In 1983 and 1984 the United States was in the midst of a cyclical recovery that was being sold to the public as if it were permanent prosperity. While the selling job was successful, the “stable expansion” promised by the administration looks more and more like a mirage. Almost as President Reagan’s election tide rolled in, a tide of the bad economic news rolled out.

Just after the election Budget Director Stockman opened his books and revealed that the budget deficits were going to be much larger than was previously supposed. The Commerce Department revised real economic growth estimates downward to a 1.9 percent rate in the third quarter. If growth were to take place at such a pace through 1985, the budget deficits would be even larger than the new higher deficits being forecast by Mr. Stockman. With such a growth rate little further decline in unemployment could be expected. While 7 to 7.5 percent unemployment is better than the almost 11 percent of 1982, it is a dismal rate if one remembers that such rates were seen only in the depths of a recession. What used to be our worst unemployment rates are now our best unemployment rates.

Growth slowed in the third quarter of 1984 because American firms were not competitive on world markets. Two thirds of what Americans bought was produced abroad. Foreign sales to the US actually rose at a 5.7 percent rate. If America had been running a balance in its balance of trade, instead of a $130 billion deficit, three million Americans would be working who are not working in 1984. Those three million extra jobs would have brought the US a long way back toward a more reasonable level of unemployment.

While an inflation rate that seems to have stabilized around 4 percent looks good in relation to the recent past, it is also dismal viewed from a somewhat longer perspective. Such rates in the late 1960s produced demands for wage and price controls to end what was then viewed as an intolerable rate of inflation. Such rates are now the best that the economy can achieve right after it has been put through the wringer of almost 11 percent unemployment and four years of no growth from the first of 1979 through the fourth quarter of 1982. Historically, 4 percent is a very high jumping-off point from which to commence the next round of inflation.

Moreover the clouds of just such an inflationary shock are piling up. Funds must be borrowed from abroad to finance America’s trade deficit. At some point the lending stops. No country can forever borrow to finance a deficit in its balance of payments. No one knows when the end will come, but it will come. And when the lending stops, the dollar falls.

Econometricians estimate that the dollar would need to fall about 30 percent to restore balance to America’s balance of trade. But such a fall means that the price of imports measured in dollars would rise approximately 30 percent. Since America imports 12 percent of GNP, simple multiplication reveals that if 12 percent of GNP becomes 30 percent more expensive, then the inflation rate would rise by 3.6 percentage points.

In addition there would be an indirect inflationary effect. Part of today’s relatively low inflation is produced by competitive pressures from abroad. American firms cannot raise prices without losing a share of the market. If imports were to go up in price because of a falling dollar, however, American-made goods competing with imports would also likely go up in price. This indirect inflationary effect of a falling dollar could easily be as large as the direct effect.

In view of the intrinsic instability created by the trade deficit, a rapidly falling dollar could act like the lightning that precedes a thunderstorm; inflation could well return to double-digit rates. What would happen then? The Federal Reserve, we may expect, will return to its habit of fighting inflation by deliberately creating recessions—a recovery in that case becomes a “contraction.”

The lack of competitiveness in international markets produces unemployment and the prospect of an inflationary shock, but it is a symptom of a much more fundamental disease—a very anemic rate of growth in US productivity. This general failure of American industry to increase per capita output has been strangely neglected in recent public discussion. Prominent politicians do not say, and do not seem to know, that during the six years from 1977 through 1983, productivity in American manufacturing grew one-half as fast as that in Germany, one-third as fast as that in France, and less than one-third as fast as that in Japan. Outside of manufacturing, the American performance was even worse.

During the 1984 election the Reagan administration pointed to the rebound of productivity in 1983 as evidence of a new long-term trend. In fact the 1983 results were nothing but the standard cyclical swing: during a recovery, labor costs are spread across more units of output and productivity temporarily grows. Right after the election the Labor Department announced that American productivity was again falling. At the end of the third quarter of 1984 the average American firm was a little less efficient than it had been at the beginning of the quarter. What had been anemic growth was now actual decline.


Nothing is more important than restoring growth in American productivity, yet nothing is being done to do so. Nothing could help more to restore productivity growth than the elimination of frequent and persistent recessions; but nothing is being done to change the structural characteristics of the economy so as to get America off its up-and-down roller coaster. Why?

