Abstract
THIS PAPER describes the nature of cost inflation in industrial economies pursuing full-employment policies. Its purpose is to indicate why the underlying tendencies to internally generated cost inflation are likely to be stronger in some economies than in others, when a certain moderate rate of unemployment is maintained, and to draw some conclusions concerning the scope for incomes policy. The strength of upward pressure on costs in an economy is broadly determined by the relation between the rates at which money wages and productivity tend to rise at a given level of unemployment. It will be argued that the rate at which money wages rise is likely to be influenced (1) by the underlying flexibility of the labor market situation and (2) by the degree of “organization” which prevails in product and labor markets. Such features of the labor market as the rate of growth of the population of working age, possibilities of increasing participation in the active labor force, and the scope for reducing underemployment in technically backward sectors, such as peasant agriculture and small-scale distribution, condition the bargaining power of the unions and determine the likelihood of wages and prices being bid up by scarcities. The degree of “organization,” which comprehends such features of the economy as the extent of unionization of labor, the extent to which the pricing policies of employers are deliberately or spontaneously coordinated, and the extent to which firms are subject to competition from foreign concerns whose costs do not move in line with domestic wage increases, strongly influence the likelihood of widespread cost pressure eventuating in an economy. Productivity growth is also likely to vary considerably as between countries, depending on the state of technology and the resources becoming available for investment. The economies which, by reason of the circumstances just described, are most likely to experience quite rapid increases in money wages are probably not those in which the underlying conditions are most conducive to rapid increases in productivity.
THIS PAPER describes the nature of cost inflation in industrial economies pursuing full-employment policies. Its purpose is to indicate why the underlying tendencies to internally generated cost inflation are likely to be stronger in some economies than in others, when a certain moderate rate of unemployment is maintained, and to draw some conclusions concerning the scope for incomes policy. The strength of upward pressure on costs in an economy is broadly determined by the relation between the rates at which money wages and productivity tend to rise at a given level of unemployment. It will be argued that the rate at which money wages rise is likely to be influenced (1) by the underlying flexibility of the labor market situation and (2) by the degree of “organization” which prevails in product and labor markets. Such features of the labor market as the rate of growth of the population of working age, possibilities of increasing participation in the active labor force, and the scope for reducing underemployment in technically backward sectors, such as peasant agriculture and small-scale distribution, condition the bargaining power of the unions and determine the likelihood of wages and prices being bid up by scarcities. The degree of “organization,” which comprehends such features of the economy as the extent of unionization of labor, the extent to which the pricing policies of employers are deliberately or spontaneously coordinated, and the extent to which firms are subject to competition from foreign concerns whose costs do not move in line with domestic wage increases, strongly influence the likelihood of widespread cost pressure eventuating in an economy. Productivity growth is also likely to vary considerably as between countries, depending on the state of technology and the resources becoming available for investment. The economies which, by reason of the circumstances just described, are most likely to experience quite rapid increases in money wages are probably not those in which the underlying conditions are most conducive to rapid increases in productivity.
I. The Process of Cost Inflation
The Nature of Wage Bargaining by Unions
Entrepreneurs and unions have come to assume that, by and large, rising productivity in the economy should be reflected in rising money wages and stable prices rather than in stable money wages and falling prices.1 Union members expect to secure increases in money wages over time, and will press such claims unless their bargaining position is too weak to permit them to do so. In general, union members expect some percentage increase in money wages each year, and expect this percentage to exceed any rise in the cost of living. Even when the cost of living is stable, there is some minimum range of increase which is considered the least worth bargaining over.2 However, the aggressiveness with which union members are prepared to press for wage increases tends to diminish above a certain customary range.3 Wage cost inflation is, therefore, less likely when productivity growth is very rapid and exceeds that range, than when productivity growth is slight and falls short of the percentage increase in money wages which union members have come to expect each year.
It is difficult to organize wage earners in sectors such as peasant agriculture, small-scale manufacturing and handicrafts, distribution, and personal services, where the typical concerns are small family businesses employing few, if any, wage earners and little capital equipment. Even if a union were to be successfully organized, in these circumstances its power to raise wages would be circumscribed by the considerable scope for substituting alternative labor (of the entrepreneur and his dependents) or additional capital for wage labor: faced with unemployment, many wage earners themselves would be likely to swell the ranks of family helpers or the self-employed.4
It is usually easier to organize wage earners in industries where the freedom of entry of new enterprises is restricted by the need for substantial capital or technical know-how. This is so both because such industries are often characterized by the concentration of considerable numbers of wage earners in relatively few concerns and because capital-intensive techniques imply more or less fixed manning requirements in the short to medium term. Thus, it is as a rule easier for labor to restrain wage competition in those sectors where employers are most favorably placed to restrain price competition. Since it is more difficult for entrepreneurs or labor to restrain entry and competition in the sectors characterized by large numbers of small-scale producers, the population dependent on those sectors tends to increase and per capita incomes there tend to decline relative to those in the more technically developed sectors when there is persistent unemployment in the more developed sectors or when aggregate demand is tending to rise more slowly than the potential labor force. Owing to this process, weak demand for labor relative to the growth in labor supply in the past tends to be reflected in current weakness of labor organization and union bargaining power and a low level of output per worker over the economy as a whole.
Generally speaking, union members expect their union to secure larger than usual increases when bargaining conditions are exceptionally good. However, they also expect the union to ensure that their earnings rise in line with those of other wage earners. Thus, if any important union succeeds in exploiting an exceptionally favorable bargaining position to secure an unusually large increase in wages and benefits, its success tends to raise the scale of wage claims elsewhere, even if bargaining conditions for other unions remain unchanged. Increases in earnings under existing wage agreements, such as those resulting from higher overtime pay when activity increases, or from higher piecework earnings when productivity rises, may have a similar effect in stimulating wage claims from other groups of workers.
Since the object of a trade union is to advance the interests of its members, the coverage of the different unions and the area to which individual wage bargains relate strongly influence the considerations that are taken into account in pressing wage claims. Competitive union behavior is a potent factor in cost inflation. Just as the individual seller in a market comprising many small producers cannot hope to prevent prices from falling by holding back supplies, so a union which is too small to influence the general movement of wages cannot hope by its forbearance to prevent wages from rising. The small union has no reason to sacrifice the specific interests of its members because the wage increase that it secures may give rise to claims elsewhere, the combined result of which may be to force up the cost of living or to create a threat of unemployment. “Even if each unionist believed that the effect of a series of wage increases in different industries would be nullified by the resulting inflation, it would still be good policy for each union to try to get its increases first and make them bigger than the average.”5 On the other hand, where the individual unions are very large or the union movement is strongly coordinated and the central organization exercises considerable control over the wage claims of individual unions, attention is more likely to be paid to the fact that wages generally cannot rise much faster than productivity without causing price inflation, probable repercussions on the balance of payments, and adverse effects on employment in export and import-competing sectors.
The pattern of wage negotiations can be broadly characterized as highly coordinated in the Scandinavian countries; coordinated in Austria, Germany, and, at least until recently, the Netherlands; and competitive in the United Kingdom, Belgium, France, and Italy. In the United States, Canada, and Japan, where wage bargaining takes place predominantly at the plant level, negotiations are more strongly influenced by the “capacity to pay” of particular enterprises, and hence decisions to press particular wage claims are reached somewhat more independently than in European countries. These differences seem to stem mainly from differences in the context in which unions developed. In general, the unions tend to be organized in a few large units with strong central direction, often with powerful central coordinating federations, in those countries where they developed in a context of large firms or of strongly coordinated resistance to wage claims by the employers.6 In other words, the character of union structure seems to be determined largely by the character of the employers’ organizations. On the whole, owing partly to less widespread unionization, but also partly to geographical and institutional factors tending to create distinct regional labor markets, there is a real contrast between the United States and Canada, on the one hand, and most European countries, on the other, in the extent to which wages differ and move differently in various regions.7 There is a similar but less marked contrast between the highly unified labor markets of the United Kingdom and the Netherlands and the regional differentiation of Italy and France.
Negotiated wage increases and wage drift
In considering how increases in wage costs come about, it is necessary to distinguish between (1) negotiated wage increases, which essentially represent a compromise, determined by the bargaining strength of the two sides, between a price at which unions would wish to supply their members’ labor and a price at which entrepreneurs hope to employ that labor; and (2) other increases in effective rates of pay per unit of labor brought about by arrangements outside the procedures of collective wage bargaining between the union and several employers in an industry, which give rise to so-called wage drift.
Negotiated Wage Increases in a Particular Industry
For the sake of simplicity, it may be useful initially to assume that all wages and salaries are in practice determined by collective agreement between the union and employers in the industry. It should be noted that though this simplifying assumption implies that wages are not bid up by individual employers competing for labor, the employers in any particular industry faced with highly favorable demand conditions as a group might concede, or even propose, an increase in wages to the union in the course of wage negotiations in order to attract additional manpower into the industry.
Influence of degree of unionization
It is clear that the attitude of both unions and employers toward wage claims will be strongly conditioned by the extent to which they expect the prospective wage increase to apply to all firms in the industry. The unions are likely to be less aggressive in pressing wage claims if they believe that the wage increases may not apply to all the workers actually employed or potentially available for employment in an industry (or craft); and the employers, if they hold similar views, will be less permissive in granting wage increases. Hence, if there is a large pool of unorganized low-wage manpower in the economy and the possibility exists of shifting production to a rural area and paying lower wages there, the bargaining power of unions in the established industrial areas will be somewhat circumscribed. Moreover, even if the unions in the existing industrial areas do succeed in pushing up wages, the influence of this on costs in the economy as a whole will be mitigated insofar as the effect is to encourage the transfer of production to regions where wages are lower. The possibility of substituting lower paid women workers for men may also have similar effects in restraining wage pressures in the countries where only a small proportion of the potential female labor force has been employed and organized.
The influence of unemployment (or disguised unemployment) is strongly conditioned by the effective degree of unionization. If a union is able to compel all employers in an industry8 to pay collectively negotiated rates, the existence even of massive long-term unemployment outside the industry need not influence the attitude of the union and its members toward putting forward wage claims. Thus, if an industry is fully and effectively unionized, an increase in unemployment outside the industry may have little influence on the attitude of workers in that industry concerning the wage rates at which they are willing to supply their labor. Only when the union has reason to believe that, because of an actual or prospective weakening in sales or profits, employers in their particular industry are more determined not to raise wages and less likely to yield to union pressures (such as strikes, working to rule, and overtime bans), are wage claims likely to be moderated because the general level of unemployment has risen.
On the employers’ side, once the effective degree of unionization is such that an increase in negotiated wages will apply throughout an industry, each employer can assume that his competitors’ wage costs will rise more or less in line with his own,9 and the employers may, therefore, be less afraid to raise wages. This in turn strengthens the bargaining position of the unions.
A priori it would seem that, because of the greater likelihood of coordinated price behavior, the risk of wage increases pushing up prices is greater when wage negotiations take place on an industry-wide basis, as in many European countries, than when they take place at different times in different concerns, as in the United States and Canada. Other factors, however, such as the greater vulnerability of individual enterprises to strike pressure, work in the opposite direction. Wage claims may have more effect in squeezing profits when negotiations take place with individual concerns. The typical sequence may then be for the union first to press a wage claim against a highly profitable concern, using “capacity to pay” arguments; if the concern accedes, it may not be inclined to raise prices before the union has used this wage settlement as a lever to secure wage increases generally throughout the industry. In such a case, there will be a lag between the initial wage increase and the raising of prices, and wage increases will have some temporary effect in squeezing profit margins, particularly in the most profitable concerns.
Influence of extent of “organization” among employers
The extent to which the price behavior of entrepreneurs is deliberately, or spontaneously, coordinated strongly influences the danger of cost pressure. This complex subject cannot be adequately treated within this study, but the importance of the degree of certainty attaching to entrepreneurs’ expectations concerning the probable price behavior of their competitors should be stressed. If some firms are actively seeking to expand their shares of the market by their pricing policies, the risk that they may not pass on higher wage costs in higher prices will be clear to other concerns; and firms generally will be less prepared to accede to union pressure than in a more stable market situation. When a more stable situation exists in an imperfectly competitive market comprising many or few firms, individual employers who are faced with a wage claim which, if successful, will apply to all wage earners (or all wage or salary earners of a certain type) within the industry, can foresee the prospective price behavior of their competitors quite clearly, since they can expect the behavior of other entrepreneurs to be conditioned by much the same considerations as their own. Consequently, the risks for the individual employer of allowing wages (or other monopolistically determined costs10) to rise are smaller, and resistance to granting wage claims (or to paying higher prices for other items entering into production) is likely to be weaker, than in more fluid market conditions.
