UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of competition in macro models Shaffer, Marvin 1974

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

Item Metadata

Download

Media
831-UBC_1974_A1 S53.pdf [ 7.55MB ]
Metadata
JSON: 831-1.0093527.json
JSON-LD: 831-1.0093527-ld.json
RDF/XML (Pretty): 831-1.0093527-rdf.xml
RDF/JSON: 831-1.0093527-rdf.json
Turtle: 831-1.0093527-turtle.txt
N-Triples: 831-1.0093527-rdf-ntriples.txt
Original Record: 831-1.0093527-source.json
Full Text
831-1.0093527-fulltext.txt
Citation
831-1.0093527.ris

Full Text

T H E R O L E O F C O M P E T I T I O N IN M A C R O M O D E L S by M A R V I N S H A F F E R B . A . , M c G i l l U n i v e r s i t y , 1970 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in the Department of E c o n o m i c s We accept this thesis as conforming to the r equ i red standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A June, 1974 "In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representative. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. " Marvin Shatter ABSTRACT In this study a macro model is outlined in which non-competitive as well as competitive product price behaviour can be in-corporated and the macro implications compared. A variety of non-competitive pricing models are con-sidered and on the basis of these (and in conjunction with a chapter link-ing the micro to the macro analysis), it is argued that non-competitive pricing is potentially important in a macro model. It can generate macro relative share, price, and output behaviour different from that generated by price - equals - marginal cost competitive behaviour. In the empirical investigations of Canadian manufactur-ing industries, evidence of non-competitive price behaviour is found. The pro-cyclical standard competitive behaviour of gross-margins (mark-ups on variable costs) is not generally observed. Combining the theoretical with the empirical results, it is concluded that non-competitive pricing is not only potentially import-ant, but of actual importance because of the empirical significance of non-competitive price behaviour. The importance of this for macro policy is briefly considered in the final chapter. T A B L E O F C O N T E N T S P a g e A B S T R A C T i i T A B L E O F C O N T E N T S i i i L I S T O F T A B L E S v L I S T O F F I G U R E S v i i L I S T O F G R A P H S v i i i C H A P T E R I - I N T R O D U C T I O N 1 O r g a n i z a t i o n and S u m m a r y of the A r g u m e n t 3 C H A P T E R II - T H E M A C R O M O D E L 7 T h e A g g r e g a t e Supp ly F u n c t i o n 8 T h e A g g r e g a t e D e m a n d F u n c t i o n 12 C H A P T E R III - T H E M A C R O G R O S S M A R G I N 30 C H A P T E R I V - M I C R O M A R G I N B E H A V I O U R 49 M i c r o G r o s s - M a r g i n B e h a v i o u r i n C l a s s i c a l C o m p e t i t i v e M a r k e t s 50 G r o s s - M a r g i n B e h a v i o u r i n N o n - c o m p e t i t i v e M a r k e t s 56 1. Sho r t R u n P r o f i t M a x i m i z a t i o n 57 Z. P r o f i t M a x i m i z a t i o n w i t h F e a r o f E n t r y 61 3. M a n a g e r i a l U t i l i t y M a x i m i z a t i o n 70 4 . B e h a v i o u r i s t C o n s i d e r a t i o n s 74 5. O l i g o p o l i s t i c C o n s i d e r a t i o n s 77 S u m m a r y 82 C H A P T E R V - P R E V I O U S E M P I R I C A L S T U D I E S 8 5 O n the C y c l i c a l B e h a v i o u r o f M a n u f a c t u r i n g G r o s s - M a r g i n s 86 O n C y c l i c a l N e t P r o f i t M a r g i n B e h a v i o u r 100 C H A P T E R V I - T H E E M P I R I C A L S T U D I E S : T H E D A T A A N D T H E I R R E L A T I O N T O T H E T H E O R Y 101 D a t a R e q u i r e m e n t s and S o u r c e s 101 (1) G r o s s - M a r g i n D a t a 101 (2) R e f e r e n c e C y c l e T u r n i n g P o i n t s 104 i i i T A B L E O F C O N T E N T S (Cont inued) P a g e (3) I n d u s t r y S p e c i f i c C y c l e s 106 (4) C o n c e n t r a t i o n D a t a 108 (5) N e t P r o f i t M a r g i n D a t a 109 R e l a t i n g the D a t a to T h e o r e t i c a l P r o p o s i t i o n s and C o n c e p t s I l l (1) T h e U s e of A n n u a l D a t a I l l (2) T h e U s e o f I n d u s t r y D a t a 112 (3) R e a l i z a t i o n o f E x p e c t a t i o n s and In ten t ions 112 C H A P T E R V I I - T H E E M P I R I C A L S T U D I E S : T H E T E S T S A N D R E S U L T S 115 O n the C y c l i c a l B e h a v i o u r of G r o s s M a r g i n s 115 O n the R o l e o f C o n c e n t r a t i o n 130 O n T a r g e t - R a t e P r i c i n g 137 S u m m a r y o f the E m p i r i c a l R e s u l t s 148 C H A P T E R VI I I - C O N C L U S I O N 150 B I B L I O G R A P H Y 156 A P P E N D I C E S : A P P E N D I X A - E L A S T I C I T Y O F D E M A N D A N D T A R G E T R A T E M A R G I N B E H A V I O U R 164 A P P E N D I X B - E M P L O Y M E N T M U L T I P L I E R A N D P R I C E E F F E C T O F A C H A N G E I N E X O G E N E O U S D E M A N D 167 i v LIST OF T A B L E S Ta'ble^. Page 3. 1 Macro Margin x Component Regressions (Y = a +• bX) 38 3. 2 Micro Margin Component x Other Component Regressions (Y = a +- bX) 39 7. 1 Industry Gross Margin Behaviour During Upswings and Downswings 118 7. 2 Industry Gross Margin Behaviour Over Entire Cycles 119 7.3 Probability of Table 7. 2 Results 121 7.4 Average Gross-Margin Behaviour 121 7. 5 Change in Macro Margin and Weighted Average of Changes in Micro Margins Over Reference Periods 123 7. 6 Annual Weighted Averages of Changes of Micro Margins 125 7.7 Group Average Changes of the Margins 128 7.8 Gross-Margin - Industry Specific Output and AVC Regression Results 128 7.9 The Correlation Tests 132 7. 10 The T-Tests 133 7.11 1948 S. I. C. National Market Tests 135 7. 12 Industry Specific Output Changes: Concentration Effect Regressions 138 7. 13 Industry Net Profit Margin Behaviour During Upswings and Downswings 141 v L I S T O F T A B L E S (Cont inued) T a b l e P a g e 7. 14 I n d u s t r y N e t P r o f i t M a r g i n B e h a v i o u r O v e r E n t i r e C y c l e s 142 7. 15 A v e r a g e o f C h a n g e s o f Ne t P r o f i t M a r g i n s . . . 14Z 7 . 1 6 C o n c e n t r a t i o n C o e f f i c i e n t s 147 7 .17 C a p i t a l - O u t p u t R a t i o C o e f f i c i e n t 147 v i L I S T O F F I G U R E S F i g u r e Page 2. 1 S t a n d a r d Supply C u r v e 9 2. 2 T r a n s f o r m e d Supply C u r v e 9 2.3 K e y n e s 1 Standard Supply C u r v e 10 2.4 K e y n e s ' T r a n s f o r m e d Supply C u r v e 10 2. 5 M a r g i n a l C o s t of C a p i t a l and M a r g i n a l E f f i c i e n c y of Investment Schedules 18 4. 1 A L i n e a r C o s t F u n c t i o n 54 4.2 The 'Sylos P o s t u l a t e 1 65 4. 3 T a r g e t P r o f i t s and P r o f i t C u r v e 65 v i i L I S T O F G R A P H S G r a p h Page 3.1 A n n u a l Changes 1948 S. I. C. (1954-59) 41 3.Z A n n u a l Changes 1948 S. I. C. (1954-59) 42 3.3 A n n u a l Changes 1948 S.I.C. (1954-59) 43 3.4 A n n u a l Changes 1948 S. I. C. (1954-59) 44 3.5 A n n u a l Changes I960 S. I. C. (1957-69) 45 3.6 A n n u a l Changes I960 S.I.C. (1957-69) 46 3.7 A n n u a l Changes I960 S. I. C. (1957-69) 47 3.8 A n n u a l Changes I960 S. I. C. (1957-69) 48 6. 1 M a n u f a c t u r i n g S e c t o r Output and A v e r a g e V a r i a b l e C o s t Indexes 107 v i i i C H A P T E R I I N T R O D U C T I O N In the m i d 1930's l a i s s e z - f a i r e e c o n o m i c s was d e a l t two s e v e r e b l o w s . On the one hand K e y n e s 1 t h e o r y of e f f e c t i v e demand gave a c a d e m i c r e s p e c t a b i l i t y to notions of unemployment e q u i l i b r i u m . ^  On the other hand Means' i d e n t i f i c a t i o n of a d m i n i s t e r e d p r i c e s s t i r r e d doubts about the f l e x i b i l i t y of p r i c e s and the c o - o r d i n a t i n g r o l e of the p r i c e s y s t e m i t s e l f . ^  T h e s e two i s s u e s have often been l i n k e d o r even equated w i t h one another. M a n y have m a i n t a i n e d that some s o r t of p r i c e o r wage i n f l e x i b i l i t y u n d e r l i e s K e y n e s ' t h e o r y of e f f e c t i v e demand. However, c a r e f u l r e a d i n g of The G e n e r a l T h e o r y p r o v i d e s ample evidence that th i s i s not the case. W i t h r e g a r d to wages, K e y n e s devoted an e n t i r e chapter 4 to d i s c u s s the i m p l i c a t i o n s of c o m p e t i t i v e wage f l e x i b i l i t y . W i t h r e g a r d to p r i c e s , K e y n e s e x p l i c i t l y adopted the c l a s s i c a l c o m p e t i t i v e a s s u m p t i o n 5 that p r i c e s equal m a r g i n a l c o s t s . K e y n e s was aware of the c o m p e t i t i v e a s s u m p t i o n s under-1 J.M. K e y n e s (1936). 2 G a r d i n e r M e ans (1935); i d e m (1939). 3 G a r d i n e r A c k l e y (1961, p. 417). 4 J.M. K e y n e s (1936, C h a p t e r 19). In thi s r e g a r d A x e l L e i j o n h u f v u d (1968, p. 37) has w r i t t e n ; "But thi s s t r o n g a s s u m p t i o n (wage r i g i d i t y ) i s not n e c e s s a r y i n o r d e r to e x p l a i n s y s t e m b e h a v i o u r of the Keyne-s i a n k i n d . It i s j u s t s u f f i c i e n t to give up the e q u a l l y s t r o n g a ssump-t i o n of instantaneous p r i c e a d j u s t m e n t s . " 5 J.M. Keynes (1936, p. 17). 2 lying his analysis. Ironically, although criticized for adopting price and wage inflexibility, Keynes himself was uneasy for not adopting such inflexibility. Means writes: "when I talked with him (Keynes) in the summer of 1939 he repeatedly referred to the unreality of the classical assumption of flexible prices which he had made in The General Theory. 1 Keynes was aware that his competitive assumptions were important with respect to the predicted behaviour of some macro variables. For example, in his article, "Relative Movements of Real Wages and Out-7 put", Keynes explicitly recognized that monopolistic price behaviour could be contributing to the procyclical real-wage behaviour observed by g Dunlop and by Tarshis. On the basis of the classical competitive assumption that prices equal marginal costs, Keynes had argued in The General Theory that counter-cyclical real-wage behaviour was likely to be observed. The broad aim of this study is to consider in detail the importance of this competitive assumption (prices equal marginal costs) in Keynes' theory of effective demand. More precisely, the aim is to consider in detail the implications and importance of introducing non-competitive, as opposed to competitive price assumptions in the theory of effective demand. 6 Gardiner Means (1959i p. 453). 7 J.M. Keynes (1939). 8 John Dunlop (1938); L. Tarshis (1938). 9 J.M. Keynes (1936, p. 10). 3 In accordance with this aim, at the theoretical level I present a macro framework in which non-competitive as well as compe-titive micro price behaviour can be incorporated and their macro im-plications compared. The specific question I consider is whether non-competitive price behaviour can generate macro behaviour different from that generated by competitive price behaviour. At the empirical level I examine Canadian manufacturing industries, over the period 1954-69, to determine whether a macro model incorporating non-competitive price behaviour better conforms to the data than a model assuming competitive behaviour. This investigation entails examining the micro and weighted-micro, i.e., macro, behaviour of these manufacturing industries. Organization and Summary of the Argument Keynes argued that even with competitive wage and price flexibility unemployment may emerge and remain. Means' point is that many prices are not flexible. In this thesis I argue that applied macro analysis requires a synthesis of Keynes and Means. Non-competitive (administered) price behaviour must be incorporated into macro analysis in order to understand and to predict macro phenomena. The argument is developed as follows. In Chapter II a macro model is outlined. A macro gross-margin is defined (overhead plus profits over variable costs) and it is through this variable that non-competitive price behaviour can enter the macro model. It is determined that ii_ the macro gross-margin behaves 4 differently with underlying non-competitive product markets (than with competitive ones), then different macro phenomena result. Relative shares, the price level, and the level of output behave differently. In. Chapter III the macro gross-margin is examined. The macro gross-margin is a weighted average of micro gross-margins. I argue that different micro gross-margin behaviour will be reflected in different macro margin behaviour. Because of definitional differences between the macro and micro gross-margins and because of general aggregation effects, this point is not as obvious as it may appear. How-ever, it is essential in providing the rationale for examining micro gross-margin behaviour. In Chapter IV micro gross-margin behaviour in compe-titive and non-competitive markets is examined. It is found that at the very least there is little reason to assume non-competitive margins will behave exactly like competitive margins. More important, it is argued that there are strong reasons for expecting a qualitative difference bet-ween their behaviour. While standard competitive models (standard in the sense of employing standard production functions) predict pro-cyclical gross-margin behaviour, non-competitive models in many cases predict counter-cyclical behaviour. Chapters II, III and IV constitute the theoretical part of the thesis. Going backwards through the argument they state: (1) Non-competitive micro gross-margins do not behave exactly the same as and perhaps do not even change in 5 the same direction as standard competitive micro gross-margins. (2) Therefore, a macro gross-margin based on non-compe-titive micro margins would behave differently from a macro gross-margin based on competitive micro margins. (3) Therefore, a macro model with underlying non-compe-titive markets would behave differently from a macro model with underlying competitive markets. In sum, in the theoretical chapters I argue that the competitive assumption is important. It is not, in the econometric jargon, a robust assumption. In the empirical chapters three questions are examined. F i r s t and foremost, do micro and weighted-micro margins of Canadian manufacturing industries behave in an unambiguously non-competitive fashion? That is, do they tend to behave counter-cyclically? It is found that this is the case. This is a very important result. Not only is non-competitive behaviour hypothetically important in a macro model, it is of actual importance because it is more consistent with observed margin behaviour. The second and third empirical questions arise from attempts to explain the first result. I examine two hypotheses: first, that the counter-cyclical behaviour is related to concentration, and second, that the counter-cyclical behaviour is caused by annual target-6 rate-of-profits price policies. The results here are negative. The non-competitive behaviour is not a product of concentration. This I believe is due to the fact that the administration of prices does not require mono-poly in the anti-trust sense. In addition, the counter-cyclical behaviour is not caused by an annual target-rate model. A different pricing model must be at work. The empirical work is presented in Chapters V, VI, and VII. In Chapter V, previous studies related to the three empirical quest-ions are examined. In Chapter VI the data requirements and sources, and the relation of the data to the theoretical concepts are discussed. In Chapter VII the tests and their results are presented. In the final chapter the implications for macro policy are considered. 