Employment and Unemployment
Employment and Unemployment
Employment and Unemployment
The commercialization and industrialization of a large part of the world during the past several centuries have involved a radical change in the way in which human labor is directed toward productive ends. Two centuries ago, or more, most of the world’s work force tilled the soil, as is still largely true except in North America and Europe. This work was performed in good part under conditions of status, in which a man was tied to a particular place and particular job by institutional arrangements which offered an individual and his family a sense of stability and security.
Today, in the more developed countries and in the commercial-industrial sectors of the less developed economies, command over labor is acquired in the market place. The interaction of demand and supply in the labor market determines, within limits, the level of wages and wage differentials, the volume of employment—and the volume of unemployment. As the relative importance of agriculture and other forms of self-employment has declined, the interplay of demand and supply, in the economy as a whole and in labor markets in particular, has come to be crucial in determining both long-term trends and short-run fluctuations in the volume of employment and unemployment.
The personal insecurity and widespread hard-ship that frequently resulted from the impersonal market determination of the level of employment began to influence government policies in the most industrialized countries in the latter part of the nineteenth century and became increasingly influential in the twentieth (Beveridge 1909). But intense world-wide concern with the welfare implications of widespread unemployment did not come until the great depression of the 1930s. Out of the shock of that experience and the economic planning that followed came the modern concern with maintaining the level of employment. This development was greatly stimulated by the “Keynesian revolution” in economic thinking, which dates from the mid-1930s.
Today “full employment” is an almost universally espoused goal. Virtually every advanced country that depends to a significant degree on private markets consciously formulates and tries to implement an employment policy. This was not true before the 1930s. The Beveridge Report of 1944, Full Employment in a Free Society, represents a landmark in this development. For the United States a comparable landmark is the Employment Act of 1946. The international concern at the end of World War ii with the need to maintain a high and stable level of employment is suggested by the pledge, made by the governments subscribing to the United Nations Charter, to take action to promote “higher standards of living, full employment, and conditions of economic and social progress and development” (United Nations . . . 1949, p. 5).
The goal of full employment may conflict with other economic goals. Thus, full employment may not be compatible with the desired degree of stability in the price level, and at times governments have felt compelled to sacrifice some employment in order to correct a disequilibrium in the balance of payments. The question of how to reconcile these goals was being debated vigorously in the Western world in the mid-1960s.
Table 1 offers some historical background on the changing pattern of employment in broad industrial sectors. In the nineteenth century the shift away from agriculture had been most marked in the United Kingdom, and in the first decade of the twentieth century agriculture already accounted for less than 15 per cent of the British labor force. In the other leading industrial countries the agricultural sector of the labor force had already fallen below 50 per cent by the beginning of the twentieth century.
The shift of labor out of farming continued steadily during the first half of the twentieth century and has, indeed, accelerated since the end of World War H (Twentieth Century Fund 1961, pp. 72-74). The movement has been not only toward what is loosely called “industry” (manufacturing, mining, construction, and electric power and gas) but also into the trade and service sectors. This latter movement has been particularly marked in the United States. It is also noteworthy that the relative contribution of industry to total employment actually declined in the United States from 1950 to 1962. Involved here is not only the changing pattern of demand but also a more rapid rise
|Table 1 — Percentage distribution of labor force by industrial sector in selected countries, 1910-1962|
|a. Agriculture includes forestry and fishing.|
|b. Industry includes manufacturing, mining, construction, gas, electricity, and water .|
|c. Service includes trade, banking and finance, transport and communication, and all other services (including government).|
|Sources.1910-1950 data from Kuznets 1957, appendix, table 4; 1962 data from Organization for Economic Cooperation and Development 1963.|
|Table 2 — Percentage distribution of total employment by occupational classification in the United States, 1900-1963|
|* Includes farmers and farm laborers.|
|Sources: for 1900-1950, U.S. Bureau of the Census 1960; for 1963, U.S. President 1964.|
in labor productivity in manufacturing than in the trade and service sectors.
Another change in the pattern of employment involves a gradual shift to white-collar occupations. This change has been particularly notable in the United States, as indicated in Table 2.
Other changes in the composition of employment can be mentioned only in passing; for example, the decline in the importance of self-employment, the rise in the participation of women in the non-agricultural labor force, an upward trend in the age at which children leave school and enter the labor force, and earlier retirement for older workers [see LABOR FORCE]. These trends are expected to continue into the future.
The historical record. During the decade from the mid-1950s to the mid-1960s, unemployment in western Europe was at a lower level than in any earlier peacetime decade in this century. Data on unemployment for selected countries since 1913 are presented in Table 3. By the standards of recent experience, unemployment was high even in the 1920s in a number of countries, notably Germany and the United Kingdom. The catastrophe of the great depression is suggested by the soaring unemployment rates during the 1930s. In the worst year of the depression, a sixth of the German labor force was unemployed, an eighth of the British, virtually a fifth of the Canadian, and as much as a quarter of the American.