While a stable market situation may apply in different circumstances, a priori such a situation would appear more likely to exist in the more highly developed industrial economies where the scope for expansion of the advanced sectors, and for advances in technique and changes in the scale of production, is relatively small. Unless quite marked fluctuations in demand are commonly experienced, firms in general are unlikely to maintain large reserves of unutilized capacity. The firm which employs price competition in order to expand its share of the market is, therefore, presumably seeking to increase its capacity and probably is increasing that of the industry. Consequently, such practical difficulties as pressure on the building sector that interferes with investment schemes, or difficulties in recruiting more labor, are liable to restrict the extent of price competition in an economy. Moreover, the gains from expanding capacity and securing a larger share of the total market will appear more certain to the firm if the industry as a whole is tending to expand. Thus, to risk a generalization, entrepreneurs in an economy undergoing rapid development of the industrial sector in a technologically “backward” setting (as, for instance, in Japan and Italy) may be more independent in their behavior and hence more inclined to resist union wage claims than employers in an economy in which the room for expansion of technically advanced production is smaller (as in the United Kingdom and Sweden).
If a stable market situation exists in an industry, entrepreneurs can expect that prices will continue to be determined on some conventional cost-plus basis because each concern is mindful of the lowering of profits per unit of output that could result from the resumption of price competition. No firm can afford to raise prices alone, and no firm is willing to initiate price reductions alone for fear of unleashing a price war. But, so long as the demand curve for the industry is somewhat inelastic, it is possible for all firms acting in unison to count on raising prices without experiencing a drop of the same proportion in quantities sold. In stable markets, an increase in negotiated wages throughout an industry not only raises costs throughout the industry but, by doing so, increases the possibility of raising prices, since it creates a situation in which all firms are inclined to raise prices more or less simultaneously. It should be noted that price increases for basic products which constitute an important element in the cost of the products of another industry or industries are likely to have an exactly analogous effect in this respect.
In the modern world, many circumstances restrict freedom of entry into various sectors or occupations and enable existing producers to keep up prices and realize a certain habitual margin of profit per unit of output. Imperfect knowledge on the part of both producers and consumers, limited mobility of factors, technological progress and limited diffusion of the latest know-how, scarcity of skills, important economies of scale even for very large enterprises, and scope for product differentiation by advertising are all important in this connection.11
Since it is in practice difficult for new firms to enter an industry, it cannot be assumed that the output of the industry will be expanded so long as any additional profit can be made by doing so. The existing rate of profit does not therefore represent the minimum return required to induce producers to supply the actual (or planned) level of output.12 Hence, a raising of wage costs by union action need not involve a reduction in present (or planned) output whenever demand for the industry’s output is not perfectly inelastic.
The factors which tend to restrict entry of new firms into certain industries, of course, militate against the equalization of rates of return to capital and management in different industries and restrict alternative employment opportunities for management and capital. Consequently, it does not follow that higher wages obtained at the expense of profits and remuneration of management in a particular industry will necessarily reduce the returns to capital and management below those elsewhere, and provide inducement to enterpreneurs to contract the industry. Nor does it follow that a reduction in returns below those obtaining elsewhere would quickly lead to reduced employment in the industry in question, since it is difficult for management and capital employed in a particular industry to shift out of that industry into more profitable fields. Eventually, if the rates of return were to remain below those which could be obtained in accessible alternative openings for management and capital over a prolonged period, the lower returns might lead to disinvestment and reduced willingness of entrepreneurs to enter, or to remain in, the industry. However, such an outcome would depend on a number of factors, such as whether returns to capital and management in other industries had also been squeezed by rising wage costs in the meantime or whether higher wage costs throughout the economy had been accommodated by monetary policy, making it possible for firms generally to raise their prices.
In an advanced economy, demand for an industry’s output is usually somewhat inelastic. The bounds of an industry are extended by mergers between the producers of goods that constitute alternative objects of expenditure; and the elasticity of demand is also reduced by product differentiation, advertising, and the like. The existence of imperfect competition in other product markets, or in the supply of labor and other factors to industries producing competing goods, renders demand for each industry’s output somewhat inelastic. This is so because the elasticity of demand for the output of any industry is determined not only by the extent to which a rise in the price of the product tends to cause buyers to switch to other goods, but also by the response of producers of those other goods to increased demand for their products. (If, for instance, there were to be no increase in the supply of the other goods for which demand had tended to increase, their prices would merely tend to rise until buyers were satisfied with about the same pattern of expenditure as before,13 and the demand for the original industry’s product would be almost or completely inelastic.) Thus, when a union succeeds in pushing up costs to all existing or potential producers in an industry, producers may find that higher wages can be passed on to the consumer with little or no reduction in the total output of the industry, provided that higher prices in this industry induce equivalent increases in prices or costs, rather than expansion of output, in other sectors. When this tends to occur, the widespread existence of monopolistic conditions in product and labor markets in effect renders the general price level unstable in the face of upward cost pressure originating in a single industry.
It follows from this description that price behavior in a single industry cannot be fully understood in isolation, without taking into account the behavior of management and labor in other sectors and the degree of control exercised by the monetary and fiscal authorities in the interest of maintaining price stability. The nature of generalized cost inflation in the industrial economy as a whole is discussed in Section II.
Influence of openness of the economy
It may be useful to consider how the openness of the economy influences wage bargaining within particular industries. Where the production of certain products is concentrated in the hands of a few very large concerns in several countries, producers may be able to reach market-sharing arrangements without fear of attracting new entrants. In general, however, a stable market situation is much more likely to exist among firms in a single country than among firms in different countries. A principal reason for this is that changes in wages and other costs are more likely to be experienced in common by firms within an industry in one country than by firms in different countries. The relative unit costs of firms in different countries are more liable to change over time, both as a result of differences in the bargaining power of the unions and as a result of changes in national policies affecting the pressure of demand on national resources or the conditions for foreign trade (such as budgetary and monetary policies, tariffs, and exchange rates). Thus, the inducement for firms in different countries to seek to expand their share of an international market by lowering prices is liable to vary considerably over time, and may sometimes be very strong.
Entrepreneurs in export-oriented industries, or in industries selling in a home market where imports are of major importance, are aware that their foreign competitors’ costs are not raised by domestic wage increases. Hence managements are more likely to resist union wage demands, since they have less assurance of being able to raise prices in line with rising costs. There is, consequently, less danger of internally generated cost inflation in small countries, such as the Netherlands and Sweden, where very large concerns are inevitably dependent on exports and even those firms which depend mainly on the home market are generally competing with more or less close substitutes produced abroad, than in larger countries. It is possible, however, that the exceptionally large size of the United States tends to reduce the risk of cost inflation, by allowing regional conditions to have more influence on wage rates14 and by reducing the tendency toward “organized” price behavior in the home market.
The Impact of wage Drift15
In practice, wage and salary payments are not determined exclusively by collective agreements between unions and employers, and increases in effective rates of pay per unit of labor also result from arrangements outside the recognized procedures of collective bargaining between the union and employers. It is customary to regard wage drift as reflecting demand pull, but though the extent of wage drift is likely to be related to the scarcity of manpower available to those sectors where collective bargaining applies, wage drift is not occasioned solely by pressure of demand for labor. It is apt to occur whenever systems of payments by results are applied, as a consequence of learning on the job and improvements in the organization of production after piece rates are established. Furthermore, the most progressive enterprises in each industry, which are successful in achieving above-average increases in productivity, frequently are willing to pay higher wages than those stipulated in the collective agreement for the industry (or to grant increases in wages before the agreement is revised), in order to facilitate the introduction of new methods, to reduce labor turnover, or to command the pick of available manpower.
At any given level of aggregate demand, wage drift will usually be greatest in those industries, or economies, where systems of payment by results are most prevalent,16 and where the actual process of wage determination in the individual plant or enterprise is least closely regulated by the procedures of collective bargaining. Since the majority of union members cannot raise their earnings by increasing effort per unit of time, the timing and scale of a union’s wage claims tend to be strongly influenced by changes in average earnings per unit of time in similar occupations. Thus, regardless of whether increases in pieceworkers’ average earnings per hour (or per week) are earned by greater labor input or justified by higher output per hour, they are liable to give rise to claims on behalf of timeworkers elsewhere. The classic example of this process is to be found in the postwar history of wage negotiations in the engineering and shipbuilding industries in the United Kingdom. All the authorities on the subject emphasize the vital role of workshop bargaining over piece rates in providing the stimulus for claims elsewhere in the plant or elsewhere in the industry.17 Recent empirical studies of wage drift have emphasized the difficulties that incentive schemes may cause.18
The character of union organization at the plant or local level has an important bearing on the likelihood of wage drift. If all the workers in a plant are organized in one union with a single representative, the wage structure will tend to be formalized, and the employer will be conscious that increases in certain workers’ earnings will quickly give rise to claims by other workers. The risk of cost pressure originating in wage drift may be reduced by a tradition (such as exists in the Netherlands19 and Germany20) of systematic job evaluation recognized by the unions and of written agreements covering many aspects of rate and earnings adjustments, resulting in an accepted concept of an orderly wage structure for the plant (or industry) as a whole.21 Wage increases for particular workers or groups are likely to be granted less lightly where earning differentials are explicitly fixed than where they are not fixed, but strong pressures for other workers’ earnings to be brought into line are subsequently liable to develop. In sum, the risk of wage drift occurring is reduced when it can be foreseen to have unfavorable repercussions for the employer concerned.
Clearly, the extent of pressure on negotiated wage rates arising from wage drift is likely to be related to the tightness of the underlying labor market situation in the economy.22 Wages actually paid are obviously more likely to rise faster than is provided for in collective agreements if employers are competing for scarce labor. The employers’ attitude toward the labor market situation, conditioned by past experience and knowledge of such factors as population trends, is therefore of considerable importance in influencing the likelihood of wage drift. If the labor force is growing rapidly and there are many new entrants, or if there are large numbers of workers in low-productivity sectors willing to take up industrial employment at going wage rates, the tendency for employers to offer higher wages to overcome local labor shortages may only appear when reported unemployment has almost disappeared, if then. To the extent that employers are able to hire more workers at going wage rates instead of offering higher wages or having recourse to overtime, one major cause of wage claims by other groups of workers in the same plants or industries, i.e., to maintain their relative position in the earnings scale, is avoided.23 If, on the other hand, large employers are inclined to regard the available labor supply as a more or less fixed pool from which they must secure an increasing share in order to expand production, they are likely to fight for scarce supplies of labor with wage incentives of all kinds.
Wage drift is liable to be an important factor in cost inflation because the organization and the behavior of unions are such that increases in wages offered to certain workers serve to trigger claims for increases in negotiated wages by other groups. Though the scale of wage drift depends on employers’ attitudes and decisions, it strongly influences the decisions of unions and their members on whether to press for increases in negotiated wage rates. This process is of considerable importance since it means that wage increases which employers in one industry are willing to offer, because of rising productivity or increasing demand, are liable to result in cost pressure from unions in other industries.24
How the Spread of Wage Increases Influences Business Expectations
Both the danger that a successful wage claim in one sector will set off widespread cost pressure throughout the economy and the risk that general cost pressure may arise from wage drift in particular industries depend on the extent of unionization in the economy and the bargaining strength of the unions in various sectors. Each of these processes will, of course, pose a less serious problem in an economy where wage comparisons tend to be confined to a single enterprise, a particular locality or region, or a group of related industries, than in an economy where each union is concerned to maintain its position in the national scale of earnings. It is no accident that the problem of wage drift has commanded so much attention in Sweden, where the system of wage negotiations is highly centralized and there is a long tradition of solidarity in the labor movement and coordination of incomes in different sectors, and in the United Kingdom, where wage bargaining usually centers around a set of so-called league tables showing relative wages in different occupations at various dates.