7 C H A P T E R II T H E M A C R O M O D E L The m a c r o m o d e l I develop i s a s h o r t r u n m o d e l of e f f e c t i v e demand. Its purpose i s to dem o n s t r a t e the r o l e of c o m p e t i t i o n i n a m a c r o m o d e l - to d e m o n s t r a t e how n o n - c o m p e t i t i v e as opposed to c o m p e t i t i v e p r i c e b e h a v i o u r w o u l d enter the m a c r o f r a m e w o r k and would a f f e c t m a c r o phenomena. The m o d e l does not p r e s e n t a c o m p r e h e n s i v e v i e w of how a c o m p l e x economy a c t u a l l y w o r k s . ^ O n l y s u f f i c i e n t d e t a i l to draw out a s p e c t s of the r o l e of c o m p e t i t i o n i s o f f e r e d . N o n e t h e l e s s , I b e l i e v e that i m p o r t a n t i m p l i c a t i o n s a r e to be d e r i v e d f r o m t h i s m o d e l about the r e a l w o r k i n g s of a c o m p l e x economy and would have to be i n -c o r p o r a t e d into any m o r e complete and s o p h i s t i c a t e d m o d e l . The m o d e l i s a g e n e r a l i z a t i o n of a s i m p l i f i e d v e r s i o n of K e y n e s ' a n a l y s i s . It i s i n f a c t v e r y s i m i l a r to those m o d e l s c u r r e n t l y l a b e l l e d N e o - K e y n e s i a n . These i n c l u d e m o d e l s by K a l e c k i , J o a n R o b i n s o n , and s e v e r a l o t h e r s i n the C a m b r i d g e (England) t r a d i t i o n . ^  C o m m o n to these m o d e l s and to the one I develop i s the c r i t i c a l r o l e of the g r o s s - m a r g i n ( K a l e c k i ' s degree of monopoly)-^ i n the The m o s t i m p o r t a n t o m i s s i o n w i t h r e g a r d to the C a n a d i a n economy i s the r o l e of i n t e r n a t i o n a l t r a d e and c a p i t a l f l o w s . It s h o u l d a l s o be ) noted that there i s no e x p l i c i t t r e a t m e n t of the m o n e t a r y s e c t o r . 2 M. K a l e c k i (1964); i d e m (1930); Joan R o b i n s o n (1962); N. K a l d o r (1959); D.J. H a r r i s (1972); G. C. H a r c o u r t (1972, pp. 210-14); S. W e i n t r a u b (1969); M. M o o r e (1967). F o r a s u c c i n c t e x p o s i t i o n of Robinson's m o d e l , see A. A s i m a k o p o l o u s (1969); i d e m (1970). It s h o u l d be noted that w h i l e s i m i l a r to these m o d e l s , my m o d e l d i f f e r s i n that i t i s d e v e l o p e d s p e c i f i c a l l y to c o m p a r e h y p o t h e t i c a l c o m p e t i t i v e and non-c o m p e t i t i v e e c o n o m i e s , not to a s s e r t what i s . 8 determination of the short period equilibrium price level, level of out-put, and distribution between wage and non-wage income. In my analysis the gross-margin is the key variable through which non-compe«-titive price behaviour enters and affects the macro system. The essence of any model of effective demand is that the levels of employment and output are determined by the intersection of the aggregate demand and aggregate supply functions. This intersect-ion corresponds to a level of employment that generates an expected ex-penditure (aggregate demand) equal to an amount which is just sufficient to induce the entrepreneurs to maintain that level of employment (aggre-gate supply). The Aggregate Supply Function The aggregate supply function relates the employment of N men to the expected proceeds net of user costs "which will just make it worth the while of the entrepreneurs to give that employment".^ It is the aggregation of supply curves which have been transformed to relate 3 Kalecki's degree of monopoly is defined as the ratio of value of output over variable costs (wages plus material costs). Because in Keynes' analysis material costs are treated implicitly, in my generalization of his analysis I define the macro gross-margin slightly differently as the ratio of value added minus wages to wages ( W * k ) # Although this does necessitate a careful treatment of w. L the relation between the macro and micro gross-margins (see Chapter 3), no substantive issue is involved here. 4 J.M. Keynes (1936, p. 41). 9 revenues to employment (Figure 2.2) offered instead of price to quan-tity produced (Figure 2. 1). Pi Pi L Figure 2. 1 Standard Supply Curve (for a given price re-lates how much quan-tity produced) Figure 2.2 Transformed Supply Curve (for given reve-nues relates how much quantity produced and therefore labour hired) Keynes 1 aggregate supply function was constructed on the basis of competitive markets. In terms of a standard supply curve he assumed that output would be produced up to the point where marginal costs would equal the expected price received (Figure 2. 3). In terms of his transformed supply curve he assumed that labour would be hired up to the point where the output produced by that labour multiplied by marginal costs at that level of output would equal the expected revenues received (Figure 2.4). Thus in Keynes 1 analysis aggregate supply at a given level of employment (i. e. , the expected proceeds net of user costs 'which will just make it worth the while of the entrepreneurs to give that 10 Figure 2.3 Keynes' Standard Figure 2.4 Keynes' Transformed Supply  Supply Curve Curve employment') equals the output produced by that level of employment multiplied by the marginal costs at that level of output. Being an aggre-gate function the output and marginal costs variables are weighted averages, the weights given by the distribution of employment among industries. In my model the aggregate supply function is stated more generally in order to allow for non-competitive behaviour. This general-ized aggregate supply function is based on a value of output curve sug-gested by Asimakopolous in an unpublished paper, "Aggregate Supply and Aggregate Demand Curves! 1. Aggregate supply at a given level of employment is simply defined as the output produced by that level of 5 employment multiplied by the supply price at that level of output. The 5 By supply price I mean value added per unit. Because aggregate supply is defined net of user costs, my supply price concept must be net of unit material costs. 11 supply price equals one plus the gross-margin times unit wage costs at that level of employment and output. 6 S(L) = p«q with p = (1 + m) w q/L where: S(L') - aggregate supply at a given level of employment p - supply price at that level of employment q - output produced by that level of employment m - gross (unit overhead plus profit)-margin q/L - unit (variable) wage costs This statement of the aggregate supply function is valid with competitive or non-competitive markets. With competitive markets micro gross-margins are those which are obtained when price equals marginal cost. They are determined by technical conditions. With non-competitive markets micro gross-margins are those which are decided upon by the price makers. They are determined by policy. In either of these cases (or in a mixed economy) the macro gross-margin is simply the weighted average of these micro-determined micro margins (see Chapter III). The macro gross-margin, given the output per function ^ Asimakopolous 1 aggregate supply concept was less generally defined with price equals to one plus a fixed mapfe-up all over constant marginal costs. 12 (technical conditions) and money wages, determines aggregate supply. Thus, this aggregate supply function simply states that to hire a given amount of labour and produce the corresponding output expected revenues have to be of such an amount that expected gross-margins are: (a) in competitive markets, the gross-margins which are obtained when price equals marginal costs, or (b) in non-competitive markets, the gross-margins which are set by policy. As a weighted average of micro-determined micro gross-margins, the macro gross-margin can arise from non-competitive as well as competitive markets. The important point in this formulation is that through the macro gross-margin, non-competitive markets can enter and affect the macro system. To the extent underlying non-com-petitive markets cause the macro margin to behave differently (this issue is the subject of Chapters III and IV), aggregate supply w i l l behave differently. It w i l l be seen in the following discussion that aggregate demand w i l l behave differently as well. The Aggregate Demand Function The aggregate demand function "relates the employment of N men to the 'proceeds which entrepreneurs expect to receive from the 7 employment of N men'." In a closed private economy i t is a function A. Asimakopolous (1972, p. 5). 13 of those v a r i a b l e s w h i c h d e t e r m i n e the l e v e l s of c o n s u m p t i o n and i n v e s t -ment at any given l e v e l of employment and in c o m e . In N e o - K e y n e s i a n f a s h i o n , I assume that the l e v e l of con-sumption depends on the d i s t r i b u t i o n of income between wage and non-wage i n c o m e as w e l l as the l e v e l of income. I assume d i f f e r e n t s a v i n g p r o p e n s i t i e s out of wage and non-wage in c o m e . E v i d e n c e of d i f f e r e n t s a v i n g p r o p e n s i t i e s was found by o B u r m e i s t e r and Taubman i n a r e c e n t study of U.S. data. E v e n i g n o r i n g the c o r p o r a t e c a s h f l o w component of non-wage i n c o m e ( r e t a i n e d e a r n -i n g s p l u s d e p r e c i a t i o n ) , they c o n s i s t e n t l y found the sa v i n g r a t i o out of non-wage income to be double that out of wage income (using a v a r i e t y of d i f f e r e n t d e f i n i t i o n s of wage and non-wage income). A s they noted, had c o r p o r a t e c a s h f l o w been i n c l u d e d i n the study the d i f f e r e n c e between the s a v i n g r a t i o s would have been m u c h g r e a t e r . B y d e f i n i t i o n , a l l c a s h fl o w i s saved. A s a r g u e d below, there a r e s t r o n g r e a s o n s f o r both the sa v i n g p r o p e n s i t y out of wage i n c o m e and out of non-wage i n c o m e to v a r y p r o -c y c l i c a l l y . T h i s c o n t r i b u t e s to a p r o - c y c l i c a l v a r i a t i o n of the o v e r a l l p r o p e n s i t y to save independent of the c y c l i c a l v a r i a t i o n of r e l a t i v e s h a r e s . T h e r e f o r e , I assume that the c y c l i c a l b e h a v i o u r of i n c o m e i s another d e t e r m i n a n t of the l e v e l of c o n s u m p t i o n . E i t h e r a permanent income or an i r r e v e r s i b l e c o n s u m p t i o n p a t t e r n h y p o t h e s i s w o u l d p r e d i c t a p r o - c y c l i c a l s a v i n g p r o p e n s i t y out of 8 E d w i n B u r m e i s t e r and P a u l T aubman (1969). 1 4 wage inc o m e . F e a r of the t r a n s i t o r y n a t u r e of changes i n i n c o m e o r the i n e r t i a of past c o n s u m p t i o n habits causes a l a g g e d adjustment of r e a l con-s u m p t i o n to r e a l i n c o m e . The p ermanent income o r i r r e v e r s i b l e c o n s u m p t i o n h y p o t h e s i s c o u l d a l s o apply to n o n - c o r p o r a t e s a v i n g out of non-wage in c o m e . However, m o r e i m p o r t a n t r e g a r d i n g the s a v i n g p r o p e n s i t y out of non-wage i n c o m e i s the c y c l i c a l b e h a v i o u r of the components of non-wage in c o m e . U n l i k e wage income, non-wage i n c o m e i s not even a r e l a t i v e l y homogeneous ca t e g o r y . It i n c l u d e s o v e r h e a d s a l a r i e s , r e n t i e r i n c o m e ( r e n t s , i n t e r e s t , and d i v i d e n d s ) , u n i n c o r p o r a t e d b u s i n e s s i n c o m e , and c o r p o r a t e c a s h f l o w ( r e t a i n e d e a r n i n g s plus d e p r e c i a t i o n ) . B e c a u s e the saving p r o p e n s i t i e s out of these components a r e v e r y d i f f e r e n t , as the components' s h a r e s of non-wage income change, the o v e r a l l p r o p e n s i t y to save out of non-wage income w i l l change. M o s t i m p o r t a n t i n thi s r e g a r d i s the sh a r e of c o r p o r a t e c a s h flo w . A l l r e t a i n e d e a r n i n g s and d e p r e c i a t i o n a r e saved w h i l e o n l y a f r a c -t i o n of the other components a r e saved. A s h i f t i n f a v o u r of c o r p o r a t e c a s h f l o w w i l l r a i s e the p r o p e n s i t y to save out of non-wage income. It has been found that c o r p o r a t e c a s h flow's s h a r e of non-wage income o v a r i e s p r o - c y c l i c a l l y , thereby, c o n t r i b u t i n g to a p r o - c y c l i c a l v a r i a t i o n of the p r o p e n s i t y to save out of non-wage in c o m e . The r e l a t i v e l y c o u n t e r -c y c l i c a l b e h a v i o u r of r e n t i e r income's s h a r e i s due to the lagged a d j u s t -ment of d i v i d e n d s to c a s h flow as w e l l as the o v e r h e a d n a t u r e of i n t e r e s t 9 M i c h a e l E v a n s (1969, p. 288). 15 and r e n t . ^  The s h a r e of o v e r h e a d s a l a r i e s v a r i e s c o u n t e r - c y c l i c a l l y s i m p l y because of i t s o v e r h e a d n a t u r e . In sum, as w e l l as on the l e v e l of inco m e , the l e v e l of con-s u m p t i o n depends on the d i s t r i b u t i o n between wage and non-wage i n c o m e and on the c y c l i c a l b e h a v i o u r of inc o m e . A g r e a t e r s h a r e of non-wage income r e d u c e s consumption. A l s o , independently of the c y c l i c a l b e h a v i o u r of the d i s t r i b u t i o n between wage and non-wage i n c o m e , the p r o p e n s i t y to consume tends to f a l l i n upswings and r i s e i n downswings. The d i s t r i b u t i o n between wage and non-wage income can be r e p r e s e n t e d by the g r o s s - m a r g i n v a r i a b l e u s e d i n the supply p r i c e equation. The g r o s s - m a r g i n i n fact equals the r a t i o of non-wage to wage in c o m e . w p -m = q/L where: w q/L p» q - w Li w* L m - g r o s s - m a r g i n p - p r i c e (value added p e r unit) w unit wage c o s t s q/L w L - wage i n c o m e p- q - w L - non-wage income 10 Ibi d . , (p. Z83-6). On the lagged adjustment of d i v i d e n d s , a l s o see John L i n t n e r (1956). 