, years. In any case, as Ira Magaziner's oft-cited report, America's Choice: High Skills or Low Wages! Report of the Commission on the Skills of the American Workforce, reminds us, high-quality labor is in record low demand by a private sector that has found the low-skill, low-wage recipe to be at least as profitable, and certainly less risky and expensive, than a high-skill, high-wage one.

Here again, faster demand growth is the essential missing ingredient. Increased domestic demand would tighten labor markets, driving up wages. Employers train workers that they have to pay well, and keep paying them well to avoid losing them to competitors. Without strong demand, there is no way to reconcile higher productivity and more jobs.

For the short term, we need massive public works spending to stimulate aggregate demand and aggregate investment. This will require a willingness to stand up to contractionist demands for budget-balancing and inflation-fighting. Once the economy is recovering, steps must then be taken to keep the increased demand from leaking out of the country through the purchase of imports. This will require a much tougher trade policy, either to protect domestic production or, at the least, to assure reciprocal market access so that roughly as much foreign demand leaks in as domestic demand leaks out.

If robust demand growth can be restored and, through trade and procurement policy, substantially reserved for domestic products (or based on strict reciprocity with other nations pursuing similar demand policies), then it begins to makes sense to also worry about longer-term supply side initiatives. But these should focus on strategies for upgrading the skill and pay level of jobs--rather than on presenting "better" employees for the shopping pleasure of stateless private corporations.

It is in this context, and not just as a matter of equity, that income distribution matters: our increasingly unequal division of the pie not only sucks in luxury imports but also shifts domestic demand from mass-produced goods made by productive high-wage workers to niche goods made by lower-wage workers. To call for an agile, flexible, lean, post-mass production supply side is to celebrate (albeit generally in ignorance) the inherited Republican income distribution. Of course, a combination of high-growth macroeconomic policy and strong unions could raise wages in niche production as well. But needless to say, the U.S. has neither.


To know where good jobs come from one must pay some attention to where they are--and aren't. We know that formerly stable industrial oligopolies were the source of most good jobs that didn't require a college education. We know that they and their suppliers were able to reap "increasing returns" by tapping rich metropolitan skilled labor markets, and using a high-wage, high-productivity, high-investment recipe. We know too that, over the past 20 years, such firms have faced stiff competition from smaller companies that use a much less pleasant recipe. Since the pay and productivity gaps between small and large firms are growing, and since real manufacturing investment and worker earnings are lower than 20 years ago, we know that these low-wage, low-investment firms are winning in the marketplace.

Nurturing metropolitan manufacturing clusters and designing policies that express the goal of having more of our high-wage, high-productivity firms succeed should be the hallmarks of a "good jobs policy" (which is also a better label than "industrial policy"). This has implications for policy on labor law, trade and foreign direct investment, training and education, the cities, defense conversion, investment incentives, health care, and more. As of now, those that understand this are in a distinct minority within Clinton administration policy circles. As a result, labor law reform is too often seen as a special-interest union issue, trade policy as a sop to labor and a few hard-hit industries, training and education as addressing spot shortages on behalf of employers, defense conversion as a retraining or distress issue, and investment tax credits as a way to reduce the cost of capital.

If market-driven training won't get us a high-skill, high-wage work force whose higher productivity begets rising wages and a growing economy, what could? Besides programs to boost demand sharply, I would offer these elements:

Labor law reform. Far from being a "labor" issue, it is, in fact, essential to stem the loss of good union jobs to both union-busting and union-avoiding low-wage competitors. (See Richard Rothstein's "New Bargain or No Bargain?" p. 32.) Union-busting and union avoidance are temptations that block the path to a high-wage economy.

More activist trade policy. Without trade reciprocity, much of higher demand just "leaks out"and sucks in imports. The firms that offer good jobs along with the unions that represent many of their workers should upgrade skills and job content (see below). As for foreign direct investment in the U.S., public policy should encourage such investment only when it promotes high-wage, high-value-added domestic jobs, and only when it supplements rather than displaces existing capacity. Federal policy could pre-empt the current self-defeating bidding war among the states to subsidize newly built facilities.

Training and education. These should provide workers with the skills they need to get jobs with high-wage firms. At the same time, policy should make it costly for employers to pursue the low-wage path. Training incentives should aim to reduce the after-tax cost of training high-wage workers but sharply increase it for employers of low-wage ones. For example, a training levy could tax only on the first $10,000 per year of payroll per employee; or employers could gain a credit or subsidy only for training that upgrades pay and skill levels, as the successful California Employment Training Panel program does. Low-wage employers are the problem, not a constituency to be served.

Conversion. Defense conversion should focus on helping high-wage defense subcontractors convert to making products now made by low-wage commercial firms. Otherwise, these defense shops will compete with high-wage commercial shops, starting a wage-cutting war for business--the worst outcome at all. The engineering talents of the defense-industrial base should be focused on inventing new, higher technology "recipes" for redesigning and making what are now industrial commodity goods. It's a tall order, but with trade and procurement policies that reduce the pressure from imports, large corporations ought to be more willing to look at suppliers that compete on engineering, quality, and delivery rather than just low price. Idled defense industry engineers also should be tapped as field agents for the growing network of manufacturing extension agencies.

Tax policy. Investment tax credits, if enacted, should disproportionately benefit creation of high-wage jobs. Why should the Treasury forego needed revenues to subsidize the creation of bad jobs? One approach would be to make the full credit available only to firms with annual hourly worker wages exceeding, say, 125 percent of the average. Using annual rather than hourly wages as the trigger also helps deny the credit to firms that rely heavily on part-time labor.

Health care reform. Finally, health care reform should have three goals, of which the Clinton team's calls for full coverage and cost containment cover only two. The third goal is to overcome the huge disadvantage that multi-payer, employer-financed health care places on mature firms, which provide most of the good jobs. Older companies that have a high ratio of retirees to active workers are only helped by a system that severs the link between employment and medical care. If "managed competition" permits large companies to opt out, those that do stay in the main pool would be doomed to higher costs. A government-as-single-payer system would socialize fully what might be termed the costs of maturity. If the Clintons follow a managed competition model, it should address the burden now carried by mature firms with enormous retiree health costs--and do so without degrading the health insurance of retirees.

Unlike the Republican economics it replaces, Clintonomics does not proceed from deeply held principles but instead from good intentions combined with questionable intuition. Clintonomics risks failure unless its sponsors rethink many of its elements. At the same time, the administration's fundamental pragmatism should make its adherents open to friendly criticism. In sum, the terrain remains contestable; but time is short.

You may also like