The probable spread of upward wage movements has an important influence on the extent to which managements are prepared to accede to union wage claims and to tolerate, or even initiate, wage drift, since it influences their views concerning the risk they take if they let costs rise.25 So long as increases in wages and other money incomes are confined to a few sectors of the economy, they will of course not tend to color the views of producers generally concerning the possibility of selling the same quantities of output at higher prices. Increases in the purchasing power of those dependent on certain sectors will be, in part if not entirely, offset by reductions in the real purchasing power of the population dependent on other sectors of the economy. But the wider the spread of an increase in wages and the more nearly the country in question approximates a closed economy, the greater the influence which the increase in wages is liable to have on the prospective level of aggregate demand. In this respect, too, the danger of internally generated cost inflation is likely to be less pronounced in small countries where one third to one half of all goods and services are sold abroad, than in larger countries where the bulk of all goods and services produced is sold in the home market.26
Insofar as a country approximates a closed economy, the aggregate demand and supply curves are not independent of each other but are closely related. Where virtually all wages are subject to collective bargaining and self-employment has been reduced to relatively small proportions (as in the United Kingdom), a situation may arise where
the coverage of the collective bargain has been tacitly extended from the industry to the whole economy in the sense that an “annual round” of wage settlements has become an accepted institution—settlements coming at much the same time are made for much the same amount. This spontaneous coordination extends beyond the bounds of each separate industry the assurance that other firms’ costs are being raised equally.27
This context influences employers’ attitudes, making them more ready to accede to wage claims unless they are primarily selling abroad. As Jack Downie noted,
If any particular set of employers know that the rate of wage increase they grant will be generalized throughout the economy, they will not be restrained from granting it by the knowledge that it will be necessary to raise prices to compensate for it.28
Of course, as Downie went on to say, if the authorities were able to keep the money supply constant, the rise in prices would eventually be halted through lower pressure of demand on resources and through unemployment of labor and capacity caused by rising interest rates and scarcity of credit. But in order to make it impossible for entrepreneurs to count on being able to raise their prices in those economies where the institutional factors are strongly conducive to the rapid spread of wage increases, the authorities would have to pursue an extremely active monetary (or fiscal) policy, deliberately cutting back aggregate purchasing power at the first sign of any large wage increase. Such a policy is hardly likely to be politically feasible in the situation described, and even if it were politically feasible, it would probably not be practicable or desirable for other reasons.
It is true that the likelihood of generalized cost pressure developing is strongly influenced by employers’ expectations, which are conditioned by the extent to which upward movements of costs and prices have in the past been accommodated by the monetary authorities. However, if the government is committed to preventing more than minimal unemployment, employers may reasonably suppose that the authorities will permit and, if need be, promote the expansion of the monetary circulation necessary to allow production to be maintained at higher prices.29
The analysis in this section has indicated that where there are a number of firms in an imperfectly competitive market, the danger of cost pressure eventuating is essentially the combined result of the degree of “organization” both of labor and of enterpreneurs—that is to say, the effective unionization of workers and the extent to which employers are able to maintain, or push up, prices by reason of individual monopoly power or by open or tacit coordination of their pricing policies. In practice, wage costs tend to rise while there are still unemployed resources in the economy, not solely because of pressure by the unions but also because an increase in wages throughout an industry provides a favorable situation for entrepreneurs to raise their prices more or less simultaneously. Employers may be prepared to let wage costs rise if their effective joint monopoly power enables them to maintain profits.30 Consequently, the role of monopolistic pricing in inducing cost inflation is not confined to the efforts of producers to raise profit margins;31 its importance derives from the power of groups of entrepreneurs (matching that of the unions to maintain wages) to maintain existing profit margins, if need be by raising prices, when there are still unemployed resources in the economy.32
When “organization” extends to many industries and sectors of the economy, an increase in wages or prices in one industry may set in motion the process of cost-push in other industries, either by encouraging entrepreneurs in competing industries to seek higher returns for management and capital by charging the higher prices which the market will now bear, or by stimulating union pressure for wage increases in other sectors. The power of entrepreneurs in each industry jointly to raise prices without fear of the consequences is substantially increased, since they can foresee increases in prices for other products and in aggregate monetary demand.
II. Effectiveness of Aggregate Demand Policy in Countering Cost Inflation
Nature of Generalized Cost Inflation
In a world of imperfect competition, it is unreasonable to suppose that whenever prices rise there is “excess demand,”33 and that unless “excess demand” exists, entrepreneurs are prevented from raising their prices by the authorities’ control over the monetary situation. Obviously, in each economy there is some level of aggregate monetary demand so high that market prices would rise solely because of scarcities, even if there were no cost-push by entrepreneurs or unions; conversely, there is some level of aggregate demand so low that entrepreneurs and union members would be unable, or unwilling, to push up or even maintain prices and wages by concerted action, since they would be competing for such restricted markets and employment openings. However, owing to the scope for monopolistic behavior, prices and wages are apt to rise before there is a situation of excess demand, as entrepreneurs and unions take advantage of a strengthening sellers’ market to raise their respective prices; and some upward pressures on wages and/or prices may persist even at a low level of output and employment.
Policy is therefore usually concerned with a range of demand within which increases in the general price level are not solely due to pressure of demand or to cost-push.34 At any given level of activity within this range, there will be some greater or lesser tendency for prices and wages to move upward, depending on (1) the pressure of demand upon existing capacity and labor resources, and hence the extent to which scarcity of particular outputs or factors of production causes bidding up of certain prices or wages, and (2) the degree of “organization” prevailing in product and labor markets, and hence the extent to which entrepreneurs and union members are able and inclined to push up prices and/or wages autonomously in particular sectors or to ensure that their earnings rise in line with increases occurring elsewhere in the economy.
Whenever a comparatively high level of activity prevails, the authorities will probably be able to reduce the rate of inflation, at least temporarily, by lowering the level of aggregate demand and so creating a greater margin of spare capacity and unemployed manpower. Although a moderate reduction in the level of demand that is not expected to be permanent is unlikely to produce a permanent change in the pricing policy of firms, employers will be less prepared for the time being to raise prices in the context of rising inventories and declining utilization of capacity. Wage increases may be postponed both because of greater resistance on the part of employers and because the bargaining position of the unions is weaker as the weapons of strike and working to rule lose much of their effectiveness in this context. If any union succeeds in pressing a substantial wage claim, there will be less likelihood than usual of this giving rise to wage claims in other sectors; and if any industry raises its prices, causing demand to be diverted to other goods and services, the price increase will be less likely to induce other increases owing to the underutilization of capacity created by the change in aggregate demand policy.
Interpretation of the Phillips Curve Analysis
It is, therefore, to be expected that the rate of increase of money wages will vary with the level of activity and of unemployment in the course of cyclical fluctuations, since wage increases can be postponed or accelerated and so are likely to be bunched in the most favorable periods. Such a relation has in fact been found to exist in the well-known Phillips curve analysis35 and other empirical studies.36 There is, however, a difficulty in interpreting these results. It is commonly suggested that the relation observed between the rate at which money wages actually increased and the rate of unemployment in periods when the level of unemployment was subject to marked short-term fluctuations can be taken to indicate the rate at which money wages would be likely to increase if a certain level of unemployment were to be maintained continuously. Professor Paish’s recent study37 provides an admirably clear statement of this thesis. However, to take Professor Paish’s example, it seems questionable whether the rate of increase in wages observed to be associated in the United Kingdom with an unemployment rate of 2¼ per cent, occurring for short periods when unemployment was usually much lower, provides a valid indication of the rate at which wages would tend to rise if unemployment were maintained at 2¼ per cent of the labor force for a longer period. The rate of increase might well prove substantially higher in these circumstances, and perhaps not very much lower than the average rate of increase over the postwar period, when unemployment averaged 1.6 per cent.
An accurate assessment of the “trade off” between maintaining a higher level of unemployment than has been customary in some European countries in recent years, and achieving a greater degree of price stability, would clearly be very useful to policymakers. As Pierre Massé remarked about the experience of France, if a 5 per cent annual growth rate could have been achieved with price stability, it would probably have been preferable to a growth rate of 5½ per cent followed by a check to expansion. However, he doubted whether lasting price stability could be achieved by such a small safety margin. The attempt to secure stability by restraining aggregate demand was, he believed, more likely to result in a relatively low rate of growth and quite significant unemployment; and he doubted whether such a solution would be tolerated for long in France.38
Unfortunately, the Phillips curve type of analysis does not provide an adequate measure of the relation between unemployment and cost pressure. As has been seen, such statistical studies are likely to overstate the sensitivity of the rate of increase in money wages to the level of unemployment over the medium to long term. Furthermore, such studies fail to take into account the influence of the level of aggregate demand on the rate of growth of productivity. Even if the rate of increase in money wages were to be reduced by maintaining a somewhat higher level of unemployment, upward pressure on costs would not be similarly reduced if productivity growth also was sensitive to the level of activity and to demand in the economy both in the short and longer run.
Feasibility of Effective Aggregate Demand Policy
In practice, even in the short run, the feasibility of countering cost inflation by monetary and fiscal policies depends on the importance attached by the electorate (or those who ultimately decide government policy) to the maintenance of high employment in the economy. If the cost of attaining greater price stability by restraining domestic demand is held to be too high by those with a voice in political decisions, restraint will be regarded as a temporary measure and the policies applied will have only a limited effect on the attitudes of management and unions.39 Employers accustomed to difficulties of recruitment will be disinclined to lay off workers because of a temporary reduction in demand. Employment and the wage bill will then be rather insensitive to reductions in aggregate demand, whereas productivity and profits will be highly vulnerable.40 If it appears easier to make up the investment later rather than to restore a depleted labor force (particularly should the firm acquire the reputation of a “bad employer”), employers may prefer to postpone investment rather than to cut down the wage bill by layoffs; employers in a squeeze because of tighter credit conditions and declining profits may even choose to cut back investment rather than reject wage claims.41 Reducing the level of aggregate demand will then bear more heavily upon investment than upon consumption.
The authorities can prevent very rapid cost inflation by avoiding a situation of severe pressure on the labor market. However, below that degree of excessive pressure on labor resources, in the long run the tendency for costs to rise in the organized sectors of the economy would probably not be much reduced by continuously maintaining a somewhat lower level of demand relative to labor resources. Many of the forces restricting the entry of new firms or the employment of nonunion labor in unionized industries would not be much less powerful merely because the rate of unemployment was maintained at, say, 3–4 per cent instead of 1–2 per cent in an economy with little self-employment; or because the self-employed population in agriculture was declining more slowly in an economy with a large peasant agriculture. In many industries a stable market situation might not be much less likely to exist just on this account; the size of the industry would be adapted to the habitual level of demand; the demand curve for the industry as a whole would still be somewhat inelastic; and the inducement for producers to raise prices in line with wages in order not to risk erosion of profit margins would still apply. In those cases where the greater availability of labor did encourage increased competition, the result might well be the appearance (or survival) of a greater number of small concerns using labor-intensive rather than advanced techniques.
Should aggregate demand policies result in the maintenance of a considerable margin of unutilized capacity intermittently over a period of years, this situation would seem likely to reduce the incentives for investment in the economy, especially in those sectors subject to fluctuations in demand, and hence to slow down the growth in capacity and the speed with which improvements in technique resulting in higher productivity are taken up. Thus, although the bargaining power of the unions and the tendencies to wage drift would probably be somewhat weaker than if the industrial sector were expanding more rapidly, the growth of productivity would probably also be lower. Even if prices were to rise less rapidly in these circumstances, the gain achieved would not be easy to assess. Greater observed price stability associated with less advance in technology is not necessarily preferable to an apparently faster rise of prices associated with more improvement of existing products and greater development of new goods and services (both of which are notoriously difficult to measure statistically).42
The observations made above suggest that it may be misleading to assume that a country can, in some meaningful sense, always count on achieving a lower rate of cost inflation and a strengthening of its international competitive position merely by maintaining a lower level of demand pressure and somewhat higher unemployment.
III. Factors Influencing Cost Inflation in Particular Countries
Sections I and II of this paper have sought to describe the nature of cost inflation in an industrial country. An attempt can now be made to see why some countries may be less likely than others to encounter internally generated cost inflation.
Upward pressure on costs in an industrial country can be regarded as roughly determined by (1) the rate at which money wages tend to rise when a certain level of unemployment is maintained in the economy—that is to say, the relationship which the Phillips curve unfortunately fails to measure, (2) the rate of growth of productivity, and (3) the level of unemployment which the authorities choose to maintain.43
Relation Between the Unemployment Rate and Wage Increases
Other things being equal, in any economy a certain rate of increase (or decline) in money wages and prices is likely to be associated with a given level of activity and rate of unemployment. However, the rate at which wages will tend to rise when a certain proportion of the labor force is unemployed is likely to differ considerably in different countries. As has been seen, the tendency for wages to move upward will be influenced by (1) the pressure of demand on supply in the labor market, which conditions the bargaining power of the labor unions and determines the likelihood of prices and wages being bid up as a result of scarcities, and (2) the degree of “organization” which prevails in product and labor markets.