16 For purposes of this exposition, the cyclical behaviour of income can be represented by the ratio of current to trend income. Thus, the level of consumption can be written as a function of the level of income, of the gross-margin, and of the ratio of current to trend income, negatively related to the last two variables.''"''" w h e r e : C Y m Y / Y t r = C ( Y , m, Y / Y t r ) C l > 0 C2 < 0 C 3 < o - level of consumption - level of income - gross margin - ratio of current to trend income 11 S t r i c t l y , the consumption function i s developed algebraically as follows: C w + C nw B u t A n d = c w . w L + c n w . ( Y - w L ) c w = c w ( Y / Y t r ) cnw = c n w ( Y / Y t r ) wL = 1 • Y 1 + m Y - w L = m 1 + m (the level of consumption equals the sum of consumption from wage income plus consumption from non-wage income ) . (equals the sum of the average propensity to consume out of wage income times the level of wage income plus the average propensity to consume out of non-wage income times the level of non-wage income) . (Both average propensities are functions of the ratio of current to trend income ) (wage income equals one over one plus the gross-margin times income; non-wage income equals the gross-margin over one plus the gross-margin times income) 17 Unlike consumption, investment is not a well-explained 12 variable. Its susceptibility to the 'state of animal s p i r i t s ' has not and perhaps cannot be captured well. Nevertheless, simple consid-erations in the theory of investment outlined below suggest that an endogeneous component of investment may be explained by the same variables which determine consumption. It should be noted that no attempt is made here to consider a l l possible determinants of investment. A large part of investment remains exogeneous in this model. However, the role of the gross-margin and of the ratio of current to trend income is considered so that the impact of these variables on aggregate demand is not distorted by considering them as determinants of consumption and ignoring their simultaneous effect on investment. 11 12 C = [ c w ( Y / Y t r ) . + c n w (Y/Ytr) ^ ] • Y Cl= "ff = C w ( Y / Y t r ) - ^ + c n w (Y/Ytr) ^ C 2 = Trrl =[aW 2 ( cnw (Y/Ytr) - c w (Y/Ytr) )} • Y < 0 w i t h c n w < c w 0 w i t h c w ' ( Y / Y t r ) < 0; c n w ' (Y/Ytr)< 0. J.M. Keynes (1936, pp. 161-2). "Most, probably, of our decisions to do something positive, the f u l l consequences of which w i l l be drawn out over many days to come, can only be taken as a result of animal spi r i t s a spontaneous urge to action rather than inaction, and not as the out-come of a weighted average of quantitative benefits multiplied by quantitative probabilities .... Thus, i f animal spirits are dimmed and the spontaneous optimism falters, leaving us to depend on nothing but a mathematical expectation, enterprise w i l l fade and die; - though fears of loss may have a basis no more reasonable than hopes of profits had before." 18 It is generally held that the level of investment in a given period is determined by the intersection of the marginal efficiency of investment and marginal cost of capital schedules. 1^ The marginal efficiency of investment schedule reflects the discount rate required to equalize the expected stream of profits from the investment to the cost of the investment. The marginal cost of capital schedule relates the marginal cost of capital to the amount of funds required. This schedule (a supply of capital to the time schedule) is composed of three segments: the cost from cash flow (an opportunity cost), the cost from bond issues (the long run rate of interest plus imputed costs regarding earnings per share), and the cost from new equity issues (the imputed cost of re-ducing earnings per share). ^ % YIELD M6I MCC A - cash flow B - bond issues C - new equity issues 'Figure 2. 5 Marginal Cost of Capital and Marginal  Efficiency of Investment Schedules 13 14 Clearly, the level of investment will be greater the higher the For a detailed treatment see J. Duesenberry (1958, Ch. 3-4). The perfectly competitive capital markets underlying the analysis of Modigliani and Miller (1958) is not assumed here. 19 marginal efficiency of investment curve and/or the lower the marginal cost of capital curve. Determining the position of the marginal efficiency of investment curve is the level of income, the extent of capacity utiliza-tion, the average age of existing equipment, the price of capital goods relative to consumer goods as well as expectations concerning future income and prices. Determining the position of the marginal cost of capital schedule is the extent of cash flow, the long run rate of interest and various financial constraint variables (leverage, etc. ) reflecting imputed costs of bond and new equity issues. The role of the gross-margin in determining investment is 1 5 through its effect on cash flows. A higher margin, cet. par. , entails greater cash flows, lower cost of capital, and therefore greater invest-ment. Thus, a higher margin, while reducing the consumption, tends to stimulate investment. This margin-investment relation reduces the impact a higher margin has on aggregate demand. However, I assume that it does not eliminate or reverse it. In other words, I assume that the effect the gross-margin has on consumption outweighs the effect on investment. This assumption (certainly universal in Neo-Keynesian literature) rests on empirical estimates of the elasticity of investment with respect to cash flows. After a summary of various studies Evans reports that this elas-~15 Income and the ratio of current to trend income are held constant. It should be noted that the comparative statics refer to the implica-tion of a higher margin (comparing two economies), not to the impli-cations of a rising margin (in one given economy). 20 t i c i t y i s i n the range of 1/4 to 1/2. 1 D S i m p l e c a l c u l a t i o n s show on the other hand that the e l a s t i c i t y of s a v i n g w i t h r e s p e c t to c a s h f l o w s i s g r e a t e r . ^ M o r e o v e r , u n l i k e the i n v e s t m e n t e l a s t i c i t y , the e l a s t i c i t y of s a v i n g w i t h r e s p e c t to c a s h flows u n d e r e s t i m a t e s the f u l l e f f e c t a h i g h e r m a r g i n has on s a v i n g . W h i l e t h i s e l a s t i c i t y c a p t u r e s the ef f e c t of a s h i f t f r o m wages to c a s h flow, i t i g n o r e s the g r e a t e r s a v i n g c a u s e d by a s h i f t f r o m wages to r e n t i e r i n c o m e . The r o l e of the r a t i o of c u r r e n t to t r e n d i ncome i n d e t e r m i n -i n g i n v e s t m e n t i s c o mplex. The movement of t h i s v a r i a b l e s e r v e s as a p r o x y f o r the movement of s e v e r a l other v a r i a b l e s w h i c h a f f e c t i n v e s t -ment. F o r example, given the l a g of c a p i t a l f o r m a t i o n to inc o m e , as in c o m e r i s e s r e l a t i v e to tr e n d , c a p a c i t y u t i l i z a t i o n r i s e s , t h e r e b y s t i m u -l a t i n g i n v e s t m e n t . A l s o , as d i s c u s s e d e a r l i e r , as income r i s e s r e l a t i v e to t r e n d , the share of c a s h f l o w s out of non-wage i n c o m e r i s e s . T h i s a l s o s t i m u l a t e s i n v e s t m e n t . F o r the m o s t p a r t , the r a t i o of c u r r e n t to t r e n d income i s p o s i t i v e l y r e l a t e d to i n v e s t m e n t . L i k e the g r o s s - m a r g i n , M i c h a e l E v a n s (1969, p. 138). The r e s u l t s E v a n s r e p o r t s a r e r e p r e -s e n t a t i v e of s e v e r a l s t u d i e s . T hey a r e not a l l b a s e d on the same con-c e p t u a l f r a m e w o r k . The e l a s t i c i t y of s a v i n g w i t h r e s p e c t to c a s h f l o w equals the i n c r e a s e i n s a v i n g c a u s e d by a one unit i n c r e a s e of c a s h flow at the expense of wage i n c o m e (since i t i s the e l a s t i c i t y a s s o c i a t e d w i t h a s h i f t f r o m wage to non-wage i n c o m e that I am i n t e r e s t e d in) m u l t i p l i e d by the s h a r e of c a s h fl o w out of t o t a l p r i v a t e s a v i n g (e = .AiL _c_ \ W i t h the saving A c s ' & p r o p e n s i t y out of c a s h f l o w equal to one and the m a r g i n a l p r o p e n s i t y to save out of wage income c e r t a i n l y no g r e a t e r than .2 (see B u r -m e i s t e r and Taubman, footnote 7), a one unit i n c r e a s e of c a s h f l o w at the expense of wages w i l l e n t a i l an i n c r e a s e of s a v i n g of at l e a s t .8. W i t h the r a t i o of c o r p o r a t e c a s h f l o w to g r o s s p r i v a t e s a v i n g w e l l o v e r .7 (the a v e r a g e of the 1969-72 r a t i o s was .8057, c a l c u l a t e d f r o m T a b l e 21 the effect of t h i s v a r i a b l e on i n v e s t m e n t i s opposite to i t s effect on con-sumption. W h i l e the i n v e s t m e n t e f f e c t may dominate the c o n s u m p t i o n effect d u r i n g c e r t a i n phases of the b u s i n e s s c y c l e , I assume that t h i s i s not the case w e l l into booms o r s l u m p s . I a s s u m e that w e l l into a boom, f o r example, the m a r g i n a l e f f i c i e n c y of i n v e s t m e n t stops r i s i n g w i t h r i s i n g i n c o m e r e l a t i v e to t r e n d . A s i n v e s t m e n t p r o j e c t s i n i t i a t e d e a r l y i n the u p t u r n a r e completed, c a p a c i t y u t i l i z a t i o n stops r i s i n g . A l s o , the f u r t h e r into a boom the g r e a t e r a r e e x p e c t a t i o n s that a downturn i s a p p r o a c h i n g . The opposite holds w e l l into a s l u m p as c a p i t a l i s u s e d up and e x p e c t a t i o n s of an upswing grow. T h i s a s s u m p t i o n states that w e l l into booms r i s i n g i n c o m e r e l a t i v e to t r e n d has a negative e f f e c t on aggregate demand (a p o s i t i v e e f f e c t w e l l into s l u m p s ) . The purpose of th i s a s s u m p t i o n i s to p r o v i d e some endogeneous s t a b i l i t y to the m o d e l . A s d i s c u s s e d l a t e r , without s u c h s t a b i l i t y exogeneous f o r c e s c o u l d be r e q u i r e d to p r e v e n t e x p l o s i v e g rowth or d e c l i n e . It should be noted that th i s t r e a t m e n t i s adopted f r o m 18 a m o d e l of the trade c y c l e suggested by K a l d o r . To s u m m a r i z e t h i s d i s c u s s i o n of aggregate demand, both the l e v e l of c o n s u m p t i o n and i n v e s t m e n t a r e f u n c t i o n s of the g r o s s - m a r g i n and the r a t i o of c u r r e n t to t r e n d income, as w e l l as the l e v e l of i n c o m e . 17 ' '. " 8 of N a t i o n a l Income and E x p e n d i t u r e A c c o u n t s , S t a t i s t i c s Canada 13-001, F o u r t h Q u a r t e r , 1972), the e l a s t i c i t y of s a v i n g w i t h r e s p e c t to c a s h flows must be w e l l o v e r . 8 x . 7 = . 56. 18 N. K a l d o r (I960). 22 It is assumed that the consumption effect of the gross-margin dominates the investment effect and, therefore, a higher margin reduces aggregate demand. It is also assumed that the consumption effect of a higher ratio of current to trend income dominates the investment effect well into booms or slumps and, therefore, at these times a higher ratio of current to trend income reduces aggregate demand. D = C + I + I = C (Y, m, Y/Ytr) + I(Y, m, Y/Ytr) + I = f (Y, m, Y/Ytr, I) . _ 5 f 19 f l - 3 Y > 0 f 3 = 3 (Y/Ytr) " J (Y/Ytr) + (Y/Ytr) ^ 0 w e l 1 ° ' ° v booms or slumps f , = > 0 where: D - level of aggregate demand C - level of consumption I - level of endogeneous investment I - level of exogeneous investment The standard Keynesian s t a b i l i t y condition, f j < 1, is assumed. 23 As stated at the outset, aggregate supply and aggregate demand form the essence of the model. A l l that remains i s the s p e c i f i -cation of units i n which values are measured. Following Keynes I use wage un i t s . In other words, values are deflated by the movement of the l e v e l of money wages. I assume that the only role of money wages i s to e s t a b l i s h the money value of a l l the v a r i a b l e s . In other words, I assume that the movement of money wages does not a f f e c t the determinants of aggregate demand or aggregate supply measured i n wage units and , therefore, does not a f f e c t the l e v e l of employment, of output, and of p r i c e measured i n wage u n i t s . In defence of t h i s treatment of money wages, a treatment suggested by Weintraub i n C l a s s i c a l Keynesianism, Monetary Theory and  the Price Level, I argue as follows. A c a r e f u l analysis (such as Keynes' chapter nineteen of The  General Theory) would d i s c l o s e many e f f e c t s that changing money wages could have on aggregate demand and supply measured i n wage un i t s . I f 24 the change was r e l a t i v e to a change i n the m o n e y s u p p l y the re c o u l d be an i n t e r e s t r a t e and t h e r e f o r e an i n v e s t m e n t e f fec t . T h e r e c o u l d be d e b t o r - c r e d i t o r d i s t r i b u t i o n a l e f fec ts w h i c h c o u l d ac t on i n v e s t m e n t and c o n s u m p t i o n . T h e r e c o u l d be e x p e c t a t i o n a l ef fec ts r e g a r d i n g the m o v e -m e n t o f i n t e r e s t r a t e s , p r i c e s and wages w h i c h c o u l d ac t on i n v e s t m e n t and c o n s u m p t i o n . In m y m o d e l t he r e c o u l d be ef fec ts on the m a r g i n s set b y p r i c e m a k e r s and , t h e r e f o r e , on agg rega t e s u p p l y (see p . 67 b e l o w ) . H o w e v e r , these effects do not a l l w o r k i n the s ame d i r e c t i o n and i n t u i t i o n cannot l e a d to an a s s u m p t i o n r e g a r d i n g the m o s t p r o b a b l e net e f fec t . In o t h e r w o r d s , to a s s u m e no net ' r e a l ' effect i s no m o r e i n c o r r e c t than to a s s u m e an effect one w a y o r the o t h e r . M o r e i m p o r t a n t , th i s a s s u m p t i o n i s not c r i t i c a l to m y a n a l y s i s . I focus on the b e h a v i o u r of c e r t a i n m a c r o v a r i a b l e s o v e r a b u s i n e s s c y c l e . It i s a s s u m e d that th i s b u s i n e s s c y c l e i s g e n e r a t e d b y s h o c k s of exogeneous i n v e s t m e n t as m o d i f i e d b y endogeneous c o n s u m p t i o n and i n v e s t m e n t . W h e t h e r the c o u r s e o f the b u s i n e s s c y c l e i s a l s o s l i g h t l y m o d i f i e d b y the ef fec ts o f a g e n e r a l m o v e m e n t i n the wage l e v e l i s e s s e n t i a l l y i r r e l e v a n t . W i t h the above d i s c u s s i o n i n m i n d , the equa t ions of the m a c r o m o d e l c a n now be l i s t e d . (1) A g g r e g a t e s u p p l y m e a s u r e d i n wage un i t s equa l s p r i c e m e a s u r e d i n wage u n i t s t i m e s output . Sw = P w - q (2) P r i c e m e a s u r e d i n wage un i t s equa l s one p l u s the g r o s s - m a r g i n t i m e s un i t wage c o s t s d i v i d e d b y m o n e y w a g e s , equa l s one p l u s the g r o s s -25 margin divided by output per man (1+m) w Pw = q/L = 1+m q/L (3) Output per man is a function of the number employed. a = g(D (4) Aggregate demand measured in wage units i s a function of income measured in wage units, of the margin, of the ratio of current to trend income, and of exogeneous investment (measured in wage units). f ! > 0 f 2 < 0 D w = f (Y w, m, Y/Ytr • I) (well into < 0 booms or slumps) f 4 y o (5) The short period equilibrium condition is aggregate supply equals aggregate demand. Sw = Dw (6) The price level measured in money equals the price level measured in wage units times the level of money wages. p = pw • w These are six equations in eleven variables: Yw = Sw, pw, q, m, w, q/L, L, Dw, Y/Ytr, I, p . Output, output per man, and the level of employment, related by the identity, q = L»q/L, in fact only represent two independent variables. Money wages and exogeneous investment are determined outside the model. The ratio of current to trend income, in 26 terms of the exposition of the short period equilibrium i s also an exogeneous variable. Finally, the gross-margin is a weighted sum of micro gross-margins and i t s behaviour depends upon the micro behaviour. For the time being the gross-margin is treated as a shift parameter. This closes the model. The six equation system simplifies into the following two equations determining the level of employment and the price level. (1*) f(Yw, m, Y/Ytr, T) pwq 1+m ,q (2*) q/L (1+m) L But Yw = Sw = (1+m)-L f((l+m)-L, m, Y/Ytr, I) = (l+m)-L p = pw•w 1+m (Eqs. 1,4,5) (Eq. 2) q/L 1+m g(L) . w • w (Eq. 6) (Eq. 2) (Eq. 3) Totally differentiating with respect to the price level, the level of employment and the margin: (1*) fifLdm + (1+m) dL] + f^dm = Ldm + (1+m) dL (2*) d p = " g i l l ? - W * g ' ( L ) d L + gTLT d m Rewriting in matrix notation: -(l-fl)(l+m) 0 dL (1+m)-w , , . , _ g ( L ) ^ g ( L ) _ dp _ _ The comparative static results are: dL _ ( l - f i ) - L - f 2 L-f2 1 - f l dm - (1-fi) (1+m) - (1+m) ( (1-fi) -L-f2)dm dm w L g (L) The sign of this derivative i s negative. With 0 ^  f ]_< l (see footnote 19), and with f2 < u ' the numerator is positive, the denominator 27 negative. Thus a higher margin, cet.par. would entail lower employment. Intuitively, a higher margin reduces aggregate demand with the shift to non-f2 1+m f o wage income (the ^ effect). This is magnified by a multiplier effect on aggregate demand ( j — • Also, there is the aggregate supply effect - (1+m) With a larger margin a given amount of expected revenues w i l l be accompanied by a smaller offer of employment. 