Work-sharing approaches: past and present.
Initial policy development on the concept in the Unitied States began in 1974, in the Office of the Secretary of Labor, as the recession of that year worsened. However, work sharing, or reduced hours of work without the short time benefit, is not new--there was extensive experience with it in the Great Depression. Although work sharing in the Great Depression involved a much different set of economic circumstances than modern-day recessions, it is useful to understand the Nation's early experience with work sharing because it has left an emotional legacy of ambivalence that affects even today's perceptions of short-time compensation. A comparative view
For example, one feeling expressed is that work sharing was tried by President Herbert Hoover and is no better an idea now as short-time compensation than it was then as work sharing. The comparison is instructive. Critics felt that work sharing under Hoover represented an attempt to avoid fiscal or monetary Federal intervention as well as to avoid public assistance. Instead, voluntary employer action was encouraged in the form of work sharing, not only to spread the work but to do so without cutting hourly wages. (Hoover felt wage cutting would compound the problem.)
Such work sharing (usually imposed by the employer) subsequently came to be seen by labor as a poor alternative President Franklin D. Roosevelt's later New Deal measures. Short-time compensation, the current form of work sharing, is a supplement, not a replacement, for macroeconomic policy, transfer payments, and social insurance. There are other differences. President Hoover's work sharing, sometimes used through the early New Deal years, often involved working half time simply because output was so low. Short-time compensation does not permit employees to work fewer than 3 days a week and has typically involved 4 days. Work sharing was oten at poverty-level weekly earnings: there was no minimium wage. Indusrial wages are incomparably higher today. Work sharing was often in unorganized plants: the National Labor Relations Act had not yet been enacted, so unions had minimal power. Today, roughly half of all manufacturing sites, where work sharing has its greatest potential, are organized, and unions would have to agree to short-time compensation. And, of course, the Hoover approach did not include partial unemployment insurance, as does the current concept of short-time compensation.
Despite these differences, work sharing under President Hoover did save jobs. It seems certain that manufacturing employment might have dropped more than it did in the short term if the workweek had not been sharply reduced from 44.2 to 38.3 hours during 1929--32. This was a 13-percent drop, accompanying a 33-percent drop in employment. Because total production decreased by 48 percent, it seems evident that a larger downward adjustment of labor than 33 percent was needed in one form or another. In his memoirs, President Hoove said 2 million workers had been helped by either work sharing or private relief by employers. Of course, weekly hours would have dropped regardless of President Hoover's efforts. However, it is unlikely that hours would have dropped so sharply. The lower fixed costs of that period facilitated work sharing, of course.
The fact that Federal-State unemployment insurance did not exist at that time not only had dire human consequences but also precluded the countercyclical use of unemployment insurance to offset part of the purchasing power lost by both the fully and partially unemployed. Second depression effort
The second big work-sharing effort came in mid-1933, 6 months into President Roosevelt's New Deal. The success of voluntary, private work sharing in providing some visible relief had led to demands for more of the same but without weekly pay reductions.
Where President Hoover had tried to prevent the loss of some jobs by persuading industry leaders to cut hours, President Roosevelt tried, with some success, to reemploy many of those who had lost jobs by cutting hours still further and estblishing minimum wages. His goal was to increase purchasing power while spreading the increased work--in a deflationary, not inflationary, economy.
The National Industrial Recovery Act, enacted in 1933, was an attempt to increase production, prices, and employment by increasing labor protction and reducing price-cutting competition. The act created the National Recovery Administration and the Public Works Administration. The act lasted only 2 years, because it was ruled unconstitutional in 1935. Under the National Recovery Administration (NRA), which administered part of the law, business adopted voluntary codes, including minimum wages and maximum hours. These foreshadowed the Fair Labor Standards Act of 1938.
The NRA helped decrease the workweek to 34.6 hours in 1934--it was now 22 percent below pre-1929 levels. This figure reflected a reduction in hours in many low-wage, soft-goods firms from 50 to 60 hours to 40 to 48 hours, and even fewer in some higher-wage, durable goods industries. Higher hourly productivity from less fatigued workers, more efficient use of workers, and increased plant utilization (for example, two 8-hour shifts, rather than one 10-hour shift as output expanded) accompanied these hours cuts.
The ratio of jobs to production was increased in part because of NRA workweek reductions, which took effect in mid-1933. The ratio of employment of production was quite low in the pre-NRA upturn of March through June 1933 and much higher in a similar upturn in early 1934.