Elements of flexibility in the labor market
A certain unemployment rate may be associated in different economies with widely differing degrees of tension in the labor market. Unemployment does not constitute the principal element of flexibility in an economy where large numbers of young persons are joining the labor force each year; nor in one where there are large numbers of persons willing to transfer from technically backward, low-income sectors to better paid employment or where there are large numbers of men (or women) willing to enter or rejoin the labor force if more jobs become available. The important differences between various industrial countries with regard to such elements of flexibility in the labor market are illustrated in Table 1.
Factors Influencing the Degree of Pressure in the Labor Market1
Countries are ranked in descending order according as the feature of the labor market situation illustrated appears to create less likelihood of cost pressure. The three groupings in each column roughly distinguish those countries in which the degree of pressure in the labor market resulting from each feature would seem to be great, moderate, or slight.
Population aged 15–64.
Activity rates appear higher in agriculture than in other sectors because the concept of employment in agriculture differs from that in other sectors. Consequently, it is necessary to exclude agriculture in order to have a more or less comparable measure of the activity rates in the different countries. Figures show the percentage of all men or women, outside agriculture, who are members of the active labor force.
Adjusted to exclude self-employed persons and unpaid family help in agriculture, i.e., the figures show the average number of unemployed in 1960–64 as a per cent of the labor force outside agriculture plus wage earners in agriculture.
1960–63.
1963.
Factors Influencing the Degree of Pressure in the Labor Market1
| Average Annual Growth of Potential Labor Force 1960–642 | Employment in Agriculture as Per Cent of Potential Labor Force, 1964 | Activity Rates Outside Agriculture, 19643 | Average Unemployment Rate, Adjusted, 1960–644 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Men | Women | |||||||||||||
| Austria | 0.0 | United | United | Sweden | 52 | Germany | 0.8 | |||||||
| Belgium | 0.1 | Kingdom | 2.7 | Kingdom | 97 | United | Netherlands | 0.8 | ||||||
| Belgium | 3.9 | Denmark | 96 | Kingdom | 51 | |||||||||
| United States | 4.1 | Germany | 95 | |||||||||||
| Netherlands | 5.6 | Austria | 94 | |||||||||||
| Canada | 5.6 | |||||||||||||
| Italy | 0.8 | Germany6 | 8.5 | Sweden | 92 | Denmark | 48 | Austria | 1.2 | |||||
| United | Sweden | 9.2 | Netherlands | 91 | Austria | 47 | Japan | 1.2 | ||||||
| Kingdom | 0.8 | Norway | 10.4 | Norway | 91 | Japan | 45 | France | 1.3 | |||||
| Norway5 | 0.9 | United States | 90 | Germany | 44 | Norway | 1.3 | |||||||
| Sweden | 0.9 | United States | 43 | United | ||||||||||
| Denmark | 1.1 | Kingdom | 1.5 | |||||||||||
| Sweden | 2.0 | |||||||||||||
| Denmark | 2.0 | |||||||||||||
| Belgium | 2.4 | |||||||||||||
| United States | 1.6 | France | 12.1 | Canada | 88 | France | 37 | Italy | 3.6 | |||||
| France | 1.7 | Denmark | 12.9 | France | 88 | Belgium | 37 | United States | 5.8 | |||||
| Canada | 1.9 | Italy | 14.5 | Japan | 87 | Canada | 34 | Canada | 6.4 | |||||
| Netherlands | 1.9 | Austria | 15.2 | Belgium | 87 | Norway | 34 | |||||||
| Germany | 2.0 | Japan | 18.4 | Italy | 85 | Netherlands | 27 | |||||||
| Japan | 2.4 | Italy | 24 | |||||||||||
Countries are ranked in descending order according as the feature of the labor market situation illustrated appears to create less likelihood of cost pressure. The three groupings in each column roughly distinguish those countries in which the degree of pressure in the labor market resulting from each feature would seem to be great, moderate, or slight.
Population aged 15–64.
Activity rates appear higher in agriculture than in other sectors because the concept of employment in agriculture differs from that in other sectors. Consequently, it is necessary to exclude agriculture in order to have a more or less comparable measure of the activity rates in the different countries. Figures show the percentage of all men or women, outside agriculture, who are members of the active labor force.
Adjusted to exclude self-employed persons and unpaid family help in agriculture, i.e., the figures show the average number of unemployed in 1960–64 as a per cent of the labor force outside agriculture plus wage earners in agriculture.
1960–63.
1963.
Factors Influencing the Degree of Pressure in the Labor Market1
| Average Annual Growth of Potential Labor Force 1960–642 | Employment in Agriculture as Per Cent of Potential Labor Force, 1964 | Activity Rates Outside Agriculture, 19643 | Average Unemployment Rate, Adjusted, 1960–644 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Men | Women | |||||||||||||
| Austria | 0.0 | United | United | Sweden | 52 | Germany | 0.8 | |||||||
| Belgium | 0.1 | Kingdom | 2.7 | Kingdom | 97 | United | Netherlands | 0.8 | ||||||
| Belgium | 3.9 | Denmark | 96 | Kingdom | 51 | |||||||||
| United States | 4.1 | Germany | 95 | |||||||||||
| Netherlands | 5.6 | Austria | 94 | |||||||||||
| Canada | 5.6 | |||||||||||||
| Italy | 0.8 | Germany6 | 8.5 | Sweden | 92 | Denmark | 48 | Austria | 1.2 | |||||
| United | Sweden | 9.2 | Netherlands | 91 | Austria | 47 | Japan | 1.2 | ||||||
| Kingdom | 0.8 | Norway | 10.4 | Norway | 91 | Japan | 45 | France | 1.3 | |||||
| Norway5 | 0.9 | United States | 90 | Germany | 44 | Norway | 1.3 | |||||||
| Sweden | 0.9 | United States | 43 | United | ||||||||||
| Denmark | 1.1 | Kingdom | 1.5 | |||||||||||
| Sweden | 2.0 | |||||||||||||
| Denmark | 2.0 | |||||||||||||
| Belgium | 2.4 | |||||||||||||
| United States | 1.6 | France | 12.1 | Canada | 88 | France | 37 | Italy | 3.6 | |||||
| France | 1.7 | Denmark | 12.9 | France | 88 | Belgium | 37 | United States | 5.8 | |||||
| Canada | 1.9 | Italy | 14.5 | Japan | 87 | Canada | 34 | Canada | 6.4 | |||||
| Netherlands | 1.9 | Austria | 15.2 | Belgium | 87 | Norway | 34 | |||||||
| Germany | 2.0 | Japan | 18.4 | Italy | 85 | Netherlands | 27 | |||||||
| Japan | 2.4 | Italy | 24 | |||||||||||
Countries are ranked in descending order according as the feature of the labor market situation illustrated appears to create less likelihood of cost pressure. The three groupings in each column roughly distinguish those countries in which the degree of pressure in the labor market resulting from each feature would seem to be great, moderate, or slight.
Population aged 15–64.
Activity rates appear higher in agriculture than in other sectors because the concept of employment in agriculture differs from that in other sectors. Consequently, it is necessary to exclude agriculture in order to have a more or less comparable measure of the activity rates in the different countries. Figures show the percentage of all men or women, outside agriculture, who are members of the active labor force.
Adjusted to exclude self-employed persons and unpaid family help in agriculture, i.e., the figures show the average number of unemployed in 1960–64 as a per cent of the labor force outside agriculture plus wage earners in agriculture.
1960–63.
1963.
In recent years the growth of the population of working age has amounted each year to about 2 per cent of the labor force in Canada, the Netherlands, Germany,44 and Japan. It is now about that rate in France and the United States. The growth rate has recently been only about half as great in Denmark, Italy, Norway, Sweden, and the United Kingdom, and it has been negligible in Austria and Belgium.
The reduction of the agricultural labor force has in recent years been at an annual rate approaching or exceeding 1 per cent of the active labor force in Canada, Denmark, France, Germany, Italy, and Japan, whereas it has been of minor importance in Belgium, the United Kingdom, and the United States.
The proportion of men of working age outside agriculture who were not members of the active labor force in 1964 ranged from as high as 12–15 per cent in Belgium, Canada, France, Italy, and Japan, to no more than 3–6 per cent in Austria, Denmark, Germany, and the United Kingdom. Similarly, more than 70 per cent of all women of working age outside agriculture were not members of the active labor force in Italy and the Netherlands, more than 50 per cent in Austria, Denmark, Germany, Japan, and the United States, and less than 50 per cent in Sweden and the United Kingdom. These differences are, of course, related to differences in the age structure of the population. (They are, for example, associated with differences in the proportion of the adult population which is still undergoing full-time education, or which has retired, or which, in the case of women, is occupied in bringing up children.) Nevertheless, they represent a real element of greater or lesser flexibility in the labor market.
The data in Table 1 suggest that in certain industrial economies, notably Japan, France, Italy, Canada, the Netherlands, and the United States in approximately that order, there is likely to be a considerable element of flexibility in the labor market, even when a low rate of unemployment is maintained. At the other end of the scale, the United Kingdom, Germany, and Sweden would appear to be the countries more likely to encounter severe tension in the labor market at a low level of unemployment. Austria, Belgium, Denmark, and Norway appear to be in an intermediate position.
It should be noted that the reduction of employment in technically backward sectors need not raise the production costs there. The largely unorganized workers in those sectors will not be in a much better position to secure increases in hourly wages until marked shortages of labor are experienced—unless there should happen to be a great increase in demand for the output of the sectors resulting in upward pressure on prices. As the number of persons dependent on agriculture or retail distribution is reduced, those remaining may well be ready to work more intensively, or for longer hours, in return for higher per capita incomes which do not raise unit labor costs. If unit labor costs do rise, there is likely to be scope for reducing labor inputs by rationalizing the scale of production, so as to permit more efficient use of labor and greater mechanization. Indeed, a rise in hourly labor costs to a level more nearly commensurate with that in the advanced sectors is actually needed in order to stimulate the employment of more capital and so to bring the productivity of labor more into line with that in the advanced industrial areas of the economy.
Marked increases in prices and wages in the less advanced sectors frequently stem from sharply increased demands upon those sectors following rapid development of the industrial sectors; or from deliberate intervention by the authorities to ensure that incomes of politically powerful groups, such as farmers, increase in line with those in the advanced sectors. Often the inability of the less advanced sectors to meet greatly increased demands from other sectors would not be alleviated even if the existing small-scale enterprises were to retain more labor. In these circumstances, the appropriate remedy must be to encourage rationalization and possibly greater dependence on imports. When the authorities intervene to support agricultural incomes, a continuing reduction of the agricultural labor force may actually be necessary if inflationary increases in the cost of food are to be avoided.45
“Organization” of product and labor markets
Important differences also exist with respect to the degree of “organization” of product and labor markets in the various industrial countries, but these are much more difficult to measure. Some rough indicators can be found concerning the coverage and strength of labor unions, and the effect of certain structural factors conditioning their bargaining power; but there is little basis on which to judge how far employers in each country are in a position to determine the prices at which they sell their output, or are subject to price competition. One element of some importance in this connection, namely, the openness of the economy, can, however, be roughly measured.
The proportion of incomes in each economy potentially the subject of union pressure for wage increases would seem to be related to the proportion of wage and salary earners in the labor force. However, it may be best to consider the relative scale of self-employment in agriculture as a special factor since government fixing of agricultural prices in most industrial countries renders the position of farmers’ organizations vis-à-vis the authorities in some ways analogous to that of a union vis-à-vis the employers. Then, because it is difficult for the unions to organize workers or to press successful wage claims in sectors comprising numerous small family enterprises, the ratio of wage and salary earners to workers on own account (employers, self-employed, and unpaid family help) may provide one rough indicator of the scope for union pressure for wage increases in the nonagricultural sectors of various economies. (The ratio may indeed constitute something of an index of the degree of “organization” likely to exist on the side both of employers and of labor in different economies.) The figures given in Table 2 suggest that, on this account, the scope for union action may be greater in the United Kingdom and Sweden, and lower in Italy and Japan, than in other industrial countries.