2- ^ - | 1 ' t l , W - ^ ) ' ^ w •«•»•> < <i-*i>-i-'2> dm - ( 1 - f l ) (1+m) The sign of this derivative is ambiguous. If output per man f a l l s as employment increases (g 1 (L) <, 0) , then the numerator represents the sum of two terms of opposite sign: f u r > 0 a n d I m ? g , ( L ) ( L " i - f i ) < 0 The problem here is that the reduction of employment associated with the higher margin entails a higher output per man. This works in the opposite direction to the effect of the margin on the price. Only by assuming a horizontal or positively sloped output per man function (g 1 ( L ) ^ 0) does a 20 higher margin unambiguously entail a hxgher price level. As suggested earlier, the gross margin i s the key variable through which non-competitive price behaviour can affect macro pheno-20 A horizontal output per man function (note that per man refers to production, notproduction plus overhead labour) does seem reasonable in light of empirical studies. See the discussion on cost curves p. 53 below. It should be noted that a horizontal output per man curve is more than sufficient for a positive price effect. The minimum condition required is ' (L) ^  - 9 (^ ) f 2 (L-]IT> 28 mena. These comparative static results clearly indicate that if under-lying non-competitive markets generate a macro margin different from that generated by competitive markets, the price level and level of employment would be different. Also, relative shares would be different. To say the macro margin is different is equivalent to saying relative shares are different. It is often argued that although these may be interesting static results, they do not relate to the dynamic behaviour of a macro system. In other words, to the extent that underlying monopoly, for example, causes the macro margin to be higher than otherwise, the price level may be higher and output lower. However, it would be argued, this does not mean that with an unchanging industrial structure price and output 21 behaviour would be different. ' Of course, this argument is valid only if monopoly factors relate only to the level and not to the behaviour of the margin. If on the other hand (and as I argue later) underlying non-competitive markets tend to generate a macro margin more counter-cyclical than otherwise, over the cycle a non-competitively generated macro margin would behave dif-ferently from a competitively generated margin. This in turn would entail a difference in the cyclical behaviour of other macro variables. Given some exogeneous investment behaviour, the movement of the price level, level of output, and relative shares would be different. See, e.g., the excerpt from the Prices and Incomes Summary Report, quoted p. 150 below. 29 A s investment and income started to r i s e , a m a r g i n moving lower than otherwise (more coun te r - cyc l i ca l l y ) would be associa ted with a l eve l of output moving higher, a p r i ce leve l moving lower (assuming g ' (L) ^ 0), and labour ' s share moving higher than otherwise . The opposite would occur as investment and income started to f a l l ; With the consumption effects of a higher m a r g i n outweighing the investment effects ( f 2 ^ 0), a more c o u n t e r - c y c l i c a l ma rg in c o n t r i -butes to the ins tab i l i ty of the macro sys tem. Indeed, a sufficiently c o u n t e r - c y c l i c a l m a r g i n would necessitate that the ra t io of current to t rend income var iable act to prevent explosive growth or decl ine . He re in l i es the s ignif icance of the assumption that the consumption effects of this var iable outweigh the investment effects (f3 < 0) w e l l into booms or s lumps . In this chapter a macro model was out l ined. It was developed as a s imple model of effective demand, genera l ized in one key respect . Aggregate supply i s stated more genera l ly in t e rms of a g ross -marg in to al low for non-competi t ive as w e l l as competi t ive p r i ce behaviour . Th i s g ross -marg in , which also enters the aggregate demand function, was seen to have a profound effect on the mac ro sys tem. In the next two chapters the manner in which non-compet i t ive markets can generate macro m a r g i n behaviour different f rom that generated by competit ive markets w i l l be examined. 30 C H A P T E R III THE MACRO GROSS MARGIN The macro gross-margin is a weighted average of micro gross-margins. In the last chapter it was determined that if the macro gross-margin behaves differently, different macro phenomena result. Given some exogeneous investment behaviour, the different gross-margin behaviour entails different relative share, price, and output behaviour. The critical question, of course, is whether non-competitive price behaviour does generate macro margin behaviour different from that generated by competitive price behaviour. The next two chapters deal with this question. In this chapter the relationship between the macro and micro margins is clarified. The purpose here is to deter-mine to what extent the macro margin is explained by micro margin be-haviour. In the following chapter a theoretical examination of competitive and non-competitive micro margin behaviour is undertaken. Together, the two chapters determine any theoretical reasons there may be for asserting that underlying non-competitive price behaviour does indeed generate different macro margin behaviour. For the theoretical and statistical analysis micro margins are defined as industry value of shipments minus wage and material costs ,, ,. . , , , , . , t /rv_. VSi - (Mi +• wiLi) , all divided by wage and material costs ( m i = -) The y S Mi f WiLi ;* macro margin on the other hand was defined in the last chapter as value added minus wages all divided by wages (m = — ™ ' ), There are two w L 31 important factors to consider when relating micro margins to the macro margin. Fi r s t , in the denominator of the micro margins are wage and material costs, whereas in the denominator of the macro margin there are only wage costs. Micro margins are defined over wage and material costs not for any special theoretical reason, but simply because firms generally plan in terms of total variable costs, not one component or the other. ^  The macro margin is defined over wages in accord with Keynes 1 implicit treatment of material costs. As in Keynes 1 analysis this implicit treatment of material costs serves to simplify the exposition of the macro model. Second, micro margins are defined as gross profits (profits plus overhead) from value of shipments, whereas the macro margin is defined as gross profits from value of output. Micro margins are defined in terms of value of shipments on the assumption that firms' plans relate to realized profits from sales rather than potential profits from output. The macro margin is defined in terms of value of output following standard macro procedure. With these two factors in mind, I derive the algebraic relation between the macro and micro margins as follows. See N.. Chamberlain (1962, Chapter 8) on the critical role of total variable costs in the planning process. 32 m w h e r e V A - w- L w L V O - M • w . L w L A l + V S - M - w . L w L A l , V S - M - w . L M + w - L w- L M f w L w- L A I f ^ m i ( M i + w L i ) ^ M + w «L w L i M +• w L w« L A i . f ) i l m i t X i w« L i M i f w L i -y M +• w . L , . JVll + W L I -v cX i = — — r ; A M +• W' L ' ' 1 w . L m - m a c r o m a r g i n V A - v a l u e added w L - wage b i l l V O - v a l u e o f output M - m a t e r i a l c o s t s A I - change i n i n v e n t o r i e s m i - m i c r o m a r g i n , i n d u s t r y i M i - m i c r o m a t e r i a l s c o s t , i n d . i w L i - m i c r o wage b i l l , i n d u s t r y i T h u s , m a c r o m a r g i n b e h a v i o u r ( m = A + • \ • X m i <X i) w L i d e p e n d s n o t o n l y on the b e h a v i o u r of the m i c r o m a r g i n s ( m i ) , but a l s o on the b e h a v i o u r of t h r ee o the r f a c t o r s : (1) i n d u s t r y w e i g h t s , i . e . , the s h a r e of v a r i a b l e , , . M i f w - L i c o s t s ( c< i = , — \ . M +• w L / ' 33 (2) the m a c r o r a t i o of wage and m a t e r i a l c o s t s , , \ M +• w L to wage c o s t s ( A = - ): (3) the m a c r o r a t i o of change i n i n v e n t o r i e s to wage c o s t s ( ^-L ). w- L T h e r e f o r e , d i f f e r e n t m i c r o m a r g i n b e h a v i o u r w i l l be r e f l e c t e d i n d i f f e r e n t m a c r o m a r g i n b e h a v i o u r o n l y i f w i t h the d i f f e r e n t m i c r o m a r g i n b e h a v i o u r 2 th e r e a r e not co n c o m i t t a n t o f f s e t t i n g m o v e m e n ts of the other three f a c t o r s . On the b a s i s of t h e o r e t i c a l c o n s i d e r a t i o n s no d e f i n i t e c o n c l u -s i o n can be d e r i v e d i n t h i s r e g a r d . D i f f e r e n t m i c r o m a r g i n b e h a v i o u r c o u l d be a c c o m p a n i e d by movement of the other v a r i a b l e s w h i c h r e i n f o r c e d or o f f s e t the m i c r o m a r g i n e f f e c t on the m a c r o m a r g i n . F o r example, suppose d u r i n g a downswing m i c r o m a r g i n s r o s e h i g h e r than o t h e r w i s e because of n o n - c o m p e t i t i v e p r i c i n g . (1) If i t i s o n l y some i n d u s t r i e s w h i c h e x h i b i t h i g h e r m i c r o m a r g i n s , the r e l a t i v e i n c r e a s e c o u l d e n t a i l a r e l a t i v e d e c r e a s e i n p r o d u c t i o n and, t h e r e f o r e , d e c r e a s e i n the quantity of m a t e r i a l s and la b o u r used. T h i s would c o n t r i b u t e to a r e l a t i v e d e c r e a s e i n the weights of these i n d u s t r i e s . On the other hand, where i m p e r f e c t i o n s i n the pr o d u c t m a r k e t s l e a d to h i g h e r than o t h e r w i s e m a r g i n s , there a r e a l s o l i k e l y to be i m p e r f e c t i o n s i n the f a c t o r m a r k e t s l e a d i n g to hi g h e r p r i c e s of la b o u r and m a t e r i a l used. ^  T h i s w o u l d c o n t r i b u t e to a r e l a t i v e 2 G.C. A r c h i b a l d (1955) has e x a m i n e d a l l t h r ee of these f a c t o r s , empha-s i z i n g the r o l e of i n v e n t o r i e s . R. Solow (1958) and J . Dunlop (1950) have e x a m i n e d i n p a r t i c u l a r the r o l e of s h i f t i n g weights. The s e m i n a l w o r k i n t h i s r e g a r d i s by G a r b a r i n o (1950). 34 increase in the weights of these industries. (Z) If it is for the most part final-goods industries which exhibit higher micro margin behaviour and if this movement contributes to a greater wage movement, the ratio of materials plus wages to wages would be lower. On the other hand, if it is in material goods industries that, margins are higher, the ratio of materials plus wages to wages would be higher. (3) Finally, to the extent that higher margins cause sales to fall lower than otherwise, the ratio of inventory change to wages could be affected as follows: insofar as intended inventory levels are deter-mined primarily by sales, intended inventory change over wages could move lower (more negative) than otherwise; on the other hand, insofar as intended inventory levels are determined primarily by desired minimum production levels, intended inventory change over wages could move higher than otherwise. Although precluded in my theoretical analysis by the equality of aggregate demand and aggre-gate supply, actual (ex-post)values of inventory change over wages may also reflect unintended inventory change. If higher margins and lower sales increase the tendency to overestimate sales, unin-tended inventory change over wages would move higher than other-wise. Thus, theoretically and quite plausibly, the three other factors determining the macro margin could reinforce or offset the effect of micro margin behaviour. Like most aggregation problems, the 35 problem here reduces to an empirical question. Is micro margin be-haviour in fact sufficiently independent of the behaviour of the other three factors to have an independent effect on macro margin behaviour? To pursue this empirical question, Canadian manufacturing industries Census data (described in Chapter VI) were examined. First, changes of the macro margin were broken up algebraically into the sum of the four component changes: a weighted average of the changes of the micro margins; a weighted average of the changes of the industry weights; a weighted change of the ratio of materials plus wages to wages; and the change of the ratio of inventory change to wages. This breakdown was derived as follows. The macro margin is: m = — rf A 4> mi <x 1 W' L l Mi 4- w. L i o< 1 = M f w- L M +• wL A = wL A change in the mac ro m a r g i n , therefore, equals the sum of the change of the f i r s t t e rm ( A -—-4~) and the change of the second t e rm ° w L ° ( A X ? m i - rXi* ). In turn, the change of the second t e rm can be broken up (though not uniquely) into the sum of weighted changes of each of the fac-to rs : m i , c<i, A . This breakdown can be done in s ix different w a y s . ^ A A Z m i - o U = A Z m i ^ c * ^ . - A Q Z m y ^ 4 This is the three variable analogue of the Laspeyre versus Paasche measures. (1) = A Q 2 o ( i 0 (A mi) + A Q S m i ^ A c X i ) + 2 m i x c X ^ A A (2) = A Q 2T OCij ( A m i ) f A Q 2 ' m i 0 (AcX i ) + 2 c X i j A A (3) = A Q X ( X i Q ( A m i ) t (AcXi) -t- 2 m ^ c X ^ A A (4) = A S o ( i 0 ( A mi) f A S - .mi (AcXi) + S m i Q c X i 0 A A (5) = A X o d j ( A m i ) t m i Q ( A o ( i ) + - S m i Q i X i 0 A A (6) = A X 0 ( i ( A mi ) f A Q X m i Q (AoCi) + 2 m i Q c X i 0 A A where A m i = m i , - m i n ; A o ( i = 0(i, - o ( i n ; A A= A, - A n Thus, w i t h the change of the f i r s t t e r m and the b r e a k d o w n of the second t e r m , a change i n the m a c r o m a r g i n can be seen as the sum of changes i n each of i t s component f a c t o r s . A n n u a l changes of the m a c r o m a r g i n and i t s four components w e r e c a l c u l a t e d and w e r e e x a m i n e d i n r e l a t i o n to one another. B e c a u s e the m i c r o m a r g i n , i n d u s t r y weight, and r a t i o of m a t e r i a l p l u s wages to wages component changes cannot be d e t e r m i n e d u n i q u e l y , m a x i m u m and m i n i m u m v a l u e s of these changes w e r e c a l c u l a t e d f r o m the s i x f o r m u l a e . The annual change c a l c u l a t i o n s w e r e done o v e r two time p e r i o d s : 1954 to 1959 (on the b a s i s of the 1948 S. I. C. ) and 1958 to 1969 (on the b a s i s of the I960 S. I. C. ). A n n u a l changes of the m a c r o m a r g i n a r e p l o t t e d a l o n g s i d e annual changes of e a c h of the components i n the graphs at the end of th i s c h a p t e r . The m o s t s t r i k i n g f e a t u r e of these graphs i s the p a r a l l e l b e h a v i o u r of the annual change i n the m a c r o m a r g i n and the change of the m i c r o m a r g i n component, i . e . , the weighted a v e r a g e of the changes 37 of the micro margins. This is true over both time periods, though the parallel behaviour is stronger for the 48 S.I.C. period. The graphs suggest that micro margin behaviour in fact is 4 reflected in the behaviour of the macro margin. Simple correlation tests confirm this observation. • Changes in the macro margin were re-gressed against each of the component changes, minimum and maximum values. The micro margin component (and only this component) was found to be statistically significant (at the 95% confidence level) in ex-plaining the macro margin behaviour. This held true for both maximum and minimum value s, over both the 48 S. I. C. and 60 S.I.C. periods (see Table 3. 