UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The theory and measurement of structural unemployment Penz, G. Peter 1968

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1969_A8 P45.pdf [ 5.88MB ]
Metadata
JSON: 831-1.0104159.json
JSON-LD: 831-1.0104159-ld.json
RDF/XML (Pretty): 831-1.0104159-rdf.xml
RDF/JSON: 831-1.0104159-rdf.json
Turtle: 831-1.0104159-turtle.txt
N-Triples: 831-1.0104159-rdf-ntriples.txt
Original Record: 831-1.0104159-source.json
Full Text
831-1.0104159-fulltext.txt
Citation
831-1.0104159.ris

Full Text

THE THEORY AND MEASUREMENT OF STRUCTURAL UNEMPLOYMENT by G. PETER PENZ B.A., University of British Columbia, 1963 A THESIS. SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of ECONOMICS We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December, 1968 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h C olumbia, I ag r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and Study. I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Depa r tmen t o f Economics The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Date December 9» 1968 ABSTRACT The purpose of this thesis is to develop a theoretical framework which could then be used to measure structural unemployment. This i s done by f i r s t surveying the relevant literature, then developing a theoretical model for the measurement of structural unemployment, and f i n a l l y applying this model to Canadian data. In the survey of the relevant literature the various approaches are categorized into the causal, the structural maladjustment and the policy approaches. The causal approach involves explaining structural unemployment in terms of the causes of labour displacement. This i s considered inadequate because i t ignores problems in the labour market adjustment process, whose function i s to wipe out the imbalances created by structural dislocations. These problems are considered by the structural maladjustment approach, together with the symptoms of structural maladjustment. It analyzes the forces promoting and impeding the labour market adjustment process. However, these forces are at present not quantifiable. The symptoms of structural maladjustment, on the other hand, are. A favourite method of determining changes i n structural maladjustment has been the analysis of the structure of unemployment. This thesis, however, supports the contention that this method is generally misleading. Analyses of long-duration unemployment are also considered not to be useful, but a framework involving the relationship of unemployment to vacancies i s seen as f r u i t f u l . The policy approach i s concerned with the relationship of unemploy-ment and inflation. According to this,approach, the degree of structural maladjustment i s indicated by the distance of the inflation-unemployment function from the origin. However, there are problems involved in using it. to measure structural unemployment, primarily because of the impurities involved in the relationship. i i The next step i s to develop a model which does not depend on proxies for labour demand, but uses variables directly related to the labour market, and has a theoretical rationale. This model must separate the effects of aggregate demand and of structural imbalances on unemployment. It does this by determining the cyclical relationship between the unemploy-ment and vacancy rates and attributing changes which cannot be explained by this relationship to changes i n the level of structural imbalances. . This model i s then applied to Canadian data. Before that can be done, however, the vacancy rate has to be derived from N.E.S. vacancy data. The ratio of actual vacancies to N.E.S. vacancies i s estimated on the basis of the ratio of total hirings to N.E.S. placements. Using the vacancy rate thus estimated, several forms of the relationship between the unemployment and vacancy, rates are empirically tested. The results indicate that very l i t t l e of the changes in the total unemployment rate are attributable to changes in structural imbalances. Variability in unemployment i s largely caused by variability i n aggregate demand. However, there appears to have been some upward trend i n the structural unemployment rate (defined as the unemployment rate that would prevail i f aggregate labour demand were equal to aggregate labour supply), from 3 per cent i n the early 1950's to nearly k per cent i n the 1960's. These results suffer from the uncertainty involved in the estimated vacancy rate, but an analysis of changes in the ratio of total hirings to N.E.S. placements, which was used in the estimationj supports the findings concerning the structural unemployment trend. TABLE OF CONTENTS Page ABSTRACT • i TABLE OF CONTENTS i i i i LIST OF TABLES .... v LIST OF FIGURES .. v i ACKNOWLEDGMENT... • ' v i i CHAPTER I - INTRODUCTION 1 1. The Controversy about Structural Unemployment 1 2. The Purpose of the Thesis 2 3. The Outline 2 FOOTNOTES . . . h CHAPTER II - A SURVEY OF THE THEORY OF STRUCTURAL UNEMPLOYMENT.... 5 A. The Causal Approach 5 1. Types of Structural Changes. 5 2. Technological Changes and Structural Displacement.... 7 3> Evaluation 11 B. The Structural Maladjustment Approach. 13 1. Definitions 13 2. The Labour Market Adjustment Mechanism Ik 3. Manifestations of Structural Maladjustment...... 22 k. Evaluation 35 C. The Policy Approach 36 1. The Policy Implications of Berman's Bottleneck Model. 36 2. Lipsey's Trade-Off Model.. 37 iv TABLE OF CONTENTS - (Continued) Page 3. The Policy Implications V? k. Evaluation V? FOOTNOTES..... ^9 CHAPTER III - A THEORETICAL MODEL OF STRUCTURAL DISEQUILIBRIUM FOR THE^ - MEASUREMENT OF STRUCTURAL UNEMPLOYMENT % 1. Conceptual Definitions.... 57 2. Refinement of the Measure of Structural Unemployment. 60 3. The Relevant Definition of a Job Vacancy 61 if. The Model... 67 5. Summary 75 FOOTNOTES ... 78 CHAPTER IV - APPLICATION OF THE STRUCTURAL DISEQUILIBRIUM MODEL TO CANADIAN DATA 80 1. The Data Problems. 80 2. The Estimation of the Structural Unemployment Rate... 8k FOOTNOTES 96 CHAPTER V - CONCLUSION 97 1. Summary of the Theory... 97 2. The Empirical Analysis 98 3. Policy Implications......... 99 k. Further Research. 101 FOOTNOTES 102 BIBLIOGRAPHY 103 LIST OF TABLES Page Table I - Ratio of Total Hirings to N.E.S. Placements... 83 Table II - Estimates of the Structural Unemployment Rate 92 i LIST OF FIGURES Page Figure 2.1 - Berman's bottleneck model: the employment expansion path and the s k i l l endowment constraints 23 Figure 2.2 - The Dow-Dicks-Mireaux system of functions: (a) the unemployment-vacancy functions; (b) the unemployment-demand and vacancy-demand functions 27 Figure 2.3 - Price inflation and unemployment: the behavioural trade-off functions and the maximum-inflation constraint 38 Figure 2.h - Price inflation and unemployment: the behavioural trade-off functions and the policy-makers' preference functions ^1 Figure 3-1 - An i l l u s t r a t i o n of the effect of accumulating labour bottlenecks in different industries on the relationships between overall labour demand (D), on the one hand, and unemployment (U) and vacancies (V), respectively, on the other.... 73 Figure 3.2 - The cyclical relationship of the unemployment rate, u, and the vacancy rate, v, (a) to relative labour demand, d, and (b) to each other 76 Figure ^.1 - The effect of minimizing the residuals in the unemployment-vacancy relationship in the cases of (a) a hyperbolic function and (b) a parabola rotated by 86 Figure 5.1 - Quarterly series of total, structural and demand-deficiency unemployment rates, based on the rotated parabola estimate 100 ACKNOWLEDGMENT I am grateful to Dr. J. Tait Montague for assuming the i n i t i a l responsibilities as supervisor and to Dr. John Vanderkamp, who subsequently assumed these responsibilities, for his advice and encouragement during the somewhat d i f f i c u l t non-resident phase in the writing of the thesis. CHAPTER I INTRODUCTION , 1. The Controversy about S t r u c t u r a l Unemployment In the f i r s t h a l f of t h i s decade a controversy raged over whether the sharp increase i n unemployment i n North America between 1956 to 1961 has p r i m a r i l y or at l e a s t p a r t l y represented an increase i n s t r u c t u r a l unemploy-ment, or whether i t has been p r i m a r i l y or wholly due to a deficiency i n aggregate demand. One of the major d i f f i c u l t i e s i n t h i s controversy has been that the s t r u c t u r a l i s t p o s i t i o n seems to have emerged from p o l i t i c a l polemics and has r a r e l y been formulated i n rigorous economic terms by a proponent of t h i s position.' 1' This has l e d to a v a r i e t y of i n t e r p r e t a t i o n s of "the s t r u c t u r a l i s t hypothesis". Aside from the problem of i n t e r p r e t a t i o n , there has been the problem of t e s t i n g the hypothesis. The t e s t s have taken the form of analyses of p r o d u c t i v i t y changes and t h e i r d i s t r i b u t i o n among the various sectors of the economy, of the d i s t r i b u t i o n of unemployment among the various groups of the labour force and changes i n i t , of changes i n long-duration unemployment, 2 of i n d i c a t o r s of job vacancies and of wage and p r i c e i n f l a t i o n . From nearly a l l of these t e s t s the conclusion was drawn that there was no s i g n i f i c a n t increase i n s t r u c t u r a l unemployment during t h i s period. However, as Richard G. Lipsey has demonstrated,"^ many of these empirical t e s t s have not been f u l l y thought out at the t h e o r e t i c a l l e v e l . While they often presented i n t e r e s t i n g f a c t s , the l i n k s to the conclusions which were drawn from the findings were often missing. Lipsey, therefore, concluded that a formal theory of s t r u c t u r a l unemployment i s required. 2 2. The Purpose of the Thesis The purpose of t h i s t h e s i s i s to t r y to develop a t h e o r e t i c a l framework which can then be used to measure s t r u c t u r a l unemployment. Before that i s done, however, the t h e o r e t i c a l work that has already been done oh the subject of s t r u c t u r a l unemployment and on aspects r e l a t e d to i t i s drawn together. T h i s survey w i l l a l s o r e f e r to e m p i r i c a l t e s t s that have been undertaken, but not f o r t h e i r r e s u l t s , r a t h e r f o r t h e i r t h e o r e t i c a l assumptions. These must be sound before the r e s u l t s can be t r u s t e d to provide r e l i a b l e q u a n t i t a t i v e feed-back f o r the theory of s t r u c t u r a l unemployment, and t h i s requirement has not been met i n much of the work i n t h i s area. From t h i s survey the most promising l i n e of model-building i s then s e l e c t e d and f u r t h e r developed to make the measurement of s t r u c t u r a l unemployment p o s s i b l e . This model i s f i n a l l y a p p l i e d to Canadian data to determine the l e v e l and trend of s t r u c t u r a l unemployment i n t h i s country. 3. The O u t l i n e The survey of the t h e o r e t i c a l work i n the area of s t r u c t u r a l unemployment .in Chapter I I i s d i v i d e d i n t o three p a r t s . In the f i r s t p art the causal approach to the determination of s t r u c t u r a l unemployment i s presented. I t contains analyses of the f a c t o r s u n d e r l y i n g s t r u c t u r a l displacement. In the second p a r t , analyses of the f a c t o r s promoting and i n h i b i t i n g s t r u c t u r a l adjustment are presented as w e l l as the theory u n d e r l y i n g some of the t e s t s of the l e v e l or tren d of the s t r u c t u r a l d i s e q u i l i b r i u m or maladjustment that i s the net e f f e c t of these f a c t o r s . The t h i r d p a r t i s cat e g o r i z e d as the p o l i c y approach, according to which s t r u c t u r a l unemployment i s assessed i n terms of the r e l a t i o n s h i p between unemployment and i n f l a t i o n . This r e l a t i o n s h i p i s s i g n i f i c a n t because 3 economic policy objectives demand the minimization of both variables, but reducing one has been observed to result i n increasing the other, and this trade-off relationship i s at least partly due to structural imbalances. In Chapter III the concepts of structural equilibrium and disequilibrium are c l a r i f i e d and the Dow-Dicks-Mireaux framework, whose central variables are unemployment and vacancies, i s used to analyze the interaction of changes in aggregate demand and of structural imbalances. This analysis makes i t possible to separate the effects of these two factors on unemployment and therefore to obtain a measure of structural unemployment. In Chapter IV this measure of structural unemployment i s applied to Canadian data. Since the vacancy statistics are very inadequate, an attempt i s made to derive an improved estimate from them. Empirical tests are then used to obtain the relationship between the unemployment and vacancy rates which are necessary to fi n a l l y obtain the structural unemployment rate. Finally, the r e l i a b i l i t y of the results i s discussed. Chapter V contains a brief review of the theory and the empirical analysis. The policy implications of the results and the direction for further research are also discussed i n i t . FOOTNOTES TO CHAPTER I The only exception seems to be.Charles C. Killingsworth, "Automation, Jobs, and Manpower", The Nation's Manpower Revolution, Hearings before the Subcommittee on Labor and Public Welfare, U.S. Senate, 88th Congress, 1st session, part 5, 1963, PP- 1461-83. See, for example, U.S. Congress, Joint Economic Committee, Subcommittee on Economic Statistics, Higher Unemployment Rates, 1957-60: Structural  Transformation or Inadequate Demand, 8?th Congress, 1st session, 1961; Lowell E. Gallaway, "Labour Mobility, Resource Allocation, and Structural Unemployment", American Economic Review, Vol. 53, No. 4, Sept. 1963, pp. 694-716; Walter W. Heller, "The Administration's Fiscal Policy", Unemployment and the American Economy, ed. Arthur M. Ross, John Wiley & Sons, Inc., New York, (1964), pp. 93-115; Otto Eckstein, "Aggregate Demand and the Current Unemployment Problem", Unemployment and the  American Economy, pp. 116 f f . ; R.A. Gordon, "Has Structural Unemployment Worsened?", Industrial Relations: A Journal of Economy and Society, Institute of Industrial Relations, University of California, Berkeley, Vol. 3, No. 3, May 1964, pp. 53-78; N.J. Simler, "Long-Term Unemployment, the Structural Hypothesis, and Public Policy", American Economic Review, Vol. 54, No. 6, Dec. 1964, pp. 985-IOOI; Frank T. Denton and Sylvia Ostry, An Analysis of Postwar Unemployment, Economic Council of Canada, Staff Study No. 5, Queen's Printer, Ottawa, Dec. 1964; John Vanderkamp, "An Application of Lipsey's Concept of Structural Unemployment", Review of  Economic Studies, July 1966, pp. 221-5. Richard G. Lipsey, "Structural and Deficient-Demand Unemployment Reconsidered", Employment Policy and the Labor Market, ed. A.M. Ross, University of California Press, 1965, pp. 210-55. After the completion of the survey chapter, an article entitled "Structural Unemployment" by John W.L. Winder was published in A.M. Kruger and N.M. Meltz (eds.), The Canadian Labour Market: Readings i n Manpower  Economics,.Centre for Industrial Relations, University of Toronto, 1968, pp. 135-220. It, too, i s a survey of the literature, but a more comprehensive one than the survey chapter in this thesis. However, i t seemed to me to be somewhat less c r i t i c a l and there are certain differences in emphasis between the two surveys, with the Winder article stressing the trade-off approach.and ignoring the Dow-Dicks-Mireaux model. CHAPTER II A SURVEY OF THE THEORY OF STRUCTURAL UNEMPLOYMENT There have been basically three approaches to the theory of structural unemployment: (l) the causal approach, (2) the structural maladjustment approach, and (3) the policy-oriented approach. The causal approach focuses on the different types of structural change and their effect on structural unemployment. The structural maladjustment approach concerns itself with the adjustment mechanism of the labour market which is responsible for absorbing structural changes and with the manifestations of its shortcomings. The policy-oriented approach defines and analyzes structural unemployment in terms of the effectiveness of various policies. The following survey is divided up according to these three approaches. A. THE CAUSAL APPROACH The causal approach, on the whole, has consisted of loose descriptions.of the structural changes in the economy that might be respon-sible for structural unemployment rather than rigorous theoretical analyses of the relationship between structural changes and structural unemployment. The.causal approach is exemplified by Paul Casselman's succinct, but very general definition: Structural unemployment may be defined as unemployment resulting from changes in the economic structure and in the economic environment.! 1. Types of Structural Changes What is meant by such structural changes has been spelt out in more detail in the lengthy definition given by H.D. Woods and Sylvia Ostry. 6 S t r u c t u r a l unemployment...is that unemployment which stems from: (1) Major s h i f t s i n consumer demand a r i s i n g from the c r e a t i o n of new products, spontaneous or induced changes i n t a s t e , the growth of competition of an imported commodity, e t c . , which reduce job o p p o r t u n i t i e s f o r workers i n a s p e c i f i c i n d u s t r y or group of i n d u s t r i e s , s p e c i f i c l o c a l areas or r e g i o n s . (2) T e c h n o l o g i c a l changes which i n v o l v e the s u b s t i t u t i o n of c a p i t a l f o r labour w i t h i n a given i n d u s t r y or make redundant or obsoles-cent s p e c i f i c s k i l l s or products. A major t e c h n o l o g i c a l change may introduce an e n t i r e l y new i n d u s t r y which g r a d u a l l y destroys an e s t a b l i s h e d i n d u s t r y — m o t o r cars and the c a r r i a g e and wagon i n d u s t r y ; e l e c t r i c l i g h t and o i l lamps; r e f r i g e r a t o r s and i c e boxes, e t c . In such cases the r e s u l t i n g unemployment would a r i s e from a combination of (2) and ( l ) . (3) The exhaustion of n a t u r a l resources i n a given area. (k) Changes i n the o r g a n i z a t i o n or ownership of i n d u s t r y that r e s u l t i n the c l o s i n g down of c e r t a i n p l a n t s f o r reasons other than those above.2 Presumably the l a s t two types of s t r u c t u r a l change r e s u l t i n the geographic concentration of labour displacement and, t h e r e f o r e , i n s t r u c t u r a l unemployment. Casselman suggested three other p o s s i b l e causes of s t r u c t u r a l unemployment: (1) government p o l i c y , (2) war or the t h r e a t of war, and (3) the r a t e of p o p u l a t i o n growth."^ There I s no doubt that government p o l i c y , which.may have to observe c r i t e r i a other than the smooth t r a n s f e r of labour without unemployment, can introduce sharp changes i n the types of labour demanded. Increased t a x a t i o n to pay f o r higher expenditures f o r s o c i a l c a p i t a l , s o c i a l s e r v i c e s and m i l i t a r y expenditures may l e a d to a s h i f t i n demand away from the p r i v a t e goods and s e r v i c e s s e c t o r . Changes i n the production f o r m i l i t a r y purposes have probably been s p e c i f i c a l l y mentioned because of the magnitude of s t r u c t u r a l changes i n v o l v e d i n disarmament and rearmament. Changes i n the r a t e of p o p u l a t i o n growth, i f 7 sudden as i n the case of the "baby boom", can create inbalances on the supply side by changing the r a t i o of labour market entrants to experienced workers. 