The record of the 1950s and 1960s provides a highly encouraging contrast. Since the mid-1950s unemployment rates in western Europe have generally been below 4 per cent, and since 1960 they have typically been below 3 per cent. In contrast, unemployment in the United States, which in the first postwar decade tended to run at a rate of 4 per cent or less except during brief cyclical recessions, failed to fall below 5 per cent in any year after 1957 (through 1964).
This is not to say that European countries have
|Table 3 - Unemployment rates in selected countries, 1913–1962° (per cent of the labor force)|
|a. The unemployment rates shown for the various countries do not all reflect precisely the same coverage and definitions, although some limited standardization has been applied to the figures. Thus, the figures shown here do not agree completely with those in Table 4, where, for a more limited period, complete comparability with American definitions and coverage has been attempted.|
|b. Not available.|
|Sources: For all countries except United States, 1913-1960 from Maddison 1964, table E-l; 1961-1962 adapted from Organization for Economic Cooperation and Development 1963; for United States, 1913-1960 from Lebergott 1964, p. 512; for 1961-1962, from U.S. President 1964, p. 195.|
|Year||France||Germany (later F.R.G.)||Italy||Sweden||United Kingdom||Canada||United States|
not had pockets of serious unemployment with which to deal. Various countries continue to have their depressed regions—the Italian Mezzogiorno and Northern Ireland, for example—and differentially high unemployment rates may exist in particular countries for teen-agers, the older age groups, or the least skilled.
The relatively high unemployment that persisted in the United States after the mid-1950s gave rise to a vigorous debate as to its causes and appropriate cures (Ross 1964). On the one hand have been those who argued that the high level of unemployment reflected primarily a deficiency of aggregate demand. The argument has run that a higher and more rapidly rising level of total demand, supported by appropriate monetary-fiscal policies, could increase employment to the point where unemployment would again be brought down to a rate of 4 per cent or less.
Opposing this view has been the structuralist argument that the main problem has been not a lack of jobs in the aggregate but a failure of the labor supply to adjust sufficiently to the rapidly changing pattern of employment opportunities (U.S. Congress, Senate . . . 1963, part 5, pp. 1461-1499). The changing pattern of the demand for labor—resulting from automation and other technological change and from the shift in the composition of output toward services—has greatly increased the demand for highly educated and white-collar workers and reduced the demand for blue-collar, unskilled, and poorly educated workers. And, the argument runs, it has not been possible to make the necessary adjustments in labor supply. As a result, the structuralist position has maintained, merely expanding aggregate demand would lead to labor shortages in occupations in which unemployment rates were already relatively low without materially reducing the rates among the groups that had the highest unemployment.
In the early 1960s the consensus among those who had studied the problem was that the high level of unemployment in the United States was primarily the result of the failure of aggregate demand to expand rapidly enough. There was general agreement also, however, that the heterogeneous character of the American labor force created a persistent structural problem of poor employment opportunities for the underprivileged parts of the labor force (Ross 1964; Gordon 1964). There was increasing recognition on all sides that there was need for a vigorous and comprehensive manpower policy to expedite the absorption into gainful employment of the least qualified groups in the labor force (U.S. Senate … 1964).
When is a person unemployed? There is no single unambiguous definition of unemployment. Different purposes call for different definitions. Further, the definitions that are implicit in the official figures published in different countries may vary because of the way in which unemployment statistics are compiled.
The difficulties involved in arriving at a generally accepted definition of unemployment can be illustrated by the household survey utilized in the United States. Information is secured for each person in the household aged 14 and over. For those for whom it is reported that they did not work for pay outside the home at least one hour during the reference week, the question is then asked: “Was … looking for work?” An affirmative reply settles the matter; the person is listed as unemployed. A change in this procedure was under consideration in 1966.
This procedure raises the following definitional questions, among others:
(1) Should unemployment be measured in hours or persons? If a person wanting to work full-time can find only a half-time job, should he be counted as half unemployed? Similarly, if an unemployed person seeks only a part-time job, should he be counted as wholly unemployed? In all countries, the official figures refer to persons wholly unemployed, that is, persons who are not working at all and are seeking full-time or part-time work. But some data on part-time unemployment are available, and the American government now publishes a monthly figure on “percent of labor force time lost” through unemployment.
(2) What should a person have done to look for work, and on what terms should he be willing to accept a job, if he is to be counted as unemployed? What sorts of overt action—such as registering at an employment office or applying personally at a factory or office—should he have taken? And how recently should he have taken such action? Further, if he is willing to work only at his former wage while a job is available to him at a lower wage, should he be counted as unemployed? These are questions that have no single and obvious answer.