Thus, in May 1933, 2 months after a sharp upturn in production, the ratio of jobs to output was about .78. In February 1934, 2 months after an upturn in production, this ratio was about .93. This 19-percent increase in the number of jobs created per unit of output was due in good part to the 12-percent decrease in average weekly hours during this period. (The output levels were also about the same in February 1934 and May 1933.)
Although it is a subject of debate, some economic historians credit the NRA with significant job creation due to work sharing, even while faulting it on other economic and constitutional grounds. Work sharing phased out
However, the work-sharing effect faded as recovery continued. Because of weak enforcement of the NRA, its many exemptions, and finally its demise in 1935, average weekly hours had moved back up to 38.6 by 1937, reflecting hours well above 40 in some firms and much lower in others. Partly as a result, the accelerating increases in output between 1934 and 1937 were accompanied by decelerating increases in employment. While rapid increases in wages may have been one reason behind this increasing gap between output growth and employment growth, longer hours also seem to have contributed.
By 1937, output was back up to its 1929 level, and employment was almost so; however, because of the steady growth of the labor force, about 21 percent of the nonfarm labor force were still unemployed in 1937. In the 1937-38 "Roosevelt" depression, weekly hours again dropped sharply. This occurred once again in 1948-46, as the United States demobilized. Since 1945, work sharing on such a national scale has not been used.
The Fair Labor Standards Act was not passed until 1938; like the NRA, it contained a work sharing measure in the form of an overtime penalty for weekly hours more than 40. The original 1938 ceiling was 44 hours; the 40-hour week was not phased in until 1940. The effect of work sharing was submerged by the oncoming full employment of World War II.
If work sharing had some beneficial effects during the Depression, why are there some negative memories of it, even under President Roosevelt? One reason is that neither the Hoover nor Roosevelt administrations used modern-day fiscal and monetary measures in a consistent way to deal with the massive unemployment they faced; as a result, work sharing in the 1930's was given a role it could not fulfill.
To some, President Hoover's work sharing attempts also symbolized cuts in earnings and the failure of voluntary, private sector-oriented policies to deal with the Great Depression; under the The early New Deal, work sharing symbolized to some the unconstitutional and big-business oriented approach of the NRA codes. Moreover, most of the early major experiments in work sharing occurred before passage of the Social Security Act of 1935, which brought with it mandatory unemployment insurance, and the 1935 Wagner Act (the National Labor Relations Act), which gave unions a legal framework for organization (although the NRA also provided the right to organize). With strong unions came strong seniority systems, not only to protect workers against arbitrary dismissal by employers, but to protect them against unilaterally imposed work sharing, for it was the practice of many employers not to guarantee a steady amount of work from 1 week to the next. Employees often showed up at their jobs only to be told there was no work that day. With unemployment insurance came the assurance that low-seniority workers would not starve if they were laid off, and that work sharing, which only "spread the misery," would no longer be needed.
The current use of work sharing, on a micro rather than a macro scale, is taking place within a framework of basic protections for workers, unlike earlier efforts. However, the full economic effects of short-time compensation, which is a preventive rather than a reemployment measure, in a completely different economy, more than 50 years later, have yet to be determined. The revival of work sharing
With the relatively low unemployment rates of the post-World War II era, work sharing was rarely discussed or used. It was not until the 1974-75 recession, at the time of the steepest downturn since the Depression, that work sharing began to be considered again.
In a paper in 1976, I wrote:
Two major conclusions can be drawn from a comparison of 1974-75 European with U.S. experience: (1) The portion of unemployment that takes the form of part-time unemployment is higher than in the United States. The result would seem to be decreased social costs, increased purchasing power and greater equity, compared to the United States. (2) The number of U.S. workers who were put on part-time unemployment even in the absence of partial [unemployment insurance] benefits is nevertheless not trivial. This suggests that the potential for more work sharing is significant if European-type incentives were instituted.
These conclusions had been reinforced by the New York City Conference on "Alternatives to Layoffs" held in April 1975: representatives of labor, management, and academia reviewed alternatives besides work sharing and found them wanting. Some firms reported mandatory cuts in pay, but there was resistance by labor. Cutting health and welfare benefits was ruled out. Voluntary furloughs were found effective by some firms, but they appealed mainly to younger, education-minded workers; older workers nearing retirement; and some working mothers. There was also disillusionment about early retirement, due to inflation.
Work sharing was found to be more effective than these other alternatives. However, the case studies presented at the conference showed that work sharing without government incentives was usually atypical.