2. Technological Change and S t r u c t u r a l Displacement  The form of s t r u c t u r a l change which i s most frequently r e f e r r e d to as being responsible for s t r u c t u r a l unemployment i s technological change. I t w i l l therefore be u s e f u l to see what r e l a t i o n between technological change and s t r u c t u r a l unemployment i s presumed to e x i s t according to contemporary l i t e r a t u r e . a. Microeconomic e f f e c t s of p r o d u c t i v i t y change J.W. Knowles and E.D. Kalacheck have presented i n t h e i r empirical study of the s t r u c t u r a l unemployment controversy a b r i e f t h e o r e t i c a l analysis of the d i f f e r e n t forms of p r o d u c t i v i t y change and t h e i r impact on unemploy-k ment at the microeconomic l e v e l . The analysis r e f e r s not only to the r e s u l t i n g s t r u c t u r a l displacement but also to the creation of job openings by technological change: Job opportunities are continuously being c u r t a i l e d i n technolo-g i c a l l y unprogressive i n d u s t r i e s , and i n those t e c h n o l o g i c a l l y advancing i n d u s t r i e s , the demand for whose product i s i n e l a s t i c with respect to p r i c e . At the same time, job opportunities are being created i n i n d u s t r i e s blessed with the junction of rapid p r o d u c t i v i t y increases and p r i c e - e l a s t i c demands, and i n other i n d u s t r i e s whose demands are highly e l a s t i c with respect to income.5 Knowles and Kalacheck have i d e n t i f i e d four major types of p r o d u c t i v i t y increases: ( l ) capacity expansion, (2) modernization, (3) geographic s h i f t s , and (k) the c l o s i n g of obsolete f a c i l i t i e s . The impact on labour displacement d i f f e r s with the type. 8 Technological change incorporated in capacity expansion in order to meet growing demand w i l l lead to no direct labour displacement. However, to the extent that part of the expansion in demand i s at the expense of competing firms, there w i l l be some indirect displacement effect. When technological change i s introduced to modernize the plant without expanding i t , the number of lay-offs at the particular plant may be quite significant. If there i s no increase in output, the modernizing plant w i l l have to dismiss superfluous workers. Even i f i t does increase output, i t may s t i l l have to dismiss unsuitable workers, while hiring (possibly less) workers more suitable to the new production techniques; at the same time, lay-offs w i l l occur at competing plants. If the increasing of productivity requires a shift i n the location of the plant, this may result in the separation of the whole work force of a plant. The actual extent of the displacement effect w i l l depend on the number of workers who move with the plant. In the case where a plant has become obsolete and submarginal, the immediate displacement effect i s 100 per cent. To; what extent labour dispacement w i l l lead to unemployment depends on the labour market conditions. Since "modernization and the closing of obsolete f a c i l i t i e s become more significant contributors /to productivity advance_s7 during periods when the capacity u t i l i z a t i o n ratios are low and aggregate demand i s growing at a slow rate",^ these two factors are l i k e l y to also contribute to unemployment. Capacity expansion, on the other hand, generally occurs during periods of high labour demand so that lay-offs are f a i r l y easily reabsorbed in the employed labour force. The length of unemployment w i l l depend on the concentration of lay-offs and on the types of labour l a i d off. The closing of plants leads to large-scale lay-offs in one area, including older workers with low 9 mobility and f l e x i b i l i t y . This is l i k e l y to result in longer unemployment periods than plant modernization, where lay-offs are more limited and where seniority rules are l i k e l y to protect the older workers, b. Macroeconomic effects of productivity change In discussing the basic assumptions of the structuralist argument that the increases in unemployment after 1957 have been primarily structural, Knowles and Kalacheck identified three ways in which technological change could possibly lead to a rise in structural unemployment in the economy at large: (l) an acceleration of the overall rate of productivity change, (2) an increased concentration of productivity gains in a select group of industries, and (3) a change in the qualitative impact of productivity 7 increases on the occupational and s k i l l structure of the demand for labour. The effect of productivity changes on employment w i l l depend on the distribution of price changes and price e l a s t i c i t i e s . Greater than average productivity increases i n a particular establishment or industry w i l l lead to the displacement of labour, i f the demand for i t s product or service i s inelastic with respect to price, or i f reductions in relative costs per unit are not passed on to the consumer in the form of quality improvements or commensurate declines i n relative price. Employment w i l l rise in establishments or industries with greater than average productivity increases i f demand i s price-elastic, and i f prices are reduced. However, labour displacements may then occur in less technolo-gically progressive industries producing substitute goods. Workers losing specific jobs w i l l experience a certain number of weeks of unemployment while hunting for a new job. Consequently, taking a l l possible combinations of these events into account, i t is often assumed that a l l other things being equal, the higher the increase in output per man-hour, the higher the unemployment rate.8 The sectoral concentration of productivity increases can lead to higher unemployment i f the productivity increases are concentrated in sectors (l) where the price elasticity of demand i s low or the benefits of the productivity increases are not passed on to the consumer, and (2) where demand i s sensitive to changes in price, but increases in output occurred at the expense of closely competitive industries producing substitute goods. io If this i s the actual situation, the duration of unemployment may also be expected to lengthen since the concentration of lay-offs w i l l affect high-seniority workers with low mobility and high-surplus labour market areas 9 may develop. The distribution of income e l a s t i c i t i e s , too, w i l l determine the actual rate of displacement. If those industries which experience the greatest productivity gains also have high income el a s t i c i t i e s of demand, the rate of displacement may actually be quite low. On the other hand, i f the employment effect of increased income i s concentrated in the industries with negligible productivity changes, the displacement effect w i l l not be reduced. Charles C. Killingsworth has argued that this latter case in effect represents the situation in the North American economy. When a labor-saving invention is introduced i n an industry which i s i n i t s rapid growth s t a g e — i t s adolescence—the invention may help to spur further rapid growth, especially through price cuts, and total employment in the industry may increase substantially. This i s the historical pattern which prompts many people to argue that "machines make jobs". But the fact i s that when an industry has reached maturity—for example, when there i s already one car for each three people—it just i s not possible to achieve further dramatic increases in sales, even with the largest price cuts within the realm of reason. The improved productivity made possible by labor-saving machines simply enables the industry to keep up with the normal growth of the market while employing fewer production workers.10 Killingsworth then proceeded to argue that the mass-producing consumer goods industries, which are most affected by automation today, are faced with relatively saturated markets, while most of the growth occurs i n the technologically less affected service industries. From the high technolo-gical displacement effect and the relatively low employment effect of increased income i t follows that there i s a relatively high rate of lay-offs in these industries. There are certain barriers to the immediate re-employment of those l a i d off so that technological or structural 11 unemployment results. Killingsworth 1s case i s plausible, but i t s quantitative significance for unemployment has not been demonstrated, nor was evidence provided for the assertion that the industries in which automation i s concentrated face saturated markets. According to the foregoing analysis, i f productivity increases were evenly distributed among a l l industries and enterprises and i f the price e l a s t i c i t i e s , the cross e l a s t i c i t i e s and the income e l a s t i c i t i e s for a l l were the same, there should be no structural or technological lay-offs. However, technological change may result in new and different requirements for labour. Knowles and Kalacheck refer to i t as the qualitative impact 12 of technological change. New s k i l l s and occupations w i l l experience increasing demand, while obsolete types of labour are l a i d off. The unemployment which such changes i n the occupational and s k i l l structure of demand precipitates i s often regarded as the most serious, since i t requires the conversion of the labour services that a worker can offer before he i s re-employed. It i s li k e l y to be of particularly long duration. In addition, Richard G. Lipsey has argued that technological change can lead tb a redistribution of income which i n turn can affect the level of employment. If i t redistributes income from wage-earners i n general to profit-earners and from unskilled workers i n particular to ski l l e d workers, and i f the beneficiaries of technological change have a lower propensity to 13 consume than the unskilled workers, then aggregate demand w i l l be reduced. Unemployment which i s created by this process must, however, be regarded as demand-deficiency unemployment, since i t can be eliminated by stimulating aggregate demand. 3. Evaluation The major merit of the causal approach to the determination of structural unemployment i s that i t puts i t s finger on those forces which 12 bring about labour displacement but which do not emerge from a reduction in aggregate demand. It points out that the dynamism of an economy can lead to lay-offs and unemployment. On the other hand, the causal approach has the serious shortcoming of neglecting the relationship between structural displacement and structural unemployment. It does not analyze the forces determining the rate of hiring, which i s just as important as the rate of displacement in the determination Ik of structural unemployment. In other words, structural adjustment, which has the opposite effect to structural displacement, i s ignored. There.seems to be only the implicit assumption that the labour market has a limited capacity for adjustment and, when overloaded by structural changes, permits a significant amount of non-demand-deficiency unemployment to deposit i t s e l f . On the basis of this assumption, we should expect structural unemployment to increase during a period of rapid growth, since i t i s l i k e l y to be accompanied by rapid technological change and shifts in consumer demand. Yet, the Canadian Senate's Committee on Manpower and Employment contended that maladjustments increase during periods of slow 15 growth. This was attributed to a "lagging rate of adjustment". The implicit assumption in this case seems to be that structural changes are more or less independent of the rate of growth and that the labour market's ab i l i t y to adjust supply to demand i s adversely affected by slow growth. This may be the case i f the mobility of labour i s reduced under such conditions. Conversely, the adjustment capacity of the labour market may be higher during a period of rapid and extensive structural changes when a high pressure of labour demand may be acting as a stimulant to retraining and labour mobility. 13 A l l this points to the question of how effective the adjustment mechanism of the labour market i s . This is discussed under the structural maladjustment approach to the theory of structural unemployment. B. THE STRUCTURAL MALADJUSTMENT APPROACH 1. Definitions The essence of the structural maladjustment approach i s given by i t s definitions of structural unemployment. While they generally also refer to structural changes in the economy as the causes of unemployment, they concentrate on the imperfections and r i g i d i t i e s i n the labour market's mechanism of adjustment between supply and demand and the manifestations thereof. The following three complementary definitions are examples of this approach. In the study of Frank T. Denton and Sylvia Ostry, structural unemployment has been defined as "long-duration unemployment arising from structural changes in the character of demand for labour which require transformation of the labour supply, a time-consuming process".^ A similar, but less precise, definition has been offered by the Senate Committee on Manpower and Employment: Structural unemployment " i s attributed to the failure of the economy to adjust at a sufficient rate to changing circumstances".^ The definition provided by Pierre-Paul Proulx i s more static. It does not expl i c i t l y refer to structural changes and the adjustment process, but focuses on the manifestations of structural imbalances at any one point of time. It describes structural unemployment as "long lasting unemployment due to a mismatching between the education, training, s k i l l s , locations, age l8 and sex of work seekers and the requirements of employers". These definitions are certainly more useful for the analysis of structural unemployment in that they refer not only to the disequilibrating Ik e l e m e n t s o f s t r u c t u r a l c h a n g e , b u t a l s o t o e q u i l i b r a t i n g e l e m e n t s i n t h e a d j u s t m e n t m e c h a n i s m o f t h e l a b o u r m a r k e t . One m i g h t s a y t h a t t h e y c o n c e r n t h e m s e l v e s w i t h t h e " n e t d i s e q u i l i b r i u m " r e m a i n i n g a f t e r t h e a d j u s t m e n t m e c h a n i s m h a s d i g e s t e d a s m u c h o f t h e d i s e q u i l i b r a t i h g e f f e c t s i t h a s b e e n f e d a s i t i s a b l e t o . P r o u l x ' s d e f i n i t i o n i s p a r t i c u l a r l y h e l p f u l f o r t w o r e a s o n s : (1) I t c l e a r l y c o n c e p t u a l i z e s t h e n e t d i s e q u i l i b r i u m o f t h e l a b o u r m a r k e t . (2) I f t h e n e c e s s a r y s t a t i s t i c s a r e a v a i l a b l e , i t c a n b e r e a d i l y q u a n t i f i e d . One o b j e c t i o n , h o w e v e r , m u s t b e r a i s e d t o t h e r e f e r e n c e t o l o n g - d u r a t i o n u n e m p l o y m e n t i n t h e D e n t o n - O s t r y a n d t h e P r o u l x d e f i n i t i o n s . S t r u c t u r a l u n e m p l o y m e n t , may j u s t a s w e l l b e o f s h o r t d u r a t i o n , i f t h e r e i s a h i g h r a t e o f s t r u c t u r a l d i s p l a c e m e n t b u t a t t h e s a m e t i m e a r e l a t i v e l y r a p i d r a t e o f l a b o u r r e a b s o r p t i o n . T h o s e d i s p l a c e d w i l l b e u n e m p l o y e d f o r s o m e f a i r l y s h o r t p e r i o d b e f o r e r e - e m p l o y e d e l s e w h e r e . T h e y a r e , n e v e r t h e -19 l e s s , s t r u c t u r a l l y u n e m p l o y e d . 2. T h e L a b o u r M a r k e t A d j u s t m e n t M e c h a n i s m T h e c o n c e p t o f m i s m a t c h i n g , a s u s e d b y P r o u l x , i s b a s e d o n t h e a s s u m p t i o n t h a t t h e s u b s t i t u t a b i l i t y o f d i f f e r e n t t y p e s o f l a b o u r i s i m p e r f e c t , l i m i t e d o r n o n - e x i s t e n t . I n t h e e x t r e m e c a s e , t h e s u p p l i e s o f t h e d i f f e r e n t t y p e s o f l a b o u r a r e f i x e d a n d t h e p r o p o r t i o n s i n w h i c h t h e s e t y p e s o f l a b o u r a r e d e m a n d e d a r e r i g i d . H o w e v e r , t h i s a s s u m p t i o n o f r i g i d c o m p l e m e n t a r i t y a n d c o m p l e t e n o n - s u b s t i t u t a b i l i t y d o e s n o t c o r r e s p o n d t o t h e r e a l w o r l d . A s a m a t t e r o f f a c t , i t i s p r o b a b l y m o r e i l l u m i n a t i n g t o r e g a r d s t r u c t u r a l i m b a l a n c e s a s d u e t o l a g s a n d i m p e r f e c t i o n s i n t h e p r o c e s s e s o f s u b s t i t u t i o n b o t h o n t h e d e m a n d a n d s u p p l y s i d e s o f t h e l a b o u r m a r k e t . 15 a. A neoclassical sketch of the adjustment mechanism Before entering into the discussion of the barriers and frictions in the substitution process, i t is perhaps best to review the actual workings of the substitution process as conceived in theory. This i s provided by the neoclassical theory of the factor market. Substitution is responsible for equilibrium between factor demand and factor supply. If there i s a shortage of factor A and an excess of factor B, the price of A w i l l increase relative to the price of B. This leads to three forms of substitution: (1) The prices of goods and services whose production and rendering involve a high u t i l i z a t i o n of scarce factor A and a low u t i l i z a t i o n of surplus factor B w i l l rise relative to the prices of goods and services with a factor mix emphasizing surplus factor B and de-emphasizing scarce factor A. Consequently, consumers w i l l substitute a goods-and-services mix which requires a higher u t i l i z a t i o n of B and a lower u t i l i z a t i o n of A for the current consumption mix. (2) In response to the changed structure of returns for factor services, factor owners and factor producers w i l l substitute scarce factor A for surplus factor B in the mix of factors supplied. As the price for B declines, a smaller quantity of B w i l l be supplied. Conversely, as the price for A rises, a larger.quantity of A w i l l be supplied. (5) Because of the changed structure of factor prices, producers w i l l substitute technological processes involving a higher u t i l i z a t i o n of B and a lower u t i l i z a t i o n of A for prevailing processes. Different types of labour can be regarded as different factors of production. If, for example, as a result of technological change skilled labour forms a bottleneck and there is unemployment among unskilled labour, the wage rate of skilled labour relative to that of unskilled labour w i l l 16 rise. The prices of goods and services with a relatively high u t i l i z a t i o n of skilled labour, such as the production of radios, w i l l rise relative to those of goods and services with a relatively high u t i l i z a t i o n of unskilled labour, such as clothing. Thus more clothing and less radios w i l l be bought. In the meantime, the producers of radios w i l l investigate and possibly introduce production techniques using less skilled labour and more unskilled labour. Finally, labour i t s e l f w i l l convert i t s unskilled services into skilled services through education and retraining. These forces can, according to neoclassical theory, be expected to work toward an equilibrium in the demand and supply of different types of labour, b. Factors impeding adjustment (i) Non-economic impediments: If the neoclassical adjustment mechanism explains the elimination of labour bottlenecks and structural unemployment, then their persistence must be explained in terms of imperfec-tions in this adjustment mechanism. Lowell E. Gallaway, in his analysis of the intra-factor allocation of the labour market, has l i s t e d five barriers to the mobility of labour necessary for the equalization of wage rates and has analyzed them with respect to their effect on unemployment. The imperfections are: (1) the existence of non-economic barriers to mobility of workers; (2) the existence of positive private economic costs associated with the movement of labor from sector to sector; (3) non-homogeneity of the labor units involved; (k) a failure of workers to maximize their u t i l i t y function; and/or (5) differences i n workers' preference functions.20 According to Gallaway, only non-economic impediments to wage equalization may also cause structural unemployment. They are the non-economic barriers to mobility and non-maximization of workers' u t i l i t y . Non-economic barriers to mobility are those that cannot be explained in terms of market forces. They may keep unemployed labour either from f i l l i n g 17 job openings or from bidding down wages to expand employment opportunities. The same occurs when there i s non-maximization of workers' u t i l i t y , which means that the actions of the workers are inconsistent with their respective 21 (subjectively determined) leisure-income preference functions. In addition, nonhomogeneity of labour may lead to unemployment. It "may be either of a type that completely excludes a worker from certain labor markets (such as a worker of below-normal intelligence being excluded from becoming a member of the medical profession) or of a type that can be overcome at some positive private opportunity cost (such as the cost of 22 education)". Gallaway claimed that in the case of opportunity costs no involuntary unemployment w i l l be involved, while complete exclusion may bring about unemployment. Another barrier to the prevention and elimination of structural unemployment i s insufficient knowledge about the available job vacancies. It i s by no means necessary to assume perfect knowledge as a condition of structural equilibrium (although i t i s for the optimum allocation of labour), but merely enough information so that every unemployed knows of at least one acceptable vacancy within his reach—assuming there i s no demand-deficiency unemployment. ( i i ) Economic impediments: According to Gallaway, economic barriers do not lead to unemployment. . If the barriers to mobility are economic i n origin (i.e., generated by private opportunity costs associated with labor market transfer), an equilibrium may be reached which reflects the objective opportunity costs implicit in these barriers. For example, the cost involved in moving from one geographic area to another or the cost involved in acquiring the s k i l l s necessary to enable a worker to move from one sector to another may justify the existence of a wage differential even though workers otherwise move with complete freedom from sector to sector. Under these V circumstances no involuntary unemployment would exist. 18 However, risks and uncertainty may provide effective economic barriers to the movement of unemployed labour into unfilled vacancies, as Ernst W. Stromsdorfer has pointed out. Uncertainty may occur regarding the availability of jobs, their remuneration and the accompanying psychic income in a new location or for a new s k i l l . It may affect the decision to retrain or to move and i s li k e l y to act as a deterrent to mobility. In addition, the existence of positive private opportunity costs of transfer or retraining may not be overcome i f the opportunity costs exceed the mearis that the unemployed has available to pay for the improvement, conversion or 25 transfer of the labour services he can offer. Furthermore, Richard Q. Lipsey has pointed out that unused factors might not be absorbed i f the price which makes i t employable i s below the subsistence level or i f i t i s hot technically possible or economically worthwhile to adjust production processes so that a l l available factors are used. The latter instance i s quite possible i f the new combination of factors requires s c i e n t i f i c and technological research, which, however, 26 yields less returns than research in other areas. ( i i i ) The time required for adjustment: It must have become obvious by now that the various barriers referred to are effective only for a certain period of time. The uncertainty factor, for example, w i l l be effective only in the short run, while the d i f f i c u l t i e s in changing to production processes with the appropriate factor proportions are li k e l y to persist for a f a i r l y lengthy period. Given a certain structural shock which upsets the labour market equilibrium, a l l barriers are li k e l y to become negligible in the very long run as they are worn down or circumvented by the forces of adjustment mentioned above, i.e., substitution in consumption, factor input and factor 19 supply. It i s , therefore, the time element in adjustment, or rather, the rapidity with which adjustment takes place, which ultimately determines the 27 effectiveness of the adjustment mechanism. c. Factors f a c i l i t a t i n g short-run adjustment Adjustment in the labour market i s in the short run fac i l i t a t e d 28 by several factors, some of which have been discussed by Walter Y. Oi, 29 M.W. Reder and Stromsdorfer. They involve incentives to workers to move and to retrain and to employers to provide retraining as well as to modify their demands with respect to the qualifications of job applicants. (i) Incentives to the workers: The incentives to workers to convert their services (for example, geographically by moving from one place to another, or occupationally by acquiring additional s k i l l s ) are provided by the structure of wage differentials, i.e., the difference between what he earns now and what he could earn after the conversion of his labour services. In the case of the unemployed, i t i s the difference between the unemployment insurance payments and the prospective post-conversion earnings. The cost of conversion has been treated by Oi and Stromsdorfer as an investment expenditure. Stromsdorfer has put i t succinctly as follows: In making the decision to invest or not, the worker...must compare two alternative streams of expected income. The f i r s t i s the one the worker receives now that his s k i l l s are relatively obsolete; the other i s the one which can potentially be gained i f the worker...undertakes the time, expense, and risk of investing. The decision to invest w i l l be made on the basis of the greater income stream, discounted to the present and summed. An act of mobility or immobility w i l l occur.30 Thus, the incentive for mobility w i l l be greater, the greater the differen-t i a l in earnings i s , the lower the costs of conversion are, the longer the period of expected future income i s , and the lower the uncertainty of the 20 calculations i s . (This, by the way, provides a purely economic explanation for the low mobility of older workers.) The importance of psychic income and institutional factors, however, are not considered here. ( i i ) Incentives to employers for the conversion of labour services: The labour market provides incentives for the conversion of labour services not only to workers, but also to employers. Oi, in his theory of wages and employment based on the treatment of labour as a quasi-fixed factor, has pointed out that the employer can invest in his employees in the form of retraining. He w i l l be maximizing expected returns as long as the present value of the expected increase in the marginal revenue product of labour attributable to the retraining i s at least as high as the present value of future wage payments plus the cost of retraining. Thus a worker retrained at company's expense w i l l receive a wage lower than his marginal revenue product."