(3) A question frequently asked is whether one should include among the unemployed those who are only marginally in the labor force—for example, boys and girls in school seeking part-time jobs or housewives who wish to work to supplement the family income or enjoy a more varied life. Those who would exclude such persons are in effect suggesting that hard-ship be the primary criterion in defining who is unemployed. Quite clearly, unemployment does not imply the same degree of hard ship to all persons seeking jobs at a particular time. The answer here seems fairly clear-cut. An inclusive definition of unemployment should be used by those who compile the data, but the latter should be published in sufficient detail—for example, by age, sex, and marital status—so that the user can combine the subtotals in whatever way best fits the particular definition he wants to use.
(4) The question of marginal attachment to the labor force can cut two ways. The official count of the unemployed may include some persons who are only marginally attached to the labor force. But it also excludes an unknown number who would like to work and would look for work under more favorable conditions. This is not a serious issue if jobs are plentiful, but it does assume importance when unemployment is at a high level. Such a situation of more or less forced withdrawal from the labor force is frequently characterized as one of disguised unemployment.
Disguised unemployment was certainly of some importance in the United States after 1957, when the national unemployment rate remained above 5 per cent. This is an area in which it is difficult to obtain reliable figures. But there has been a growing body of evidence that the lack of jobs created disguised unemployment among women, older persons, the unskilled, those with the least education, and nonwhites (U.S. President 1964, pp. 30-31). Or to put the matter in other terms, labor-force participation rates among these groups were depressed by relatively high unemployment rates.
(5) Some other questions that arise in defining and measuring unemployment can only be mentioned. What should be done about those who are presumably unemployable because of physical or psychological handicaps? Should any attempt be made to measure “underemployment,” that is, the extent to which persons are employed at jobs that call for less than their highest current level of skill? While it would certainly be desirable to have even partial measures of such underemployment, the obstacles in the way of developing such data are virtually insuperable.
The problem of underemployment is most frequently discussed in another connection—the very low average productivity and virtually zero marginal productivity of the agricultural populations in some of the least developed parts of the world. Here the pressure of population on natural resources is such that, given the state of production techniques, a moderate decline in the agricultural work force would lead to little if any reduction in total output. This condition is sometimes said to create “disguised unemployment,” a term that we have used to mean withdrawal from the labor force because of the scarcity of jobs.
Methods and problems of measurement . It has already been noted that unemployment is measured in different ways in different countries. Systems of compiling unemployment data can be grouped under five headings: (1) sample surveys of the labor force; (2) compulsory unemployment-insurance statistics; (3) unemployment-relief data; (4) trade union records of unemployment; and ( 5 ) registrations at government employment offices.
Of these, the first and the last are the most common (U.S. President’s Committee … 1962, appendix A). In 1964 sample surveys were the source of the official data on unemployment in Canada, Japan, and the United States; and such surveys have been used as a supplementary source of information in a number of other countries—for example, France, Germany, Italy, and Sweden. Employment-office registrations were the primary official source in France, Germany, Great Britain, Italy, Sweden, and elsewhere.
Section A of Table 4 presents the official unemployment rates for a number of countries for the years 1960-1962. Differences in definition and method of measurement, however, mean that a number of adjustments are necessary to make these
|Table 4 – Unemployment rates in selected countries, as published and after adjustment to United States definitions, 7960-7962 (per cent of the labor force)|
|A. OFFICIAL FIGURES AS PUBLISHED||B. ADJUSTED TO U.S. DEFINITIONS|
|a. Some figures for these years are preliminary.|
|b. Not available.|
|Source: Myers 1964, p. 174.|
figures fully comparable. Among the points of difference are the following:
(1) Treatment of unpaid family workers. Japan, for example, includes in the labor force unpaid family workers who worked as much as one hour during the survey week. The United States includes them only if they worked 15 hours.
(2) The treatment of persons on temporary lay off and of those waiting to start a new job. Before 1957 these groups were counted as employed in the United States; they are now treated as unemployed. In Japan, they are considered to be employed.
(3) The age at which children are included in the labor force.
(4) The period to which the data apply. In the United States, for example, the period is a week, and anyone working even one hour during the week is counted as employed. In Germany and Great Britain, on the other hand, the reference period is a day. A person not working on that day is counted as unemployed, although he may have worked on one or more other days during that week.
( 5 ) The various unemployed groups that may be excluded from the unemployed. It is fairly common to exclude the self-employed, who are included in the United States. If the data come from unemployment-insurance records, those ineligible for coverage would be excluded. If the figures are based on registrations at employment offices, persons who are seeking jobs but do not register will be excluded.
(6) The computation of the unemployment rate. To compute an unemployment rate it is necessary to divide the number unemployed by some figure that includes the employed. In the United States, the denominator is taken to be the total civilian labor force. The self-employed are included. In a number of countries, the unemployment rate represents only the percentage of wage-and-salary workers who are unemployed.