Relative Importance of Wage and Salaried Employment and Self-Employment, 19641
Countries are ranked in descending order according as the feature of the labor market situation illustrated appears to create less likelihood of cost pressure. The three groupings roughly distinguish those countries in which the degree of pressure in the labor market resulting from each feature would seem to be great, moderate, or slight.
Including all unemployed.
Employers, self-employed, and unpaid family help.
The figures for the United Kingdom, France, Belgium, and Austria are author’s estimates.
The figures for the United Kingdom, Denmark, Norway, France, and Belgium are partly based on author’s estimates indicated in footnotes 4, 6, and 7.
The figures have been adjusted to include estimates of unpaid family help, which is not covered in the national sources.
All figures are author’s estimates.
1963 data.
Relative Importance of Wage and Salaried Employment and Self-Employment, 19641
| Percentage of Civilian Labor Force | |||||||
|---|---|---|---|---|---|---|---|
| Workers on Own Account3 | |||||||
| Wage and salary earners2 | Other sectors | Ratio of Wage to Workers and Salary Earners on Own Account Outside Agriculture5 | |||||
| All sectors | Agriculture | Total | Unpaid family help4 | ||||
| United Kingdom6 | 91.3 | 8.7 | 2.0 | 6.7 | 1.7 | United Kingdom | 13 |
| United States | 86.4 | 13.6 | 4.3 | 9.3 | 0.8 | Sweden | 12 |
| Sweden | 85.0 | 15.0 | 8.1 | 6.9 | 0.6 | ||
| Canada | 83.5 | 16.5 | 7.6 | 8.9 | 0.9 | Canada | 9 |
| United States | 8 | ||||||
| Germany | 79.7 | 20.3 | 10.1 | 10.2 | 2.4 | Netherlands | 8 |
| Netherlands7 | 79.0 | 21.0 | 9.0 | 12.0 | … | Denmark | 8 |
| Denmark | 76.2 | 23.8 | 15.3 | 8.5 | … | Germany | 7 |
| Norway | 74.0 | 26.0 | 14.8 | 11.2 | 0.8 | Austria | 7 |
| France | 73.8 | 26.2 | 11.5 | 14.6 | 3.7 | Norway | 6 |
| Belgium | 73.3 | 26.7 | 5.3 | 21.4 | 6.9 | France | 6 |
| Austria8 | 73.0 | 27.0 | 17.0 | 10.0 | 2.4 | Belgium | 6 |
| Italy | 65.1 | 34.9 | 14.7 | 20.2 | 4.4 | Italy | 3 |
| Japan | 57.3 | 42.7 | 24.7 | 17.9 | 3.0 | Japan | 3 |
Countries are ranked in descending order according as the feature of the labor market situation illustrated appears to create less likelihood of cost pressure. The three groupings roughly distinguish those countries in which the degree of pressure in the labor market resulting from each feature would seem to be great, moderate, or slight.
Including all unemployed.
Employers, self-employed, and unpaid family help.
The figures for the United Kingdom, France, Belgium, and Austria are author’s estimates.
The figures for the United Kingdom, Denmark, Norway, France, and Belgium are partly based on author’s estimates indicated in footnotes 4, 6, and 7.
The figures have been adjusted to include estimates of unpaid family help, which is not covered in the national sources.
All figures are author’s estimates.
1963 data.
Relative Importance of Wage and Salaried Employment and Self-Employment, 19641
| Percentage of Civilian Labor Force | |||||||
|---|---|---|---|---|---|---|---|
| Workers on Own Account3 | |||||||
| Wage and salary earners2 | Other sectors | Ratio of Wage to Workers and Salary Earners on Own Account Outside Agriculture5 | |||||
| All sectors | Agriculture | Total | Unpaid family help4 | ||||
| United Kingdom6 | 91.3 | 8.7 | 2.0 | 6.7 | 1.7 | United Kingdom | 13 |
| United States | 86.4 | 13.6 | 4.3 | 9.3 | 0.8 | Sweden | 12 |
| Sweden | 85.0 | 15.0 | 8.1 | 6.9 | 0.6 | ||
| Canada | 83.5 | 16.5 | 7.6 | 8.9 | 0.9 | Canada | 9 |
| United States | 8 | ||||||
| Germany | 79.7 | 20.3 | 10.1 | 10.2 | 2.4 | Netherlands | 8 |
| Netherlands7 | 79.0 | 21.0 | 9.0 | 12.0 | … | Denmark | 8 |
| Denmark | 76.2 | 23.8 | 15.3 | 8.5 | … | Germany | 7 |
| Norway | 74.0 | 26.0 | 14.8 | 11.2 | 0.8 | Austria | 7 |
| France | 73.8 | 26.2 | 11.5 | 14.6 | 3.7 | Norway | 6 |
| Belgium | 73.3 | 26.7 | 5.3 | 21.4 | 6.9 | France | 6 |
| Austria8 | 73.0 | 27.0 | 17.0 | 10.0 | 2.4 | Belgium | 6 |
| Italy | 65.1 | 34.9 | 14.7 | 20.2 | 4.4 | Italy | 3 |
| Japan | 57.3 | 42.7 | 24.7 | 17.9 | 3.0 | Japan | 3 |
Countries are ranked in descending order according as the feature of the labor market situation illustrated appears to create less likelihood of cost pressure. The three groupings roughly distinguish those countries in which the degree of pressure in the labor market resulting from each feature would seem to be great, moderate, or slight.
Including all unemployed.
Employers, self-employed, and unpaid family help.
The figures for the United Kingdom, France, Belgium, and Austria are author’s estimates.
The figures for the United Kingdom, Denmark, Norway, France, and Belgium are partly based on author’s estimates indicated in footnotes 4, 6, and 7.
The figures have been adjusted to include estimates of unpaid family help, which is not covered in the national sources.
All figures are author’s estimates.
1963 data.
The coverage and effective bargaining power of labor unions differ greatly as between industrial economies, both for historical reasons (such as the time period over which labor organization has been carried out and the legal impediments or encouragements to such organization) and because the bargaining power of unions is conditioned by the extent of coordinated resistance to wage claims on the part of employers. Some of these differences in the wage bargaining situation are summarized in Table 3.
Indicators of Structural and Institutional Factors Influencing the Likelihood of Cost Inflation
The figures show the average of exports and imports of goods and services expressed as a percentage of gross national product in 1964.
The characterization is based on the description given in the accompanying text.
The characterization is based on the description given in the first part of Section I of the text.
The ratio of paid-up union members to wage earners in each country is shown below. However, as noted in the text, these ratios are liable to be influenced by marked differences in the proportion of wage earners who actively support the unions but fail to pay their dues. This proportion is probably higher in Italy, the United Kingdom, and France than in the other countries.
| Sweden | 67 | Netherlands | 41 | Canada | 27 |
| Austria | 62 | Germany | 37 | United States | 25 |
| Norway | 62 | Japan | 33 | France | 21 |
| Denmark | 61 | ||||
| Belgium | 59 | ||||
| Italy | 49 | ||||
| United Kingdom | 37 |
| Sweden | 67 | Netherlands | 41 | Canada | 27 |
| Austria | 62 | Germany | 37 | United States | 25 |
| Norway | 62 | Japan | 33 | France | 21 |
| Denmark | 61 | ||||
| Belgium | 59 | ||||
| Italy | 49 | ||||
| United Kingdom | 37 |
The characterization is based on the summarization of Table 1 in the text above.
The characterization is based on the three groupings shown in the last column of Table 2.
The coordination of both employers and unions in wage bargaining has weakened considerably in the Netherlands in recent years.
Indicators of Structural and Institutional Factors Influencing the Likelihood of Cost Inflation
| Character of Wage Bargaining by | Scope for Union Pressure on Account of | |||||||
|---|---|---|---|---|---|---|---|---|
| Openness of Economy1 | Employers’ federations2 | Unions3 | Degree of Unionization4 | Flexibility of labor market5 | Structure of employment6 | |||
| I. | Countries of uncoordinated wage bargaining | |||||||
| United Kingdom | Slight | 20 | Uncoordinated | Competitive | Great | Great | Great | |
| France | Slight | 14 | Uncoordinated | Competitive | Slight | Slight | Medium | |
| United States | Slight | 5 | Uncoordinated | Independent | Slight | Slight | Medium | |
| Canada | Medium | 25 | Uncoordinated | Independent | Slight | Slight | Medium | |
| Belgium | Great | 37 | Uncoordinated | Competitive | Great | Medium | Medium | |
| II. | Countries of coordinated wage bargaining | |||||||
| Germany | Slight | 19 | Coordinated | Coordinated | Medium | Great | Medium | |
| Sweden | Medium | 26 | Coordinated | Coordinated | Great | Great | Great | |
| Austria | Medium | 26 | Coordinated | Coordinated | Great | Medium | Medium | |
| Denmark | Great | 31 | Coordinated | Coordinated | Great | Medium | Medium | |
| Norway | Great | 42 | Coordinated | Coordinated | Great | Medium | Medium | |
| Netherlands | Great | 47 | Coordinated7 | Coordinated7 | Medium | Medium | Medium | |
| III. | Countries where the scope for union action is limited by the labor market situation | |||||||
| Italy | Slight | 17 | Uncoordinated | Competitive | Great | Very slight | Slight | |
| Japan | Slight | 12 | Uncoordinated | Independent | Medium | Very slight | Slight | |
The figures show the average of exports and imports of goods and services expressed as a percentage of gross national product in 1964.
The characterization is based on the description given in the accompanying text.
The characterization is based on the description given in the first part of Section I of the text.
The ratio of paid-up union members to wage earners in each country is shown below. However, as noted in the text, these ratios are liable to be influenced by marked differences in the proportion of wage earners who actively support the unions but fail to pay their dues. This proportion is probably higher in Italy, the United Kingdom, and France than in the other countries.
| Sweden | 67 | Netherlands | 41 | Canada | 27 |
| Austria | 62 | Germany | 37 | United States | 25 |
| Norway | 62 | Japan | 33 | France | 21 |
| Denmark | 61 | ||||
| Belgium | 59 | ||||
| Italy | 49 | ||||
| United Kingdom | 37 |
The characterization is based on the summarization of Table 1 in the text above.
The characterization is based on the three groupings shown in the last column of Table 2.
The coordination of both employers and unions in wage bargaining has weakened considerably in the Netherlands in recent years.
Indicators of Structural and Institutional Factors Influencing the Likelihood of Cost Inflation
| Character of Wage Bargaining by | Scope for Union Pressure on Account of | |||||||
|---|---|---|---|---|---|---|---|---|
| Openness of Economy1 | Employers’ federations2 | Unions3 | Degree of Unionization4 | Flexibility of labor market5 | Structure of employment6 | |||
| I. | Countries of uncoordinated wage bargaining | |||||||
| United Kingdom | Slight | 20 | Uncoordinated | Competitive | Great | Great | Great | |
| France | Slight | 14 | Uncoordinated | Competitive | Slight | Slight | Medium | |
| United States | Slight | 5 | Uncoordinated | Independent | Slight | Slight | Medium | |
| Canada | Medium | 25 | Uncoordinated | Independent | Slight | Slight | Medium | |
| Belgium | Great | 37 | Uncoordinated | Competitive | Great | Medium | Medium | |
| II. | Countries of coordinated wage bargaining | |||||||
| Germany | Slight | 19 | Coordinated | Coordinated | Medium | Great | Medium | |
| Sweden | Medium | 26 | Coordinated | Coordinated | Great | Great | Great | |
| Austria | Medium | 26 | Coordinated | Coordinated | Great | Medium | Medium | |
| Denmark | Great | 31 | Coordinated | Coordinated | Great | Medium | Medium | |
| Norway | Great | 42 | Coordinated | Coordinated | Great | Medium | Medium | |
| Netherlands | Great | 47 | Coordinated7 | Coordinated7 | Medium | Medium | Medium | |
| III. | Countries where the scope for union action is limited by the labor market situation | |||||||
| Italy | Slight | 17 | Uncoordinated | Competitive | Great | Very slight | Slight | |
| Japan | Slight | 12 | Uncoordinated | Independent | Medium | Very slight | Slight | |
The figures show the average of exports and imports of goods and services expressed as a percentage of gross national product in 1964.
The characterization is based on the description given in the accompanying text.
The characterization is based on the description given in the first part of Section I of the text.