^ Oi mentioned, however, that this applies only to specific training, i.e., training which i s not l i k e l y to be applicable to jobs in other firms. Otherwise the worker could subsequently bid up his wage by threatening to 32 switch jobs or actually doing so. Consequently, retraining of a more general nature w i l l not be undertaken by the firm without immediate recompensation by the worker or the government. ( i i i ) Modification of job qualifications: Not only w i l l employers accept inferior workers in order to train them, but under shortage conditions they w i l l actually operate with inferior labour. Reder has conceived of a trade-off between the quality of the hired worker and the period of search for a worker during which the job remains vacant. Since job vacancies represent forfeited returns to the employer, he w i l l try to minimize the vacancy period as well as maximize labour quality. During a time of labour shortage the same quality of labour can only be found after a longer vacancy 21 period than under slacker conditions. Consequently, in order to keep his vacancy period down, the employer i s li k e l y to make some compromise in labour q u a l i t y . ^ This assumes, of course, that such substitutability i s technically feasible even i n the short run. d. The implications for the rate of adjustment What does the neoclassical theory together with the elaborations and qualifications presented above t e l l us about the rate of adjustment of the labour market, i.e., the rate at which unemployed labour i s reabsorbed? Actually, i t t e l l s us very l i t t l e . What i t does do i s point out the forces at work, those which push toward the equilibrium and those which push away from i t . The relative strength and speed of these forces, however, are not analyzed. In general, the rate of adjustment, i n neoclassical terms, depends on the rates, with respect to time, of consumer substitution, substitution among the labour services supplied by the workers and substitution among the labour services demanded by employers, in response to imbalances between the structures of labour demand and supply. The rate of consumer substitution w i l l depend largely on the f l e x i b i l i t y of prices and on the extent and rapidity of the response of demand to changing prices. The rate of substitu-tion of one kind of labour service for another by workers w i l l be mostly determined by the f l e x i b i l i t y of wages and the movement and conversion of labour in response to changing employment opportunities and wages. These in turn depend on the knowledge of supply conditions by employers and the knowledge of vacancies and the wage structure by workers, as well as the certainty about such knowledge. Institutional barriers, lack of knowledge and uncertainty reduce the rate of labour substitution. On the other hand, i f Reder's theory of the "ladder effect" i s widely applicable, the possibility of upgrading and downgrading required job qualifications should 22 greatly f a c i l i t a t e labour substitution. Adjusting changes in the employers' requirements regarding worker qualifications w i l l depend in the short run on f l e x i b i l i t y in the techniques and organization of production and in the long run on the availability of alternative techniques. Before a microeconomic approach such as the one just presented can lead to a predictive theory of the rate of adjustment of the labour market, the strength of the equilibrating forces as well as their time lags have to be assessed together with the barriers to them. 3. Manifestations of Structural Maladjustment Since the different adjustment-promoting and adjustment-impeding forces described in the previous section cannot be individually assessed as to strength and speed, attempts have been made to assess the "net disequilibrium" at any one point of time. What i s meant by net disequilib-rium or maladjustment or structural imbalance i s shown by Barbara R. Berman's model.^ a. A model of structural imbalance This model i s based on the extreme assumption that the supplies of different types of labour are fixed and the proportions in which these types of labour are demanded are r i g i d . Such a situation i s presented in Figure 2.1. Two complementary, non-substitutable types of labour are assumed to be involved: skilled and unskilled labour. The respective supplies of each are represented by the s k i l l endowment point E. The employment of each at different levels of production i s given by the employment expansion path 35 ABC. As production expands, employment w i l l be pushed to B. At this point a bottleneck in the supply of skilled labour appears, since further expansion cannot take place without increasing the supply of skilled labour. Figure 2.1 - Berman's bottleneck model: the employment expansion path and the s k i l l endowment constraints 2h The only unemployment that persists now i s unemployment among unskilled workers, which amounts to BE. This can be called the level of structural unemployment. • If there were substitutability on the supply side, i.e., unskilled labour could readily convert i t s e l f into skilled labour, the s k i l l endowment point would move toward C. If there were substitutability on the demand side, i.e., i f employers could easily adjust their production processes to substitute unskilled labour for skilled labour and would be faced with the incentives to do so, then the employment expansion path would move to the l e f t toward E. F u l l employment requires that the s k i l l endowment point l i e s on the employment expansion path. Berman seems to have referred to the employment expansion path as 37 a locus of the ex post composition of employment. However, i t can also be used as a labour demand expansion path i n the ex ante sense. We can then use i t to compare the structure of labour demand with the structure of labour supply. The demand composition point represents the structure of demand and the s k i l l endowment point the structure of supply. The level of full-employment demand i s given by C. In the aggregate i t i s equal to labour supply E, since at C labour demand D = ON + FC + OM - FE and FC = FE so that D = ON + OM = S, i.e., supply. However, the structure of demand is different from the structure of supply. The demand for ski l l e d labour exceeds the supply by FC, while the demand for unskilled labour f a l l s short of supply by the same amount, i.e., FE. As has been stated above, the level of structural unemployment is BE. It i s not FE, which i s the number of unskilled workers whose labour services would have to be converted into skilled labour to bring about f u l l 25 employment. The difference between BE and FE is explained by the fact that BF unskilled labour i s unemployed, not because of inadequate demand or the possession of inappropriate s k i l l s , but because of the unavailability of complementary skilled labour: This model can be expanded to accommodate any number of types of labour. (The theoretically optimum classification of types of labour would be determined by the pattern of labour complementarity.) In the extreme case of absolute complementarity among the different types of labour, the level of unemployment reaches i t s floor as soon as the very f i r s t labour bottleneck i s reached. Thus, a bottleneck in the supply of inspectors of electronic equipment would indicate that unemployment could be reduced no further. While this assumption of absolute complementarity i s not very r e a l i s t i c , the model does explain the appearance of labour bottlenecks before unemployment has been eliminated and the imbalance in the structure of labour supply and the structure of employment (or demand). According to Berman's model, the frictional-structural unemployment rate can be estimated to be the minimum unemployment rate at each cycle peak. However, we would have to assume that there i s at least full-employment aggregate demand at each cycle peak. What i s more serious, i t i s based on the unrealistic assumption of r i g i d complementarity and complete non-substitutability of labour. There are, aside from the policy-oriented approach, three other methods by which i t i s attempted to assess structural unemployment without the extreme assumption of absolute non-substitutability. They are the level of job vacancies concurrent with unemployment, unemployment differentials among the different industries, occupations, regions, etc., and the rate of long-duration unemployment. 26 -b. Job vacancies According to the Berman model, there would be no direct functional relationship between unfilled job vacancies and unemployment. As long as labour demand i s B or less, there w i l l be no vacancies. Once i t increases beyond B, unemployment w i l l remain constant' at i t s structural level, while vacancies w i l l steadily increase. At full-employment demand, the number of vacancies (FC + FB) w i l l be equal to the number of unemployed (BF + FE). As demand increases further, vacancies w i l l increase correspondingly, while unemployment remains at BE. J.C.R. Dow and L.A. Dicks-Mireaux, however, have empirically demonstrated that there i s a recognizable inverse relationship between 39 unfilled job vacancies and unemployment. The shape of the unemployment-vacancy function i s given in Figure 2.2(a). The functional link between the two variables i s the demand for labour. The relationship between labour demand, on the one hand, and unemployment and vacancies, respectively, on 40 the other, i s given in Figure 2.2(b). i i The shape of the functions U-^i and V^V^ is based on the following assumptions: (1) As demand increases from deficient demand to adequate demand (d* = 0), unemployment decreases as rapidly as demand increases. In other words, the number of unemployed decreases by as much as the number of workers demanded increases. (2) As demand increases further and comes to be in excess, unemploy-ment i s reduced at a decreasing rate. This might be attributed to the fact that the pressure of demand becomes increasingly f u t i l e as i t runs into the increasingly severe unemployability of the remaining unemployed. (3) As excess demand shrinks, the decrease i n vacancies i s as rapid as the decrease i n demand. 27 Figure 2.2 The Dow-Dicks-Mireaux system of functions: (a) the unemploymentrvacancy functions; (b) the unemployment-demand and vacancy-demand functions (k) As demand f a l l s below d =0, vacancies are reduced at a decreasing rate. This might be attributed to the persisting d i f f i c u l t y in f i l l i n g highly specialized vacancies even during periods of relatively high unemployment. The only explanation that Dow and Dicks-Mireaux give for the asymptotic t a i l s of the unemployment and vacancy functions i n quadrants II and IV i s that unemployment and vacancies "cannot shrink below zero" and therefore "must be supposed to become decreasingly sensitive" as they approach zero. At the point where unemployment i s equal to the vacancies, there is neither net excess demand nor net deficient demand (d =0). This i s a definition, not a conclusion. Furthermore, "maladjustment" i s defined as the amount of unemployment at this level of demand. It i s this maladjustment, or the structural imbalance of the labour market, which accounts for the concurrent existence of unemployment and vacancies. If there were no structural imbalances at a l l , i f the labour market adjustment mechanism operated perfectly, then the unemployment func-tion would be MOA and the vacancy function BON, where MN i s the 45° l i n e . Structural unemployment can be taken to be equal to Dow's and Dicks-Mireaux's "maladjustment" measure. However, i t could be measured at only one level of net demand for labour, i.e., d =0. When there i s excess demand, unemployment w i l l be reduced below i t s structural level. Thus, according to this model, non-demand-deficiency unemployment can vary with the pressure of excess demand. If the degree of maladjustment increases, the unemployment and vacancy functions i n Figure 2.2(b) w i l l shift outward. Assuming that the degree of maladjustment increases from OK = OL to OP = OQ, then the functions w i l l shift to U^ U^  and ^2^2'' r e s P e c t i v e l y « ^he unemployment-vacancies function F F in Figure 2.2(a) w i l l shift to F F . Thus a 29 non-cyclical change i n unemployment w i l l be reflected in a shift of the unemployment-vacancy function, while a cyclical change involves a movement along the function. The d i f f i c u l t i e s with this approach are (l) the general inadequacy of vacancy data and (2) the fact that the hyperbolic shape of the unemploy-ment-vacancy function i s assumed rather than obtained by deductive or inductive analysis. Nevertheless, because i t uses variables directly related to the labour market rather than proxies and because i t provides a measure of structural unemployment that can be applied at any phase of the cycle, i t deserves further development. In Chapter III a theoretical explanation for the function w i l l be given and in Chapter IV i t i s applied to Canadian data after the vacancy rate has been estimated, c. The structure of. unemployment (i) The basic argument: The comparison over time of unemployment rates of the different groups of the labour force, e.g., industries, occupations, regions, education levels and age-sex groups, has been a favourite technique of the participants in the structural-unemployment controversy for ascertaining changes in structural unemployment. This approach can be divided into two methods: (l) the comparison of the unemployment rates of specific groups, and (2) the comparison of the overall dispersion of unemployment rates among the various segments of the labour force. The structuralist hypothesis has been interpreted by several-economists to mean that structural unemployment i s concentrated in (l) blue-collar occupations, (2) goods-producing or manufacturing industries, (3) groups with relatively l i t t l e schooling, (k) higher age groups, and (5) depressed areas. It i s generally implied that changes in structural unemployment can be detected from deviations of the respective unemployment 30 rates from their normal relationships to overall unemployment. If the unemployment rates are above the rates expected on the basis of the rela-tionship to the overall unemployment rate, i t i s considered to indicate an increase in structural unemployment, and i f below, then a decrease. This conclusion, however, requires the restrictive assumption that increases in structural unemployment are more concentrated i n the above labour force groups than are increases in demand-deficiency unemployment. The alternative approach has been to ascertain whether the correlation between the unemployment rates of the various labour force groups has deteriorated over time, which, according to Gallaway, would indicate an increase in structural unemployment. While this approach avoids the need for the explicit specification of the structurally disadvan-taged groups, i t requires the assumption that the distribution of increases (or decreases) in structural unemployment among the labour force groups are significantly different from the distribution of increases (or decreases) in demand-deficiency unemployment. ( i i ) The structure of unemployment and the cycle: It has been empirically established that the unemployment impact of cyclical fluctua-tions i s not randomly distributed among the various labour force groups. The differences in the impact can be largely explained by two phenomena: (l) Since the short-run income el a s t i c i t i e s of a l l goods and services are not the same, there w i l l be a differential impact of the cycle on the various industries. (2) The firms w i l l not lay off different types of workers in proportion to the composition of their work forces. The reasoning for the second point has been developed by Oi and Reder. Oi has argued that there are certain fixed costs involved in the employment of labour. They arise from the costs of hiring and of training. The expected marginal revenue product of labour has to cover not only the 31 wage r a t e but a l s o the a m o r t i z a t i o n of the i n i t i a l f i x e d c o s t s , or the p e r i o d i c r e n t , as Oi c a l l e d them. Since i n the short run the p e r i o d i c rent does not enter i n t o employment d e c i s i o n s , a d e c l i n e i n the product p r i c e and th e r e f o r e i n the marginal revenue product does not l e a d to l a y - o f f s u n t i l i t i s greater than the p e r i o d i c r e n t . Consequently, the greater the r a t i o of p e r i o d i c r e n t to wage r a t e , the more i n s e n s i t i v e employment w i l l be to p r i c e changes.^ Since abundant types of labour, such as u n s k i l l e d workers, presumably i n v o l v e r e l a t i v e l y lower f i x e d employment costs than scarce types of labour, such as s k i l l e d workers, t h e i r employment w i l l be more s e n s i t i v e to d e c l i n e s i n product demand than the employment of scarce labour. On t h i s b a s i s , c y c l i c a l unemployment w i l l be predominantly composed of labour which has l i t t l e s p e c i f i c t r a i n i n g and i s the most r e a d i l y a v a i l a b l e . Reder has advanced a s i m i l a r argument by r e f e r r i n g to the "ladder e f f e c t " : "...as o v e r a l l unemployment r i s e s , the unemployed from higher occupational s t r a t a back down i n t o the u n s k i l l e d ranks d i s p l a c i n g u n s k i l l e d workers, i n c r e a s i n g the r e l a t i v e unemployment r a t e of the u n s k i l l e d and 51 reducing t h a t of the s k i l l e d . " This works i n reverse when unemployment, f a l l s . Workers w i l l be re-employed i n more or l e s s the reverse order i n which they were l a i d o f f . However, having e s t a b l i s h e d the c y c l i c a l p a t t e r n of the s t r u c t u r e of unemployment, t h i s does not mean that the s t r u c t u r e of unemployment r e s u l t i n g from demand d e f i c i e n c y can be d i s t i n g u i s h e d from the s t r u c t u r e of unemployment r e s u l t i n g from s t r u c t u r a l changes i n the economy. Lipsey has made t h i s p o i n t very s t r o n g l y and has argued that i t i s only i n the p e r i o d or recovery that n o n - c y c l i c a l changes i n the s t r u c t u r e of labour demand become apparent i n the new h i r i n g p a t t e r n , s i n c e s t r u c t u r a l l y d i s p l a c e d jobs 52 w i l l not re-emerge w i t h the expansion. 32 If, for example, structural changes have the same impact on the structure of unemployment as demand deficiency, the two effects cannot be separated. As a matter of fact, i t i s conceivable that structural changes may affect the structure of unemployment in an equalizing way, which would then have to be interpreted as a reduction i n structural unemployment, whereas i n effect an increase took place, d. Long-duration unemployment A number of definitions have explicitly defined structural unemployment to be long-duration unemployment, i.e., the proportion of the labour force which has been unemployed for at least some lengthy period of time. The implication of such a definition i s that either the causes of mismatching of demand and supply necessarily lead to only long-duration unemployment or short-duration unemployment associated with the mismatching 53 of demand and supply i s arbitrarily defined as fr i c t i o n a l unemployment. (i) Random distribution of unemployment durations: Not a l l long-duration unemployment i s necessarily structural. Whether or not long-duration unemployment i s to be expected as a natural concomitant of any type of unemployment depends on the assumptions about the flow into and out of unemployment. For example, i f the movement out of unemployment followed the fi r s t - i n - f i r s t - o u t system, a l l unemployed would experience approximately the same length of unemployment. Berman adopted the more r e a l i s t i c assumption "that, once in the state of unemployment, people get to go through the exits according to a stochastic process. We can imagine a lottery in which for any period there are q U winning tickets out of a total of U tickets and our assumptions means 5k that everyone has an equal chance of getting a winning ticket of U / U " . On the basis of this assumption, the expected average duration of unemploy-ment (or the proportion of unemployed who have been unemployed for at least 33 some given period) w i l l be greater the lower the turnover rate of the unemployed i s . By comparing the distribution of the duration of unemploy-ment which i s to be expected on the basis of the turnover rate (or the unemployment rate, since the turnover rate and the unemployment rate have 55 been found to be highly correlated) with the actual distribution we can assess how far reality deviates from the assumption of homogeneity within the labour force and among the unemployed. ( i i ) The relative deterioration of the s k i l l s of the long-duration  unemployed: N.J. Simler has come up with a theory of long-duration unemploy-ment which rejects Berman's assumption of a random movement of persons out of the unemployed group. His argument is the following: A hard core of long-duration unemployed workers w i l l emerge i f the probability of re-employment i s a decreasing function of the dura-tion of unemployment. Among other reasons, this w i l l be the case i f the s k i l l level of unemployed workers i s also a decreasing function of the duration of unemployment, and i f the structure of wages f a i l s to adapt to the changing structure of s k i l l s ; or i f the unemployed workers' s k i l l s remain intact but do not advance with the increasing level of s k i l l s of the employed labor force; or i f the wage structure f a i l s to adjust to the changing s k i l l structure. Either way, the gap between the potential productivity of the unemployed and the actual productivity of employed widens with time; and, therefore, the probability of re-employment diminishes with time. Under these conditions, the change of a hard core of unemployed workers developing becomes the greater, the longer the unemployment rate fluctuates between U^ /the usual recession rate/and ^2. /^he recession rate during slow long-run growthy7, for with each succeeding cycle the flow into the category of very-long-term unemployment exceeds the flow out of i t . 