In 1962 the U.S. Bureau of Labor Statistics made the first intensive attempt to standardize the unemployment figures in a number of countries, the basis of comparison being the definitions in use in the United States. The results are shown in Section B of Table 4. Of the countries listed, the largest relative adjustments were in the British, French, and Italian figures—the first two upward and the last downward. Even after these adjustments the unemployment rates of all the other countries except Canada were below that of the United States for each of the years 1960-1962.
The kinds of unemployment . For both analytical and policy reasons, it would be extremely useful to be able to distinguish among various kinds of unemployment. In the classifications that have been attempted, the categories that one most frequently encounters bear the titles “frictional,” “seasonal,” “cyclical” (or “deficiency-of-demand”), and “structural” unemployment (U.S. Congress, Joint Economic Committee 1961).
Virtually all attempts at classification seek to take account of three sets of factors that tend to create a varying gap between the size of the labor force and total employment. These factors may be summarized as follows:
1. Frictional and seasonal unemployment. Even with a satisfactory level of aggregate demand and a homogeneous labor force, movement to a new job takes time. Hence, under the best of circumstances, there will be a minimum “float” of workers in the process of moving to new jobs. This can be called “minimum frictional unemployment.” Such frictional unemployment is assumed to be balanced by an equal or larger number of job vacancies. If the labor force is not perfectly homogeneous, this frictional minimum may be different for different segments of the working population.
Seasonal unemployment is frequently included in this notion of frictional unemployment. Even if seasonal workers withdraw from the labor force or find other work in the off season, the frictions and time involved in such movements create a seasonal pattern in total unemployment. The seasonal pattern of unemployment does not remain constant over the years, and it is related, in ways which are still not well understood, to changes in the level of aggregate demand.
Seasonal unemployment creates a variety of problems, not the least of which is the problem of measurement. Considerable research has been aimed at improving existing methods of adjusting unemployment figures for seasonal variation (U.S. President’s Committee … 1962, chapter 6). And various measures by both government and private employers have had some effect in dampening the amplitude of seasonal movements in unemployment in particular industries. An interesting example was the effect of Schlechtwettergeld (bad-weather compensation) in reducing recorded seasonal unemployment in the German construction industry (Germany [Federal Republic] … 1962, pp. 56-57).
In Sweden in the early 1960s unemployment in midwinter tended to run three to four times as high as in midsummer. It is not surprising that such countries as Sweden and Norway have devoted considerable effort to reducing these wide seasonal swings. In the United States in 1964 the official seasonal index of the unemployment rate for experienced wage and salaried workers ranged from 84.1 in October to 122.9 in February (see “Rates of Unemployment” 1964). The seasonal amplitude in Canada was still wider.
2. Cyclical unemployment. Next is the set of factors implied by what formerly was called “cyclical” and is now frequently termed “deficiency-of-demand” unemployment. What is implied here is that the total demand for goods and services, given existent wage rates and labor productivity, is not sufficient to generate jobs for all those who want to work (after appropriate adjustment for minimum frictional unemployment). The number of job vacancies open in the economy as a whole is significantly less than the total number of people seeking work. This is the kind of unemployment that was so predominant during the great depression, and it is to the prevention of this kind of unemployment in particular that so-called fullemployment policies are addressed.
3. Structural unemployment. The term “structural” unemployment is most frequently associated with the third set of factors. While this term has been used in different ways, structural unemployment almost always implies the following:
(a) There are particular sectors of the labor force from which workers cannot easily and quickly move into other sectors in search of jobs.
(b) In some or all of these sectors with impaired mobility, unemployment significantly exceeds available vacancies. And because of inadequate mobility, labor supply does not easily adjust to the inadequate level of demand. Hence, unemployment rates are higher in these sectors than in the economy as a whole, and such differentially high unemployment rates tend to persist for relatively long periods.
(c) There may be insufficient demand for particular types of labor for a number of reasons. Three in particular might be cited. First, the demand for particular skills may be reduced because of technological change or a shift in the pattern of demand. Discussions of structural unemployment in the United States in the early 1960s emphasized this factor particularly. Second, there may be a shift of economic activity out of a geographical region not matched by a comparable exodus of workers. This leads to the problem of “depressed areas.” And third, there may be an influx of workers—of a particular type or into a particular region —at such a rate that they cannot be quickly absorbed into jobs. The influx of refugees into West Germany after World War n led to considerable debate regarding the size of the structural unemployment problem in that country (Germany [Federal Republic] … 1952).
The three kinds of causes of structural unemployment cited in the preceding paragraph all have to do with identifiable shifts in the pattern of demand for or supply of labor, to which the economy finds it difficult to adjust because of some degree of labor immobility. It is also possible that demand and supply for a particular kind of labor may have been out of balance as far back as our records go. Immobility, however it came to exist, may perpetuate such sectoral imbalances and create differentially high unemployment rates that continue indefinitely. Such more or less permanent immobility may be associated with a variety of institutional factors. Thus, we expect to find that unemployment rates are higher among the least skilled and least educated. We should also expect that unemployment rates would be higher at the extremes of the age distribution than in the prime working ages. In the United States unemployment rates have been higher among Negroes than among whites since unemployment data were first recorded [see DISCRIMINATION, ECONOMIC].