The ratio of paid-up union members to wage earners in each country is shown below. However, as noted in the text, these ratios are liable to be influenced by marked differences in the proportion of wage earners who actively support the unions but fail to pay their dues. This proportion is probably higher in Italy, the United Kingdom, and France than in the other countries.
| Sweden | 67 | Netherlands | 41 | Canada | 27 |
| Austria | 62 | Germany | 37 | United States | 25 |
| Norway | 62 | Japan | 33 | France | 21 |
| Denmark | 61 | ||||
| Belgium | 59 | ||||
| Italy | 49 | ||||
| United Kingdom | 37 |
| Sweden | 67 | Netherlands | 41 | Canada | 27 |
| Austria | 62 | Germany | 37 | United States | 25 |
| Norway | 62 | Japan | 33 | France | 21 |
| Denmark | 61 | ||||
| Belgium | 59 | ||||
| Italy | 49 | ||||
| United Kingdom | 37 |
The characterization is based on the summarization of Table 1 in the text above.
The characterization is based on the three groupings shown in the last column of Table 2.
The coordination of both employers and unions in wage bargaining has weakened considerably in the Netherlands in recent years.
The effective coverage of the unions is not very satisfactorily indicated by the number of paid-up union members, owing to marked differences in the proportion of wage earners who actively support unions but fail to pay their dues. The paid-up union membership tends to be higher in a strongly coordinated labor movement or where the unions administer benefits such as pensions or insurance. Allowing for this factor, it is probably safe to conclude that the effective coverage of the unions is not much lower in Italy or the United Kingdom than in the Scandinavian countries, Belgium, and Austria, that it is somewhat lower in Germany, the Netherlands, and Japan, and lower again in Canada, France, and the United States (see Table 3, footnote 4).
The pressure for wage increases actually exerted by the unions is conditioned both by the extent of coordination between employers in wage bargaining and the relative openness of the economy. These factors are not unrelated. Broadly speaking, strongly coordinated employers’ federations came into being in those economies where the leading enterprises were exposed to foreign competition when they first encountered widespread pressure for wage increases from the developing unions. Thus, in five of the six smallest and most open industrial economies—and in only one large economy, Germany—powerful coordinated employers’ federations face coordinated labor unions, whose structure was dictated by the need to bargain with highly centralized employers’ organizations; on the other hand, in six of the seven bigger industrial economies, largely uncoordinated labor unions face uncoordinated employers’ organizations.
The former group includes, besides Germany, the three Scandinavian countries, Austria, and the Netherlands, but excludes Belgium. It appears that the important factors encouraging a high degree of coordination among employers in these countries were their relatively late industrial development and need to challenge established industrial centers such as the United Kingdom, Belgium, and France; the greater predominance of large-scale enterprises by this time; and the wider threat posed by industrial unions than by the small-craft unions originally established in the older centers.
Despite their dependence on foreign markets, industrialists in the United Kingdom and Belgium did not have the same inducement to coordinate their wage bargaining at the time when unions were developing because they possessed a commanding technical leadership and because union pressure for wage increases developed gradually in a highly localized and uncoordinated manner. In France and the United States, industries developed to supply a large and protected home market, and the development of the union movement in these countries—as also in Canada and Japan—was not such as to pose a serious threat of concerted pressure for wage increases. In France, the threat of union pressure was tempered by the fragmentation of the unions and their weak legal status vis-à-vis the individual employer. In North America and Japan, the employers had little stimulus to coordinate their opposition to the unions, since for historical and legal reasons, the unions were organized to bargain with individual enterprises.
Japan and Italy are probably best regarded as countries where, for various reasons, concerted union pressure for wage increases is only now becoming a possibility. Effective union action was, broadly speaking, prohibited by law during the prewar periods when much industrialization was taking place in these countries; and since the war the bargaining power of the unions has been restricted by the large number of new entrants to the industrial labor force from sectors of low-income self-employment. Somewhat the same position existed during the 1950’s in Germany, as the unions were only gradually rebuilding their organization and finances after having been disbanded by law from 1933 to 1945, and there was over this period a very large influx of refugees.
It is now possible to sum up this analysis as follows. In five of the seven larger industrial economies, there exists a considerable element of flexibility in the underlying labor market situation even at a low level of unemployment. This is especially true of Japan, Italy, and France; it is also true to a lesser extent of Canada and the United States. The effective degree of unionization among wage earners is comparatively slight in three of these countries (France, Canada, and the United States). In Japan and Italy, the unions cover a higher proportion of wage earners but their bargaining power is circumscribed by the exceptionally large proportion of self-employed workers in the labor force.
In the other two large economies, Germany and the United Kingdom, and also in Sweden, a greater degree of tension in the labor market is likely to exist at a low level of unemployment. In Germany and Sweden, but not in the United Kingdom so far (if one excludes the present emergency wage freeze imposed by the authorities), coordinated wage bargaining by the employers and unions tends to lessen tendencies to competitive wage claims.
In the other smaller industrial economies, the degree of tension in the labor market at a low level of unemployment is probably less than in Sweden, or than in Germany and the United Kingdom. The openness of these economies limits the possibility of upward movements of costs in excess of those experienced in other countries. Except in Belgium, this fact has been reflected in coordinated wage bargaining by the employers leading to coordinated wage claims from the unions. In Belgium, unions and employers’ organizations still remain highly fragmented and uncoordinated, as they have historically always been. Until a few years ago, however, the bargaining power of unions was limited by relatively heavy unemployment.
It is difficult to assign weights to the various factors covered in Table 3, but the information given there suggests that the underlying tendencies to cost inflation are likely to be greater in Belgium, France, Germany, Sweden, and the United Kingdom than in Austria, Canada, Denmark, Italy, the Netherlands, Norway, or the United States. The tendencies to cost inflation are probably least powerful in Italy and Japan.
Productivity Growth
Much more study is needed of the factors determining the growth of productivity in the different industrial countries, and of the relation between the demand for labor and the growth of productivity within an economy. It seems reasonable to assume that at any period there is some potential range of productivity growth in each industrial country and that the actual growth of productivity is related to the strength of internal and external demands upon the economy’s resources. The potential range of productivity growth is likely to vary as between countries, depending on the existing state of technology. Thus, in economies where industrial techniques have lagged, it may be possible to achieve rapid increases in per capita output by the application of advanced techniques already in use elsewhere, provided that adequate resources can be secured for investment. As against this, the more advanced economies may be better placed to apply the latest and most complex techniques (such as automation). In general, the most highly advanced economies with the highest per capita income may be expected to dispose of greater resources for investment and hence to be in a better position to achieve gains in productivity. However, this need not be the case in a world of unfettered international trade and foreign investment. If there is great scope for the adoption of modern techniques in the less organized economies, and if the level of hourly wages is relatively low at the ruling exchange rates and remains so owing to the less effective organization and weaker bargaining power of the unions in those economies, enterprises may be able to realize very high rates of profit on exports with which to finance a rapid expansion of modern industrial capacity, and high rates of profit may encourage substantial investment of foreign capital.
The relative importance of domestic and export markets to producers in the economy is very likely to influence the relation between the level of demand maintained within the economy and the growth of productivity in that economy. For enterprises predominantly selling abroad, a lowering of domestic demand will have comparatively slight effect on total sales while possibly moderating the rise in costs relative to that occurring in other countries; thus profits and so resources available for investment will probably tend to increase. For enterprises predominantly selling in the home market, however, the lowering of domestic demand, besides reducing costs, is likely to result in smaller total sales and lower total profits (since export sales are a minor part of total output, the increased profits from expanded exports are unlikely to counterbalance the reduction in profits in the home market). Investment and the growth of productivity are therefore liable to be reduced when a lower level of demand is maintained.
Level of Unemployment Maintained
In practice, the level of unemployment maintained in an economy tends to be determined by such factors as the occupational distribution of the population, the extent of unionization, and the nature of political representation in the country, rather than to be “chosen” by the authorities. The problem of cost inflation is thus compounded by the fact that the criteria of full employment are most stringent in those economies where the labor force is most fully organized and the danger of generalized cost pressure at a given level of unemployment is greatest. In those economies, the authorities are likely to be under considerable pressure to maintain a lower margin of unutilized labor resources and to give the full-employment objective more weight in relation to other policy objectives, such as price stability, than is the practice in economies where the labor unions are less powerful.
Tendencies to Cost Inflation in Different Countries
The foregoing description suggests that the level of unemployment which is likely to be maintained and the rates at which money wages and productivity tend to rise when a certain level of unemployment is maintained are not independent and that their influence is likely to be mutually reinforcing. Thus, in one economy there may be a willingness on the part of the electorate to tolerate a relatively high degree of unemployment; a relatively slight tendency for wages to increase at a given level of unemployment; and a potentially high rate of productivity growth. In another economy, the electorate may be strongly disinclined to tolerate the maintenance of more than minimal unemployment; wages may tend to increase more rapidly at a given level of unemployment; and the potential rate of productivity growth may be relatively low.
The description given above prompts certain tentative conclusions concerning the relation which is likely to exist in different economies between the rate of increase in wages and the level of unemployment maintained over a period of years. In a relatively closed and highly “organized” economy—that is to say, one in which there are powerful, widely based labor unions and rather stable conditions of imperfect competition in many product markets—there will probably be some level of activity at which favorable demand conditions in particular sectors, reflected in increasing profits and, probably, in wage drift, will tend to induce quite large wage increases over a broad area of the economy. This will cause unavoidable pressure on prices in the sectors where productivity growth is lower than average, and induce further wage claims stemming from increases in the cost of living. In other words, above some critical level of demand pressure, there will be a sharply increasing tendency to price and wage inflation. Below this critical level, however, there may be a range of levels of activity within which the rate of wage increase is not very sensitive to differences in the levels of activity and unemployment, and is mainly determined by the strength of autonomous cost-push in particular sectors of the economy. A curve of this nature is indicated by A-A in Chart 1.
Relation Between Rate of Increase in Money Wages and Rate of Unemployment Maintained over a Number of Years in a Highly “Organized” Economy (a) and in a less “Organized” Economy (b)
Citation: IMF Staff Papers 1967, 001; 10.5089/9781451956191.024.A006
Note: The darker portion of each curve represents the range of unemployment rates which is “politically tolerable” to the electorate.Relation Between Rate of Increase in Money Wages and Rate of Unemployment Maintained over a Number of Years in a Highly “Organized” Economy (a) and in a less “Organized” Economy (b)
Citation: IMF Staff Papers 1967, 001; 10.5089/9781451956191.024.A006
Note: The darker portion of each curve represents the range of unemployment rates which is “politically tolerable” to the electorate.Relation Between Rate of Increase in Money Wages and Rate of Unemployment Maintained over a Number of Years in a Highly “Organized” Economy (a) and in a less “Organized” Economy (b)
Citation: IMF Staff Papers 1967, 001; 10.5089/9781451956191.024.A006
Note: The darker portion of each curve represents the range of unemployment rates which is “politically tolerable” to the electorate.In a less industrialized and less highly “organized” economy (i.e., one in which greater competition prevails in product markets and the unions are confined to a narrow area of the economy), increasing profits and wage drift in particular sectors will have less influence on the rate at which wages and prices rise in the economy as a whole. Owing to a lower degree of organization of wage earners, the rate at which wages tend to rise will be more sensitive to the level of unemployment and underemployment maintained in the economy. A curve of this nature is indicated by the curve B-B in Chart 1.
If, as has been suggested, the former economy also has the lower political tolerance for unemployment, the authorities there may be virtually precluded from consistently maintaining a level of unemployment at which money wages do not rise faster than those in the less highly “organized” economy. The competitive position of the more “organized” economy need not be weakened as a result of this situation, provided that the rate of growth of productivity is also higher than in the less “organized” economy at each level of unemployment. However, as has been seen above, there seems no reason to believe that this must necessarily be the case. If not, such a situation may pose balance of payments problems for the more highly “organized” economies. Nevertheless, the strengthening of the competitive position of the less industrialized countries vis-à-vis the more highly “organized” economies is desirable from an international standpoint, since it tends to promote fuller employment in the industrial world as a whole and to narrow disparities between average per capita income in industrial countries, whereas the strengthening or maintenance of the competitive position of the countries with the highest per capita income would almost certainly tend to widen such disparities.
The modifications to Duesenberry’s framework of ideas suggested here serve to reinforce his contention that tendencies to payments disequilibria of a long-term character are more or less inherent under a system of fixed exchange rates, given the objectives of national economic policies and certain dynamic aspects of the international economy. It would, however, seem more difficult for any country to check the growth in its deficit and improve its competitive position in the longer run by increasing the level of unemployment, than Duesenberry’s analysis might suggest. It might perhaps be more realistic to regard some maximum level of unemployment in the medium term as a politically determined constraint upon economic policymakers, which in practice has priority over other policy objectives.