56 The groups which Simler regarded as most vulnerable to s k i l l decay are the older workers and those with relatively l i t t l e education. The latter, for example, would have just as much chance of getting l a i d off as any other labour force group. "But when aggregate demand expands and unemployment f a l l s , they have less chance of becoming re-employed than other workers, and their prospects diminish the longer they remain unemployed." 3k Simler applies t h i s imbalance also to older workers by assuming "that l a y - o f f s among workers of equal s k i l l are made i n a random fashion and 58 not on a l a s t - i n , f i r s t - o u t b a s i s " . F i r s t of a l l , t h i s contradicts the conclusion to be derived from Reder's more reasonable assumption regarding the ladder e f f e c t , which does mean a l a s t - i n - f i r s t - o u t approach. Secondly, i t makes the unreasonable assumption that older workers are randomly d i s t r i b u t e d among the d i f f e r e n t s k i l l s . In r e a l i t y older workers are more protected against unemployment than younger workers. However, Simler's contention that older workers, once unemployed, have l e s s chance of being re-employed seems r e a l i s t i c . This might be explained on the basis of higher r e t r a i n i n g costs for older workers than for younger workers to achieve the same' increment i n the marginal labour product and possibly of the shorter time horizon of the investment i n older workers. At any rate, the r e l a t i v e d e t e r i o r a t i o n of the s k i l l s of the unemployed provides one reason for why actual long-duration unemployment i s higher than i s expected on the basis of Berman's assumption of random re-employment. I t explains how "persistent and long-term, or s t r u c t u r a l , unemployment—as d i s t i n c t from temporary and short-term, or f r i c t i o n a l , unemployment—can emerge i n the f i r s t place..., can increase r e l a t i v e to a l l other unemployment..., and can become incr e a s i n g l y concentrated among older workers...without there having occurred a s t r u c t u r a l change of any 59 kind whatsoever". Thus a r i s e i n long-duration unemployment at a constant (high) l e v e l of unemployment may be non-structural i n the causal sense. But i t would be s t r u c t u r a l i n the d e s c r i p t i v e sense, since the r e l a t i v e deteriora-t i o n of the s k i l l s of the long-duration unemployed means that the demand and supply structures w i l l be increasingly mismatched. Whether i t can be described as structural from the point of view of the appropriate policy action w i l l emerge from the following section. In conclusion i t can be said that the proportion of long-duration unemployment i n total unemployment may reflect the level of structural unemployment, but i t w i l l also vary with demand conditions. Only after the cyclical pattern in this ratio has been removed can i t be regarded as an index of structural unemployment. Even then i t does not actually specify the proportion in the unemployment rate which is to be regarded as structural. k. Evaluation The structural maladjustment approach i s the necessary complement to the causal approach in the theoretical illumination of the structural unemployment problem. It shows how the market's equilibrium mechanism absorbs disequilibrium created by structural shocks and the resulting displacement of labour. It also indicates the factors that impede this process of structural adjustment. On the other hand, the methods that have been developed to assess the level or trend of structural unemployment leave much to be desired. Berman's bottleneck model i s to be seen as an illuminating device rather than a method for measuring structural unemployment. The analysis of changes in the structure of unemployment turned out to be of doubtful use i n the assessment of the trend in structural unemployment because of the problems of distinguishing between the structure of demand-deficiency unemployment and that of structural unemployment. As a matter of fact, i t was demonstrated that under certain conditions the effects of an increase in structural unemployment would be interpreted as a reduction in structural 36 unemployment by t h i s method. A s i m i l a r problem emerges i n the use of the r e l a t i o n s h i p of l o n g - d u r a t i o n to o v e r a l l unemployment, si n c e l o n g - d u r a t i o n unemployment can as w e l l stem from prolonged demand d e f i c i e n c y as from severe s t r u c t u r a l d i s l o c a t i o n . The one approach which merits f u r t h e r a t t e n t i o n i s the one i n v o l v i n g the use of job vacancy data. I t i s derived d i r e c t l y from the concept of s t r u c t u r a l labour imbalances. I t s important shortcoming from the p o i n t of view of theory i s the l a c k of an explanation about the u n d e r l y i n g processes. Nevertheless, i n the model i n Chapter I I I t h i s approach i s used as the b a s i c framework and i t i s attempted to e l i m i n a t e t h i s shortcoming. C. THE POLICY APPROACH The t h i r d type of t h e o r i e s of s t r u c t u r a l unemployment i s the p o l i c y - o r i e n t e d group. I t s use has been j u s t i f i e d by Berman as f o l l o w s : Since t h e . s t r u c t u r a l unemployment controversy i s (or should be) b a s i c a l l y a p o l i c y controversy, i t would seem n a t u r a l to construct a d e f i n i t i o n i n terms of the r e s u l t s of p o l i c y a c t i o n s . We cah then t r y to p r e d i c t what e f f e c t s v a r i o u s p o l i c y a c t i o n s might have, and our p r e d i c t i o n s w i l l at one and the same time serve to measure s t r u c t u r a l unemployment and l i g h t our way along the optimum p o l i c y path.6° 1. The P o l i c y I m p l i c a t i o n s of Berman's Bottleneck Model On t h i s b a s i s , Berman proceeded to define the d i f f e r e n t types of unemployment i n terms of the p o l i c y measures by which they can be e l i m i n a t e d . Her d e f i n i t i o n s may be presented as f o l l o w s : = demand-deficiency unemployment, which can be e l i m i n a t e d by monetary and f i s c a l p o l i c i e s . U = s t r u c t u r a l unemployment, that which can be e l i m i n a t e d by r e t r a i n i n g and other manpower programs ( a f t e r demand-d e f i c i e n c y unemployment has been eliminated) p l u s the unemployables. U = the remaining unemployment, which i s what Berman c a l l s "rock bottom" f r i c t i o n a l unemployment.6l 37 In Figure 2.1 on page 23i which depicts Berman's model, these components can he identified. Structural unemployment is BE, as has been explained above. The remaining unemployment, for example, at A, AG + GB, is demand-deficiency unemployment. (Frictional unemployment is not shown in the diagram.) B is the point where the structural labour bottleneck is encountered, after which fiscal and monetary policies cannot reduce unemployment any further. In order to get an orderly expansion beyond B, the sk i l l mix of the economy needs to be relocated from point E to a point closer to the employment expansion path. Since retraining one person subtracts One from the unskilled labour force and adds one to the skilled, a retraining program will cause the sk i l l endowment point to migrate in a 45-degree line. Thus, retraining EF people will enable the economy to expand toward point C, and cause additional hiring of BF unretrained unskilled people, as well as FC (= EF) retrained newly skilled workers. Retraining may thus have a "multiplier" effect on employment. Additional demand of g (G.N.P. - G.N.P., ) is , however, also a necessary condition. C D The major policy question that arises with respect to this model is: At what point must the stimulation of aggregate demand be regarded as becoming ineffective in reducing unemployment? Berman's answer was that this occurs when a bottleneck appears in "some significant sectors of the labor market". How is such a bottleneck to be recognized? The signal for i t , Berman suggested, is a heightening of inflationary tendencies. However, she warned that i t must be possible to analytically extract price movements caused by other pressures. She then left this problem with the assumption. that "we know how to avoid error in the recognition of labor bottlenecks".^ 2. Lipsey's"Trade-Off Model a. Model and definition Inflation as a symptom of labour bottlenecks has been used by Gk Lipsey to construct his inflation-unemployment trade-off model. In i Figure 2.3 the trade-off function RR describes the behavioural relationship 38 p Figure 2.3 - Price inflation and unemployment: the behavioural trade-off functions and the maximum-inflation constraint 39 between p r i c e changes and unemployment. I t "shows combinations of U and P which can be a t t a i n e d by v a r y i n g the l e v e l of aggregate demand; g e n e r a l l y , the higher the l e v e l of aggregate demand i s , the lower w i l l be the l e v e l of 6 5 unemployment, but the higher w i l l be the r a t e of p r i c e i n f l a t i o n " . I f the policy-makers take the view that unemployment should be minimized w i t h i n the c o n s t r a i n t of some maximum r a t e of i n f l a t i o n , we can, according to L i p s e y , d i s t i n g u i s h between demand-deficiency unemployment, on the one hand, and s t r u c t u r a l and f r i c t i o n a l unemployment, on the other. i -Given the beha v i o u r a l f u n c t i o n RR and a maximum acceptable r a t e of i n f l a -t i o n of OA, the lowest p o s s i b l e l e v e l of unemployment i s OD. This i s the l e v e l of s t r u c t u r a l and f r i c t i o n a l unemployment. Any unemployment i n excess of t h a t l e v e l i s a t t r i b u t e d to inadequate demand. Demand s t i m u l a t i o n can reduce unemployment only to OD without pushing i n f l a t i o n beyond the maximum acceptable r a t e . To reduce unemployment below OD, s e l e c t i v e measures would have to be introduced. F i r s t of a l l , p u r e l y from the p o i n t of view of the maximization of n a t i o n a l product, a l l such measures w i l l be a p p l i e d as lon g as they y i e l d a net r e t u r n according to standard c o s t - b e n e f i t a n a l y s i s (where the costs are those of the scheme and the b e n e f i t s are the discounted value of the increase i n output as a r e s u l t of the scheme). This w i l l s h i f t the t r a d e - o f f f u n c t i o n to the l e f t , to R^R^, where, at the maximum acceptable r a t e of i n f l a t i o n , unemployment w i l l now, a f t e r the i n t r o d u c t i o n of s e l e c t i v e measures, be OE, i n s t e a d of OD. I f there are s o c i a l b e n e f i t s to be derived from a f u r t h e r r e d u c t i o n i n unemployment, a d d i t i o n a l p o l i c i e s can be introduced, y i e l d i n g a net r e t u r n i n terms economic and s o c i a l b e n e f i t s over the costs to s o c i e t y . These w i l l s h i f t the t r a d e - o f f f u n c t i o n to ^ ^ 2 ' •'•eav^nS a c o r e °f unemployment, whose e l i m i n a t i o n by f u r t h e r manpower p o l i c i e s i s not warranted on the,basis of economic and s o c i a l c o s t - b e n e f i t a n a l y s i s . In other words, RoRp represents the curve where the marginal social returns equal the marginal social costs of the manpower policies. In this case, OF i s fri c t i o n a l unemployment and FD i s structural unemployment. It is apparent that Lipsey's definition differs from Berman's in so far as the unemployables are part of f r i c t i o n a l unemployment according to Lipsey, whereas according to Berman they are part of structural unemployment. However, instead of laying down a maximum inflation constraint, i t i s preferable from the point of view of policy-making to think i n terms of a trade-off between different goals, namely f u l l employment and price s t a b i l i t y . This means that policy-makers are simultaneously minimizing unemployment and inflation. An example of the relative preferences of policy-makers between the two goals i s given by the preference functions 1^, I,,, etc. in Figure 2.4, each of which gives the locus of combinations which for the policy-makers i s of equal value in terms of the social costs of inflation and unemployment. The closer that the preference function i s to the origin, the better the goal of minimizing unemployment and inflation w i l l be satisfied. The concave shape of the functions i s possible i f policy-makers are relatively more concerned about unemployment at low levels of inflation and about inflation at low levels of unemployment. The social costs of unemployment and inflation can be minimized in the short run only within the constraint of the behavioural trade-off function RR . They w i l l be minimized at the point where the lowest preference function, or policy trade-off function, can be obtained by moving along RR . This i s the usual tangency point. Here demand-deficiency unemployment i s unemployment i n excess  of OD. By applying selective policies so that the economic cost-benefit test i s satisfied, unemployment can be reduced to OE. Inflation, too, i s Figure 2.k Price inflation and unemployment: the behavioural trade-off functions and the policy-makers' preference functions h2 r e d u c e d f r o m OA t o O B . On t h e b a s i s o f c r i t e r i a o t h e r t h a n t h e m a x i m i z a t i o n o f n a t i o n a l p r o d u c t t h e s e p o l i c i e s w i l l b e p u s h e d f u r t h e r , s o t h a t u n e m p l o y -m e n t w i l l b e c o m p r e s s e d t o OF a n d i n f l a t i o n t o O C . S t r u c t u r a l u n e m p l o y m e n t t h e n i s FD a n d f r i c t i o n a l u n e m p l o y m e n t i s O F . A s L i p s e y h a s p o i n t e d o u t , " s t r u c t u r a l a n d f r i c t i o n a l u n e m p l o y m e n t 66 s h a d e i n t o e a c h o t h e r w i t h n o c l e a r b o u n d a r y s e p a r a t i n g t h e m " . I f we c o n c e i v e o f f r i c t i o n a l u n e m p l o y m e n t i n t h e w i d e s t s e n s e , i . e . , a s u n e m p l o y -67 m e n t w h i c h i s " t h e p r o d u c t o f i m p e r f e c t i o n s i n t h e l a b o r m a r k e t " , t h e n " s t r u c t u r a l u n e m p l o y m e n t i s t h a t p a r t o f f r i c t i o n a l u n e m p l o y m e n t w h i c h i s n o t a c c e p t a b l e e i t h e r b e c a u s e t h e r e w o u l d b e a n e t m o n e y g a i n i n r e m o v i n g i t o r b e c a u s e t h e s o c i a l g a i n s o f r e m o v i n g i t a r e j u d g e d t o o u t w e i g h t h e n e t m o n e y c o s t o f s o d o i n g " . ^ T h e c o n c e p t o f u n a c c e p t a b i l i t y p o i n t s t o t h e t w o s u b j e c t i v e e l e m e n t s i n t h e a n a l y s i s : ( l ) t h e p o l i c y p r e f e r e n c e f u n c t i o n s , a n d (2) t h e e v a l u a t i o n o f t h e s o c i a l b e n e f i t s f r o m r e d u c i n g u n e m p l o y m e n t . T h e a n a l y t i c a l e c o n o m i s t , t h e r e f o r e , m u s t e i t h e r k n o w w h a t t h e p o l i c y p r e f e r e n c e f u n c t i o n s a r e , o r h e c a n p r e s e n t t h e p o l i c y - m a k e r s w i t h t h e RR a n d R^R^ f u n c t i o n s a n d d a t a o n t h e r e t u r n s i n t e r m s o f r e - e m p l o y m e n t f r o m i n v e s t m e n t i n m a n p o w e r a n d o t h e r s e l e c t i v e p o l i c i e s a n d t h e n o b s e r v e t h e p o l i c y - m a k e r s ' r e a c t i o n b e f o r e h e c a n d e t e r m i n e t h e l e v e l s o f t h e d i f f e r e n t t y p e s o f u n e m p l o y m e n t , b . T h e t h e o r e t i c a l b a s i s T h e b e h a v i o u r a l t r a d e - o f f f u n c t i o n RR i s d e r i v e d f r o m t h e P h i l l i p s c u r v e , w h i c h d e s c r i b e s t h e r e l a t i o n s h i p b e t w e e n t h e r a t e o f c h a n g e o f w a g e s a n d t h e l e v e l o f u n e m p l o y m e n t . T h i s d e r i v a t i o n i s b a s e d o n t h e a s s u m p t i o n t h a t p r i c e i n f l a t i o n i s c l o s e l y c o r r e l a t e d w i t h t h e r a t e o f c h a n g e o f w a g e s . 4 3 Thus the behavioural trade-off function p = f x(u) (2.1) where p = price inflation and u = unemployment rate, is based on,the Phillips function w = f (u) ( 2 . 2 ) where w = rate of change of wages, and the assumption that f 3(w) ( 2 . 3 ) Equation ( 2 . 2 ) was originally formulated and tested by 69 A.W. Ph i l l i p s . The level of unemployment was regarded as a measure of the demand for labour relative to i t s supply. The function was expected to take the form of a curve for the following reasons: When the demand for labour i s high and there are very few unemployed we should expect employers to bid wage rates up quite rapidly, each firm and each industry being continually tempted to offer a l i t t l e above the prevailing rates to attract the most suitable labour from other firms and industries. On the other hand i t appears that workers are reluctant to offer their services at less than the prevailing rates when the demand for labour i s low and unemployment is high so that wage rates f a l l only very slowly. The relation between unemployment and the rate of change of wage rates i s therefore l i k e l y to be highly non-linear.70 The function describing this relationship i s similar in shape to the RR function i n Figures 2 . 3 and 2 . 4 describing the relationship between price inflation and the unemployment rate. This means that any demand-pull i effects on the price level must be simultaneously reflected in the wage level. i i i . • • . \ c. Impurities in the inflation-unemployment relationship Impurities in the inflation-unemployment relationship may be due to disturbances in equation (2.2) or in equation (2.3). Such disturbances in the Phillips curve may result from cost-push pressure on the wage level or, alternatively, from wage restraint. One attempt to deal with this problem was to take account at least of those changes in the cost-push pressure on the wage level which followed a cyclical pattern but were not related to the current level of labour demand. This involved introducing the rate of change in unemployment as an additional independent variable into the Phillips equation. It is based on the argument that the bidding up of wages by employers and the pressure exerted by unions w i l l also be influenced by expectations about future demand conditions and that the rate of change in unemployment i s a good indicator of these expectations. Empirically this variable helped to explain the 7 1 "loops" that were observed in the Phillips relationship. Lipsey argued that these "loops" were not an expectations effect but the result of aggregating data from various labour markets with different demand 72 -conditions. S.F. Kaliski, however, pointed out that neither hypothesis 73 could be rejected. While the determination of the effect of the rate of change of unemployment on the rate of change of wages may remove some of the disturbances of the Phillips relationship resulting from varying cost-push pressures on the wage level, i t cannot eliminate a l l of them. Impurities in equation (2.3) can arise from cost-push and demand-pull factors. The relationship between price inflation and wage changes can be disturbed by such non-wage cost-push factors as changing profit-sales ratios and changing import-price inflation. On the demand-pull side, non-labour shortages of factor inputs may put pressure on the price level without affecting the wage level to a comparable degree. 4 5 These impurities in Lipsey's trade-off function do not destroy i t s meaningfulness. They merely make i t clear that i t relates two variables which interact only in an indirect way and are, therefore, not entirely harmonious. 3. The Policy Implications a. Policy sequence According to Lipsey, " i n a perfect world, in which policy-makers acted rationally, we would behave as i f we were deficient-demand theorists and increase aggregate demand unti l the limit set by acceptable price rises was reached; we would then a l l behave as structuralists and consider how the 74 remaining unemployment could be removed". This approach has been c r i t i c i z e d from two angles: (1) As Berman has pointed out, after a cyclical expansion has encountered serious labour bottlenecks, selective policies have to be 7 5 applied simultaneously with expansionary policies. Otherwise the s k i l l endowment point i n Figure 2 .1 i s shifted from E to C by selective policies, but the employment composition point i s s t i l l at B. Therefore, sufficient demand must be assured while selective policies make additional expansion possible. (2) Proulx has argued that selective policies are appropriate even before the labour bottleneck has been reached, since there w i l l be a lag before they become effective. This i s particularly significant when "the composition of the marginal demand for labour i s so different from the composition of the average demand for labour, and so d i f f i c u l t to predict therefrom,...that i t i s desirable to stress general education, or technical and vocational education for groups or families of occupations (time 7 6 consuming processes) as against stop-gap and short-run training programs". kG b. Trade-off goals Instead of using the trade-off model with only the policy goals of minimizing unemployment and minimizing inflation, Proulx has suggested 7 7 that the maximization of growth be introduced into the model. "A belief that manpower can make a significant contribution to economic growth coupled with a belief that selective policies would not be very effective anti-cyclical weapons may prompt rational policy-makers who weight growth more heavily than stabilization to opt i n favour of a continuous use of selective 7 8 policies alongside aggregative public policies." While i t i s sound policy to use manpower programs not merely to lower the unemployment-inflation trade-off function, but also to promote growth, this does not necessarily mean that Lipsey's model should be extended to include growth as another trade-off goal. To justify such an extension would require an argument that the rate of growth i s significantly affected either by the economy's particular position on the unemployment-inflation trade-off function or by the degree of structural disequilibrium as 7 9 indicated by the position of this trade-off function. Proulx did not attempt to provide such an argument. Without i t , the introduction of the growth goal i s an unjustified complication of the model. To the extent that selective policies promote growth by reducing unemployment, they are included in Lipsey's model i n the f i r s t improved trade-off function, R-jR-^ * To the extent that manpower and other selective policies promote growth without affecting unemployment, e.g., retraining the employed, their economic justification can be analyzed independently and they can be introduced in addition to the policies justified on the basis of the unemployment analysis. One other goal which R.A. Gordon has suggested as important for the maximization of national economic welfare i s the balance of payments k7 equilibrium. However, in the form presented by Gordon, i t does not complicate Lipsey's trade-off model. He regards i t as a constraint rather than an optimizable variable, so that the only implication that the balance-of-payments equilibrium has for the trade-off model i s that the minimization of unemployment or inflation has to be suspended while this constraint i s 80 being violated. It does not involve an additional trade-off goal. 4. Evaluation The policy-oriented approach to the measurement of structural unemployment has two important merits: (l) It provides an absolute measure of structural unemployment without requiring vacancy data. ( 2 ) It demonstrates how the level of labour demand, on the one hand, and structural imbalances, on the otherj affect two variables, both of which are to be minimized according to p o l i t i c a l c r i t e r i a . On the other hand, i t has certain drawbacks. Since inflation and unemployment are causally related only i n an indirect way, this relationship is l i k e l y to contain significant impurities. The relationship between labour demand and the rate of change of wages may not be constant over time and may suffer from disturbances due to wage-push effects. Introducing the rate of change of unemployment is not l i k e l y to adequately take account of them. The relationship between the rate of change of wages and price inflation, too, suffers from disturbances due to varying rates of import-price inflation, profit-sales ratios and shortages of production inputs other than 8l labour. Even i f the effects of variables such as productivity growth and import prices are assured, as in Vanderkamp's empirical model, distortions s t i l l remain because the structural displacement effects due to productivity changes would thereby be unjustifiably removed. Finally, when this model i s applied to regression analysis, the least-squares principle results in a bias according to which as much as possible of the variations in unemploy-ment are attributed to cyclical factors and as l i t t l e as possible to the unstated structural factors. 