These structural differentials will be the more marked the more heterogeneous is the labor force —and the measure of heterogeneity in this context is intersectoral immobility. These unemployment differentials are particularly marked, and the labor force is particularly heterogeneous, in the United States.
These differentially high unemployment rates may contain a frictional as well as a deficient-demand element. Among construction workers, for example, there may be ample vacancies, but the element of weather plus the need to move from job to job will cause frictional unemployment to be higher than in most other industries. Over and above such differentially high frictional unemployment, there may also be a shortage of jobs in relation to supply of this particular kind of labor.
It might be added that it is not easy in fact to differentiate between unemployment due to an economy-wide deficiency of demand and that due to structural factors in the sense defined above. For one thing, variations in aggregate demand affect different groups differently. For another, even the most persistent structural reasons for differential-employment opportunities tend to weaken in very tight labor markets; the extreme labor tightness during World War n illustrates this. Or, to cite
|Table 5 – Unemployment rates for different sectors of the labor force, United States, 1948–1963 (per cent of sectoral labor force)|
|a. The figures for 1948-1956 in these categories have been adjusted to account for a moderate change in definitions beginning in 1957.|
|b. The figures for 1948-1956 in these categories have not been so adjusted.|
|c. Not available.|
|Sources: “Rates of Unemployment” 1964, pp. 34-35; “Unemployment Rate” 1966; U.S. President 1964, p. 201; 1966, pp. 166-170.|
|All civilian Workersa||4.6||5.7||5.5||4.2||2.9||3.8|
|By age and sexa|
|Men 20 years and over||3.2||4.5||4.7||3.4||2.5||c|
|Women 20 years and over||4.5||5.4||5.2||4.2||2.9||c|
|Both sexes, 14-19 years||13.6||15.6||13.2||10.4||7.1||c|
|Professional, technical only||1.5||1.8||1.7||1||0.9||1.7|
|Unskilled laborers only||8.4||12.1||12.4||8.2||6.1||7.5|
another example, the structural unemployment resulting from the influx of refugees into West Germany largely disappeared in the very tight labor market of the late 1950s and early 1960s.
Table 5 gives some notion of the heterogeneity of the American labor force and of the resulting wide spread in unemployment rates for different parts of the working population. Thus, the unemployment rate for teen-agers has tended to run about 2.5 times the national rate. The rate for white-collar workers has tended to be about half the national rate, while the incidence of unemployment among unskilled laborers is more than twice as heavy as for the labor force as a whole. Similarly, the unemployment rate for nonwhites is about twice the national average. It is hardly necessary to elaborate upon the social and political, as well as economic, consequences of these high unemployment rates. This is particularly so with respect to what is referred to as “the Negro problem” in the United States.
In European countries the labor force is more homogeneous. Also, far fewer data are available on differential unemployment rates by age, occupation, etc., than is the case for the United States. As noted previously, however, every country has some relatively depressed areas, and a regional breakdown of the unemployed is available for a number of countries. Thus, in the United Kingdom in 1962 regional unemployment rates varied from 1.3 per cent in London and southeastern England to 3.7 per cent in Scotland and to no less than 7.5 per cent in Northern Ireland.
A high level of employment—or “full employment”-is now an accepted goal of national policy in virtually all of the economically advanced countries of the world. The acceptance of this goal immediately raises two questions. How is the goal of full employment to be defined And how is this goal related to other economic objectives with which it may sometimes conflict—for example, stability of the price level and equilibrium in a country’s balance of payments?
A variety of definitions of full employment has been offered. One made famous by Sir William Beveridge (1944) involves “having always more vacant jobs than unemployed men.” The criterion that vacancies should exceed unemployment has frequently been criticized as inflationary, and a common alternative definition runs in terms of approximate equality between vacancies and unemployment.
In recent years, with growing sensitivity to the potential conflict between the goals of full employment and price stability, there has been a tendency to define the employment objective in an alternative way: as the lowest level to which unemployment can be pushed by an expansion of aggregate demand without bringing about an unacceptable rise in the price level. Here, the definition makes it explicit that the employment target depends on the nature of the trade-off between a fall in unemployment and the associated rise (if any) in the price level.
Most definitions of full employment have sought to describe “a situation in which unemployment does not exceed the minimum allowances that must be made for the effects of frictional and seasonal factors” (United Nations … 1949, p. 13). The continued high level of unemployment in the United States between 1957 and 1965, however, raised the question as to how full employment should be defined if a significant amount of structural unemployment exists.