1). Consistent with this is the finding that the annual changes of the micro margin component are not significantly correlated with annual changes of the other three components (at the 95% confidence level). The behaviour of the micro margin component is independent of the other three components (see Table 3. 2). The purpose of this chapter has been to determine whether different micro margin behaviour will be reflected in different behaviour of the macro margin. On the basis of empirical tests, I conclude that the behaviour of the micro margins is indeed sufficiently independent of the other three factors determining the macro margin to have an indepen-dent and observable effect on the macro margin. This means that if micro margins move more counter-cyclically because of non-competitive micro price behaviour, one can assert that this will entail the macro 38 T A B L E 3. 1 M A C R O M A R G I N x C O M P O N E N T S R E G R E S S I O N S (Y = a f b X ) X b Prob(b=0) R 2 Sample P e r i o d A n n u a l Changes A n n u a l Changes M a c r o M a r g i n M i c r o M a r g i n ( M i n . ) " M i c r o M a r g i n (Max. ) " Ind. Weights ( M i n . ) " Ind. Weights (Max. ) 11 Mat. +• Wages O v e r Wages ( M i n . ) " Mat. +• Wages O v e r Wages (Max. ) 8211 027 8177 .023 -2.852 -2.542 . 7034 . 6986 In v e n t o r y Change O v e r Wage -. 7831 A n n u a l Changes A n n u a l Changes M a c r o M a r g i n M i c r o M a r g i n ( M i n . ) " M i c r o M a r g i n (Max. ) " Ind. Weights (Min.) " Ind. Weights (Max. ) " ( M f w L ) / w L (M i n . ) '» ( M + w L ) / w L (Max. ) " Inv e n t o r y Change O v e r Wages . 6821 •2.686 • 2.404 . 9088 430 . 530 714 710 724 6847 .023 . 018 . 062 . 161 . 055 9137 .051 857 48 S. I. C. 871 218 145 050 051 046 415 60 S.I.C. 443 2792 580 302 185 316 326 033 39 T A B L E 3. 2 M I C R O M A R G I N C O M P O N E N T x O T H E R C O M P O N E N T  R E G R E S S I O N S (Y = a+bX) S a m p l e Y X b Prob(b=0) R Z P e r i o d A n n u a l C h a n g e s A n n u a l C h a n g e M i c r o M a r g i n Ind . W e i g h t ( M i n . ) ( M i n . ) - 4 . 0 7 . 2 9 4 5 . 3 4 9 48 S . I . C . " Ind . W e i g h t ( M a x . ) - 4 . 2 3 . 3 2 5 6 . 3 1 5 " " M-t-w L / w . L ( M i n . ) - . 0 5 0 . 9 3 1 2 . 0 0 0 2 " " M+w- L / w . L ( M a x . ) - . 0 3 6 . 9351 .0001 " " I n v e n t o r y C h a n g e O v e r W a g e s - 1 . 7 2 . 4 8 6 7 . 1 7 4 " M i c r o M a r g i n Ind . W e i g h t ( M a x . ) ( M i n . ) - 4 . 2 0 . 2 8 2 7 . 3 6 3 *' Ind . W e i g h t ( M a x . ) - 4 . 3 3 .3191 . 3 2 2 " " M - i - w L / w - L ( M i n . ) . 0 6 0 . 9 2 8 4 . 0 0 0 3 " M+w- L / w . L ( M a x . ) .071 . 9 2 4 7 . 0 0 0 4 " " I n v e n t o r y Change O v e r W a g e s - 1 . 7 5 . 4 8 4 2 . 1 7 6 " A n n u a l Change A n n u a l C h a n g e M i c r o M a r g i n Ind . W e i g h t s - . 9 9 6 . 5 0 5 . 0 4 7 60 S . I . C . ( M i n . ) ( M i n . ) " Ind . W e i g h t ( M a x . ) - . 4 5 1 . 7 8 0 . 0 0 7 " M-t-w L / w . L ( M i n . ) . 0 7 3 . 8 5 5 . 0 0 2 " " M + w L / w - L ( M a x . ) . 0 8 9 . 8 3 3 . 0 0 3 T A B L E 3.2 - (Cont inued) 4 0 Y X b P r o b ( b=0) R 2 S a m p l e P e r i o d A n n u a l Change M i c r o M a r g i n ( M i n . ) I n v e n t o r y Change O v e r Wages - . 7 1 0 . 1 0 5 . 2 3 8 6 0 S . I . C . M i c r o M a r g i n ( M a x . ) Ind . W e i g h t ( M i n . ) - 1 . 2 0 . 4 3 5 . 0 6 3 11 Ind . W e i g h t ( M a x . ) - . 5 9 2 . 7 3 3 . 0 1 2 11 n M + w L / w - L ( M i n . ) . 1 1 7 . 8 0 3 . 0 0 6 11 tt M t - w L / w L ( M a x . ) . 1 3 4 . 7 8 0 . 0 0 7 11 11 Inv . Change O v e r W a g e s - . 7 1 3 . 1 1 9 . 2 2 3 11 m a r g i n m o v i n g m o r e c o u n t e r - c y c l i c a l l y than o t h e r w i s e . T h i s i s the r a t i o n a l e for the m i c r o m a r g i n a n a l y s i s w h i c h f o l l o w s and i t s r e l a t i o n to the m a c r o m o d e l . GRAPH 3. 4 ANNUAL CHANGES m% SJ.C. 44 GRAPH 3. 5 ANNUAL CHANGES' ItkQ S.hC. (/<?S~7-£>9) 45 SIS? (.0-S1 W-4 0 6*-V I.H-l,S CS-Li U-US l*-i,7 «•»-<.? YEARS G R A P H 3. 6 ANNUAL CHANGES i960 ()95~?-£>9) 46 j I Y E A R S i I i ! ! i J 3 I GRAPH 3. 7 47 YEARS GRAPH 3. 8 48 «-57 51-5? 40-S<) »-W W-tl fcS"6W a-CS i7-4t «-67 yiTARS 49 C H A P T E R IV MICRO MARGIN BEHAVIOUR The purpose of this chapter is to compare micro gross-margin behaviour in competitive and non-competitive markets. I argue that non-competitive micro gross-margins do not behave exactly the same as and possibly do not even move in the same direction as compe-titive gross-margins during business cycle fluctuations. That non-competitive micro gross-margin behaviour differs from competitive gross-margin behaviour (the argument of this chapter) means that macro gross-margin behaviour based on non-competitive markets differs from macro gross-margin behaviour based on competi-tive markets (the argument of Chapter III). ^  This, in turn, means that a macro system based on non-competitive markets will behave differently from a macro system based on competitive markets (the argument of Chapter II). In other words, assumptions regarding competition at the micro level are important in modelling macro phenomena. In the following analysis, I examine models of a competitive market and then of non-competitive markets to determine and compare gross-margin behaviour during business cycle fluctuations. By business cycle fluctuations I mean primarily fluctuations of output. How do the gross-margins behave when output rises (upswings) and when output falls ^ It should be noted that while micro margins in Chapter III referred to the margins for each industry, in this chapter they refer to the margins of individual firms. Perversities arising from intra-industry aggre-gation are ignored. See p.112 below. 50 (downswings). However, since business cycle fluctuations are often associated with general movements of wage and material prices, I also consider how the gross-margins behave when factor prices rise and fall. The models are presented in a short run context - decisions are made with given plant and equipment. Nevertheless, where margin behaviour during business fluctuations is affected by accumulation or de-cumulation of plant and equipment, this is considered. However, only changes of the scale of plant and equipment are considered. I do not consider changes which would affect the short run cost function for any given amount of plant and equipment. Micro Gross-Margin Behaviour in Classical Competitive Markets In classical competitive markets sellers perceive horizontal demand curves. There are sufficiently many buyers and sellers of the homogeneous product that each individual seller does not perceive any effect that his own sales have on the market price. The critical feature of classical competitive markets is that the seller makes no price deci-sion. The seller can sell as much as he like's at the given market price, none at a higher price, and to no advantage at a lower price. His only decision, therefore, is with regard to the level of output. It is well-known, of course, that to maximize profits the seller will adjust output until marginal cost equals the price. The gross-margin in competitive markets, therefore, is that _ For the effect that competitive sellers in fact have on the market price see Joan Robinson (1934). 51 margin which emerges when price equals marginal cost. Specifically, it equals the ratio of marginal cost to average variable cost minus one . _ P-AVC MC-AVC MC (rn - — ^ V C ~ — A V C = AVC " ^hus the cyclical behaviour of com-petitive gross-margins depends on how the ratio of marginal cost to average variable cost varies during business cycle fluctuations. Fi r s t , it should be noted that this ratio is independent of general factor price movements (holding relative factor prices constant). Letting q=f(I) be the production function, where I represents the compo-site variable input labour and materials, and letting V=tl be the cost of the variable inputs, where t represents the cost per unit of labour and materials, then the ratio of marginal cost to average variable cost is independent of t. While this may seem trivial, it will be seen later that non-competitive margins are not always invariant to general factor price movements. MC = 11 - t 3 1 - fc - - J L -AVC d q 9 q f1 M P i — i i . _ _L _ t q = q = f/I = APj MC _ t/MPi _ A P T AVC ~ t/APj MPj Thus the cyclical behaviour of competitive gross-margins is technologically determined. It depends entirely on the properties of the cost function and underlying production function. How does the ratio of marginal cost to average variable cost vary with output? Strictly, the most general restrictions imposed on a compe-52 t i t i v e cos t func t ion (C(0) > 0; C ' (q) > 0, C"(q) > 0) do not d e t e r m i n e the s i g n of the d e r i v a t i v e o f the r a t i o of m a r g i n a l cos t to a v e r a g e v a r i a b l e cos t w i t h r e s p e c t to output . W i t h the C E S cos t f u n c t i o n s , fo r e x a m p l e , t h i s d e r i v a t i v e w i l l be p o s i t i v e w h e n the e l a s t i c i t y of s u b s t i t u t i o n b e t w e e n l a b o u r and c a p i t a l i s l e s s than one; nega t i ve when the e l a s t i c i t y i s g r e a t e r than one . 3 W i t h a C E S c o s t func t ion : C = t l + r K w h e r e q = A ( < X J ~ P + ( 1 - C < ) K ~ P \ " 1 / P w h e r e : 0 < 1 p > -1 A V C = — = t I / A J c<r p+ ( 1 - o O K - P } _ 1 / P dC dJ_ _ d C 1  M C = d i ' dq d i ' d q / d l _ M C  r ~ A V C f V [ - 1 . A J o C I - P + d - ^ K - P J - ^ - ^ - p . o C . I - P - l ) " = <XI P | <* I-P +• ( l-oC)K-PJ = o<|<* - t - ( l - pC) I P K" P j d r d r d i _ d r dq d i dq d i d q / d l = { o C U - o O - p l P ^ K - p ) . - ^ = | o C ( l - o t ) . p . l P - l . K - P } . ^ dr .". the s i g n o f depends on the s i g n o f the p a r a m e t e r p q S i n c e p = ^ (see K . J . A r r o w , _et a l . (1961)), w h e r e 6~ i s the e l a s t i c i t y o f s u b s t i t u t i o n , the d e r i v a t i v e w i l l be p o s i t i v e w i t h (5" <C 1, z e r o w i t h o - = 1 (the C o b b - D o u g l a s ca se ) , and nega t ive w i t h 6 > 1. 4 I w o u l d l i k e to thank G . C . A r c h i b a l d and P . N e h e r for c o r r e c t i n g m e on t h i s m a t t e r . 53 W h i l e no g e n e r a l statement can l o g i c a l l y be deduced, i t i s c o m m o n l y h e l d that c o m p e t i t i v e g r o s s - m a r g i n s behave p r o - c y c l i c a l l y , d M C i . e . , the d e r i v a t i v e AVC > i s p o s i t i v e . T h i s i s i m p l i c i t l y a s s u m e d dq by a l l those who expect c o m p e t i t i v e p r i c e s to f a l l r e l a t i v e to v a r i a b l e c o s t s d u r i n g a downswing. ^ I a l s o make this a s s u m p t i o n . I assume that p r o d u c t i o n f u n c t i o n s w h i c h generate p r o - c y c l i c a l c o m p e t i t i v e g r o s s - m a r g i n s a r e the s t a n d a r d c a s e . To the extent that c o m p e t i t i v e i n d u s t r i e s can be c h a r a c t e r i z e d by C E S p r o d u c t i o n f u n c t i o n s , I am a s s u m i n g that the e l a s -t i c i t y of s u b s t i t u t i o n i s l e s s than one. T h i s i s c o n s i s t e n t w i t h e m p i r i c a l e s t i m a t e s of the e l a s t i c i t y of s u b s t i t u t i o n i n C E S p r o d u c t i o n f u n c t i o n s i n both C a n a d i a n and U.S. m a n u f a c t u r i n g i n d u s t r i e s . ^ 1 In an i m p r e s s i v e n u mber of s t u d i e s , a l i n e a r t o t a l c o s t f u n c t i o n , e n t a i l i n g constant m a r g i n a l and average v a r i a b l e c o s t s up to c a p a c i t y , has been w i d e l y o b s e r v e d . In a 1970 a r t i c l e , Y o r d o n r e p o r t s that: " A l m o s t a l l s t a t i s t i c a l s t u d i e s of i n d u s t r i a l cost f u n c t i o n s r e p o r t constant m a r g i n a l cost o v e r a wide range of output". 7 Cfc,e.g., J . Dunlop (1938), A. C. N e a l (1942), R. R u g g l e s (1955), B. L a d e n (1972), B. S c a r f e (1972). K . J . A r r o w , et a l . (1961, p. 230); Y. K o t o w i t z (1968). W.J. Y o r d o n (1970, p. 59). Other s t u d i e s r e p o r t i n g l i n e a r cost f u n c t i o n s i n c l u d e F. R. O l i v e r (1968); J . J o h n s t o n (I960, p. 168); A . P . S l o a n (1964, p. 145);P.W.S. A n d r e w s (1949, pp. 102-3); J o e l Dean (1952, p. 208). 54 The varying proportions assumption underlying the tradi-tional U-shaped marginal cost curve is not appropriate for most indust-r i a l processes. Plant and equipment while fixed in terms of cost is not fixed in terms of physical use. Its use can be and is varied in propor-tion to the input of labour and materials. • capac ity Figure 4. 1 A Linear Cost Function With such a cost function marginal cost rises vertically at capacity (Figure 4. 1). It is clear that with this cost function too, the competitive margin will vary pro-cyclically. "The industry always produces close to capacity and has to sell the output for whatever it will Q bring in the market". In upswings higher demand entails higher prices (intersecting at a higher point on the vertical part of the marginal cost curve) while average variable cost is fixed at the capacity level. This means a higher gross-margin results. In downswings lower demand M. Moore (1970, p. 15). That competitive markets are described in this manner is further evidence of the belief that cost functions are linear up to capacity. 