1 2 3 k 5. 6 7 8 9 io n 12 13 Ik k9 FOOTNOTES TO CHAPTER II Paul H. Casselman, The Economics of Employment and Unemployment, Public Affairs Press, Washington, D.C, 1965, p. 115* H.D. Woods and Sylvia Ostry, Labour Policy and Labour•Economics i n Canada, Macmill of Canada, Toronto, 1962, pp. 377-8. Casselman, op. c i t . , pp. 116-7. United States, Congress, Joint Economic Committee, Subcommittee on Economic Statistics, Higher Unemployment Rates, 1957-60: Structural  Transformation or Inadequate Demand, 87th Congress, 1st session, 196l, pp. k2-7. Ibid., p. 42. Ibid., p. kk. Ibid., pp. 9-12. ' . • Ibid., p. 10. Actually this statement i s not entirely correct. Productivity increases do not need to be above average to lead to labour displacement. If demand i s price-inelastic, even a small productivity-increase can lead to an employment reduction. Ibid., pp. 10-11. Charles C. Killingsworth, "Automation, Jobs, and Manpower", The Nation's  Manpower Revolution, Hearings before the Subcommittee on Labour and Public Welfare, U.S. Senate, 88th Congress, p. 1468. Ibid., pp. 1468-9. U.S. Joint Economic Committee, op. c i t . , p. 11. Richard G. Lipsey, "Structural and Deficient Demand Unemployment Reconsidered", Employment Policy and the Labor Market, ed. Arthur M. Ross, University of California Press,. 1965, pp. 248-9. Lipsey has also argued that structural shifts due to technological change and other factors can set up cyclical' swings i n aggregate demand and income by redistributing income between groups with different lags in their consumption behaviour. Knowles and Kalacheck (who, by the way, came to the conclusion that the hypothesis according to which structural unemployment has been on the increase i s incorrect) looked also at the labour supply side of the structural unemployment question. They considered the "propensity of unemployed workers to seek jobs in other occupations, industries, and geographic areas" as the fourth macroeconomic determinant of the level of structural unemployment. (See U.S. Joint Economic Committee, op. c i t . , p. 12.) It was not included in the above discussion because i t does not f a l l into the causal approach as circumscribed by the definitions at the beginning of this section. 50 15 Canada, Senate, Special Committee on Manpower and Employment, Final  Report, Queen's Printer, 1 9 6 l , p. 2 . Frank T. Denton and Sylvia Ostry, An Analysis of Post-War Unemployment, Economic Council of Canada, Staff Study No. 3 , Ottawa, Queen's Printer, Dec. 1964, p. 2 . 17 Senate Committee on Manpower and Employment, op. c i t . , p. 10. l 8 Pierre-Paul Proulx, "The Composition of Unemployment in Canada", Employment, Unemployment and Manpower, McGill University, Industrial Relations Centre, 15th Annual Conference, June 1964, p. 37 . 19 The duration of unemployment w i l l be further discussed below. 20 Lowell E. Gallaway, "Labour Mobility, Resource Allocation, and Structural Unemployment", American Economic Review, Vol. 5 3 , No. 4 , Sept. 1963, p. 696. 2 1 I b i d-> PP' 696-7 . 2 2 Ibid., p. 697. 2 5 Ibid., p. 696. 24 Ernst W. Stromsdorfer, "Labor Force Adjustment to Structural Displacement in a Local Labor Market", Industrial and Labor Relations Review, Vol. 18, No. 2, Jan. 1965, p. 152. 25 It may be argued that a person wishing to finance a retraining course and cover his family's l i v i n g expenses in the meantime can always take up a loan. However, banks usually regard unemployed workers with obsolete s k i l l s as poor risks. 26 Lipsey, Employment Policy and the Labor Market, pp. 253-4. 27 Gallaway seems to imply that lags in the adjustment process are a basic part of an efficient system of factor allocation. See Gallaway, op. c i t . , pp. 699-700. But surely the lengths of the lags w i l l determine the degree of disequilibrium that the system w i l l permit. Consequently, the longer the adjustment lags are, the more inefficient the allocation system must be. 28 Walter Y. Oi, "Labor as a Quasi-Fixed Factor", Journal of P o l i t i c a l  Economy, Vol. 70, No. 6, Dec. 1962, pp. 538-55-29 M.W. Reder, "Wage Structure and Structural Unemployment", Review of  Economic Studies, Vol. 31 , No. 8 8 , Oct. 1964, pp. 309-22. Stromsdorfer, op. c i t . , p. 152. Oi, op. c i t . , pp. 540. 32 T Loc. c i t . 51 Reder, op. c i t . , pp. 310-11. Knowles and Kalacheck have given a plausible example of such a compromise. "If, for instance, the growth in the supply of engineers is less rapid than the growth in demand at going wage rates, the use of engineers i s economized, by providing the average engineer with a greater number of less skilled assistants." (U.S. Joint Economic Committee, op. c i t . , p..11, n. k.) Barbara R. Berman, "An Approach to an Absolute Measure of Structural Unemployment", Employment Policy and the Labor Market, ed. Arthur M. Ross, University of California Press, 1965, pp. 258-60. The shape of the employment expansion path i s derived from Berman's assumption that "the marginal propensity to hire unskilled labour i s higher than the average propensity to hire unskilled labour and the marginal propensity to hire skilled labour i s lower than the average". See ibid., p. 259. However, for the explanation of the structural imbalance between the structure of employment and the structure of supply the shape of the employment expansion path i s not important. What is essential i s that the employment expansion path is r i g i d and that i t does not pass through the s k i l l endowment point. Ibid., pp. 258-9. Frictional unemployment, which, according to Berman, consists of the unemployables and "rock bottom" unemployment, i s ignored in the model. The precise demarcation between fr i c t i o n a l and structural unemployment i s discussed below, under the policy-oriented approach. Each type of labour i s l i k e l y to have i t s share of f r i c t i o n a l unemployment. See ibid., pp. 258-60. S.F. Kaliski recently developed this approach into an empirical test of the existence of structural unemployment. ("Structural Unemployment in Canada: The Occupational Dimension", paper prepared for the Annual Meeting of the Canadian Economics Association, Calgary, June 1968, mimeographed.) Finding the high-skill occupations to have a low elasticity of employment with respect to output, he concluded that "expansions of the magnitude observed i n the postwar period can take place without creating much additional demand for highly skilled workers" (p. 30) and therefore "there is not l i k e l y to be any occupational structural unemployment /which i s defined as unemployment which arises because workers in complementary occupations are unavailable (p. 12^7 at the level of aggregation considered" (p. 31). He drew this conclusion in spite of the fact that he had recognized that " i t i s in practice unlikely that s k i l l bottlenecks w i l l constitute an absolute bar to any further expansion of output beyond the point of ' f u l l demand'. They are more li k e l y to permit expansion only at increasing cost resulting in price and possibly wage inflation" (p. 10). He ignored completely the fact that this inflation w i l l trigger deflationary policies which bring to a halt the further reduction of unemployment. This paper i s a good il l u s t r a t i o n of the p i t f a l l s of the Berman approach. 52 39 J.C.R. Dow and L.A. Dicks-Mireaux, "The Excess Demand for Labour: A Study of Conditions in Great Britain, 1946-56", Oxford Economic Papers, New Series, Vol. 10, No. 1, Feb. 1958, pp. 1-33. 40 4 Ibid., p. 21. The pressure of labour demand d i s measured as follows: d* = -u + >f uv where u > v d*. = v - uv where u < v See ibid., p. 22. The asterik has been used to distinguish this measure of labour demand from the one employed in Chapter III. 4l Ibid., p. 20. 42 Loc. c i t . 43 Dow and Dicks-Mireaux assume that the unemployment-vacancies function i s a rectangular hyperbola, so that m = >/uv. This makes i t possible to measure maladjustment m at any level of demand. See ibid., p. 22. 44 U.S. Joint Economic Committee, op. c i t . ; Walter W. Heller, "The Administra-tion's Fiscal Policy", Unemployment and the American Economy, New York, John Wiley & Sons, (1964), p. 97; R.A. Gordon, "Has Structural Unemploy-ment Worsened?", Industrial Relations: A Journal of Economy and Society, Institute of Industrial Relations, University of California, Vol. 3i No. 3, May 1964, pp. 6l-7. 45 U.S. Joint Economic Committee, op. c i t . ; Heller, op. c i t . ; R.A. Gordon, Industrial Relations, May 1964, pp. 61-7; Denton and Ostry, op. c i t . , p. 13. 46 • Killingsworth, op. c i t . , pp. 1471-7• 47 N.J. Simler, "Long-Term Unemployment, the Structural Hypothesis, and Public Policy", American Economic Review, Vol. 54, No. 6, Dec. 1964, p. 990-1. 48 Otto Eckstein, "Aggregate Demand and the Current Unemployment Problem", Unemployment and the American Economy, John Wiley & Sons, New York, (1964), p. 119; Denton and Ostry, opT c i t . , p. 13. 49 Gallaway, op. c i t . , pp. 710-2. The dispersion approach used in Denton and Ostry, op. c i t . , pp. 9-12, and U.S. Joint Economic Committee, op.cit., pp. 20-1, f a l l s somewhere between these two approaches in so far as i t does not require the specification of the structurally disadvantaged groups, but contains the assumption that an increase in structural unemployment leads to a dispersion of the unemployment rates in the various labour force groups which i s greater than that warranted by the prevailing overall unemployment rate. 5 0 Oi, op. c i t . , pp. 539-43. ^ Reder, op. c i t . , p. 319. 53 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 Lipsey, Employment Policy and the Labor Market, pp. 249-252. This latter interpretation applies to Simler's approach. See Simler, op. c i t . , p. 998. It is not clear which interpretation applies to the Denton-Ostry approach. In their definition they described structural unemployment to be of long duration. Subsequently, however, a l l that they asserted was that, according to the structuralist view, "workers who have been displaced by structural change w i l l experience greater-than-average d i f f i c u l t y in shifting into new employment, and hence experience a longer-than-average period of job-seeking. ...thus, one symptom of growing structural unemployment should be an increase i n the average duration of unemployment." (See Denton and Ostry, op. c i t . , pp. 15-6. I t a l i c s added.) A l l that this means i s that a greater proportion of the structurally unemployed w i l l be long-duration unemployed than of the demand-deficiency unemployed. It does not mean that a l l structural unemployment i s long-duration unemployment. Berman, op. c i t . , p. 263. Ibid, p. 268. Simler, op. c i t . , p. 997. An additional reason for the reduced probability of re-employment for the long-duration unemployed might be a decline i n morale. Ibid., p. 998. Loc. c i t . Loc. c i t . Berman, op. c i t . , p. 256. Ibid., pp. 256-7. It i s not clear whether seasonal unemployment i s to be broken down into the given components or whether i t is to be regarded as a separate component. The omission of seasonal unemployment suggests that Berman had the f i r s t alternative in mind. Ibid., pp. 259-60. Ibid., pp. 256-7. Lipsey, Employment Policy and the Labor Market, pp. 210-6. Ibid., pp. 211-2. Ibid., p. 215. R.C. Wilcock and W.H. Franke, Unwanted Workers, Collier-Macmillan, London, 1963, p. 6. L.G. Reynolds has also defined f r i c t i o n a l unemployment in this way. See Reynolds, Labor Economics and Labor Relations, Prentice-Hall, 1964, pp. 3^6-7. This approach inevitably makes structural unemployment a sub-type of f r i c t i o n a l unemployment. 54 68 69 70 71 72 73 74 75 76 77 78 79 Lipsey, Employment Policy and the Labor Market, p.215« Italics removed. A.W. P h i l l i p s , "The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957", Economica, New Series, Vol. 25, No. 99, Nov. 1958, pp. 283-99. Ibid., p. 283. Subsequently Lipsey developed a more rigorously formulated framework for the relationship, providing a justification for the shape of the function. ("The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1862-1957: A Further Analysis", Economica, Feb. I960, pp. 12-19.) This justification was attached by Bernard Corry and David Laidler ("The Phillips Relation: A Theoretical Explanation", Economica, May 1967, pp. 189-197) who argued that i t was possible to have a positive relation between unemployment and wage changes in the excess-demand phase of the cycle. John Vanderkamp ("The Phillips Relation: A Theoretical Explanation - A Comment", Economica, May I968, pp. 179-183) defended the postulated shape of the Phillips curve by strengthening Lipsey's framework. The Corry-Laidler rebuttal ("The Phillips Relation: A Theoretical Explanation - A Reply", Economica, May 1968, p. l84) promised a paper in which they hoped to show "that the Ph i l l i p s relationship would be weaker where V U / i . e . , where vacancies exceed unemployment/" but did not present a convincing defense of the possibility of part of the curve having a positive slope. Phi l l i p s , op. c i t . , p. 283. Lipsey, Economica, Feb. i960, pp. 20-23. S.F. Kaliski, "The Relation Between Unemployment and the Rate of Change of Money Wages in Canada", International Economic Review, Vol. 5, No. 1, Jan. 1964, pp. 4-5. Lipsey, Employment Policy and Labor Market, p. 217. Berman, op. c i t . , p. 260. Proulx, Structural Unemployment and Public Policy, Draft of a paper read to the Canadian Economics and P o l i t i c a l Science Association Conference, Vancouver, June 10, 1965, mimeographed, p. 29-Ibid., p. 26. Ibid., p. 30. Empirically, G.L. Reuber has observed on the basis of Canadian data that "at no point was a,satisfactory s t a t i s t i c a l relation found between increases i n productivity ^rfhich was used as the index of growth/7 and either price-level movements or the percentage level of unemployment". (G.L. Reuber, "The Objectives of Canadian Monetary Policy, 1949-61: Empirical 'Trade-Offs' and the Reaction Function of the Authorities", Journal of P o l i t i c a l Economy, Vol. 72, No. 2, April 1964, p. 113.) 55 R.A. Gordon, "Fu l l Employment as a Policy Goal", Employment Policy and  the Labor Market, ed. A.M. Ross, University of California Press, 1965, pp. 30-k. John Vanderkamp has tested the effect of import prices on inflation and has found them to be very significant. See his "Wage and Price Level Determination: An Empirical Model for Canada", Economica, May 1966, pp. 203-210, and his "An Application of Lipsey's Concept of Structural Unemployment", Review of Economic Studies, July 1966, pp. 222-223. CHAPTER III A THEORETICAL MODEL OF STRUCTURAL DISEQUILIBRIUM FOR THE MEASUREMENT OF STRUCTURAL UNEMPLOYMENT The previous chapter contains three approaches to the absolute measure of structural unemployment: (1) Berman's bottleneck model, (2) the unemployment-vacancy model by Dow and Dicks-Mireaux, and (3) Lipsey's trade-off model. Each of them has i t s merits, but also suffers from certain weaknesses. Berman's model requires that labour bottlenecks are distinctly recognizable and, furthermore, makes the highly unrealistic assumption that unemployment cannot be reduced any further once such a bottleneck has been encountered except by selective manpower policies. The model by Dow and Dicks-Mireaux lacks a rationale about the labour market mechanism underlying i t . Lipsey's trade-off model contains the impurities involved in the relationship of price inflation to labour demand. In this chapter a model w i l l be constructed which uses as pure a concept of structural imbalance as possible and which i s derived from an analysis of the underlying economic processes. To do this, structural unemployment w i l l f i r s t be formally defined, i n i t i a l l y i n terms of the mismatching of labour characteristics and subsequently i n terms of structural disequilibrium. Then the concept of structural equilibrium and disequilibrium w i l l be explained within the unemployment-vacancy framework of the model by Dow and Dicks-Mireaux, which is very useful for this purpose. Finally, the effect of fluctuations in overall- demand w i l l be analyzed. This leads to an equation whose quantification w i l l permit the measure of structural unemployment. 57 1. Conceptual Definitions a. Structural unemployment The best definition of the concept, as distinct from the measure, of structural unemployment is in terms of the maladjustment approach. Accordingly, structural unemployment i s defined as that component of unemployment which is due to a mismatching of the structure of the demand for labour with the structure of the supply of labour. In other words, the distribution of labour qualities, or characteristics, demanded by a l l the employers is different from the actual distribution of characteristics within the labour force; at the same time, the quantity of aggregate labour demand may or may not equal that of labour supply. This means that, on the one hand, there is a certain amount of unsatisfied demand for labour with certain characteristics, and, on the other, available labour with certain other characteristics cannot be fu l l y used and i s in excess supply. The term "structural imbalance" describes such situations well. Examples of structural imbalance would be unemployment among coal miners accompanied by a shortage of mining engineers; a surplus of farm labourers and, at the same time, a dearth of construction labourers; unemploy-ment among truck drivers in Halifax and abnormal over-time by truck drivers in Toronto. These examples show that the imbalance may occur i n different dimensions, such as the occupational, the industrial and the regional. Generally, i t is not just one of these dimensions which we might expect to be involved, but several of them simultaneously. So we might find that there i s unemployment among coal miners in Nova Scotia and a shortage of inspectors in the electronics industry in Ontario. This, too, would be a case of structural imbalance and structural unemployment. 58 To the extent that unemployment i s structural, i t must have a quantitatively equivalent counterpart of labour shortage in some other group of the labour force. In other words, only that part of unemployment which is matched by an equal amount of vacancies can be called structural unemployment. b. Demand-deficiency unemployment The other component of unemployment is demand-deficiency unemploy-ment. Structural and demand-deficiency unemployment are here regarded as mutually exclusive as well as exhaustive. Any other type of unemployment must be regarded either as being a sub-group of one of these main types of unemployment or as being included i n both of them in this system. The definition of demand-deficiency unemployment follows from the definition of structural unemployment. Demand-deficiency unemployment i s defined as that part of unemployment which i s due to an inadequacy i n the level of overall demand. , In other words, i t i s the difference between the number of available jobs, f i l l e d and unfilled, and the number of workers in the labour force. According to this definition, there i s no demand-deficiency unemployment as long as vacancies exceed unemployment. The difference between unemployment and vacancies, i f positive, i s the level of demand-deficiency unemployment. c. Frictional unemployment The two other types of unemployment which are conventionally referred to are fr i c t i o n a l and seasonal unemployment. According to the structural imbalance approach, the distinction between structural and fr i c t i o n a l unemployment is not very important. A distinction might be made on the basis of the duration of unemployment or of i t s susceptibility to structuralist policies. To avoid unnecessary complications, the definition of structural unemployment used here includes both short-duration 59 unemployment as well as unemployment of intermediate or long duration. It includes employables, those who can be appropriately retrained* as well as unemployables, those who cannot be retrained for available jobs. For policy purposes, these distinctions may be important. However, for the purpose of measuring non-demand-deficiency unemployment, the combined treatment of fri c t i o n a l and structural unemployment•is more convenient. Consequently, the term structural unemployment in this chapter w i l l actually refer to frictional-structural unemployment, d. Seasonal unemployment Seasonal unemployment can in the context of this analysis be regarded as part of either demand-deficiency unemployment or of structural unemployment. It i s attributable to demand deficiency to the extent that i t results from seasonal fluctuations in overall demand for goods and services and from seasonal fluctuations in the average non-labour costs (or the feasibility) of production and of the rendering of services, since they affect the level of overall labour demand. Seasonal unemployment may be structural i f during certain months of the year there are unfilled vacancies for certain groups of the labour force while there i s unemployment among other groups. For example, winter unemployment among loggers may be accompanied by unfilled job vacancies among winter resort personnel. Part of the winter unemployment would then be structural. It could be eliminated i f loggers could perform the winter jobs for which there i s insufficient labour supply. Seasonal unemployment may also be structural i f there are year-round jobs which are open and unfilled, while there i s seasonal unemployment. For the purpose of illustration, let us suppose that there are unfilled vacancies for mechanics who work the f u l l year. At the same time there are seasonally employed loggers. Let the vacancies for mechanics be equal to 6o the number of employed loggers, say x. During most of the year, there w i l l be no unemployment in these groups, while vacancies w i l l amount to x. In the winter, there w i l l be x unemployed loggers and x vacancies for mechanics. Now, i f the loggers were retrained to become mechanics, there would be x vacancies (for loggers) for only part of the year, and there would be no unemployment at any time of the year. Thus, according to our definition, seasonal unemployment i n such a context would be structural.