The problem here can be illustrated by Figures 1 and 2. In Figure 1 the job-vacancy rate is measured on the vertical axis and the unemployment rate on the horizontal. The 45° line portrays all possible situations that correspond to an equality of vacancies and unemployed. Curve AA’ represents one “structural situation.” Vacancies rise as unemployment falls (assuming general expansion with structural rigidities) and vice versa; point C represents a situation of “full” employment, in which vacancies and unemployment are equal—at an over-all unemployment rate of 3 per cent in the diagram.
Assume now that, because of structural changes, the curve shifts to BB’. Vacancies now equal unemployment
at an unemployment rate of 6 percent. If the attempt were made to force unemployment much below this by expansionary measures, labor shortages would mount while unemployment would decline only slowly and painfully, with substantial upward pressure on wages and prices. Is 6 per cent to be considered the full-employment target, including not only frictional but also structural unemployment?
The second definition cited suggests the relationships illustrated in Figure 2. Here, again, unemployment is measured along the horizontal axis, but now the rate of change in the price level is represented on the vertical axis. This is a variant of what has come to be called the “Phillips curve,” which relates wage changes rather than price changes to the level of unemployment (see Phillips 1958). The curves XX’ and YY’ describe two possible relationships between price-level changes and unemployment. The curve XX’ implies that prices can be kept stable at an unemployment rate of 3 per cent. Lower unemployment leads to price increases at an accelerating rate.
Let the situation change to that portrayed by YY’. This upward shift is most commonly associated with labor’s increased bargaining strength; but the curve might also shift because of the sort of structural change previously described, with the result that labor shortages in some sectors of the economy are associated with unemployment in excess of vacancies in other sectors. (Not only immobility between sectors but also downward wage rigidity is assumed.) As a result, in the second situation, price stability can be maintained only at an unemployment rate of 6 per cent. Is this full employment?
Since World War n it has been clear that all countries have been willing to settle for something less than absolute price stability. Probably no government would be prepared to pay a significant price in terms of higher unemployment in order to keep the rise in the price level below the rate of, say, 2 per cent per year. This upper limit on the rate of inflation that would be tolerated can be represented by the horizontal line CD in Figure 2. Then OH and OK represent the minimum levels to which unemployment could be reduced in the two cases.
This discussion suggests that the definition of full employment should be approached in two stages. First, we may define “aggregative full employment” as corresponding to that unemployment rate that meets the over-all criterion: total vacancies equal to total unemployment or the smallest amount of unemployment consistent with the desired degree of price stability.
Given the goal of aggregative full employment, which can best be implemented through the use of monetary-fiscal policy, we then need to ask the following question. By means of the various instruments of manpower policy (training and retraining programs, relocation of workers and industry, reduction of discrimination in hiring, and so on), how far is a government prepared to go in reducing the amount of structural unemployment that exists at aggregative full employment? Or, in terms of the diagrams, how far is it prepared to go in seeking to shift the BB’ and YY’ curves to the left? The answer to this question, when costs and benefits are weighed on the scales of the policy makers’ value system, yields what we may call aggregative and structural full employment. In the United States, aggregative full employment in the early 1960s was assumed to correspond to an unemployment rate of about 4 per cent, and it was believed by many that unemployment might, by means of an intensive manpower policy, be brought down to 3 per cent—at a cost that was worth incurring in terms of the benefits to be achieved (U.S. Senate … 1964).
It is one thing to conceptualize the goal of full employment; it is another thing to quantify it. Most governments have been reluctant to announce the precise unemployment rate they consider to be equivalent to full employment. One can, however, draw some inferences from recent policy actions and statements. Thus, in 1964-1965 it was reasonable to assume that the full-employment targets in the leading European countries ranged between about 1.5 and 3 per cent unemployment, using American definitions (Gordon 1965). For the United States the corresponding target figure would be about 4 per cent.
The targets cited here correspond to what has been called “aggregative full employment.” In addition, governments have various kinds of manpower policies aimed at reducing structural and frictional unemployment.
Other aggregative goals . The potential conflict between full employment and price stability has already been mentioned. There are also at least three other “aggregative” goals that need to be taken into account when a government seeks to pursue a full-employment policy. They are balance-of-pay-ments equilibrium, growth of total output, and what has come to be called “incomes policy” (chiefly keeping the rise in wages in line with the increase in productivity).
There is normally no conflict between the goals of full employment and rapid economic growth. But there is some conflict between full employment and the other goals. Involved here is a set of interrelationships that can conveniently be discussed in two parts.
First, relative price stability is a goal that is desired to some extent for its own sake. The more rapidly prices rise or fall—apart from balance-of-payments considerations—the greater is the damage to economic welfare as seen by those who frame economic policy. In recent years Germany and the United States have been the two large countries that have shown the greatest sensitivity to inflationary pressures. In most European countries, however, the trade-off between unemployment and price stability seems to be such that the governments concerned would not be willing to sacrifice much, if any, employment in order to reduce the rate of price increase—provided that (I) prices in other countries rose also, so that serious balance-of-payments problems did not arise, and (2) the rise in prices did not degenerate into runaway inflation.