55 entails lower prices and gross-margins. If the downswing is sufficiently severe output and the margin could fall to zero. However, this would 9 10 still be consistent with pro-cyclical margin behaviour. * In sum, while logically any behaviour is possible, I assume that pro-cyclical competitive gross-margin behaviour is the standard case. Such behaviour will result with CES production functions when the elasticity of substitution is less than one, and will result with a different, but widely observed case for manufacturing industries, where the cost function is linear up to capacity. It should be noted here that I interpret counter-cyclical gross-margin behaviour as evidence of non-competitive price behaviour (see Chapter VII). Of course, an alternate interpretation, that competi-tive price behaviour is at work but (what'1) term 'non-standard' cost func-tions prevail, is possible. As noted in footnote 1, there can be aggregation problems going from firm to industry behaviour. The problem is particularly acute when there are linear cost curves. In a downswing high average variable cost-low margin firms may shutdown causing the weights of these firms to fall to zero. The increase of the weights of the remaining relatively high margin firms could cause an increase in the industry gross-margin even if all firm gross-margins fell. Such perversities could only occur if a large number of firms shutdown during the down-swing and reopen in the upswing. I assume this is not generally the case. The demonstration that competitive margins are independent of factor price movements is not applicable in this case because of the discontinuous productivity curves. However, when considering factor price movements holding real demand fixed, the gross-margin remains unaffected. As money wages and material prices shift up, the price will shift up by the same percentage. 56 Gross-Margin Behaviour in Non-Competitive Markets This non-competitive category includes all markets which do not fit the classical competitive model. The critical feature here is that the firm is not a price-taker, rather it must decide upon and administer a price. This category includes monopolistic and oligopo-listic markets, and those markets which in terms of number of sellers, would be classified as competitive, but which are not competitive in the classical sense. In this latter group are markets with limited monopoly^ or overlapping oligopolies which arise because of spatial or product dif-ferentiation, and markets where despite the large number of sellers, mutual dependence is recognized and acted upon through trade associa-tions and the like. There is no single model which adequately reflects all non-competitive markets. Profit maximization in each period, suitable in classical competitive markets because there is no discretionary price policy and anything but the profit maximizing output would be very short-lived, need not be the operating assumption in non-competitive markets. Fir s t , profits in the long run may be better served by less than period-to-period profit maximization because of possible entry, or more generally, because of inter-period relationships in the firm's demand curve. Second, profits in the long or short run may not be the goal be-cause of separation of ownership and control. A managerial utility maxi-mization model may be more appropriate. Third, even if profit maxi-T l M. Moore (1970, p. 56). 57 mization is the goal it may not be pursued in the classical way because of behaviourist considerations. Necessary concrete processes of cor-porate organization may entail less than period-to-period profit maxi-mization. (Indeed, the latter is not a real concept in the behaviourist context). Fourth, profit maximization may not be pursued or achieved in the classical way because of oligopolistic considerations. Short period profits may be sacrificed to accommodate collusion and/or short period profits may be lost because of the breakdown of collusion. Models incorporating these possibilities are presented in turn. F i r s t a model of profit maximization in each period is presented. It should be kept in mind here that my goal is not to determine which is the 'most correct' of these models. My goal is to determine whether in these models the gross-margin behaviour during business fluctuations differs from that in classical competitive markets. Since all of these non-competitive models may have some validity, if it can be demonstrated that in any or all gross-margin behaviour differs from competitive gross-margin behaviour, then the theoretical argument that non-competitive gross-margin behaviour may differ from competitive gross-margin behaviour has been established. The empirical tests will then determine the significance of this theoretical point. 1. Short Run Profit Maximization To maximize profits, sellers in non-competitive markets adjust output until marginal revenue equals marginal cost and 1 2 sell that output at the corresponding price on their demand curve. 58 Changes i n the g r o s s - m a r g i n a r e d e t e r m i n e d as f o l l o w s : To m a x i m i z e p r o f i t s : M R = M C But; M R = (1 - I ) p n c ( 1 - 1 ) p = M C n -P_ - 1 m = A V C M C 1 A V C 1-1/n 1 , M C M C 1 d m = T T T A i d ( A V C } f ~Jsfcd ( 1-1/n ' A v r . A V C V , 7 7 2 ' 1-1/n V C A C (n-1)' where: n - i s the absolute value of the e l a s t i c i t y of demand w i t h r e s p e c t to p r i c e n > 1 T h e r e f o r e , i n t h i s n o n - c o m p e t i t i v e model, g r o s s - m a r g i n be-h a v i o u r depends not o n l y on the b e h a v i o u r of the r a t i o of m a r g i n a l cost to a v e r a g e v a r i a b l e cost (as i n the c o m p e t i t i v e model), but a l s o on the b e h a v i o u r of the p r i c e e l a s t i c i t y of demand. It f o l l o w s that i t i s v e r y un-l i k e l y that g r o s s - m a r g i n s i n t h i s n o n - c o m p e t i t i v e m o d e l w i l l behave 13 e x a c t l y the.same as c o m p e t i t i v e g r o s s - m a r g i n s . 12 In th i s m o d e l , and i n a l l except the l a s t ( o l i g o p o l i s t i c c o n s i d e r a t i o n s ) , i t i s a s s u m e d that each f i r m i s l i k e a m o n o p o l i s t w i t h an i s o l a t e d i n d i v i d u a l demand c u r v e . F o r a g e o m e t r i c a p p r o a c h y i e l d i n g s i m i l a r r e s u l t s see John M o o r e and L. L e v y (1955). 5 9 Even if the elasticity of demand does not vary during business fluctuations (dn = 0), a given change in the ratio of marginal cost to average variable cost will entail a proportionately smaller change in this non-competitive margin than in the competitive gross-margin. This can be demonstrated as follows: In competitive markets: M C M = ~ A V C " 1 , , , M C . D M = D ( A V C ) = — i . d ( -M£. ) (a) m M C _ 1 A V C A V C In non-competitive markets: M C 1 m = A V C _ J _ n 1 , M C V dm = 1 D ' A V C with dn=0 n 1 — = 1-1/n d ( M C } m M C _ 1 _ 1 A V C A V C ' 1-1/n 1 , , M C . , , d (^77^ ) (b) M C l A V C A V C n Since M C _ > M C ~~~T A V C A V C " 1 n dm dm  m in (a) is greater than m i n ^ A V C A V C 60 This indicates that non-competitive gross-margins will vary-less pro-cyclically (rising less in upswings; falling less in downswings) than competitive gross-margins (dn=0). In an upswing, for example, a given increase in output giving rise to a given increase in marginal cost over average variable cost will entail a smaller gross-margin increase in the non-competitive than in the competitive market. The 14 difference in behaviour is greater the smaller the elasticity of demand. 15 When the cyclical behaviour of the elasticity of demand is considered the situation is more complicated. ^  If elasticity of demand varies pro-cyclically ( - T r — - ^ 0). then the difference in behaviour is A q strengthened. Indeed, a counter-cyclical non-competitive gross-margin could be generated (falling in upswings; rising in downswings). On the other hand, if the elasticity of demand varies counter-cyclically (^r— <i 0), q then the difference in behaviour is weakened or even reversed - non-competitive margins could vary more pro-cyclically than competitive margins. Needless to say, there is no consensus among economists 14 15 16 In this comparison, the same cost function is attributed to the com-petitive or to the non-competitive market. Where non-competitive markets arise precisely because of different cost conditions, this comparison is not appropriate. The cyclical behaviour of the price elasticity of demand refers to changes because of cyclical twists of the demand curve as it shifts over the course of the cycle as well as changes because of move-ments along the demand curve during the cycle. It should be noted that with linear cost curves where output remains below capacity, the cyclical behaviour of the elasticity of demand is the sole determinant of the cyclical behaviour of the gross margin (m =-1 - 1 = _ J _ ) . 1-1/n n-1 61 r e g a r d i n g the c y c l i c a l b e h a v i o u r of the e l a s t i c i t y of demand. H a r r o d has a r g u e d that i n a s l u m p people become m o r e p r i c e c o n s c i o u s and the e l a s t i c i t y of demand tends to r i s e . Or, "as output as a whole i n c r e a s e s and i n d i v i d u a l s become m o r e affluent, t h e i r s e n s i t i v e n e s s to 1 7 p r i c e d i f f e r e n c e s d e c l i n e s " . G a l b r a i t h on the other hand has a r g u e d the opposite tendency. P u r c h a s e s , i n p a r t i c u l a r p u r c h a s e s of m o r e du r a b l e goods, a r e postponed i n a slump. The e l a s t i c i t y of demand f a l l s 18 in a s l u m p as l o w e r p r i c e s do not induce m o r e s a l e s . D e s p i t e the l a c k of concensus r e g a r d i n g the c y c l i c a l b e h a v i o u r of the e l a s t i c i t y of demand,what can be c o n c l u d e d f r o m th i s s i m p l e s h o r t r u n p r o f i t m a x i m i z i n g m o d e l i s that u n l e s s e l a s t i c i t y of demand v a r i e s s u f f i c i e n t l y c o u n t e r - c y c l i c a l l y , n o n - c o m p e t i t i v e g r o s s - m a r g i n s w i l l v a r y l e s s p r o - c y c l i c a l l y than c o m p e t i t i v e g r o s s - m a r g i n s . If e l a s t i c i t y of demand v a r i e s s u f f i c i e n t l y p r o - c y c l i c a l l y , the n o n - c o m p e t i t i v e g r o s s -m a r g i n s here w i l l even v a r y c o u n t e r - c y c l i c a l l y , opposite i n d i r e c t i o n to s t a n d a r d c o m p e t i t i v e g r o s s - m a r g i n s . Whatever the b e h a v i o u r of the e l a s t i c i t y of demand i t would be e x t r e m e l y u n l i k e l y f o r t h i s n o n - c o m p e t i -t i v e g r o s s - m a r g i n to behave e x a c t l y the same as the c o m p e t i t i v e g r o s s -m a r g i n . 2. P r o f i t M a x i m i z a t i o n w i t h F e a r of E n t r y When s e l l e r s i n n o n - c o m p e t i t i v e m a r k e t s e x p l o i t t h e i r m o n o p o l y power, they g e n e r a l l y e a r n e x c e s s p r o f i t s , p r o f i t s beyond 1 7 R.F. H a r r o d (1936. p. 21). 18 J. K. G a l b r a i t h (1936). 62 that w h i c h i s n e c e s s a r y to keep them i n the m a r k e t . B e c a u s e these e x c e s s p r o f i t s s i g n a l o t h e r s to enter the m a r k e t , s e l l e r s i n non-com-p e t i t i v e m a r k e t s w i t h any c o n c e r n f o r f u t u r e p r o f i t s m u s t be c o n c e r n e d w i t h p o t e n t i a l e n t r a n t s . S e l l e r s can enjoy r e n t s without f e a r of e n t r y up to a point d e t e r -m i n e d by the b a r r i e r s to e n t r y . B a r r i e r s to e n t r y i n c l u d e a l l those f a c t -o r s w h i c h w o u l d p r e v e n t an entrant f r o m e a r n i n g the same p r o f i t s as those a l r e a d y i n the i n d u s t r y . S p e c i f i c a l l y , these b a r r i e r s i n c l u d e : absolute c o s t advantages, where en t r a n t s w o u l d pay h i g h e r c o s t s than e x i s t i n g s e l l e r s ; p r o d u c t d i f f e r e n t i a t i o n advantages, where e n t r a n t s would r e c e i v e a l o w e r p r i c e or have g r e a t e r s a l e s p r o m o t i o n o u t l a y s than e x i s t i n g s e l -l e r s ; and e c o n o m i e s of s c a l e r e l a t i v e to the s i z e of the m a r k e t , where a f t e r - e n t r y c o s t s would be h i g h e r because of the f u r t h e r s h a r i n g of the m a r k e t . ^ 9 Studies of C a n a d i a n m a n u f a c t u r i n g i n d u s t r i e s suggest that s m a l l 2D m a r k e t s make e c o n o m i e s of s c a l e an i m p o r t a n t b a r r i e r to e n t r y . A l s o , h i g h c a p i t a l r e q u i r e m e n t s and c o n c e n t r a t i o n i n the C a n a d i a n c a p i t a l m a r k e t a r e a s o u r c e of absolute cost advantage b a r r i e r s . E n t r a n t s would have to 21 i n c u r m u c h h i g h e r c a p i t a l c o s t s than the e x i s t i n g s e l l e r s . F i n a l l y , e n t r y into C a n a d i a n m a r k e t s by f o r e i g n companies, i . e . , i m p o r t c o m p e t i t i o n , i s l i m i t e d by t a r i f f s . T a r i f f s , t h e r e f o r e , a l s o act as b a r r i e r s . If b a r r i e r s a r e s u f f i c i e n t l y g r e a t one c o u l d f i n d what B a i n c a l l s _ F o r a f u l l d i s c u s s i o n w i t h s p e c i a l r e g a r d to U.S. m a n u f a c t u r i n g i n d u s t -r i e s , see Joe B a i n (1956). 2 ^ See, e.g., S. S t y k o l t and H. E a s t m e n (1967). 21 G. R o s e n b l u t h (1961, p. 211). an e f f e c t i v e l y b l o c k a d e d s i t u a t i o n . " H e r e b a r r i e r s a r e so h i g h that p r o f i t s can be m a x i m i z e d i n each p e r i o d without f e a r of e n t r y . If b a r r i e r s a r e v e r y low one c o u l d f i n d what B a i n c a l l s an i n e f f e c t i v e l y i m p e d e d s i t u a -t i o n , where s e l l e r s m a x i m i z e s h o r t r u n p r o f i t s despite the fact that t h i s w i l l induce e n t r y . However, i f b a r r i e r s a r e m o d e r a t e or s u b s t a n t i a l one c o u l d f i n d what B a i n c a l l s an e f f e c t i v e l y i m p e d e d s i t u a t i o n , where s e l l e r s do not p r i c e at the s h o r t r u n p r o f i t m a x i m i z i n g l e v e l . It i s i n th i s s i t u a t i o n that f e a r of e n t r y i s m a n i f e s t e d i n b e h a v i o u r as s e l l e r s s a c r i f i c e some p r o f i t s i n the s h o r t r u n to p r e s e r v e at l e a s t a m o d e r a t e am o u n t i o f e x c e s s p r o f i t s i n the long r u n . To a n a l y z e the b e h a v i o u r of s e l l e r s i n t h i s l a s t s i t u a t i o n s e v e r a l a p p r o a c h e s a r e p o s s i b l e . R ecent s t u d i e s have a p p l i e d o p t i m a l c o n t r o l t h e o r y to t h i s p r o b l e m . L e t t i n g p r o b a b i l i t y of e n t r y be a f u n c t i o n of p r o f i t s , the t h e o r i s t s p ostulate that s e l l e r s a d m i n i s t e r t h e i r p r o f i t s and induce e n t r y at a r a t e w h i c h m a x i m i z e s the e x p e c t e d p r e s e n t 23 value of t h e i r c u r r e n t and future p r o f i t s . In m y a n a l y s i s , however, I f o l l o w a d i f f e r e n t a p p r o a c h f r o m the o p t i m a l c o n t r o l s t u d i e s . F o l l o w i n g the a n a l y s e s of A n d r e w s , H a r r o d , S y l o s - L a b i n i ; and B a i n , 2 4 I a ssume that s e l l e r s i n e f f e c t i v e l y i m p e d e d ~22 Joe B a i n (1956, p. 22). 