In fact, an underlying crisis for the firm--whereby its very existence was threatened--was a common theme in bringing about work sharing. This was true of Pan Am and the Washington Star (the Star did go out to business eventually). Union leadership also had to be unusually good in terms of communication with rank and file. (Once unions were convinced the crisis was real, there were often unusual efforts by union leaders to get the rank and file to discuss alternatives to layoffs in meetings and votes.)
The firms were often marked by an unusual degree of labor-management cooperation, with management often opening its books. Pan Am went "beyond union contract requirements to develop worker involvement difficult decisions."
Nor were the firms especially typical of the average work force. Highly skilled workers were often involved, such as Pan Am flight crews or Newspaper Guild members. It was in the company's interest that young, highly trained people not be lost. There was a team spirit--born of the flight cabin or city room--among the workers. Large numbers of women, many of them the family's second earners, may have also facilitated work sharing in firms such as the New York Telephone Company.
The question was how to create incentives to encourage work sharing in more typical layoff situations as well as in those with the unique chemistry described above.
The New York conference found that work sharing in the form of "a shorter workweek, or rotating and staggered shifts, or any other method by which average work hours are reduced" emerged as the "alternative to layoffs with the widest potential application to recession--based economic problems and to almost all types of business and industry." It also found, however, that work sharing is "not a panacea. Its use is limited by the necessity of providing a living wage." Thus, the conference found that anything more than a 20-percent reduction in hours would create too much hardship. Nor would it work when an entire shift must be eliminated, or conversely, only marginal reductions are contemplated--for example, work sharing would not succeed for 20 of 1,000 employees. Short-time compensation in the 1980's
The early 1980's have been a period of anti-inflationary restraint, in which planned use of macroeconomic "finetuning" through countercyclical monetary and fiscal policy has been more limited than in the past. Even those who favor micro job-creation tools, such as public jobs programs, usually advocate targeting them to the structurally unemployed--the disadvantaged and long-term unemployed. The disillusion with countercyclical public policies may argue for at least experimenting with policies for the cyclically unemployed that are rooted in the private sector and based to some extent on redistribution of employment rather than solely on countercyclical economic stimuli and public spending.
Moreover, while past efforts to deal with cyclical unemployment have included large public jobs programs, expanded budgets, tax cuts, or new investment, these solutions have not usually had an early impact on recessions or acted as preventatives. To the extent that they have been successful, it has often not been until after, rather than before, layoffs occurred. Job saving has not been a feature of such policies, as it is of short-time compensation.
Equity is the major benefit of short-time compensation. The economic and social costs of full-time unemployment are distributed more evenly across all workers in a plant (or plant unit) rather than among a small minority of workers.
Some economists have argued that it is the total decline of hours of employment that counts, not its distribution. They see work sharing as a "diversion," a waste of time and resources that could be spent on other countercyclical measures.
However, the may be ignoring the social costs of full-time unemployment, which increases the costs of public assistance, food stamps, and other transfer programs during a recession. Many studies suggest that full-time unemployment also increases the incidence of alcoholism, drug abuse, child abuse, and other social problems, which translate into additional public costs, human costs, and suffering. Distributing the same total hours of unemployment among many people on a 4-day workweek may decrease the social costs. It might also help public policy deal in a more rational way with the problem of health insurance for the unemployed, because workers now often lose their health insurance soon after layoff. And, if it were ever adopted on a wide scale, it might also redistribute work and income in a way that bolsters confidence and slows down the decline in consumption during a downturn.
Short-time compensation might also help provide a framework work for developing constructive activities, such as education and training, during a downturn. It is unrealistic to think that all workers in a work-sharing program would meaningfully enroll in education or training. However, it would be productive for some. The broad distribution of downtime among the work force would also enable employers to provide training on a part-time basis to any workers they need it, not just those laid off. (Such training would have to be voluntary on the part of workers, of course.) Public-private mechanisms under the Job Training Partnership Act might conceivably be used for potentially dislocated workers.
In general, the meshing of short-time compensation with retraining and education is an area deserving further thought. With hundreds of community colleges and technical schools now operating throughout the Nation, it is possible to imagine large numbers of workers who are put on 4-day weeks or 6-hour days for a 2- to 6-month period, using that time to attend classes. Work sharing in Germany versus the U.S.