^ 2. Refinement of the Measure of Structural Unemployment The above definition of structural unemployment i s easily 2 quantifiable, given appropriate vacancy data. The structural unemployment rate, as defined so far, i s simply the unemployment rate or the vacancy rate, whichever i s less at any point of time. This definition would make i t possible to measure the structural unemployment rate without d i f f i c u l t y . However, i t has a complicating idiosyncrasy: this measure i s liable to cycli c a l fluctuations. This can best be seen by referring to Figure 2.2(a) in the previous chapter, which represents the relationship between the unemployment and vacancy rates hypothesized by Dow and Dicks-Mireaux. The upper part of the i function F^F^ represents conditions of demand deficiency where the vacancy rate i s smaller than the unemployment rate and therefore equal to the structural unemployment rate according to the above definition. As we move from F^ toward F^, the vacancy rate and the "structural unemployment rate" increase u n t i l we reach the diagonal. At this point we enter conditions of excess demand, where the unemployment rate i s smaller than the vacancy rate and therefore represents the "structural unemployment rate". As we move i toward F the total and "structural" unemployment rates decline. 61 However, i t does not make sense to have a cyclically fluctuating structural unemployment rate. Consequently, the definition of structural unemployment must be restricted to apply only the point of full-employment demand when the unemployment and vacancy rates are equal: u = u = v when u = v (3.1) s Now nothing can be said about the level of structural unemployment when there i s deficient or excess labour demand, unti l the cyclical relation-ship between the unemployment and vacancy rates i s determined. This i s the purpose of the remainder of this Chapter. Before this relationship i s discussed however, the next section analyzes the concept of a vacancy. 3. The Relevant Definition of a Job Vacancy It i s important for this analysis that vacancies are appropriately defined. Since the structural vacancy rate must be equal to the structural unemployment rate i n this model, there must be a certain conceptual symmetry between vacancies and unemployment. Just as unemployment represents potential additional employment on the supply side, so vacancies must represent potential additional employment on the demand side. Basically a vacancy i s simply ah unfilled job opening. This concept, however, involves ambiguities with respect to (1) the efforts of the employer to f i l l the. vacancy, (2.) the time at which a job applicant could be employed, (3) the wage conditions, (4) the demand for labour which is complementary to the vacancy, and (5) the distinction between potential job additions and potential job substitutions. Since a vacancy represents potential employment and not necessarily a dire scarcity, there i s no need to impose the restrictive condition that the employer must be actively searching for a worker to f i l l the vacancy. A l l that i s necessary i s that i f a job applicant with the appropriate qualifications presents himself, he w i l l be hired. 62 On the other hand, there must be the condition that he w i l l in that case be hired right away (except where labour complementarities are involved, as w i l l be seen below) i f the vacancy i s to be part of the current unsatisfied labour demand. One problem connected with this point i s that of inflated vacancies due to the bidding of several firms for the same contracts. The jobs created by the planning for the projects w i l l be greater than f i n a l l y required, because they w i l l disappear in the unsuccessful firms. However, the permanency of a job i s not a necessary criterion in the definition of a vacancy. As long as the employers are prepared to hoard labour for the expected contracts and hire before their awarding, these jobs must be regarded as vacancies."^ The fact that the unsuccessful employers may lay off the newly hired persons or drop the vacancy soon afterwards merely means that there i s a certain v o l a t i l i t y in labour demand at the disaggregated level. At the aggregate level this effect i s not l i k e l y to be noticeable. The third problem i s that of whether or not to include job openings with sub-standard wages from the measure of vacancies. The "standard" wage may refer to either a legal minimum, a non-legal minimum determined on the basis of some social criterion or the market equilibrium level. In the case of a legal minimum, the exclusion of job openings offering wages less than this standard w i l l depend on whether i t i s comprehensively enforcible, whether i l l e g a l l y f i l l e d jobs are excluded from the employment figure and whether the employment prevented by the legal barrier i s not to be included in the measure of structural unemployment. The application of a social minimum to the definition of a vacancy requires that jobs f i l l e d at sub-standard wages are excluded from the employment and labour force figures. 6 5 The application of a standard based on the market equilibrium wage is more complex. Requiring the standard wage to represent the equilibrium wage in the present defines away a l l vacancies because the very existence of a vacancy indicates that the wage rate i s below the equilibrium level. Instead, one would have to use some concept of a "normal" wage which would bring about equilibrium after a more or less lengthy period of time has been allowed for structural adjustment. This problem has i t s counterpart in the definition of unemployment. The Report of the Committee on Unemployment Statistics stated: "Some housewives and retired men are available for work but prepared to accept work only under specific circumstances of pay and surroundings; others are eager to obtain immediate work of almost any kind. The latter are obviously unemployed; but whether the former should be classed as unemployed depends on the intensity of their desire for work and on whether the specific circumstances they have in mind are reasonably commensurate with their qualifications."^ Once i t has been determined that the unemployed are best enumerated by using the discretionary criterion of "whether the specific circumstances they have in mind are reasonably commensurate with their qualifications" rather than letting the employment status be determined by the interviewer, the enumeration of vacancies must be conducted on a parallel basis. Thus, job openings must be counted as vacancies as long as they provide benefits and conditions "commensurate" with the type of work and qualifications 5 required. Alternatively, i f unemployment is defined to include persons with wage demands so high and vacancies to include job openings with wage offers so low as to prevent the elimination of structural unemployment, even when the demand and supply structures with respect to a l l other characteristics have been perfectly matched, then this means that this remaining structural imbalance i s due to.a mismatching of the wage characteristics of demand 6k and supply.^ What i s important for the measurement of structural unemployment i s not so much the scope of the concepts of unemployment and vacancies but their symmetry with respect to each other, so that job seekers and job openings are comparable. The fourth problem in defining vacancies stems from the danger of omitting vacancies which could be f i l l e d with available labour but are not because certain complementary labour is not available. If the employment of a shift of workers i s held up because of the unavailability of a shift engineer, then not only must the job of shift engineer be regarded as a vacancy but also the jobs of the whole shift, since these represent the current potential employment. Thus any survey of job vacancies, whose data i s to be used in labour demand estimates, must ask not only for the job openings that cannot be f i l l e d but also for the additional labour that would be hired i f the d i f f i c u l t vacancies were f i l l e d . Such complementarities may exist not only within establishments but also between them. Inter-establishment complementarities may manifest themselves in bottlenecks in the inter-industry product flows for which labour bottlenecks are responsible. Thus the expansion of employment in the automobile industry may be halted in spite of brisk consumer demand, because there i s a labour bottleneck in the steel industry supplying the auto industry. Such a complementarity can exist on the input side as well as on the output side of the firm or industry i n which the labour bottleneck exists. We can therefore have the converse case where a labour bottleneck in the auto industry impedes the expansion of employment in the steel industry i n spite of growing f i n a l demand. Unless the d i f f i c u l t i e s involved in tracing these complementarities can be overcome, there w i l l be deficiencies in the quantification of the vacancy rate and relative labour demand and a 7 s t a t i s t i c a l demand creation effect associated with structural adjustment. 65 This also raises the problem of structural imbalances in the system created by bottlenecks in the supply of production inputs other than labour. It might be argued that since these bottlenecks cannot be cleared up by manpower policies, the unemployment they are responsible for must be attributed to a shortage of labour demand which then i s to be traced to a structural imbalance rather than overall demand deficiency. However, unless we want to abandon the dichotomy between aggregate-demand and structuralist policies and create the additional distinction between structuralist manpower policies and other structuralist policies, i t makes more sense to classify unemployment attributable to non-labour structural imbalances as structural unemployment, too. This i s also justified by the fact that there may be a choice as to the policy by which to reduce a structural imbalance, that Is, by changing the structure of supply either of labour or of raw materials and intermediate products or of capital and technology. For example, a shortage of plasterers and lathers might be met either by encouraging more workers to enter the occupation, or by stimulating the production and use of more semi-fabricated building materials, or by inducing the use of more labour-saving equipment. Similarly, a shortage of bookkeepers might be reduced by paying training allowances for accountant courses or by making information available on alternative accounting systems which require fewer accountants. The avail-a b i l i t y of these options may sometimes make i t d i f f i c u l t as well as pointless to determine the particular production input "responsible" for the structural g imbalance. Thus, ideally jobs which would become available i f a production input bottleneck of any kind whatsoever i s eliminated should also be considered vacancies. 66 On the basis of these problems, i t may be worthwhile to distinguish between two types of vacancies: (l) overt vacancies, and ( 2 ) latent vacancies. Overt vacancies are those which can be f i l l e d right away because they involve jobs for which the required complementary produc-tion inputs are available. Latent vacancies are those for which job seekers are available but cannot be f i l l e d because complementary production inputs are not available. From the point of view of policy, this distinction i s important. Manpower policy needs to tackle only the overt vacancies and by f i l l i n g them w i l l , with the help of policies clearing up bottlenecks in the supply of other production inputs, increase employment by the number of overt and latent vacancies. Taking into account the problem of actual data collecting, i t may not be possible to enumerate more than the overt vacancies. Perhaps the latent vacancies associated with the overt vacancies in the same establish-ments can be obtained by surveys. But i f i t i s thought that the latent vacancies resulting from bottlenecks in complementary firms are significant, 9 they have to be estimated. The f i f t h problem in defining vacancies i s that of those whose f i l l i n g would merely lead to the displacement of currently employed persons. If a company declares a job opening for a well-educated salesman who could replace one of their present salesmen who has had l i t t l e schooling, this vacancy does not represent a potential increase i n employment. Thus we have to distinguish job offers which are potential job additions from those which are potential job substitutions and exclude the latter from our vacancy measure. The appropriate definition of a vacancy can be boiled down to a job opening which could be f i l l e d immediately i f (a) the appropriate job 67 applicant presented himself or (b) the required complementary production inputs were available, and whose f i l l i n g would not involve the dismissal of a presently employed worker. k. The Model a. Purpose and outline of the model Since the models discussed in Chapter II have various distinctive weaknesses, i t i s necessary to develop a model which can satisfactorily explain the relationship between the structural unemployment rate and the relative labour demand and provide an equation that w i l l yield a formula by which the unemployment rate at any level of labour demand can be adjusted to the corresponding rate at the full-employment demand level. This i s the aim of the model which follows. Since structural unemployment represents a failure or a sluggish-ness of the equilibrating mechanism of the labour market, the model w i l l be based on the interaction of structurally disequilibrating and equilibrating effects which are exerted on and by the labour market. Consequently, the concept of structural equilibrium and disequilibrium w i l l f i r s t be analyzed. Then the basic part of the model w i l l be developed, where i t w i l l be assumed that there is continuous full-employment demand. This assumption i f fi n a l l y dropped and the implications of varying demand are investigated. In the end a model is obtained which, when applied to data, w i l l provide a formula for the measurement of structural unemployment. b. The concept of structural equilibrium and disequilibrium  Structural equilibrium in the labour market i s achieved when the structure of labour demand is matched by the structure of labour supply. Thi means that when the relative labour demand, that i s , the per cent of aggregat labour demand relative to aggregate labour supply, is 100, there i s neither 68 unemployment nor vacancies. Furthermore, i t means that when there i s excess demand, the vacancy rate i s equal to the excess relative labour demand, while unemployment i s zero. Conversely, when there i s demand-deficiency, the unemployment rate i s equal to the deficiency i n relative labour demand, with the vacancy rate equal to zero. If, however, there i s unemployment as well as vacancies, this means that the forces of substitution outlined in the previous chapter have not yet been able to convert a l l the structurally unemployed into characteristics appropriate for the vacancies or to convert a l l structural vacancies so that they w i l l f i t the characteristics of the unemployed. The labour market i s consequently in disequilibrium in the structural sense. The cause of structural disequilibrium are structural dislocations exerting themselves on the labour market, whose adjustment i s impeded by limited or lagging substitutability among different types of labour and other production inputs. A disequilibrating effect involves the conversion of employment into unemployment and vacancies. If a structural change creates a vacancy which cannot be f i l l e d immediately and at the same time makes a worker superfluous, i t has increased the structural disequilibrium by one worker. The simultaneous creation of vacancies and unemployment may not be due to any single structural change nor occur within a single firm, but emerges from the whole concurrent set of structural changes and i s dispersed over the whole economy. Disequilibrating effects would be exerted by an array of structural changes which displace workers in certain occupa-tions, industries, areas, etc., and provide potential employment in others. They axe counteracted by the equilibrating effects of structural adjustment. Structural adjustment can occur through three forms of substitution: (1) Consumers may switch from products which have scarce labour embodied in them and whose prices are therefore rising to goods and produced by more abundant types of labour and characterized by prices rising less than the general price level or even f a l l i n g . (2) Producers may adopt new production techniques which economize on scarce types of labour and perhaps u t i l i z e more extensively the abundant types of labour. They can also reduce those non-labour production inputs whose supply i s adversely affected by the labour bottleneck and increase those which are substitutable for the scarce inputs, whether labour or other. (3) The characteristics of jobs and workers may be substituted. Employers may sacrifice quality and efficiency and f i l l their vacancies with inferior labour, or, alternatively, upgrade labour by providing training. Workers may accept employment in other firms, industries and areas and retrain to enter new occupations in order to escape unemployment or to increase their earnings. New workers with more appropriate qualifications may replace retiring workers with obsolete s k i l l s . Thus, structural adjust-ment can occur through equilibrating changes in the structure of labour demand (consumer substitution, equilibrating technological change, changes in non-labour production inputs, modification of job requirements) as well as i n the structure of labour supply (labour mobility, education and training). Structural adjustment i s equilibrating in the sense that i t converts unemployment and vacancies into employment. If an unemployed worker moves from Vancouver to Toronto to f i l l a job opening there, he is substituting his geographic characteristic of being a resident of Vancouver 70 by that of being a resident of Toronto, thereby reducing unemployment in Vancouver and vacancies in Toronto and increasing overall employment. The structural disequilibrium, measured by structural unemployment or structural vacancies, has thus been reduced by one worker, c. Changing overall demand Until now relative labour demand has been assumed to remain constant. If the model i s to be applied to a cyclically fluctuating economy, this assumption must be relaxed. Let us look at the following hypothetical sequence of events: (1) The i n i t i a l position i s one of such low demand that there is only a negligible vacancy rate or none at a l l . (2) A sudden structural shock takes place during this depression. (3) There are no additional structural changes during the subsequent phases of the cycle, nor does any structural adjustment take place. In other words, there i s absolute complementarity on the labour demand side and non-substitutability on the labour supply side. In this case, as long as aggregate demand i s very low, the structural shock w i l l not be manifested by vacancies. It i s only when demand expands to an adequate level that a bottleneck w i l l occur and unemployment w i l l not decline with subsequent increases i n demand which w i l l be f u l l y represented by increases in vacancies. The employment growth is ended once the bottleneck i s encountered. This pattern occurs in reverse during the contraction. When demand f a l l s below the point where the struc-tural imbalance i s no longer a bottleneck, the burden of demand declines shifts from vacancy declines to unemployment increases. This case describes the Berman model of the problem of structural imbalance. Its striking 71 characteristic i s that demand changes exert themselves either on unemployment or on vacancies (depending on the phase of the cycle), but not on both simultaneously. To come closer to reality, the assumption of complete complemen-tarity has to be replaced by that of partial complementarity. It i s not r e a l i s t i c to assume that the very f i r s t labour bottleneck completely inhibits any further employment expansion, even i f i t i s assumed that there i s no structural adjustment. Expansion can s t i l l continue in those firms which have not run into labour bottlenecks and which do not depend on orders or supplies from firms which actually have encountered such bottlenecks. Thus, a shortage of s k i l l s in the rubber goods industry may inhibit the expansion of the automobile industry which requires tires, but w i l l not affect the food processing industry, for instance. This i s what i s meant by partial complementarity. As demand increases, the overt vacancies in the rubber goods industry and the latent vacancies in the automobile industry w i l l increase, but other industries w i l l be able to expand employment. As the labour bottlenecks increase with demand, the vacancies w i l l rise at an accelerating rate. This is so because with rising demand (l) the vacancies associated with each bottleneck w i l l increase, and (2) the number of bottlenecks i t s e l f w i l l increase. On the other side of the coin, unemployment w i l l decline at a decelerating rate, as (l) unemployment in each labour group i s diminished, and (2) the number of labour groups characterized by unemployment is reduced. This can be demonstrated by the following simple example. Let us assume that our economy consists of three industries which are sufficiently independent of each other in the inter-industry flow pattern that the continued expansion of each does not depend on the continuation of the expansion of the other two. We are also maintaining the assumption that 72 there are no structural adjustments and no additional structural dislocations. Thus, there w i l l be no mobility of labour, neither of the employed nor of the unemployed, from one industry to another. We can now plot hypothetical labour demand functions for the three industries, A, B and C, with respect to overall labour demand (D), demand being measured in terms of numbers of workers (N). (See Figure 3 . 