23 See, e.g., D a v i d B a r o n (1972); D. W. G a s k i n s (1971); F. G. P y a t t (1971); M.I. K a m i e n and N. L. S c h w a r t z (1971). A l s o , but m o r e g e n e r a l than f e a r of e n t r y see A l e x P. J a c q u e m i n (1972); H e n r y Wan J r . (1966). 24 P.W.S. A n d r e w s (1949); R.F. H a r r o d (1952); P. S y l o s - L a b i n i (1969); Joe B a i n (1956). 64 situations simply pursue an entry-preventing strategy. They do not select an optimal path of inducing entry because uncertainties surround-ing any entry can threaten the very existence of the established firms. As Harrod writes: "any experienced man of business would pronounce it most 'unsound' to make a temporary surplus profit by charging a high price at which it is known that sales are unlikely to be capable of being main-tained in the long run". ^5 Making a temporary surplus at which sales cannot be maintained is an integral part of the optimal control approach. How sellers pursue an entry-preventing strategy depends upon their perceptions of what will induce others to enter the market. Where the critical barrier is economies of scale the 'Sylos postulate' might be adopted. Assuming that potential entrants fear that the output of the existing sellers will be maintained in the face of entry, the exist-ing sellers produce a quantity (and sell at the corresponding price) such that the marginal demand curve juxtaposed to the ordinate axis is always below the long run average cost curve (Figure 4.2). Essentially, the existing sellers produce sufficient output to prevent potential entrants from generating sufficient additional market demand (by price competi-tion) to operate at a profitable scale. A different entry-preventing strategy is a target-rate-of-profits on capital policy, where the target-rate is limited to the extent 25 R.F. Harrod (1952, p. 147). 2 6 F. Modigliani (1958, p. 217). 65 IRAC BC - marginal demand curve (B C when juxtaposed to ordinate axis) q-7 - the output of existing seller s AC - the demand curve LRAC - the long run average cost curve Figure 4. 2 The 'Sylos Postulate' of the barriers to entry. This very plausible strategy, however, does not determine a unique price and output position. In the following dia-gram (Figure 4. 3), if target profits TT are desired, any point along the horizontal line segment AB is feasible and satisfactory. PROFlf CURVE Figure 4. 3 Target Profit and Profit Curve In the following analysis I assume that a high output-target-rate policy is adopted to prevent entry. The highest level of output con-66 sistent with the target-rate (position B) is selected. This strategy has the feature of combining elements of the 'Sylos postulate' with the target rate. Profits are kept low in keeping with the barriers and output is pushed high to prevent accommodating entrants at a profitable scale. The target-rate and amount of capital determine the desired level of profits. With expected output and revenues the firm can calculate what the average gross-margin over variable costs must be in order to pay the overhead and still achieve the desired level of profits. Alge-braically, m = P- q-(M+w L) M+w L TT + F M +• w L TT + F j M +- w* L q. q 1 t TT f F AVC ' q with TT = ft" = f • K 1 r-K+F m = ~AVC~ • — where: m - gross-margin p - price q - output M +• w L - variable costs TT - profits F - overhead AVC - average variable costs (in the following analysis a constant AVC function is assumed) 67 TT - target profits r - target-rate K - amount of capital Thus the margin must equal the ratio of target gross-profits (overhead plus profits) per unit to average variable costs. Clearly during upswings with rising output the gross-margin must fall given target profits ( — ^ ) < 0). 2^ During downswings the gross-margin o q must rise. This counter-cyclical behaviour is opposite to that predicted in standard competitive markets. It should be noted that the gross-margin will vary inversely with wage and material price movements. Such movements will raise average variable costs and Z^AYC i s * e S S t h a n z e r o - T o t n e extent that these factor prices vary pro-cyclically, this factor price-gross margin relationship is a further source of counter-cyclical gross-margin behaviour. In classical competitive markets, it will be recalled, gross-margin behaviour was independent of general factor price movements. Thus this target-rate model generates gross-margin behaviour strikingly different from competitive gross-margin behaviour. These results, however, are subject to four qualifications. F i r s t , where cyclical changes in the capital stock are not in-significant, a ceteris paribus output-margin relationship is not appro-27 Although I assume a constant AVC function in this analysis, it should be noted that even if AVC varies with output, the counter-cyclical result holds provided total variable costs (AVC* q) rise in upswings, and fall in downswings. 68 priate in determining the cyclical behaviour of the margin. To ensure counter-cyclical behaviour it must be assumed that in upswings (down-swings) output rises (falls) relative to the capital stock. Similarly, factor price movements must be relative to the capital stock. Second, unless sellers.-perceive unit elastic demand curves, a problem concerning the interrelationship between revenues and the margin arises. When sellers raise their margin in the face of lower demand curves, they are adjusting (raising) the share of gross-profits out of revenues. If the demand curve is not unit elastic this adjustment of the share of revenues will in itself alter revenues. In Appendix A it is demonstrated that only if the demand curve is not too elastic can the seller successfully raise the margin to maintain target profits in a down-swing. Only in such a case will the greater share of gross profits out-weigh the revenue effect of the further decline of output caused by the higher margin. However, since sellers in North American manufacturing ? 8 industries are notorious for perceiving low elasticities of demand ° and since studies of manufacturing industries report low price e l a s t i c i t i e s , 2 9 this qualification may not be as important as it may appear. Third, the target-rate objective may refer to an average rate over a period longer than typical business cycles. This would be the case if existing sellers assumed that potential entrants would consider long run averages rather than, for example, annual rates of profit. In such a situation the gross-margin would equal the standard volume (an 2 8 Nourse (1957). 29 H. Houth.hakken and L. Taylor (1970, p. 165); B.L. Scarfe (1972, pp. 145-6). 69 expected average over several years) overhead plus profits per unit divided by average variable costs. The gross-margin would then not respond to cyclical fluctuations in output. Counter-cyclical gross-margin behaviour still could result, but this would be due to pro-cyclical wage and material price behaviour, not to output fluctuations. While this is an important qualification it should be noted that Blair found support for an annual target-rate model. Examining the annual profit rates of five large U.S. corporations, 1953-68, he found that "noteworthy 30 deviations from target occurred on only a few occasions". Fourth, even with an annual target-rate objective, the target-rate itself may vary over the cycle. The barriers to entry underlying the target-rate may vary cyclically. However, where economies of scale are important (as in Canadian manufacturing industries), this would only serve to further the counter-cyclical margin behaviour. With smaller markets in slumps and, therefore, with economies of scale relative to the market that much greater, the target-rate could be higher. In booms the target-rate would be lower. This possible counter-cyclical variation in the target-rate would further the counter-cyclical gross-margin behaviour because higher target-rates entail higher gross-margins ( —— / 0 ) . In sum, though qualifications must be kept in mind, this high-volume target-rate objective derived from an entry-preventing model _ John Blair (1972, p. 491). Also see R. Hall and C. Hitch (1951) discussed on p. 79 below. 70 31 generates counter-cyclical gross-margin behaviour. It should be noted that this high-volume target-rate objective could be derived not just from fear of entry, but analogously from con-siderations of any factor which would undermine future profits. F i r m s may limit their profits to a target-rate because of fear of government intervention, fear of alienating customers or fear of irreversible wage demands. Thus any of these factors if acted upon could lead to the same counter-cyclical margin behaviour as determined in this entry-preventing model. 3. Managerial Utility Maximization It has long been recognized that modern corporations are not the traditional owner-operator proprietorships that underly the rationale for profit maximization in the long or short run. Equity financing has led to widespread separation of ownership and control. In Berle and Means* seminal study, for example, separation of ownership and control by management control or legal devices was found for over 30% of U.S. 32 non- financial assets. In the 1937 Royal Commission on Price Spreads, 3 3 broadly similar results were determined for the Canadian economy. Where managers are in control it is necessary to consider their objectives when analyzing the gross-margin behaviour of the cor-Similar conclusions were developed geometrically by H. A. Cohen (1971). Sylos-Labini (1969) also derived similar results. 3 2 A. Berle and G. Means (1932). 33 Royal Commission on Price Spreads (1937, pp. 15-18). 71 poration. Profit maximization will only be the guiding objective of the corporation if the managers adopt it. However, considerations of managers' utility functions have suggested several plausible alternatives to profit maximization. Some objectives, though different from profit maximization, demand profit maximization. For example, if M a r r i s ' growth maximi-34 35 zation is the objective, profits must be maximized in each period. The growth objective will entail different dividend policy but no different gross-margin behaviour. Gross-margin behaviour different from a profit maximizing model will only result when the objective requires that some profits be foregone. One such objective is Baumol's short run sales maximiza-tion. In pursuit of prestige or whatever, managers push sales beyond the profit maximizing amount up to a point where minimum tolerable profits are achieved. Minimum tolerable profits are determined by the rate which is required to satisfy finance needs and to appease the stock-holder s. This objective, however, is identical to a high-volume target-rate objective. The only difference from the entry-preventing model is that here the target rate is not determined by the level of barriers but by financial considerations. Thus as in the entry-preventing 34 R. M a r r i s (1964). Note that growth maximization refers to growth of assets, sales or output. In the growth maximization model these all grow at the same rate because only under such circumstances is the growth rate sustainable. 3 5 J.H. Williamson (1966). 3 6 W.J. Baumol (1958). 72 model the gross-margin will behave counter-cyclically. Only if the minimum tolerable rate of profits varies sufficiently pro-cyclically 37 would this result be denied. A more complex approach than Baumol's sales maximization 38 is an analysis offered by O.E. Williamson. He postulates a mana-gerial utility function depending on salaries, staff, discretionary invest-ment funds, and management slack absorbed as cost. Salaries and staff enhance managers' own well-being and prestige. Investment funds other than those required to maintain and augment appropriately plant capacity provide the freedom to pursue personal interests. Excess costs allow the managers to relax somewhat in the pressure toward absolute efficiency. In his model this managerial utility function is maximized subject to a profit constraint. In equilibrium, potential profits will be absorbed by excess salary and staff costs, excess production costs (due to inefficiency), reported profits for discretionary investment funds, and reported profits for taxes, dividends and internal growth requirements. Reported profits for taxes, dividends and internal growth are fixed in amount (equal to the profit constraint). The other three factors are in amounts such that the marginal satisfaction from each of them is equal. The important point is that under normal conditions reported profits do not equal the "37 Unfortunately, further speculation on the cyclical behaviour of Baumol's minimum tolerable rate is impossible because in his analysis, it is not a very well-defined variable. 3 8 O.E. Will amson (1963).73 potential level of profits. Excessive costs make up the difference. Moreover, when potential profits are higher, for example during a boom, this difference is greater. . Excessive costs increase to share in the increase of potential profits. The gross-margin equals revenues minus variable costs all divided by variable costs. Revenue behaviour in this model is the same as that in a profit maximizing model. Output and revenues are such that potential profits are maximized. (Williamson is assuming that managerial inefficiency relates only to costs, not to the determination of the optimum level of output). Cost behaviour in this model, on the other hand, is different. Excessive variable costs rise in booms and fall in slumps. Assuming this pro-cyclical variation holds true for excessive variable cost per unit and that this variation is proportionally greater than necessary unit cost fluctuation (as it would be with a linear cost function), in Williamson's model variable costs would be more pro-cyclical than in a profit maximizing model. In other words, the margin behaviour would be more counter-cyclical. The common feature of Baumol's and Williamson's models is that at least some potential profits are absorbed in excessive costs or lower prices - in Baumol's model to enjoy more sales; in Williamson's model to relax. If this causes average variable costs to vary more pro-cyclically relative to price fluctuation, the margin will vary less pro-cyclically (or more counter-cyclically) than in the short run profit maxi-mization model. Of course, as in the short run profit maximizing model, 74 there is little reason to expect gross-margin behaviour to be exactly the same as competitive gross-margin behaviour. Indeed, insofar as the short run profit maximizing model generates less pro-cyclical gross-margin behaviour than the competitive model, a managerial utility maximizing model suggests even less pro-cyclical behaviour (if not opposite, i.e., counter-cyclical behaviour). In view of the wide spec-trum of possible objectives that managers may pursue, not much else can be said. 4. Behaviourist Considerations In a fascinating overview of micro price theory, B.J. Loasby noted that in traditional theory "pricing behaviour is i r r e -levant to price theory". 