Some economists have expressed fear that use of work sharing will lead to a hoarding of underutilized labor and thus lower productivity. The following discussion suggests that short-time compensation may not only decrease layoffs but also may improve cyclical productivity. These examples are illustrative; more in-depth research is needed on these and other issues. The following tabulation shows percent changes in economic indicators for manufacturing and for the mechanical engineering sector in Germany, 1981-82:
Because the mechanical engineering sector used short-time compensation more heavily than did manufacturing industries as a whole, we would expect mechanical engineering to show a much heavier use of short weeks and, thus, less decline in employment. Indeed, while mechanical engineering reduced total hours about the same percentage as did manufacturing, it reduced employment much less than manufacturing. Average hours declined twice as much in mechanical engineering, and mainly reflect changes in weekly hours.
Output per hour (productivity) increased by the same percent in both cases bacause total hours were cut back faster than output. In mechanical engineering, this productivity increase was partly because of work sharing augmenting layoffs.
Is the job-saving effect as dramatic as it seems at first glance? Mechanical engineering has more skilled workers than the average manufacturing industry, and is less labor intensive. Cost savings from layoffs might be less feasible, making layoffs less likely. Moreover, employers face a greater risk of permanently losing skilled workers. So it is not clear that the mechanical engineering sector would have lost 2.0 percent more jobs in the absence of a 2.0-percent workweek reduction. Without short-time compensation, there might have been more hoarding of labor.
Nevertheless, the figures suggest significant job-saving effects from work sharing, without the productivity loss that hoarding of full-time, underutilized workers brings in the United States. The following tabulation shows percent changes in economic indicators for the mechanical engineering industry in Germany and its counterpart industry in the United States, nonelectrical machinery, 1974-75:
The U.S. industry did hoard more labor relative to Germany in the absence of short-time compensation. Total hours did not decline as fas as output in the United States, whereas it decreased faster than output in Germany. Part of the reason was that average weekly hours decreased by 5.3 percent in Germany, compared with a 1.8-percent decline in the United States; when combined with the reduction in full-time employment, the totals were 9.3 percent for the United States and 9.5 percent for Germany. Thus Germany reduced total hours relatively more, even though it reduced the number of employees relatively less. Because output declined approximately 4.9 percent more than total hours in the United States, but 4.3 percent less than total hours in Germany, the change in output per hour was negative (-4.9) percent for the United States and positive (4.3 percent) in Germany during the 1974-75 period.
The this United States-Germany productivity gap is "artificially" widened during a downturn is evident from the fact that the U.S. rate of productivity increase was actually higher (4.0) in the overall growth period, 1969-77, than the German rate (3.3) in an essentially comparable period, 1970-78. These are the rates that measure the real differences in technology and other efficiencies between the same industry in the two countries. The 1974-75 gap, therefore, was partly because of the added flexibility in hours cuts afforded by heavy use of short weeks. (Comparative data for the 1982 downturn is not available.)
These data suggest that work sharing may bring with it more total hours unemployment in Germany, even while decreasing layoffs, because employers can not only eliminate some jobs but also work some of the remaining employees on a part-time basis. However, some of Germany's decrease in employees should be discounted because it reflects continuation of a longer-term trend of sharply shrinking employment in manufacturing, unlike in the United States. Also, germany's lack of experience-rated tax contributions by employers (different from that in the United States) may induce some added hours of unemployment because employers do not bear the added cost. All these factors may contribute to a "surplus" or induced unemployment effects, whereby not all hours of work sharing are substituted for jobs saved, but instead may be in addition to layoffs. Nevertheless, as long as there is some appreciable effect on layoffs, the social costs of such "surplus" work sharing may be smally when compared with the benefits of lewer layoffs and higher cyclical productivity. The latter brings with it less increase in unit labor costs, and thus less increase in prices, which stimulates demand, speeding economic recovery.
This may even have implications in terms of international competition. The mechanical engineering industry is export-oriented.
Short-time compensation probably helped German manufacturers in the mechanical engineering industry to compete with U.S. manufacturers in the nonelectrical machinery industry during the 1974-75 period. As demand declined, the Germans could muster both heavy work sharing and some reduction in force to maintain productivity, allowing them to retain skilled personnel without adding to the unemployment insurance taxes. U.S. manufacturers not only faced higher unemployment taxes for whatever layoffs occurred, but also had less flexibility to maintain productivity through work sharing as a supplemental labor adjustment tool.
|Printer friendly Cite/link Email Feedback|
|Publication:||Monthly Labor Review|
|Date:||Sep 1, 1984|
|Previous Article:||Worker participation and productivity change.|
|Next Article:||Pay in data processing services by occupation and urban area.|
|A capital markets approach to mass tort bankruptcy.|
|Cognitive Approaches to Human Perception.|
|ViewpointsResign, or issue your apology!|
|Researching social change; qualitative approaches.|