A country which encounters severe cost inflation when it maintains a politically tolerable level of unemployment may seek to slow down the rate of inflation in a number of ways. The authorities may try to make it more difficult for producers to raise prices in line with wages by eliminating legal or other restraints to competition and barriers to new entrants to certain industries; they may seek to reduce obstacles to the attainment of higher productivity, such as restrictive union policies and obsolete systems of wage payments; they may seek to strengthen labor market mechanisms in order to secure greater mobility and lessen the likelihood of the bidding-up of wages; and they may try to reduce the strength of cost-push by means of incomes policy. The description of the process of cost inflation in this paper prompts certain tentative conclusions concerning the scope for incomes policy.
IV. The Scope for Incomes Policy
The achievement and maintenance of full employment, comprehending in the term the absorption of underemployment hitherto apparent as abnormally low labor productivity in certain sectors, constitutes a new situation in many industrial countries, and one that is likely to require specific measures to restrain cost inflation. The analysis in this paper suggests that tendencies to cost inflation are likely to intensify in future because (1) the proportion of incomes and prices determined in noncompetitive markets is likely to increase with increasing industrial employment, growing unionization, and the increasing scale of enterprises; (2) longer experience of a consistently high level of aggregate demand, and of low unemployment with reduced possibilities of shifting labor out of low productivity occupations, may tend to encourage a more aggressive attitude on the part of labor and a more permissive attitude on the part of employers toward wage increases, leading to a strengthening in the bargaining position of the unions; and (3) with growing international integration, the area of “organized” price behavior may extend increasingly over national boundaries to firms in different countries, so that the discipline of international competition becomes less effective.
The practicability of incomes policy—that is to say, of deliberate intervention by the authorities in the process of price formation for labor and products aimed at preventing gross (untaxed) money incomes from rising excessively in relation to the growth of national output in real terms—is likely to differ widely as between industrial countries. It may well be easier to establish an effective system for influencing the determination of incomes from employment in those countries where the underlying labor market situation and institutions are less conducive to cost inflation; and vice versa. The unions are more likely to cooperate when their position is, for various reasons, comparatively weak; and the existence of strongly centralized and coordinated institutions on the part of employers and unions, which is needed for the effective implementation of a systematic incomes policy, is itself a factor tending to lessen the severity of wage cost pressure in the economy. Thus, it would seem that if all countries were to adopt incomes policy measures in order to maintain price stability, the effect might be to increase rather than to moderate disparities in the movement of costs as between countries.
In practice, however, the inducement to adopt an incomes policy is in stronger in some economies than in others. The inducement is greatest where relative price stability is needed to facilitate expansion of employment,46 either before or when the employment objective is met, or to improve a critical balance of payments position.
In an economy where there is little possibility of expanding employment or of increasing the degree of utilization of manpower resources, and which is not in deficit in its external transactions on current account, incomes policy cannot be expected to prevent costs and prices from rising at more or less the same rate as in other countries—that is unless the external sector is of very minor importance, as in the United States. Otherwise, if labor costs are kept stable when prices in other countries are rising fairly rapidly, increasing demand and profits in the export and import-competing sectors will create pressure on the labor market and probably give rise to an increasing surplus on current account. The bidding up of wages by employers in the expanding sectors is almost bound to occur—as it did in the Netherlands in the early 1960’s. Even if the unions exercise some restraint in putting forward wage claims, as in Sweden, wages in the export and import-competing sectors will tend to drift upward more or less in line with those in competitor countries. At a certain point, prices in protected sectors, such as services and construction, will tend to be adjusted steeply upward to take advantage of the growth in money demand from other sectors and to cover increased costs resulting from heavy wage drift brought about by employers’ efforts to attract, or retain, labor.
If the authorities intend to maintain balance of payments equilibrium and a fixed exchange rate, they cannot hope to secure much greater price stability than is being achieved in other countries, unless there is scope for expanding employment47 and for improving an adverse external balance. Thus the goal of incomes policy can only be to prevent costs from rising faster than in other industrial countries. This is a fact which increases the practical problems of implementing systematic measures of incomes policy by making it difficult to state in advance any general and invariable criteria for permissible increases in money incomes. A policy aimed at securing internal price stability by preventing money incomes from rising faster than productivity is far easier to grasp than a policy which allows incomes to rise considerably faster than productivity in one period, but not in another. However, except possibly in the largest economies, the attempt to prevent money incomes from rising faster than productivity under full employment conditions is doomed to failure if wages and prices are rising rapidly in neighboring countries; and the apparent failure of incomes policy in these circumstances may later render its implementation ineffective when the international situation calls for a greater degree of restraint. It would therefore seem that active intervention by the authorities, designed to ensure that money incomes generally do not rise faster than productivity, might best be limited to those economies where there is a definite need to prevent costs and prices from rising as fast as in other countries.
The inducement to adopt an incomes policy would then arise in two very different situations: (1) where the tendencies to cost inflation are relatively moderate but the incentive to maintain the advantage is strong, owing to the need to provide for continued expansion of employment opportunities; and (2) where the tendencies to cost inflation are particularly strong and have led to deterioration of the competitive position and of the external balance, and there is a pressing need to improve the situation if the full-employment objective is to continue to be realized. In such circumstances, the need is to ensure that in the future costs and prices rise less rapidly than those elsewhere.
Obviously, situation (1) represents an easier task for incomes policy, and past experience suggests that it is more likely to be possible to secure the implementation of incomes policy in these circumstances. To the extent that certain countries successfully achieve their objectives in situation (1), the problem of countries in situation (2) of course becomes more intractable.
When it appears necessary to strengthen considerably the incentives to export and to lessen substantially the incentives to import in order to right the current account balance at a level of activity corresponding to the full-employment objective, internal costs and prices will need to be lowered substantially in relation to those abroad. This will be possible at unchanged exchange rates, provided that prices elsewhere are rising fairly rapidly and internal prices and unit wage costs can be held virtually stable (or even reduced). However, it may in practice be difficult, if not impossible, to achieve internal price stability without reducing the level of internal activity well below the desired full-employment objective, at least temporarily.
To conclude, it would seem desirable to distinguish two different situations: first, where there is a general need to counter tendencies to creeping inflation when full-employment policies are actively pursued; second, where there is a more specific need to ensure that internal costs and prices in a particular economy rise less than those in other industrial countries. One may usefully conceive of two different types of policy as suited for these two different situations. The first, which has been characterized as institutional engineering, would be generally applicable, and would cover a wide range of efforts to remedy particular features of the economy which tend to create or reinforce tendencies to cost pressure. The second would comprise more systematic attempts to implement an incomes policy with the cooperation of the parties concerned in wage negotiations and price determination, in order to keep wage and prices increases, within limits, more or less clearly stated in advance.
Inflation par la hausse des coûts et politique des revenus dans les pays industriels
Résumé
Cet article décrit la nature de l’inflation par la hausse des coûts dont souffrent les pays industriels qui poursuivent des politiques de plein emploi. Pour pouvoir comprendre la manière complexe selon laquelle les facteurs qui déterminent la force des poussées inflationnistes sur les coûts agissent les uns sur les autres dans différents pays, il est essentiel de se faire une idée relativement exacte du processus selon lequel les salaires sont déterminés dans les industries oligopolistiques dans lesquelles les salariés sont organisés en syndicats. La première section énumère les facteurs qui influencent les négociations syndicales en matière de salaires et décrit la façon dont se produisent les augmentations de salaires; elle étudie d’abord les augmentations de salaires négociées entre les divers syndicats et plusieurs employeurs et, ensuite, les autres décisions en matière de salaires qui donnent lieu à des glissements de salaires. La deuxième section discute l’influence que le niveau de la demande exerce sur l’inflation par la hausse des coûts et le succès de la politique régissant la demande globale comme instrument de contrôle de cette inflation.
La hausse des salaires est influencée 1) par la pression que la demande exerce sur l’offre sur le marché de la maind’œuvre, ce qui détermine le pouvoir de négociation des syndicats et entraîne une surenchère des prix et des salaires résultant de pénuries dans ces domaines, et 2) par la mesure dans laquelle une certaine “organisation” peut freiner le libre jeu de la concurrence sur le marché des produits et de la main-d’œuvre, ce qui permet aux employeurs et salariés de coordonner chacun dans leur domaine leurs politiques des prix. Un taux donné de chômage ira vraisemblablement de pair avec des degrés de tension divers sur les marchés nationaux de la main-d’œuvre, selon le taux de croissance de la population adulte et la marge de réduction du sous-emploi qui existe dans les secteurs techniquement arriérés ou d’augmentation du taux d’activité de la main-d’œuvre parmi la population adulte. L’étendue de “l’organisation” diffère aussi considérablement selon les pays. Il existe des différences notables entre les effectifs réels des syndicats, et leur pouvoir de négociation dépend de la structure du marché de la main-d’œuvre et du degré de coordination entre employeurs lors de négociations de salaires. Il est plus difficile de juger dans quelle mesure les employeurs de chaque pays sont à même de déterminer les prix auxquels ils vendent leur production, mais on peut mesurer un élément relativement important — c’est-à-dire la mesure dans laquelle une économie est ouverte aux influences extérieures. La troisième section s’efforce d’indiquer dans quelle mesure les pays industriels diffèrent à tous ces égards. La section finale formule certaines conclusions en ce qui concerne la portée d’une politique des revenus dans le cadre de l’analyse précédente.
La inflación provocada por el alza de los costos y la política de ingresos en los países industriales
Resumen
En este artículo se describe la naturaleza de la inflación provocada por el alza de los costos en las economías industriales que persiguen políticas de pleno empleo. Para poder comprender la acción recíproca compleja entre los diversos factores que intervienen, lo cual determina la fuerza de la presión alcista sobre los costos en los diferentes países, es esencial formarse una idea más o menos cabal de cómo se determinan los salarios en las industrias oligopolistas cuyos obreros se encuentran organizados en sindicatos. La primera parte de este artículo describe los factores que influyen en las negociaciones sindicales en materia de salarios y la forma en que se efectúan los aumentos de los mismos. En primer lugar, se analizan las negociaciones sobre los aumentos de salarios que se llevan a cabo entre los diversos sindicatos y varios patronos y, en segundo lugar, otras decisiones ajenas a esos contratos colectivos, que dan lugar a desviaciones alcistas de los salarios. En la segunda parte se analiza la influencia del nivel de la demanda sobre el ritmo de la inflación de los costos y la eficacia de una política de restringir la demanda global como medio de coartar dicha inflación.
En la tendencia a ocurrir movimientos ascendentes de salarios influyen: 1) la presión que la demanda ejerce sobre la oferta en el mercado de la mano de obra, que es un factor determinante del poder de negociación de los sindicatos y determina la probabilidad de que por la escasez de la oferta se pujen los precios y los salarios, y 2) el grado de “organización” de los mercados de productos y de mano de obra que es el factor taxativo de la competencia en esos mercados y que permite a patronos y asalariados coordinar sus respectivas políticas de precios.
El grado de tensión en un mercado nacional de mano de obra, asociado a cierta tasa de desempleo, depende del incremento que experimente la fuerza obrera como resultado del crecimiento demográfico normal, y de la posibilidad de reducir el subempleo en los sectores técnicamente atrasados o de elevar la proporción de trabajadores adultos que forman parte de la fuerza obrera activa. También difiere mucho de un país a otro dicho grado de “organización”. Hay considerables diferencias en el número de asalariados cubiertos por las negociaciones colectivas obrero-patronales, y el poder de negociación de los sindicatos viene condicionado por los aspectos estrucurales del mercado de la mano de obra y al grado de coordinación que existe entre los patronos a efectos de las negociaciones sobre salarios. Resulta más difícil juzgar hasta qué punto las empresas de cada país pueden determinar los precios de venta de su producción; pero hay un elemento de cierta importancia capaz de ser ponderado, y es el grado en que la economía está abierta al exterior. La tercera parte de este estudio trata de las diferencias que se registran en estos aspectos en los diversos países industriales. En la última parte se llega a ciertas conclusiones en cuanto a las posibilidades para una política de ingresos.
Miss Romanis, Assistant Chief of the Current Studies Division, is a graduate of Cambridge University. She was formerly on the staff of the Oxford University Institute of Statistics; Ministry of Production, London; the Economic Directorate, Organization for European Economic Cooperation, Paris; and the Economic Survey Division, United Nations, New York. She has contributed several articles to economic journals.