1 . ) The supply of labour (S) for each industry i s constant since we are assuming no structural adjustment. Unemployment i s represented by the excess of supply over demand and vacancies by the excess of demand over supply. The f i r s t labour bottleneck i n the expansion of overall demand is encountered in industry A, leading to the emergence of the f i r s t vacancies. As overall demand continues to increase, the vacancies i n A w i l l also increase* Then a bottleneck appears in industry B, which means that the rate of increase of vacancies per unit of overall demand w i l l rise from the rate of increase of demand in A per unit of overall demand to the sum of the rates of increase of demand in A and B per unit of overall demand. When the bottleneck i n C is reached, the rise in vacancies i s again accelerated. After a l l the industries have encountered bottlenecks, the acceleration w i l l come to an end and the slope of the vacancy function w i l l be one. This must be so since unemployment among those currently employable without structural adjustment has been eliminated so that, according to our definition of labour demand, the total demand for labour w i l l rise by one labour unit with each additional vacancy. Similarly, unemployment f a l l s at the same rate as that at which labour demand rises, u n t i l the f i r s t unemployment pool i s exhausted, that i s , the f i r s t labour bottleneck appears in the form of vacancies. Any further 73 Figure 3.1 - An ill u s t r a t i o n of the effects of accumulating labour bottlenecks in different industries on the relationship between overall labour demand (D), on the one hand, and unemployment (U) and vacancies (V), respectively, on the other 74 expansion of labour demand w i l l be accompanied by a diminishing rate of reduction of unemployment, u n t i l the unemployment pools of a l l currently employable labour groups have been absorbed. This basic shape of the unemployment and vacancies functions i s not changed by relaxing the assumptions about structural dislocations and adjustment. So far, i t has been assumed that after the i n i t i a l structural imbalance no further dislocations occurred and no structural adjustment took place. The effect of continuous adjustment and dislocations on the functions w i l l now be considered. The effect of structural adjustment is to accentuate the curvature of the unemployment and vacancy functions. Structural adjustment has been shown above to occur on the labour demand side through consumer substitution and technological change and on the supply side through Industrial, geographic and occupational mobility. If f i n a l demand and production techniques are responsive to labour shortages, labour demand w i l l shift from unit A i n the example to unit C and perhaps B as overall demand increases. This means that vacancies i n A w i l l increase less rapidly and unemployment in B and C w i l l decline more rapidly than i s indicated in Figure 3.1 A more responsive form of adjustment, however, i s probably labour mobility. As vacancies emerge in A, the unemployed in B and C are li k e l y to try to get jobs i n A. The less adjustment in the form of geographic movement and changes in occupation is required, the more significant w i l l be this s h i f t . Its effect i s to reduce the labour supply of C and increase that of A, which again means that the increase of vacancies in A w i l l be moderated and the decline in unemployment in C w i l l be accelerated. The result i s that the vacancy and unemployment functions intersect each other at a lower point, which means that structural unemployment i s lower. On the other hand, continuous structural dislocations w i l l work in the opposite direction. Labour demand in the tight industry w i l l be pushed up and in the slack industry down and even on the supply side there may be structural dislocations i n that there may be particularly high at t r i t i o n in the tight industry (although that i s an unlikely instance). This w i l l push up the unemployment and vacancy functions and flatten their curvature. The net result, i n an economy with many industries and a large distribution of bottlenecks, w i l l be the kind of continuous functions shown in Figure 3.2(a). When they are combined into a single function i n the unemployment-vacancy coordinate system, they emerge as the hyperbola shown in part (b) of Figure 3.2. This i s consistent with the function used by Dow and Dicks-Mireaux and the preceding discussion therefore provides a theoretical rationale for i t too. We now have the relationship u = f(v) which is of an approximately hyperbolic shape. It is not a single function, however, but a set of functions a l l of which incorporate the same relation-ship between the unemployment rate and the vacancy rate but each of which corresponds to a different level of structural imbalance, i.e. u t = f ( v t , at) (3.2) where a varies with the level of structural imbalance. The structural unemployment rate can then be obtained by solving u = f(u, at) (3-3) for u. The latter then represents the structural unemployment rate. 5. Summary The purpose of this Chapter has been to construct a model for the absolute measurement of structural unemployment which (l) employs variables 76 Figure 3.2 - The cyclical relationship of the unemployment rate, u, and the vacancy rate, v, (a) to relative labour demand, d, and (b) to each other 77 directly related to the labour market, (2) is based on a theoretical rationale, and (3) makes i t possible to compare structural unemployment under various demand conditions. The basic definition on which the model is based i s that of structural unemployment, that i s , that part of unemployment which i s quantitatively matched by vacancies. It represents the qualitative imbalance between labour demand and supply. However, the measure which derives from this definition fluctuates with the cycle in overall demand. The definition therefore had to be restricted to apply only to full-employment-demand conditions. Under other conditions certain adjustments have to be made. To determine these adjustments, the cyclical relationship between unemployment and vacancies had to be analyzed. After considering the effect of bottlenecks on cyclical expansion and contraction and of structural adjustment and dislocations, i t was concluded that the relationship i s hyperbolic. On the basis of the fact that there i s this definite relation-ship between unemployment and vacancies, the simple equations for the measurement of structural unemployment were developed. 78 FOOTNOTES TO CHAPTER III This example is an indication of the danger involved in minimizing unemployment without considering earnings. Seasonal work may well pay enough to make up for the idleness during part of the year. Nevertheless, training may be warranted to provide for seasonal labour mobility and thus maximize labour income as well as alleviate labour market shortages at least during the slack season. The fact that only inadequate vacancy data are available i n Canada does not make this approach useless. It remains useful in that (l) i t points out the importance of collecting adequate vacancy data for the measurement of structural unemployment and labour demand and (2) i t can s t i l l be used in the meantime for s t a t i s t i c a l testing (a) by making estimates of the vacancy rate on the basis of N.E.S. vacancy data or (b) by using a s t a t i s t i c a l proxy for the vacancy rate, such as the rate of change of wage rates. In Chapter IV N.E.S. vacancy data are used. See J.C.R. Dow and L.A. Dicks-Mireaux, "The Excess Demand for Labour: A Study of Conditions in Great Britain, 1946-56", Oxford Economic Papers, New Series, Vol. 10, No. 1, Feb. 1958, p. l6.n.2., for a similar position. Canada, Government, Departments of Trade and Commerce and of Labour, Committee on Unemployment Statistics, Report, Ottawa, I960, pp. 4-5. The applicability of the intensity-of-desire-for-work criterion, on the other hand, depends on the purpose of the unemployment measure. If unemployment i s to measure only unquestionably involuntary unemployment, then there are legitimate grounds for excluding unemployed persons merely prepared to accept work but not actively seeking i t , whereas i f unemploy-ment i s to measure the potential expansion of employment, they must be included, because they represent potential matches for appropriate vacancies. The appropriate structuralist measure would be a program of information by which the job seekers and the employers are made aware of the incompetitiveness of their wage demands and offers, respectively. Conversely, the s t a t i s t i c for labour demand w i l l be reduced as disequilibrating effects create more unemployment than recorded vacancies. In the determination of a shortage of non-labour production inputs, the same types of problems arise as in the definition of a job vacancy. 79 Two basic methods of estimating the latent vacancies have occurred to me, one aggregative and the other disaggregative. The aggregative method involves the separation of changes of employment due to aggregate demand from those due to changes in structural imbalances. Without going into the details and problems of such an empirical analysis, i t seems to me that this separation can be obtained on the basis of the observation that changes in aggregate demand result in changes in employment and vacancies in the same direction whereas changes in the degree of structural maladjustment result in changes in employment and vacancies in opposite directions, since i n the latter case either the employed are displaced while vacancies occur elsewhere or the unemployed find matching vacancies and thus reduce vacancies while increasing employment. The disaggregative method involves input-output analysis. Assuming that the intra-establishment effects can be directly surveyed, the remaining inter-industry effect of overt vacancies on employment can be assessed on the basis of the responsiveness of the relevant occupation-industry coefficients to the labour bottleneck. The more r i g i d these coefficients are, that i s , the more the industry output is affected by i t s overt vacancies, and the more the dependent industries are affected by the restraint on output placed on the industry with the vacancies, the greater w i l l be the employment effect. In other words, the degree of r i g i d i t y or responsiveness of the coefficients indicates the quantitative significance or insignificance, respectively, of changes in latent vacancies. CHAPTER IV APPLICATION OF THE STRUCTURAL DISEQUILIBRIUM MODEL TO CANADIAN DATA The structural disequilibrium model described in Chapter III was primarily constructed for the measurement of structural unemployment and i t s variations. To make i t applicable, however, reliable vacancy stati s t i c s are required. Unfortunately, these are not available at present for Canada. Only vacancies reported to the National Employment Service are recorded and they are not very reliable. However, they can be used to demonstrate how the model i s to be applied, and the period 1953-65 is covered for this purpose. It would, however, be unwise to use them to definitively measure structural unemployment. 1.. The Data Problems a. The statement error i n the reported vancancies It i s quite obvious from the description of the source that the number of vacancies reported to N.E.S. f a l l s very much short of the economy's total vacancies. While the estimation of the statement error i s not necessary for the determination of s t a t i s t i c a l l y significant relationships in the model, i t i s nevertheless done to get a reasonably approximate picture of overall labour shortages and surpluses, for which a vacancy rate comparable in measurement to the unemployment rate i s necessary, and to make the cardinal measure of structural unemployment r e a l i s t i c . One approach to the estimation of the statement ratio i s to use the ratio of the total hirings to those hirings which resulted from N.E.S. referrals as the statement ratio of the total vacancies to the unfilled 81 vacancies reported to N.E.S. The use of the hirings ratio i s based on the relationship between the vacancy rate, the hirings rate and the average duration of vacancies. As long as these three variables are relatively stable, the average duration of vacancies i s equal to the vacancy rate divided by the hiring rate. v where d = average duration of vacancies, and h = hiring rate. For example, i f the vacancy rate i s stable at 8 per cent of the labour force and the hiring rate in the current month is 4 per cent, new vacancies must be appearing at a rate equal to the hiring rate, that i s , 4 per cent of the labour force. Assuming, temporarily, that the duration i s the same for a l l vacancies, half of the vacancies w i l l have appeared in the previous month. The other half must have appeared in the second-last month and in the current month i s being f i l l e d and i s being replaced in the vacancy pool by the new vacancies. This means that the vacancy duration period i s two months (8 divided by 4). If the assumption of the equality of the duration periods i s relaxed* the conclusion has to be modified to refer to the average vacancy duration period. If there are significant fluctuations in any of the three variables, the above statement of the relationship i s partially upset by lags contained in the actual relationship, which, to be completely accurate, would have to be described i n terms of the emergence and f i l l i n g rates of vacancies and the relative distribution-of duration periods. However, in the context in which the relationship i s used, this problem can be ignored, as w i l l be seen below. The above relationship can be restated as v = hd V (4.2) 82 If T i s used to denote vacancies in the whole economy and N vacancies handled by N.E.S., then the vacancy statement ratio i s given by the equation V h ^ — — (k ^ ) The problem with this formula i s that there is no information about d^. To solve i t , i t i s assumed that % = *Z < ^ > Is this assumption reasonable? Employers may, on the whole, have been passing on to N.E.S. only those vacancies which are relatively d i f f i c u l t to f i l l . On the other hand, N.E.S. had immediate contact with a very large labour supply. Furthermore, jobs for highly skilled labour, which probably have longer vacancy duration periods than semi-skilled and unskilled jobs, tended to be f i l l e d outside the N.E.S. system. On the basis of these considerations i t is certainly not obvious whether the average duration of N.E.S. vacancies or that of a l l vacancies was greater. Equality between the two, on the other hand, is not obvious either. The argument for nevertheless using the assumption consists of the following points: (1) No better method was discovered. (2) Even i f the assumption of equality i s not correct, the results w i l l s t i l l be meaningful v v as long as the ratio of d^ to d^ .is more or less constant. That the ratio i s approximately constant is reasonable to assume. That the results w i l l s t i l l be meaningful in that case i s due to the fact that the most important purpose of the exercise is to discover any trend in structural unemployment rather than i t s precise level. Thus, while there may be a bias in the vacancy rate and, therefore, in the structural unemployment rate, as long as this bias is constant the trend w i l l be free from i t . 83 or With this assumption equation (4.3) becomes VN = hN V = — V T ^ N (4.5) Information about the hirings ratio has been obtained from two sources: the Survey of Hirings and Separations in Certain Industries covering about two thirds of the employed labour force and the N.E.S. statist i c s on placements. The data are provided in Table I. Table I Ratio of Total Hirings to N.E.S. Placements Monthly N.E.S. Total Hirings -Average Annual Placement Rate N.E.S. Monthly 1 N.E.S. (% of labour Placements Hiring Rate Placements force) Ratio % •000 1953 6.9 993 . 1.53 4.5 1954 6.5 862 1.31 5.0 1955 6.9 954 1.42 4.9 1956 7.4 1,047 1.51 4.9 1957 6.7 878 1.22 5.5 1958 6.0 840 1.14 5.3 1959 6.3 986 1.32 4.8 i960 6.0 958 1.25 4.8 1961 6.0 1,120 1.43 4.2 1962 6.2 1,336 1.68 3.7 1963 6.2 1,178 1.45 4.3 1964 6.4 1,241 . 1.49 4.3 1965 6.4 1,258 1.47 4.4 Average 4.7 Sources: (1) (2) D.B.S., Hiring and Separation Rates in Certain Industries, Cat. No. 72-006. Canada Department of Labour, Labour Gazette. 84 The last column in this table also provides the vacancy statement ratio. It fluctuates to a certain extent, but on the whole i t i s reasonably stable around the average of 4.7. Rather than make any pretensions to great r e l i a b i l i t y by using the most refined method which would be more sensitive to the crude assumptions underlying i t , i t i s assumed that the vacancy statement ratio has been more or less stable around 4.7. The vacancy rate computed on the basis of this method i s given in Table II toward the end of this chapter, b. The period of analysis The period of analysis that has been chosen i s 1953 to 1965-I965 was selected because in 1965 administrative changes in Canada's employ-ment service occurred which destroyed the comparability of the vacancy series over time and which were the reason for their ultimate discontinuation. 1953 was used as the starting point because before that year the Labour Force Survey was not conducted monthly, but only four times a year. The period 1953 to 1965 appeared to provide an adequate number of observations, and i t was therefore not considered worthwhile to construct seasonally adjusted quarterly series for unemployment and the labour force from four observations a year for the years before 1953«^ 12. The Estimation of the Structural Unemployment Rate a. Alternative forms of the relationship between the unemployment rate and the vacancy rate There are alternative forms Into which equation (3«2) u t = F (V a t } can be put. The following w i l l be considered. (i) Estimation without regression analysis: One method i s to assume the simple hyperbolic relationship. u = | (4.6) 85 where a can vary with uv. According to the definition of structural unemployment i n Chapter III u - u when u = v s so that in this case u = u when the two functions u = •= and u = v intersect. s v This occurs when u = so that u = /u,v. (4.7) s t The d i f f i c u l t y with this form i s that i t imposes a hyperbolic function with a very limited shape on the observed data. ( i i ) Regressed asymmetical hyperbolas: A more flexible hyperbolic function i s b2 u = a 2 + -f (4.8) Since this equation, unlike (4.6), has two unknown parameters, i t has to be submitted to regression analysis. For the subsequent estimation of. the structural unemployment rate, b^ i s kept constant and a.^ permitted to vary with the residuals, so that ( \ - v V ( ut - V2 + 4b u = - — : s — • (*+.9) s t 2 Aside from the fact that this equation i s complicated, i t suffers from a s t a t i s t i c a l distortion. This distortion results from the fact that regression analysis minimizes the vertical deviations from the least-squares curve whereas changes in structural imbalances result in shifts away from or  toward the origin. Regression analysis w i l l therefore, lead to a curve which is flatter than i t would be i f the deviations were minimized in line with the origin. This can be seen from the simplified i l l u s t r a t i o n depicted in Figure 4.1(a). Let us assume that the economy has experienced two distinct levels of structural imbalance, one low (I) and the other high ( I t ) . For each level, there are three observations, one at high demand (H), one at 86 L Figure 4.1 -. The effect of minimizing the residuals in the unemployment-vacancy relationship in the cases of (a) a hyperbolic function and (b) a parabola rotated by 45° . 87 medium demand (M) and one at low demand (L). The curve which minimizes the vertical deviations i s IV, which i s considerably flatter than III, the curve that minimizes the deviations in line with the origin and correctly adjusts any observation to i t s structural-imbalance equivalent on the diagonal. Thus, to the extent that the curvature in the true function i s significant, the estimates for the structural unemployment rate based oil equation (4.4) w i l l be high for low-demand periods and low for high-demand periods. They are also liable to be more erratic in the low-demand periods than in the high-demand periods. The same d i f f i c u l t y , except for the cyclical pattern i n the variability of the estimate, affects the hyperbola (4.10) which can be submitted to regression analysis in logarithmic form. Keeping b.,. constant and letting a., vary (proportionately) with the residuals, the 3 3 estimate of the structural unemployment rate i s given by (1 + b,) / . u = ^ X/u^v.b3 (4.11) s t V t t ( i i i ) Parabola rotated by 45°: One alternative to minimizing the vertical deviations i s to use a format which w i l l permit minimizing the diagonal deviations. This i s a reasonably good approximation to minimizing the deviations in line with the origin for the middle sector of the quadrant, but tends to excessively sharpen the curvature of the function near the axes. This format, therefore, tends to underestimate structural unemployment i n periods of very high and very low demand. In order to make this format amenable to regression analysis, the quadrant has to be rotated to the l e f t by 45°, so that the .u-axis l i e s 88 between the y-axis and the negative x-axis of the new coordinate system and the v-axis l i e s between the y-axis and the positive x-axis. (See Figure 4.1(b).) The redefinition of the coordinate system i s given by the equations v - u V + u and y = Y2 Using a quadratic parabola to approximate a hyperbola rotated by 4-5°, we get the equation y = % + \ * 2 which, in terms of u and v, becomes (v + u ) = a K + bc.(v - u ) 2 (4.12) where a^ = a^\/2 \ and b 5 = -Keeping b^ constant and letting a^ vary with the residuals, the estimate of the structural unemployment rate i s given by v. + u b u = t , t - -f(v. - u,r (4.13) 2 2 t t This again i s a symmetric function and, therefore, somewhat restrictive i n that respect. (iv) Linear asymmetric function: There i s no plausible theoretical rationale for a linear relationship between the unemployment rate and the vacancy rate. Such a function would have to be kinked at the points where i t meets the two axes. However, within a limited range of demand level i t may provide a f a i r l y good approximation of the relationship. It too i s subject to the bias resulting from the minimization of the vertical 8 9 deviations in regression analysis and i s liable to overestimate structural unemployment in low-demand periods. In high-demand periods the bias can work in either direction. The equations are u = a 6 + b^v (4.l4) U t " B 7 v t and u = * V (4.15) s t 1 " b7 (v) The results of the regression analysis: Four of the five equations considered require regression analysis for quantification. The following results were obtained: u = 0.8435 - 10.6592 I R 2 = .72 (4.16) v (0.9411) log u = 2.3927 - O.8388 log v R 2 = .68 (4.17) e (0.0806) e v + u = 7.1694 ^ -'0.0630 (v - u ) 2 R 2 = .61 (4.18) • (0.0072) u = 9.2421 - 1.5296 v R 2 = .68 (4.19) (0.1489) The regression coefficients in a l l four equations are easily significant at the .005 level. What seems curious at f i r s t glance i s the relatively steep slope of (4.19). A look at the data, however, explains the reason. The great majority of observations are in the l e f t sector of the quadrant. In only eight of the 52 observations does the vacancy rate exceed the unemployment rate. Because i t i s l i k e l y that the only reason why the linear function has such a good correlation coefficient i s that most of the observations are clustered around one flank of an essentially hyperbolic relationship but very few around the other, i t i s probably a poor basis for estimating the rate of structural unemployment. 