Some answers are to be found in Martin Weitzman’s The Share Economy and Piore and Sabel’s The Second Industrial Divide. The Reagan administration and most of the economics profession believe that social institutions and structural arrangements for productive work either don’t matter much or take care of themselves. “Get the government out of the economy and off the backs of the people.” Underlying this political slogan is the belief that competition forces the best possible institutional arrangements to the fore. “If it wasn’t efficient it wouldn’t exist.” “If it does exist, it must be efficient.” “If there was a better way to do it, that better way would automatically drive the inferior way now in use out of existence.” Therefore, societies don’t have to make deliberate social changes in the ways in which they organize themselves. Instead of trying to improve the ways that society is organized so that it works better, as President Roosevelt did during the Great Depression. Americans have merely to stand aside and “let free enterprise do its thing.”

The public will reject this argument when they see that it isn’t working. What has to be seen to be believed will, I think, soon be visible, however painfully. When recovery turns to recession, we may expect that people will want to know why a system that is supposed to work according to automatic assumptions is not working, and how that system might be changed. Then the views of Weitzman and of Piore and Sabel may get serious attention.

They all share some unfashionable premises that can be stated bluntly. Social organization matters. The most efficient forms of social organization do not automatically come forward. Societies can consciously organize themselves efficiently or inefficiently. The societies that win economically are the ones that pay attention to improving their social organization. Efficient social organization will usually beat inefficient social organization; efficent organization may often be found in Japan while inefficient organization prevails in the United States.

Weitzman examines the structural changes in social organization that he believes would be necessary to run the American economy at continuing full employment without inflation. Piore and Sabel examine shifts in modes of social organization that would be necessary to run firms more efficiently. Weitzman’s thesis should logically come first. He starts with a simple sociological observation that derives from an underlying economic reality. In capitalistic economies, corporations that sell goods or services often send presents to their customers at Christmas time; in socialist economies the customers give presents to sellers. Why? In socialist economies the problem is production. There is a scarcity of goods at current prices and buyers need to ingratiate themselves with (one might say “bribe”) sellers. In capitalistic economies the problem is sales. Since there is a surplus of goods at current prices, sellers need to ingratiate themselves with (bribe?) buyers.

The reasons for this can be stated simply. Modern advanced capitalism is marked by imperfectly competitive firms. Most firms are sufficiently large that they cannot sell all that they can produce at current prices; they must lower prices (or spend more on selling) if they are to sell more of what they produce. This means that the extra revenue (what economists call marginal revenue) from an additional unit of sales is less than the price previously received (what economists call average revenue). But if society wants firms to expand their production to full employment levels, then production costs must also fall as sales expand. If production costs do not fall as output expands, firms will not expand production since to do so would be to lose money. The falling revenue per unit of extra sales does not cover the cost incurred in producing more output.

Since labor costs typically account for 70 percent of total production costs, labor costs must fall if total production costs are to fall. In the textbook world of classical supply and demand they do so. Full employment is guaranteed by falling wages. If unemployment breaks out, unemployed workers will bid for current jobs by offering to work for less than those employed and so they drive wage rates down for those employed. With lower wage costs employers will expand production. With production expanding unemployed workers will be reabsorbed into employment. In a perfectly competitive economy in which every market worked as it is supposed to, unemployment would not exist and if it were to occur for some unknown reason it would be quickly absorbed by the expanding demand flowing from falling wages and lower production costs.


The unfortunate reality is that unemployment does exist and wages do not fall so as to reduce production costs. Despite unemployment rates ranging from 25 percent in 1933 to 17 percent in 1939 real wages rose during the Great Depression for the lucky ones who remained employed. Between 1979 and 1982 average labor compensation rose 30 percent despite an unemployment rate that almost reached 11 percent. Wages should have been falling to reduce unemployment, but they weren’t. Perhaps if the US had been willing to let the Great Depression run for more than ten years and were willing to let the recent recession run for more than four years, wage rates might have started to fall—but that remains in question. In neither case were there any signs that the wage structure was about to crack. Within any reasonable period of time, wages do not fall to the level required to produce full employment.

In view of this reality Weitzman argues that the US needs to change its social organization. It should move from the current system of fixed wages (so much per hour, week, or month) to what he calls a share economy where some portion of wages is paid as a fraction of the firm’s profits or revenue. In a share system, if revenue per unit falls as sales expand, wage costs also fall per unit produced, and the firm has an economic incentive to increase production to the level of full employment. Until the firms run out of extra workers to hire they will find that more production and sales always mean more profits. A share economy would eliminate unemployment since wages would respond as they are supposed to do and not as they have done throughout the postwar years.