This brings us to the second aspect of the potential conflict between full employment and the other aggregative goals. The strongest threat to pursuit of a vigorous full-employment policy comes from the possibility of balance-of-payments disequilibrium. If expansionary policies cause spending and prices to rise faster at home than abroad, a country may incur a deficit in its balance of payments which, if continued long enough, will force it to devalue its currency. This is an especially serious problem for countries for which international trade is particularly important. Such countries will, from time to time, be forced to pursue restrictive monetary and fiscal policies, sacrificing some employment in order to protect international monetary reserves. In the early 1960s the United States was apparently prepared to put up with unemployment considerably in excess of what was widely considered to be the full-employment target, in part because of the need to reduce the persistent deficit in its balance of payments. In 1966, the British government felt compelled to impose strong deflationary measures to protect the balance of payments.
It is in this connection that price stability plays its most important role. Price stability is a means of achieving balance-of-payments equilibrium. It is primarily a means, rather than an end in itself. It might be noted that what is important here for the balance of payments is not the behavior of the domestic price level by itself but the behavior of domestic prices relative to that of foreign prices.
As the 1960s began, there was increasing recognition on both sides of the Atlantic of the need for an incomes policy that would hold the rise in wages roughly in line with the rise in labor productivity (Organization for Economic Cooperation and Development … 1962; U.S. President 1962, pp. 185-190). The Netherlands has had an official, but only partially successful, incomes policy since the end of the war. Here again there is a potential conflict with a vigorous full-employment policy. An incomes policy is extremely difficult to apply unless a government is prepared either to exercise compulsion or to accept a higher level of unemployment than would otherwise be necessary. It is not surprising, therefore, that efforts to implement an incomes policy have not so far been very successful.
The events of the decade from the mid-1950s to the mid-1960s suggest a strong contrast between the ways in which western Europe and the United States reconciled the full-employment and other aggregative goals during this period. The United States was prepared to sacrifice a significant amount of employment in order to keep a relatively stable price level, partly for its own sake but chiefly for the sake of its balance-of-payments position. The relative emphasis put on a fairly strict interpretation of the full-employment goal was considerably stronger in Europe. (For further discussion of the issues in this section see Gordon 1965; Ross 1964, pp. 155-171; Gordon & Gordon 1966.)
Development of manpower programs . The increasing recognition of the need to formulate the goal of full employment in structural as well as aggregative terms calls for not only the appropriate use of monetary-fiscal measures but also the development of a comprehensive manpower program.
Manpower policy is concerned with the structural aspects of employment and unemployment—with facilitating the adjustment of the composition of labor supply to the changing composition of labor demand. A well-developed manpower program would include measures of the following types: (1) an active and comprehensive employment service to assist in placement, to provide vocational guidance, and possibly to collect data on vacancies; (2) an adequate system of unemployment insurance, integrated with a system of subsistence allowances for retraining and aid in the relocation of workers; (3) programs to facilitate the geographical mobility of workers; (4) heavy emphasis on training and retraining to assist workers in acquiring new skills; (5) special programs for the physically and mentally handicapped and for older workers; (6) a well-developed program of projecting the future demand for labor and prospective shortages and surpluses of particular kinds of labor, and wide dissemination of such information; (7) cooperation with the schools in the improvement of vocational education; and (8) special employment projects to assist particular regions or groups of workers in making difficult adjustments to new patterns of demand.
Sweden has a particularly well-developed manpower program that aims to be “so varied, so individualised as—in time—to fit every single person on the employment market” (Olsson 1963, p. 411). Most of the other western European countries have significant manpower programs, which vary in the aspects that have been particularly emphasized. Thus, training and retraining have been stressed in France and Italy; Sweden has emphasized regional labor mobility as well as retraining; regional industrial development has played a key role in manpower policy in the United Kingdom. All of these countries—as well as the other countries of western Europe—have also developed, to different degrees, various of the other dimensions of manpower policy previously listed. It is interesting to note also that elements of an international manpower policy have been developed within the European Common Market.
Spurred by high over-all unemployment and concerned with the especially high unemployment rates in particular segments of the labor force, the United States began to move toward a comprehensive manpower policy at the beginning of the 1960s. A decentralized employment service had existed since the 1930s, as had a system of unemployment insurance. In 1961 the Area Redevelopment Act was passed, followed in 1962 by the much more comprehensive Manpower Development and Training Act. The emphasis in the latter was almost entirely on retraining. Other measures have involved acceleration of public works, federal support of education, special provisions for the training and rehabilitation of underprivileged youth, and special measures to counter discrimination against Negroes (U.S. President 1963; 1964).