90 The fact that the data i s very asymmetrically distributed with respect to the diagonal also affects the other functions permitting asymmetry. Thus (4.16) and (4.17) must also be viewed with suspicion. This leaves as the most reliable function (4.18) which at the same time has the lowest correlation coefficient. The value of the correlation coefficient, however, does not necessarily provide the best criterion for evaluating the alternative relationships. It merely indicates which rela-tionship attributes most of the variations in the unemployment rate to the vacancy rate, representing the level of demand, and least to other factors, which have been assumed to be structural. The ideal test, involving s t a t i s t i c a l variables for structural dislocations and adjustment, was not possible because of the problem of quantification. Therefore, while the correlation coefficient i s a useful criterion, i t i s not the ultimate criterion. The problem of the asymmetry of the data i s sufficiently serious, as i s demonstrated by the way in which i t affected the regression coefficient in equation (4.l4), that the symmetry criterion must take precedence over the correlation criterion, at least i n the case where the differences between the correlation coefficients are so small. For this reason, equation (4.l8) i s considered to be the best estimate of the relationship between the unemploy-ment rate and the vacancy rate. c. Estimating the structural unemployment rate (i) The alternative estimated series: Even though equation (4.l8) i s considered to be the best approximation of the relationship between the unemployment rate and the vacancy rate, the structural unemployment rate as estimated on the basis, of a l l the alternative relationships, including equation (4.7) which did not require regression analysis, has been given in Table II. This should provide some indication of the degree of uncertainty attached to the estimated series. The uncertainty i s naturally highest when there is very low or high demand. Thus in the fourth quarter of i960 and the f i r s t quarter of 196l, a period of very low demand, the estimates range from J.6 to 4.2. It i s remarkable how close the two series based on the two symmetrical functions (4.7) and (4.l8), f i t each other. This suggests that the equation u = v/uv s " i s a good estimator which i s also simple. ( i i ) The trend: The results suggest that the structural unemploy-ment rate, that i s , the unemployment rate adjusted for excess or deficient labour demand conditions, rose from about 3.0 per cent in 1953 to about 3.6 per cent in 1956 and, after a slight reduction in 1957-58, to about 3.9 per cent in 1962. This last level was maintained f a i r l y steadily u n t i l the end of the analysis period, that i s , 1965. ( i i i ) Evaluation: It is not easy to compare these results with those of Vanderkamp which were obtained from tests of the inflation-unemployment relationship."^ The latter led to the conclusion that while there was a slight suggestion that the inflation-unemployment function has shifted outward slightly, there was no evidence that this shift was significant. There are three possible explanations for this difference in results. F i r s t of a l l , i t i s possible that the other variable that Vanderkamp introduced into his tests, the rate of change of import prices, has increased over time. . More plausible i s the possibility that the Table II Estimates of the Structural Unemployment Rate Structural Unemployment Rate Estimated on the Basis of Estimated Asymmetric Relationships .. Symmetric Relationships Unemploy- Vacancy Simple ^ Simple r . Rotated „ D / Period ment Rate Rate 2 Linear Hyperbolic Logarithmic Hyperbolic Parabola 53 I 2.8 3-4 3-2 3.1 3-1 3.1 3.1 II 2.6 3-2 3-0 2.9 2.9 2.9 2.9 III 3-0 3.0 3.0 3.0 3-0 3.0 3.0 IV 3-9 2.6 3.1 3-2 3.2 3.2 3-2 54 I 4.0 2.3 3-0 3.0 3-1 3.0 3.1 II 4.7 2.1 3-1 3.1 3.3 3.1 3.2 III 5.2 2.0 3.3 3.2 3.4 3.2 3.3 IV 5.0 2.2 3.3 3-3 3.4 3-3 3.4 55 I 4.8 2.2 3-2 3-2 3.4 3.2 3-3 II 4.4 2.4 3.2 3.2 3.3 3.2 3.3 III 4.1 2.9 3.4 3.5 3.5 3.4 3.5 IV 3.8 3-5 3.6 3.7 3.7 3.6 3.6 56 I 3.7 3.8 3.8 3.7 3.7 3.7 3.7 II 3-4 4.0 3.8 3.7 3.7 3.7 3.7 III 3.0 4.2 3-7 3.5 3.5 3-5 3.6 XV 3.1 4.1 3.7 3-5 ' 3-5 3.6 3.6 57 I 3-9 3.4 3.6 3.7 3.7 3.6 3.6 II 4.1 2.9 3.4 3.5 3.5 3.4 3-5 III 5.0 2.2 3.3 3.3 3.4 3-3 3.4 IV 6.0 1.8 3-5 3.3 3.5 3.3 3.3 VP ro Table I I - (Continued) S t r u c t u r a l Unemployment Rate Estimated on the B a s i s of Estimated Asymmetric R e l a t i o n s h i p s Symmetric R e l a t i o n s h i p s Unemploy- Vacancy Simple k Simple g Rotated , Pe r i o d ment Rate Rate2 Li n e a r Hyperbolic Logarithmic Hyperbolic Parabola 58 I 6.6 1.7 3.6 3.4 3.6 3-3 3.4-I I 7.2 1.8 3.9 4.0 3.8 3.6 3.6 I I I 7.4 1.6 3.9 3.7 3.7 3.4 3.4 IV 7.3 1.6 3.9 3.6 3.7 3.4 3.4 59 I 6.2 2.0 3.7 3-7 3.7 3-5 3.5 I I 5.9 2.1 3.6 3.7 3-7 3.5 3.5 I I I 5.6 2.1 3.5 3.5 3-6 3.4 3.5 IV 5.8 2.1 3.6 3.6 3.6 3.5 3-5 60 I 6.3 1.9 3.6 3.6 3.6 3.4 3.5 I I 6.9 2.0 3.9 4.1 3-9 3-7 3.7 I I I 7.4 1.8 4.0 4.1 3.9 3.6 3.6 IV 7.6 1.8 4.1 4.2 3.9 3.7 3.6 61 I 7.6 1.8 4.1 4.2 3.9 3.7 3.6 I I 7.5 1.8 4.1 4.1 3.9 3.7 3-6 I I I 7.0 2.1 4.0 4.4 4.0 3.8 3.8 IV 6.2 2.1 3.7 3-9 3.8 3.6 3-6 62 I 6.0 2.4 • 3.8 4.1 4.0 3.8 3.8 I I 5.7 2.6 3.8 4.2 4.0 3.8 3.8 I I I 5.8 2.7 3.9 4.3 4.1 4.0 3-9 IV 5.9 2.7 4.0 4.4 4.1 4.0 4.0 63 I 5-9 2.6 3.9 4.3 4.1 3-9 3.9 I I 5.6 2.7 3.8 4.2 4.0 3-9 3.9 I I I 5.4 2.9 3.9 4.2 4.1 4.0 4.0 IV 5.2. 3-1 3.9 4.3 4.1 4.0 4.0 Table II - (Concluded) Structural Unemployment Rate Estimated on the Basis of Estimated Asymmetric Relationships Symmetric Relationships Unemploy- Vacancy Simple ^ 5 Simple g Rotated , Period ment Rate Rate 2 Linear Hyperbolic Logarithmic Hyperbolic Parabola 64 I k.9 3-3 3.9 4.2 4.1 4.0 4.0 IT 4.8 3-1 3-8 4.0 3.9 3.9 3.9 III k.6 . 3.2 3.8 4.0 3-9 3.8 3.8 IV 4.5 3-4 3.8 4.0 4.0 3-9 3.9 65 I k.l 3.5 3.7 3.8 3.8 3.8 3.8 II 4,2 3.7 3-9 4.0 4.0 3.9 3.9 III 3-8 3-9 3.9 3.8 3.8 3.8 3.8 IV •3-5 4.1. 3-9 3.7 3.8 3.8 3.8 7 A l l data are seasonally adjusted. For estimation procedure see section 1(a) i n Chapter IV. ^ Based on equations (4.19) and (4.15).. ^ Based on equations (4.16) and (4.9). Based on equations (4.17) and (4.11). ^ Based on equations (4.6) and (4.7). n Based on equations (4.l8) and (4.13). Sources: (1) For employment, unemployment and labour force, seasonally adjusted: Canada, Dominion Bureau of Statistics, Seasonally Adjusted Labour Force Statistics, January 1953 - December 1966, Cat. No. 71-201, Queen's Printer, Ottawa. (2) For reported vacancies: Canada, Department of Labour, Labour Gazette, Queen's Printer, Ottawa. 95 r e l a t i o n s h i p between p r i c e i n f l a t i o n and the vacancy rate has changed over time. F i n a l l y , the vacancy statement r a t i o may have increased over time and l e d to a spurious r i s e i n the estimated s t r u c t u r a l unemployment rate. It i s possible that the l a t t e r explanation might apply to the increase i n the estimated s t r u c t u r a l unemployment rate from 1957 to 1962. Table I showed that the r a t i o of t o t a l h i r i n g s to N.E.S. placements declined during that period. This suggests that the proportion of the t o t a l vacancies that was reported to the N.E.S. increased and that f o r 1957-58 the vacancy rate was underestimated and for 1961-62 overestimated. Thus also the s t r u c t u r a l unemployment rate would then be underestimated i n 1957-58 and overestimated i n 1961-62. However, i f the r a t i o of h i r i n g s to N.E.S. place-ments i s r e a l l y i n d i c a t i v e of f l u c t u a t i o n s i n the vacancy statement r a t i o , the r i s e of the estimated s t r u c t u r a l unemployment rate between 1953 and 1956 would be an underestimate and there should have been a r i s e a f t e r 1962. In general, the s t r u c t u r a l unemployment rates for 1953 and I96I-65 would be overestimates, while those f o r 1954-60 would be underestimates. However, i t should be noted that the h i r i n g s - N.E.S. placement r a t i o i s more or l e s s the same for 1953 and 1963-65 while there i s a considerable difference i n the estimated s t r u c t u r a l unemployment rate for these two periods. 96 FOOTNOTES TO CHAPTER IV This decision subsequently created certain d i f f i c u l t i e s . See section c(i) below. This problem suggests that i t was a mistake to limit the period of analysis to the years since 1953. Unfortunately i t had not become apparent unt i l after the s t a t i s t i c a l work had been completed. Time limitations prevented a revision of the analysis. J. Vanderkamp, "An Application of Lipsey's Concept of Structural Unemployment", Review of Economic Studies, July 1966, pp. 221-225. CHAPTER V CONCLUSION The purpose of this thesis has been to develop a theoretical framework which could then be used to measure structural unemployment. In this f i n a l chapter the theory w i l l be briefly summarized, both the survey of the literature and the model developed in Chapter III, the problems and results of the empirical analysis recapitulated, the policy implications briefly explored and lines of further research suggested. 1. Summary of the Theory The survey chapter on the theory of structural unemployment began with describing the causal approach to the analysis of structural unemploy-ment. It was shown how structural unemployment i s explained in terms of changes in the structure of fi n a l demand and technological change and how the labour-displacement effect of the latter depended on the price and income el a s t i c i t i e s of demand of the affected industries. Since these determinants explain only the rate of structural displacement, the structural maladjustment approach, which stresses the role of structural adjustment and the factors inhibiting i t , was considered. The various kinds of substitution promoting the reabsorption of displaced labour as well as r i g i d i t i e s and lags working against i t were outlined. Dow and Dicks-Mireaux's vacancy model was then investigated and found to be of considerable value, but at the same time to be lacking in theoretical explanation. A scrutiny of the methods of determining the degree of structural imbalance by analyzing the structure of unemployment led to the 98 c o n c l u s i o n that there were unsolved problems i n the d i s t i n c t i o n between the s t r u c t u r e of demand-deficiency unemployment and that of s t r u c t u r a l unemploy-ment. Long-duration unemployment, which was used as an index of s t r u c t u r a l unemployment i n another k i n d of model, was found not to be d i r e c t l y i d e n t i f i a b l e w i t h s t r u c t u r a l unemployment. The p o l i c y approach i s concerned w i t h the r e l a t i o n s h i p of unemployment and i n f l a t i o n . I t c o n s i s t s of Lipsey's t r a d e - o f f model, according to which the degree of s t r u c t u r a l maladjustment i s i n d i c a t e d by the distance of the inflation-unemployment f u n c t i o n from the o r i g i n . However, there are problems i n v o l v e d i n u s i n g i t to measure s t r u c t u r a l unemployment because of the s u b j e c t i v e and p o t e n t i a l l y v a r y i n g policy-makers' preferences and the i m p u r i t i e s i n v o l v e d i n the inflation-unemployment r e l a t i o n s h i p . In s p i t e of data problems i n v o l v e d i n a model u s i n g the vacancy , v a r i a b l e , i t was considered worthwhile to develop a model which does not have t o depend on pr o x i e s f o r labour demand and s t r u c t u r a l imbalance, whose construct has a t h e o r e t i c a l r a t i o n a l e and which s a t i s f a c t o r i l y separates the e f f e c t s of aggregate demand and of s t r u c t u r a l imbalances on unemployment. I t c o n s i s t s of determining the c y c l i c a l r e l a t i o n s h i p between the unemployment and vacancy r a t e s and a t t r i b u t i n g changes which cannot be explained by t h i s r e l a t i o n s h i p to changes i n the l e v e l of s t r u c t u r a l imbalances. 2. The E m p i r i c a l A n a l y s i s Before t h i s model could be a p p l i e d to Canadian data, the vacancy r a t e had to be derived from N.E.S. vacancy data. On the b a s i s of a comparison of the economy's volume of h i r i n g s and the volume of N.E.S. placements, the r a t i o of a c t u a l vacancies to N.E.S. vacancies was estimated to be 4.7. This f a c t o r was used to estimate the economy's vacancy r a t e . 99 Several forms of the relationship between the unemployment and estimated vacancy rates were tested. On the basis of the biases resulting from the asymmetrical distribution of the empirical observations in functions permitting asymmetry, i t was decided that the best form was a parabola rotated by The results of the other forms were also presented and are not very different except in the period of very deficient demand. In Figure 5.1 the structural and deficient-demand components of the unemployment rate according to the rotated parabola estimate are plotted. While there has been some upward trend in the structural unemployment rate, with the biggest increases concentrated in 1955-56 and 1961-62, most of the changes i n the total unemployment rate are attributable to deficient-demand unemployment. These results suffer from the uncertainty involved in the estimated vacancy rate that was used, but the slight upward trend i n the structural unemployment rate was reinforced by a look at the trend in the ratio of hirings to N.E.S. placements. 3. Policy Implications The level of the structural unemployment rate has been estimated to be about ~5% per cent, rising from 3 per cent in the early 1950's to nearly h per cent i n the 1960's. It should be noted that this i s the unemployment rate that prevails when there Is neither excess nor deficient aggregate labour demand, that i s , when unemployment and vacancies are equal. It i s not necessarily the unemployment that prevails when there i s the optimum level of aggregate demand, which may be either excess or deficient. Since there i s relatively great concern with inflation on the part of Canadian governments, the policy-makers' preferred level has tended to be deficient rather than excess demand. Therefore, the level of unemployment which i s to be dealt with by selective manpower policies i s relatively high. 100 Figure 5-1 - Quarterly series of total, structural and demand-deficiency unemployment rates, based on the rotated parabola estimate 101 On the other hand, the increases in unemployment in 1953-54, 1957-58 and 1959-60 were almost entirely due to declines in aggregate demand and the responsibility for dealing with them lay on macroeconomic policy. Unfortunately i t s application in 1957-62 was contradictory. According to W.L. Winder, with a flexible exchange rate, f i s c a l expansion tends to raise employment but also to increase foreign competition, because interest rates rise and attract foreign capital, resulting i n an increase in the exchange rate (which makes imports cheaper and exports costlier). Both the direct and indirect effects of monetary expansion would tend to increase employment, given a flexible exchange rate. The main reliance in Canada up to 1962 should therefore have been placed on monetary policy whereas, on the contrary, expansionary f i s c a l policy was employed in a vain attempt to offset contradictionary / s i c / monetary policy.1 As a matter of fact, even f i s c a l policy at the time was expansionary only in a very hesitant way. 4. Further Research Two lines of further work in the area of structural unemployment would be particularly useful. The f i r s t one is to put the problem into a dynamic framework and to analyze and quantify the determinants of structural dislocations, on the one hand, and of structural adjustment, on the other. The second line i s to determine which kinds of selective manpower policies are most appropriate to assist the adjustment process. Both lines require a much more disaggregative approach than could be adopted in the empirical part of this thesis. FOOTNOTE TO CHAPTER V W.L. Winder, Canadian Labour Market, p. 172. BIBLIOGRAPHY Berman, Barbara R. "An Approach to an Absolute Measure of Structural Unemployment." Employment Policy and the Labor Market. Ed. Arthur M. Ross. University of California Press, 1965, pp. 256-268. Bowen, William G. and R. Albert Berry. "Unemployment Conditions and the Movements of the Money Wage Level." Review of Economics and  Statistics. Vol. 45, 1963, PP- 163-172. Canada, Department of Labour. Labour Gazette. Queen's Printer, Ottawa. Canada, Dominion Bureau of Statistics. Hiring and Separation Rates in  Certain Industries. Cat. No. 72-006, Queen's Printer, Ottawa. . Seasonally Adjusted Labour Force Statistics, January 1953-December 1966. Cat. No. 71-201, Queen's Printer, Ottawa. Canada, Senate, Special Committee on Manpower and Employment. Final Report. Queen's Printer, 1961. Casselman, Paul H. The Economics of Employment and Unemployment. Public Affairs Press, Washington, D.C, 1955. Denton, Frank T. and Sylvia Ostry. An Analysis of Postwar Unemployment. Economic Council of Canada, Staff Study No. 3» Queen's Printer, Ottawa, December 1964. Dow, J.C.R. and L.A. Dicks-Mireaux. "The Excess Demand for LabourJ A Study of Conditions in Great Britain, 1946-56." Oxford Economic  Papers. New Series, Vol. 10, No. 1, February 1958, pp. 1-33-. Eckstein, Otto. "Aggregate Demand and the Current Unemployment Problem." Unemployment and the American Economy. Ed. Arthur M. Roes. John Wiley & Sons, Inc., New York (1964), pp. 116 f f . Gallaway, Lowell E. "Labour Mobility, Resource Allocation, and Structural Unemployment." American Economic Review. Vol. 531 No. 4, September 1963, pp. 694-716. Gordon, Margaret S. "U.S. Manpower and Employment Policy." Monthly Labor  Review. U.S. Department of Labor. Vol. 87, No. 11, November 1964, . pp. 1314-1321. Gordon, R.A. "Has Structural Unemployment Worsened?" Industrial Relations:  A Journal of Economy and Society. Institute of Industrial Relations, . University of California, Berkeley. Vol. 3, No. 3> May 1964, PP. 53-78. 104 . "Full Employment as a Policy Goal." Employment Policy and, the  Labour Market. Ed. Arthur M. Ross. University of California Press, 1965, pp. 25-55. Heller, Walter W. "The Administration's Fiscal Policy." Unemployment  and the American Economy. Ed. Arthur M. Ross. John Wiley 8c Sons, Inc. New York (1964), pp. 93-115-Kaliski, S.F. "The Relation Between Unemployment and the Rate of Change of Money Wages in Canada." International Economic Review. Vol. 5> No. 1, January 1964, pp. 1-33• _. "Structural Unemployment in Canada: The Occupational Dimension." Paper prepared for the Annual Meeting of the Canadian Economics Association, Calgary, June 1968, mimeographed. Killingsworth, Charles C. "Automation, Jobs, and Manpower." The Nation's Manpower Revolution. Hearings before the Subcommittee on Labor and Public Welfare, U.S. Senate, 88th Congress, 1st session, part 5, 1963, pp. 1461-1483. Laidler, David and Bernard Corry. "The Phillips Relation: A Theoretical Explanation." Economica. New Series, Vol. 34, May 1967, pp. 189-197. . "The Phillips Relation: A Theoretical Explanation - A Reply." Economica. New Series, Vol. 35, May 1968, p. 184. Lipsey, Richard G. "The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1862-1957! A Further Analysis." Economica. New Series, Vol. 27, No. 105, February i960, pp. 1-31. . "Structural and Deficient-Demand Unemployment Reconsidered." Employment Policy and the Labor Market. Ed. Arthur M. Ross. University of California Press, 1965, pp. 210-255-Myrdal, Gunnar. Challenge to Affluence. Random House, New York (1962). Oi, Walter Y. "Labor as a Quasi-Fixed Factor." Journal of P o l i t i c a l  Economy. Vol. 70, No. 6, December 1962, pp. 538-555-Phillips, A.W. "The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957-" Economica. New Series, Vol. 25, No. 99, November 1958, pp. 283-299-Proulx, Pierre-Paul. "The Composition of Unemployment in Canada." Employment, Unemployment and Manpower. McGill University, Industrial Relations Centre, 15th Annual Conference, June 1964, pp. 36-54. . "Structural Unemployment and Public Policy." Draft of a paper read to the Canadian P o l i t i c a l Science Association Conference, Vancouver, 10 June 1965, mimeographed. Reder, M.W. "Wage Structure and Structural Unemployment." Review of Economic Studies. Vol. 31, No. 88, October 1964, pp. 309-322. Rees, Albert. ''Discussion: The Problem of Expanding Economic Activity." Unemployment and the American Economy. Ed. Arthur M. Ross. John Wiley & Sons, New York (1964), pp. .135-140. Reuber, Grant L. "The Objectives of Canadian Monetary Policy, 1946-61: Empirical 'Trade-Offs' and the Reaction Function of the Authorities." Journal of P o l i t i c a l Economy. Vol. 72, No. 2, April 1964, pp. 109-132. Reynolds, Lloyd G. Labor Economics and Labor Relations. Prentice-Hall, Inc., Englewood C l i f f s , N.J. (1964). Simler, N.J. "Long-Term Unemployment, the Structural Hypothesis, and Public Policy." American Economic Review. Vol. 54, No. 6, December 1964, pp. 985-IOOI. ~~ Solow, Robert M. The Nature and Sources of Unemployment in the United States" Wicksell Lectures 1964, Almgvist. & Wicksell, ' Uppsala (1964). Stromsdorfer, Ernst W. "Labor Force Adjustment to Structural Displacement in a Local Labor Market." Industrial and Labor Relations Review. Vol. 18, No. 2, January 1965, pp. 151-165. ' ~~ ' ~"~ Thurow, Lester C, "The Changing Structure of Unemployment: An Econometric Study." Review of Economics and Statistics. Vol* 47, No. 2, May 1965, pp. 137-149. ~~ ~ . United States, Congress, Joint Economic Committee, Subcommittee on Economic Statistics. Higher Unemployment Rates, 1957-60: Structural  Transformation or Inadequate Demand.87th Congress, 1st session, I 9 6 I . ' ~~ ' . Vanderkamp, John. "Wage and Price Level Determination:. An Empirical Model for Canada." Economica. New Series, Vol. 33» No. 130, , May 1966, pp. 194-21o\ _. ""An Application of Lipsey's Concept of Structural Unemployment." Review of Economic Studies. July 1966, pp. 221-225. • "The Phillips Relation: A Theoretical Explanation - A Comment." Economica. New Series, Vol. 35» May 1968, pp.'179-183• Wilcock, R.C. and W.H. Franke. Unwanted Workers. Collier-Macmillan, London, 1963. Winder, W.L. "Structural Unemployment." The Canadian Labour Market:  Readings in Manpower Economics. Eds. A.M. Kruger and N.M. Meltz. Centre for Industrial Relations, University of Toronto, 1968, pp. 135-220. Woods, H.D. and Sylvia Ostry.. Labour Policy and Labour Economics in'Canada. Macmillan of Canada, 1962. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0104159/manifest

Comment

Related Items