See J.R. Hicks, “The Instability of Wages,” The Three Banks Review (Edinburgh), No. 31, September 1956, pp. 3–19.
This may have an important influence on the form of wage claims: additional paid holidays or a reduction in weekly hours may appear a more substantial gain than a 1 or 2 per cent increase in wage rates.
In the sense that, if the cost of living were rising by 1 or 2 per cent a year, the rank and file would be less inclined to strike and more prepared to compromise and accept a 6 per cent rise against a claim for an 8 per cent rise when the union was in a strong bargaining position, than to accept 3 per cent against a 4 per cent claim even when the union was in a weaker position.
Differences between countries with regard to the underlying labor market situation and structure of employment are illustrated in Tables 1 and 2 and discussed in Section III, below.
Charles O. Hardy, “Prospects of Inflation in the Transition Period,” Prices, Wages, and Employment, Post-War Economic Studies, No. 4, Board of Governors of the Federal Reserve System (Washington, 1946), p. 24. See also Edward H. Chamberlin, “Labor Union Power and Cost-Inflation,” in Inflation, Proceedings of a Conference Held by the International Economic Association, edited by D.C. Hague (New York, 1962), p. 226.
Differences in the wage bargaining situation in various industrial countries are illustrated in Table 3 and discussed further in Section III, below.
This difference was clearly brought out in the study by the Organization for Economic Cooperation and Development, Wages and Labour Mobility, Economic Studies, Economic Policy Committee, Working Party 4 (Paris, 1965).
Or all employers of a particular occupational group in the case of craft unions.
Chamberlin, op. cit., p. 224.
Which may include the cost of raw materials, manufactured components, equipment, or services.
See Piero Sraffa, “The Laws of Returns Under Competitive Conditions,” The Economic Journal, Vol. XXXVI (1926), pp. 535–50.
For example, in a highly monopolistic industry, comprising a few firms exploiting a new product or technique (where it may be relatively easy to organize the workers, especially if there are heavy capital requirements or special factors anchoring production in a particular area), the union may be able to push wages above the going wage rate in other industries, somewhat reducing the exceptionally high monopoly profits of firms in the industry.
That is, unless the consequent price increase was so large as to have significant income effects or to influence the choice between spending and saving.
Regional considerations apparently have greater influence in wage determination in the United States and Canada than in many European countries, but this is in part the result of the lower degree of unionization.
In statistical studies, wage drift is usually defined as the difference between the actual rate of increase in earnings and the rate of increase which would have resulted from the provisions of the collective wage agreements concerning changes in time and piece rates, overtime rates, etc. Complications may arise in using this concept, however, when the growth of actual earnings is influenced by marked changes in the degree of utilization of labor. See Bent Hansen and Gösta Rehn, “On Wage Drift,” in 25 Economic Essays in Honour of Erik Lindahl (Stockholm, 1956), pp. 87–138. For a description of how wage drift arises and an analysis of its importance in various cases, see also E.H. Phelps Brown, “Wage Drift,” Economica, Vol. XXIX (1962), pp. 339–56, and G. Rehn, Wage Drift in Sweden, Trade Union Research Department, European Productivity Agency (Paris, 1959). The last is an English summary of Löneglidning; Rapport från expertgrupp tillsall at SAF och LO (Stockholm, 1957).
See Shirley W. Lerner, “Wage Drift, Wage Fixing and Drift Statistics,” The Manchester School of Economic and Social Studies, Vol. XXXIII (1965), pp. 155–77.
See National Incomes Commission, Agreements of November-December 1963 in the Engineering and Shipbuilding Industries, Report No. 4, Cmnd. 2583 (London, 1965); Phelps Brown, op. cit.; K.G.J.C. Knowles and D.J. Robertson, “Earnings in Engineering, 1926–1948,” Bulletin of the Oxford University Institute of Statistics, Vol. 13 (1951), pp. 179–200; and T.P. Hill and K.G.J.C. Knowles, “The Variability of Engineering Earnings,” ibid., Vol. 18 (1956), pp. 97–139.
See, for example, Denis Pym, “Is There a Future for Wage Incentive Schemes?” British Journal of Industrial Relations, Vol. II (1964), pp. 379–97.
See Martin P. Oettinger, “Nation-wide Job Evaluation in the Netherlands,” Industrial Relations (Berkeley), October 1964, pp. 45–59.
In Germany the trade unions themselves, led by the engineering and metalworking unions, are pressing for a revision of the traditional occupational classifications to make them more appropriate for modern industrial conditions. See Heinz Markmann (Senior Economist, Deutscher Gewerkschaftsbund), “Incomes Policy in Germany: A Trade Union View,” British Journal of Industrial Relations, Vol. II (1964), pp. 322–39, especially pp. 332–33.
It has been suggested that in the United Kingdom wage drift “has occurred sometimes on the initiative and generally without the active resistance of management,” and that wage drift “will be checked only as managers accept responsibility for their domestic wage structures, and move away from extemporization and its chain reactions towards, for instance, the negotiation and consistent application of works agreements” (Phelps Brown, op. cit., p. 353).
Under conditions of heavy and persistent unemployment, there may be some converse tendency to “negative wage drift,” in that collective agreements may be less effectively enforced; if employees are conscious of competing for jobs, pressure on costs may be reduced by the weaker attitude of workers toward the establishment of wage-fixing practices at the plant level (i.e., less insistence on payment of the rate for the job, “loose” piece rate fixing, and the like).
John Corina’s unpublished doctoral dissertation, Trade Unions and Wage Restraint, prepared at Nuffield College, Oxford, in 1962, clearly demonstrates the importance of this process in causing the collapse of the 1949–50 pay pause in the United Kingdom.
Increases in negotiated wages that exceed productivity gains are often regarded as evidence of cost-push. However, this view is an oversimplification, because when demand for labor is strong, increases in negotiated wages may merely serve to bring official rates more closely into line with wages already being paid by employers in the industry. This point was not made clear in the main text of The Problem of Rising Prices, by William Fellner, Milton Gilbert, Bent Hansen, Richard Kahn, Friedrich Lutz, and Pieter de Wolff, Organization for European Economic Cooperation, Paris, 1961 (hereafter cited as The Problem of Rising Prices). However, in the separate note on Wage Drift (p. 67), the OEEC experts conceded that indications of wages actually being paid, and hence of the current rate of drift, were often an important factor in wage negotiations, and advanced the view that “generally speaking the collective agreements have provided a large number of moving pegs to which the wages actually paid have remained tethered.” The analogy does not, of course, rule out the possibility that the pegs were moved precisely because the correspondence between wages actually being paid and official rates was becoming strained.
Phelps Brown, op. cit., p. 354. The individual employer “may even suspect that the general round of wage increases will generate sufficient demand to absorb an undiminished supply at higher prices” (Paul Streeten, “Wages, Prices and Productivity,” Kyklos (Basle), Vol. XV (1962), p. 725).
E. H. Phelps Brown, The Economics of Labor (New Haven and London, 1962), p. 174.
Jack Downie (late Chief Economist, Organization for Economic Cooperation and Development), “The Importance of Knowing What You Want,” in Part III, “What Can We Learn from European Experience?” of Unemployment and the American Economy [First Conference on Unemployment and the American Economy, Berkeley, California, 1963], edited by Arthur M. Ross (New York, 1964), p. 163.
Phelps Brown, loc. cit.
Profits per unit of output will rise by the same proportion as unit wage costs where oligopoly pricing is based on a percentage mark-up on costs.
As, for example, the OEEC experts imply (The Problem of Rising Prices, P. 69).
It is misleading to suggest that, if “collective bargaining results in wage increases greater than the supply and demand forces alone would have yielded” and in excess of the growth in productivity, this is the result of wage-push and not, in part, a consequence of the fact that the monopolistic power of certain producers enables them to pass on increased costs to the buyer (The Problem of Rising Prices, pp. 45–46).
“Excess demand” was defined by the OEEC experts as a “volume of aggregate demand which could not be met at existing prices without exerting undue pressure on productive resources [so that] capacity becomes strained, a general shortage of labor develops and prices and wages are bid up by buyers and employers competing for scarce resources” (The Problem of Rising Prices, p. 33).
See Lord Kahn’s evidence before the Radcliffe Committee: Committee on the Working of the Monetary System, Principal Memoranda of Evidence (London, 1960), Vol. 3, p. 142, especially pars. 35–37. Lord Kahn, a member of the OEEC expert group, disassociated himself from their analysis in terms of excess demand (The Problem of Rising Prices, footnote 1 on p. 33).
A.W. Phillips, “Employment, Inflation and Growth,” and “The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861–1957,” Economica, Vol. XXV (1958), pp. 283–99, and Vol. XXIX (1962), pp. 1–16.
Such as L.A. Dicks-Mireaux and J.C.R. Dow, “The Determinants of Wage Inflation, United Kingdom 1946–56,” Journal of the Royal Statistical Society, Vol. 122, pt. 2 (1959), pp. 145–83; Richard G. Lipsey, “The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1862–1957: A Further Analysis,” Economica, Vol. XXVII (1960), pp. 1–31; and L.R. Klein and R. J. Ball, “Some Econometrics of the Determination of Absolute Prices and Wages,” The Economic Journal, Vol. LXIX (1959), pp. 465–82. However, Rattan J. Bhatia did not find a significant relationship in his study, “Unemployment and the Rate of Change of Money Wages Rates in the United States, 1900–1958,” Economica, Vol. XXVIII (1961), pp. 286–96; and K.B. Griffin concluded that, with regard to the postwar U.S. data, “the notion of the Phillips curve can be subjected to any interpretation one desires!” (“A Note on Wages, Prices and Unemployment,” Bulletin of the Oxford University Institute of Statistics, Vol. 24 (1962), pp. 379–85).
F.W. Paish, “The Limits of Incomes Policies,” in Policy for Incomes? by F.W. Paish and Jossleyn Hennessy, Institute of Economic Affairs, Hobart Papers, No. 29 (London, 1966), pp. 11–46.
Pierre Massé, Rapport sur la politique des revenus établi à la suite de la Conférence des revenus (octobre 1963-janvier 1964), La Documentation Française, Recueils et Monographies, N° 47 (Paris, 1964), p. 7.
In such a case, the position of the authorities necessarily “contains an important element of bluff. [If] any government [tries] to give absolute priority to price stability [and] this involves heavy unemployment, it may not remain in office.”—Sir Donald MacDougall, “Inflation in the United Kingdom,” The Economic Record, Vol. XXXV (1959), p. 387.
This course of development has been well described in France—see Jean Marchai, “Les conditions de l’équilibre macroéconomique dans la stabilité des prix,” Revue Economique (Paris), N° 6 (1964), pp. 853–67—and in the United Kingdom—see J.C.R. Dow, The Management of the British Economy, National Institute of Economic and Social Research, Economic and Social Studies, XXII (London, 1964), and H.M. Treasury, Economic Report on 1964 (London, 1965), published as a supplement to Economic Trends, No. 137, March 1965.
Such behavior is not irrational if rejecting a wage claim exposes the concern to the risk of a prolonged strike and a substantial loss of profits which cannot be made up in the future, or to the possibility of a permanent reduction in its share of the market (Marchal, op. cit., p. 855; Massé, op. cit., p. 7).
See H.A. Turner and H. Zoeteweij, Prices, Wages, and Income Policies in Industrialised Market Economies, International Labor Office (Geneva, 1966), pp. 30–31.
This framework of exposition is similar to that used by Professor Duesenberry in a paper presented to the American Finance Association in December 1965: J. Duesenberry, “Domestic Policy Objectives and the Balance of Payments,” The Journal of Finance, May 1966, pp. 345–53.
The present growth rate in Germany is considerably lower owing to the low birth rate in the years immediately after the war.
Since expenditure on food does not rise in line with income as the standard of living rises, it is virtually impossible for agricultural incomes to rise as fast as those in manufacturing and services, at the same time as the relative prices of food and other consumer goods remain unchanged, unless the proportion of the population engaged in agriculture is declining.
Including the need to provide more jobs in other sectors to permit the absorption of manpower out of low-productivity employment in agriculture and services.
Even if unemployment is very low, there may be scope for expanding employment, as a result of an increasing population of working age, substantial immigration, or the possibility of reducing underemployment in backward sectors. It may be remarked that substantial immigration is more likely to be permitted in countries where the extent of unionization is not great.