By 1965 no country had yet developed a man power program “so varied, so individualised as to fit every single person on the employment market.” But movement toward this goal, more rapid and systematic in some countries than in others, was the trend. Avowed full-employment goals and systematic manpower policies did not exist in the Western world when the great depression struck. Considerable progress toward the formulation of such goals and the implementation of such policies has been made since then.
R. A. Gordon
Beveridge, William H. (1909) 1930 Unemployment: A Problem of Industry. London: Longmans.
Beveridge, William H. (1944) 1945 Full Employment in a Free Society. New York: Norton.
Employment and Unemployment: Government Policies Since 1950: I. 1956 International Labour Review 74:1-22.
Germany (Federal Republic), Bundesministerium fÜr Arbeit uno Sozialordnung 1952 Analyse der westdeutschen Arbeitslosigkeit, Statistik. Bonn: The Ministerium.
Germany (Federal Republic), Bundesanstalt fÜr Ar-beitsvekmittlung und Arbeitslosenvehsicherung 1962 Ein Jahrzehnt Bundesanstalt für Arbeitsvermitt-lung und Arbeitslosenversicherung: 1952—1962. Nuremberg: The Institute.
Gordon, R. A. 1964 Has Structural Unemployment Worsened? Industrial Relations 3:53-77.
Gordon, R. A. 1965 Full Employment as a Policy Goal. Pages 25-55 in Arthur M. Ross (editor), Employment Policy and the Labor Market. Berkeley: Univ. of California Press. → From the second Conference on Unemployment and the American Economy, held in 1964 at Boulder, Colorado.
Gordon, R. A.; and GORDON, MARGARET (editors) 1966 Prosperity and Unemployment. New York: Wiley.
Kuznets, Simon 1957 Quantitative Aspects of the Economic Growth of Nations: 2. Industrial Distribution of National Product and Labor Force. Economic Development and Cultural Change 5, no. 4, part 2.
Lebergott, Stanley 1964 Manpower in Economic Growth: The American Record Since 1800. New York: McGraw-Hill.
Maddison, Angus 1964 Economic Growth in the West: Comparative Experience in Europe and North America. New York: Twentieth Century Fund.
Myers, Robert J. 1964 Unemployment in Western Europe and the United States. Pages 172-198 in Arthur M. Ross (editor), Unemployment and the American Economy. New York: Wiley. → From the first Conference on Unemployment and the American Economy, held in 1963 at Berkeley, California.
Olsson, B. 1963 Employment Policy in Sweden. International Labour Review 87:409-434.
Organization for Economic Cooperation AND Development 1963 Manpower Statistics: 1950-1962. Paris: The Organization.
Organization for Economic Cooperation and development, economic policy committee 1962 Policies for Price Stability: A Report to the Economic Policy Committee by Its Working Party on Costs of Production and Prices. Paris: The Organization.
Phillips, A. W. 1958 The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom: 1861-1957. Economica New Series 25:283-299.
Rates of Unemployment. 1964 U.S. Bureau of Labor Statistics, Monthly Report on Labor Force, Employment, Unemployment, Hours and Earnings  January: 34-35.
Ross, Arthur M. (editor) 1964 Unemployment and the American Economy. New York: Wiley. → From the first Conference on Unemployment and the American Economy, held in 1963 at Berkeley, California.
Twentieth Century Fund 1961 Europe’s Needs and Resources: Trends and Prospects in Eighteen Countries. New York: The Fund.
Unemployment and Structural Change. 1962 International Labour Office, Studies and Reports New Series  No. 65.
Unemployment Rate. 1966 U.S. Bureau of Labor Statistics, Employment and Earnings and Monthly Report of the Labor Force  February:20-24.
United Nations, Department of Economic Affairs 1949 National and International Measures for Full Employment. Lake Success, N.Y.: United Nations.
U.S. Bureau of the Census 1960 Historical Statistics of the United States; Colonial Times to 1957: A Statistical Abstract Supplement. Washington: Government Printing Office.
U.S. Congress, Joint Economic Committee 1961 Unemployment: Terminology, Measurement and Analysis. Washington: Government Printing Office.
U.S. Congress, Senate, Committee on Labor AND Public Welfare 1963 Nation’s Manpower Revolution: Hearings Before the Subcommittee on Employment and Manpower of the Committee on Labor and Public Welfare. Washington: Government Printing Office.
U.S. President 1962 Economic Report of the President. Washington: Government Printing Office. -> Published annually since 1947.
U.S. President 1963 Manpower Report of the President. Washington: Government Printing Office.
U.S. President 1964 Manpower Report of the President. Washington: Government Printing Office.
U.S. President 1966 Manpower Report of the President. Washington: Government Printing Office.
U.S. President’s Committee to Appraise Employment and Unemployment Statistics 1962 Measuring Employment and Unemployment. Washington: Government Printing Office.
U.S. Senate, Labor and Public Welfare Committee 1964 Toward Full Employment: Proposals for Comprehensive Employment and Manpower Policy in the United States. Washington: Government Printing Office.