Similarly inflationary shocks would be mitigated. At the beginning of 1979 the US spent 2 percent of its GNP on imported oil; at the end of 1979 it spent 5 percent of its GNP for the same amount of imported oil; and during 1979 American productivity did not grow. As a result the pile of goods and services available to be divided among Americans at the end of the year was only 97 percent as large as that available at the beginning of the year. There were two ways to deliver the bad news. If all wages (and other incomes) had fallen 3 percent, there would have been a balance between American incomes and the pile of goods and services available to be divided. With wages falling 3 percent, American production costs, and hence prices, would have fallen enough to offset imported oil prices in the consumer price index. But wages did not fall 3 percent. They rose 9 percent instead as each worker attempted to keep up with inflation. This compounded the original problems and produced a 12 percent rate of inflation. While the attempt to keep up with inflation was rational for each individual worker it was socially irrational (mathematically impossible) for workers as a group. The net result was not just a higher than necessary rate of inflation but the subsequent need to have a harsh recession to push that 12 percent inflation back down to more reasonable levels.

If Weitzman’s share economy had been in place in 1979, wages would have fallen as revenue fell because of higher imported-oil prices. What should have happened would have happened. The Japanese have a share system that works according to the size of the bonuses paid out twice a year. And the Japanese, as it happens, sailed through the second OPEC oil shock with hardly a tremor of inflation. Bonuses were drastically cut, producing the equivalent of an 11 percent wage reduction.

If what Weitzman proposes seems so sensible, why don’t we try it? Part of the answer is to be found in the dominant view that social organization does not require deliberate change; but another part is to be found in self-interest and an unwillingness to confront reality. A share system evidently hurts those who remain employed during periods of high unemployment. If unemployment exists because wages on the margin are too high, those already employed will receive less under a share system than they do now. Their share of extra revenue will not be as high as the wages they now get.

As with most problems, this one becomes clearer when seen in someone else’s back yard. In Belgium every employee must by law be given six weeks of paid vacation. Such laws are fine for those employed. But they raise the costs of labor and make it unprofitable for employers to hire new employees. The workers who are unemployed pay, in a very real sense, for the six weeks of paid vacation for those employed. A share system makes life more uncertain for those who feel sure that they will not be laid off during an economic downturn. Under a share system they wouldn’t know quite so confidently just what their income would be in the year to come. Personal planning becomes marginally more difficult. Many union leaders, who would like to assure their members that life will become more agreeable each year, will resist a share system.

Under a share system, however, Americans will find it harder to kid themselves about reality. When an event like the 1979 OPEC oil shock occurs—and the fall of the dollar may produce a comparable effect—Americans will have to face the fact that their standard of living will fall as they give up more exports to get the old volume of imports. No one will be able to bail out of reality by thinking that he can get a wage hike higher than the rate of inflation.

The objections to a share economy are both minor and self-serving. A share economy would be a much better economy. It would have a higher standard of living, as well as lower unemployment and lower inflation, than we now have. Other societies have declined—Great Britain comes to mind—as they lost the ability to undertake social changes. America may be another. By doing nothing Americans may collectively choose to be among the inefficient.

Piore and Sabel start by pointing out that America’s lack of efficient economic performance cannot be blamed on any of the simple causes that are so frequently invoked. America has much lower productivity growth than many European countries and a huge deficit in its balance of payments with Europe. Yet Europeans spend more on government social programs than Americans dream of spending. America has much lower productivity growth than Japan and a huge deficit in its balance of payments with Japan. Yet Japan allows government interference in matters that Americans would not dream of. “Administrative guidance” is a way of life in Japan, where powerful corporations take advice from the ministry of trade and industry on what to produce. Most American executives would claim that such dependence is a nightmare.

The thesis of Piore and Sabel’s book is that the rest of the world is essentially beating America by doing what Americans like to talk about—being more flexible and more responsive to the competition—but have not been able to do. Americans cannot be flexible and responsive to rapid changes in world demand for products not because they are inflexible personally but because they have organized themselves socially in ways that make flexibility and responsiveness difficult. They have tied a large part of their economy to a system of mass manufacturing based on long production runs—i.e., on making much the same product much the same way over a long period—and to a legalistic division of labor in which each minute task is specified and becomes a specialty. In such a system, frequent changes of product and method become almost impossible. Japanese auto factories, for example, change from model to model, according to sales volume, much more rapidly than their American counter-parts. By cooperating with one another to help expand the production of “hot” styles Italian shoe manufacturers can respond to shifts in the market much more rapidly than their American counterparts.

With their thousands of work rules, job titles, and associated wage rates, American firms cannot move labor speedily from place to place and job to job as the Japanese can. Japanese wages are based on seniority with the firm. The workers are encouraged to learn different skills and are much more flexible in changing from one task or technique to another. With their militaristic ways of imposing decisions from the top down, American firms cannot compete with foreign firms who can tap the ideas, initiative, and experiences of their employees. Piore and Sabel find, for example, that many foreign firms have worked out decision making from the bottom up by using “quality control circles,” in which groups of workers collaborate in suggesting more efficient ways of organizing the productive process.

The economic culture in which the foreman acts as the workers’ representative to management (the typical pattern among our successful foreign competitors) is more flexible than the one where the foreman is the instrument of management for transmitting never-to-be-questioned orders to workers (the American pattern). “Locked” numerically controlled machine tools which only skilled white-collar technicians are allowed to program (the American pattern), are less capable of adjusting to bottlenecks on the shop floor than “unlocked” numerically controlled machine tools for which the blue-collar workers who operate the machines are taught to do the programming (the Japanese or West German pattern).

While none of these rigid practices is new to American industry, technology has made them into more of a handicap than they used to be. Computers, micro-processors, robots, and all that goes with them are radically changing the industrial world. They are moving production away from fixed mass manufacturing where one can compete successfully by having low-skilled workers on the assembly line turning the same screw year after year. Workers now have to be capable, for example, of readjusting the programming of robots so that they can shift from one model of car to another. And Piore and Sabel have much evidence to show that workers acting as participants are more efficient than workers acting as automatons. While American industry is rapidly rushing to purchase the hardware (robots, computers, etc.) that will allow more flexible production processes, few of these firms as yet have been willing to make the sociological changes that must go along with the hardware to make it fully effective. Old hierarchical arrangements have to be replaced with very different methods of supervising the labor force and getting them to participate in the decision-making process.

While it may be accurate in both industry and government to describe Japan and Europe pejoratively as more “bureaucratic” than America, an even more pejorative label can be attached to America—it is an overly legalistic society. “Administrative guidance,” by which government officials confer with managers and collaborate in setting new industrial trends, could indeed be a nightmare, but, as Piore and Sabel suggest, it can also be the saving element in a bureaucratic society. When such guidance works, one finds few formal rules, a need to consult before doing anything, many informal constraints, but everything is negotiable—and possible.

Legalistic societies by contrast attempt to write detailed regulations for everything (work rules, job titles, promotion up seniority ladders, who may give an order to whom, who gets the largest office, which jobs are associated with which wage rates) and then use lengthy adversarial quasi-legal procedures to settle disputes—arbitrators, special masters, trials. Very little is negotiable and much must be settled in accordance with some prior written code of rules and regulations. Much cannot be done and changing the rules is both hard and time-consuming.

One can disagree with Piore and Sabel on the degree to which consumer demands for mass-made manufactured products have been satiated. I would argue that stagnant sales are caused more by stagnant incomes than they are by satiation. When incomes rise and interest rates fall, the demand for these products seems just as strong as ever. But these products are themselves apt to be produced in more flexible ways in the future. Perhaps they also exaggerate the speed with which the shift from old-style mass production to flexible manufacturing is, or should be, taking place. One can also argue that nothing would improve American industrial performance more than an economy run without frequent recessions and therefore recommend that they take account of Weitzman’s arguments for a share economy. At the same time one can agree with their basic thesis. The American corporation has turned itself into a “do it by the rules” microcosm of our larger society, and that way of doing things just doesn’t work very well.

Still, the existing social organizations aren’t going to collapse automatically simply because they don’t work: they will go on producing a huge American product but less and less competitively. The underlying reality, which may or may not be confronted, is that the more uncertainty an economy faces, the more it needs flexibility. A world without OPEC oil shocks evidently had less need for a share economy than a world with them; just as American firms had less need for flexibility where they were faced with technically equal or inferior competitors. The world is now full of economic uncertainty, and America needs more flexibility than its institutions now permit. Whether its political and industrial leaders will be willing to recognize this remains the large, and largely unasked, question in the country today.

This Issue

February 14, 1985