3 ^ F i r m s are merely points where external forces are received and, given the objectives, acted upon. Prices emerge, but what internal behaviour or processes lie behind are ignored. The relatively new behaviourist school has attempted to change this emphasis. Recognizing that modern firms typically are large complex corporations with considerable discretionary power, pro-ponents of this school focus on the organizational structure in which, and the processes by which, decisions are made and implemented. Cyert and March^" suggest four areas in the internal workings of a corporation which must be considered in a behaviourist theory of the firm: goal formation, formation of expectations, delinea-"39 B.J. Loasby (1971, p. 884). 40 R. Cyert and J. March (1963, p. 1). 75 tion of alternatives, and implementation of the chosen alternative. Cyert and March (and others) have not, nor do I attempt to present, full theories of these different areas. Their work is mainly experimental and suggestive. Nevertheless, some considerations which arise in these areas do bear directly on gross-margin behaviour. F i r s t , regarding goal formation, organizational operations demand concreteness. As in Simon's famous example, when the number of needles is not known, orders are not given to find all the needles in 41 the haystack. A concrete goal is established. Similarly in an uncertain and continually changing economic environment, orders are not given to maximize profits. . Specific targets are chosen and sought. Generally, a budget is prepared outlining production goals, sales goals, cost breakdowns, etc., and those in charge attempt to meet 42 the requirements of the budget, not of a hypothetical maximum. Since a target-rate-of-profit underlies the budget, behaviourist considerations suggest that a version of a target-rate-of-profits model may be more realistic than a profit maximizing model. The target-rate will not be fixed. As Weston recently has emphasized, target-rates are not ends, they are part of a dynamic adap-43 tive strategy. If realized profits easily meet or exceed target-profits, barring other factors (e.g., fear of entry), the target-rate will be revised upward. Similarly, if the target-rate consistently is not 41 H. A. Simon (1959). • 42 This is emphasized in Neil Chamberlain (1962). 43 J. Weston (1972). 76 achieved, it will be revised downward. However, other behaviourist considerations suggest that the adaptive movement of the target-rate will be relatively slow. The target-rate and underlying procedures and goals are adjusted on the basis of new information. However, behaviourist theory notes that information is not just given, it must be sought. Moreover, adjustments do not simply occur, alternatives must be specified and new decisions made and implemented. Although theories of information search and change of targets and procedures are very primitive, one obvious point has been made. The impetus toward change depends on what Cyert and March term organizational slack. When business is good and the target is easily achieved, there is less pressure to seek informa-tion and implement new goals and procedures. On the other hand, when business is bad and it is difficult to achieve the target, more information will be sought and new procedures tried in an attempt to maintain the 44 target. Therefore, the target-rate and with it the realized rate of profit will not vary as much as the maximum rate of profit. In slumps the target and realized rate of profit will be very near the maximum rate of profit as all efforts are made to achieve the target. In booms on the other hand, the target and realized rate of profit will not rise as much as the hypothetical maximum because the pursuit of information and new procedures in order to achieve hypothetical maximums decline after 44 R. Cyert and March (1963, Chapter 3). 77 specified goals have been achieved. To the extent less than maximum profits in a boom are caused by higher average variable costs and/or lower prices than profit maximizing levels, the gross-margin will rise less (or fall more) than the short run profit maximizing gross margin. Starting from a smaller gross-margin in the boom and moving close to the short run profit maxi-mizing gross-margin in the slump means that this 'behaviourist' gross margin will fall less (or rise more) in the downswing. In other words, behaviourist considerations suggest less pro-cyclical (or more counter-cyclical) gross-margin behaviour than the short run profit maximizing gross-margin. Like the managerial utility maximizing model this implies a further difference from compe-titive gross-margin behaviour. 5. Oligopolistic Considerations In the preceding four non-competitive models the sellers have been considered in isolation. It is clear, however, that except in the case of pure monopoly, the margin behaviour described depends on consistent parallel action. Without consistent parallel action, radically different behaviour could result. For example, despite the counter-cyclical gross-margin behaviour predicted in a target-rate model, price breakdowns in slumps could lead to strong pro-cyclical gross-margin behaviour even with target-rate objectives. This problem of co-ordination on the part of oligopolists raises two issues. F i r s t , in what ways are objectives modified and 78 actions tempered to accommodate and facilitate consistent parallel action? And second, under what conditions are breakdowns likely to occur ? If all firms were identical in all respects there would be no problem of inconsistent objectives or misinterpreted actions. Each firm, — th of the market, would act alike and respond similarly to changing economic conditions. Different costs, market shares and time horizons, however, preclude such simple harmony. Thus firms in an oligopolistic setting must adjust their objectives and temper their actions to bring about consistent parallel action. In well-established (mature) non-competitive markets the positions of the different firms have been adjusted or have evolved into some basic harmony. Some tacit understanding emerges in order that the firms may act on the recognition of mutual dependence. Neverthe-less, the fear of being misunderstood or disagreed with always remains. Facing somewhat different market and cost conditions, individual firms fear that any individual price action will not be followed or will be con-sidered by others as a competitive challenge. Sweezy argued that such fears can cause kinked demand curve behaviour. 4^ Individual firms perceive a kink at the existing price, assuming price cuts will be followed while price increases will not be followed. Because of this, prices will be inflexible in relation to moderate changes in costs and demand. Under such conditions the gross-P. Sweezy (1939). 79 margin would vary inversely with moderate changes in average variable costs and would remain the same with moderate shifts in demand. Hall and Hitch also argued kinked demand curve behaviour However, they argued that the kink arose from profit changes, not changes in price. Because cost changes are more visible and common than changes in profit potential, price changes in response to cost changes would not be misunderstood. This would especially hold true where trade associations published common cost figures. Thus in Hall and Hitch's analysis,profit stability, not price stability would be predicted. Hall and Hitch saw this kinked demand curve as one of the 47 factors underlying full-cost pricing behaviour. In full-cost pricing, essentially the same as in target-rate pricing, a gross-margin on variable costs is established to cover overhead and desired profits. As in the target-rate model, full-cost pricing will generate counter-cyclical gross-margin behaviour especially if overhead plus profits per unit are calculated on the basis of actual or forecast output. (Hall and Hitch noted in their survey that actual or forecast output was used in over half of the firms which used full-cost pricing; standard output was used in only one-third). In any event, counter-cyclical gross-margin behaviour will result with full-cost pricing to the extent that wage and material prices vary pro-cyclically. This is because of the inverse relation between factor prices and the gross-margin with full-cost pricing. 4 6 R. Hall and C. Hitch (1951). 47 On the importance of oligopoly in generating full-cost pricing also, see R.B. Heflebower (1955). 80 Similar to the kinked demand curve, yet different and in 48 a sense more general, is Galbraith's 'catching-up' argument. Galbraith argued that firms only slowly adjust their price during business cycle fluctuations. In the context of oligopolistic considera-tions, this can be understood as the firms' waiting until everyone recognizes that a price change is appropriate. Because of this slow adaptation 'unliquidated' profits would arise in booms and be exploited in slumps. Although for different reasons, the gross-margin here would behave like the behaviourist gross-margin. To the extent that less than maximum profits in the boom are caused by too low a price (a very likely possibility if wage and material prices rise in the upswing), Galbraith's 'catching-up' argument predicts a less pro-cyclical (more counter-cyclical) gross-margin than in the simple short run profit 49 maximizing model. Sticky prices, full-cost, and 'catching-up' are the result of firms modifying their actions in light of oligopolistic considerations -they reflect attempts to accommodate the oligopoly. What remains to be examined in this section is the second issue, the probability that price breakdowns do occur despite accommodating behaviour. Price breakdowns entail lower prices and margins than otherwise, the extent depending on the intensity of the price war which results. The question here is whether the probability of breakdowns of 48 J.K. Galbraith (1957). AO, See D. McFetridge (1972, p. 189).. McGetridge argues that this 'catching-up of the margin' during a slump appears valid for Cana-dian manufacturing industries. McFetridge's results are discussed below. 81 any given intensity varies over the cycle. Only if this probability varies systematically over business cycle fluctuations will expected (in the probabilistic sense) cyclical gross-margin behaviour be affected by price breakdowns. Otherwise, the probability of price breakdowns will entail an expected gross-margin lower than a monopoly gross-margin 50 as much in a boom as in a slump, leaving cyclical behaviour unaffected. Clearly, the temptation to cut prices at the expense of others is great in a slump. When business is bad, sellers are ready to try drastic measures to improve their position. This is especially the case when a firm's solvency is at stake. On the other hand, during a slump the fear of price cutting is that much greater. Competitive price cuts will be more quickly met, with a rapidly falling price for al l . It is on the basis of this latter point that many argue co-ordinated 'monopoly-like' action is more probable in slumps than booms. Abramovitch writes, for example: "So far as these considerations go, then, there is greater likelihood in depressions that prices will ap-proximate the monopoly figure than in periods of pros-perity. Where retaliation is more certain, it is more likely to be avoided and conversely when it is less certain". ^1 50 Algebraically, Po = Pm(l-Pr) + Pb-Pr Pb = cXPm Po = P m ( l - P r ( l - <* )) with constant Pr, ^ P o = ^ P m Po Pm Po - expected oligopoly price; Pm - monopoly price; Pb - breakdown price; Pr - probability of breakdown. 51 M. Abramovitch (1938, p. 203). 82 Or, as John Robinson writes: "Since the fear of loss is more powerful than the hope of gain, the tendency toward restrictive combinations is stronger in a slump than in a boom". ^ 2 When Cyert examined this issue he found no cyclical ten-53 dency regarding price breakdowns. To the extent that Cyert's results are valid in general, the consideration of price breakdowns would not affect gross-margin behaviour. To the extent that Abramovitch and John Robinson's arguments hold true, expected oligopolistic gross-margins would be more counter-cyclical than monopoly gross-margins. Summary In this chapter gross-margin behaviour was examined in competitive and non-competitive markets. In classical competitive markets the behaviour of the gross-margin is technologically determined. I-h? what I term the standard case, pro-cyclical gross-margin behaviour results. This is certainly the case when the widely observed linear cost function applies. It is also the case with CES production functions when the elasticity of substitution is less than one (an observed attribute of CES production functions in manu-facturing industries). Because no general model of non-competitive markets exist, gross-margin behaviour in five different models was examined. In each model different assumptions or different considerations were involved. ~52 Joan Robinson (1936, p. 291). 5 3 R. Cyert (1955). 83 The five models taken together suggest the following results: It is very unlikely that non-competitive margins behave exactly the same as competitive margins. In many cases non-competitive margins will behave less pro-cyclically than competitive margins (i.e., will rise propor-tionately less in upswings, fall less in downswings). In the simple short run profit maximization model less pro-cyclical behaviour than competitive gross-margins would result unless the elasticity of demand varied sufficiently pro-cyclically. More-over, in Williamson's managerial utility maximization model, in the behaviourist model, in Galbraith's 'catching-up' argument and when the probability of price breakdowns is considered, the gross-margin could behave less pro-cyclically than the short run profit maximizing gross margin. This furthers the difference from competitive gross-margin behaviour. In many cases non-competitive gross-margins behave not only quantitatively but also qualitatively differently from standard com' petitive gross margins. In other words, non-competitive gross margins may vary counter-cyclically. This would occur in the short-run profit-maximization model if the elasticity of demand varied sufficiently pro-cyclically. More important, this would occur if an annual target-rate or full-cost pricing objective was adopted. Target-rate or full-cost pricing could arise where fear of entry affects short-run pricing, where managers pursue a 84 profit-constrained sales maximization objective or in an o l i -gopolistic setting (a. la Hall and Hitch). As stated at the outset of this chapter, the purpose here has been to establish theoretical reasons why gross-margins in non-compe-titive markets may behave differently from competitive gross-margins during business cycle fluctuations. In a non-competitive market where any one of the five non-competitive models are appropriate, gross-margin behaviour different from competitive behaviour has been established. 85 C H A P T E R V PREVIOUS EMPIRICAL STUDIES In the following chapters I discuss the results of my empirical investigations of gross-margin behaviour in Canadian manu-facturing industries. The empirical investigations deal with three is sue s. F i r s t and foremost, do gross-margins in manufacturing industries behave in an unambiguously non-competitive manner? That is, do they for the most part vary counter-cyclically? In terms of the thesis I argue, this is the critical question. This asks whether it is appropriate to assume standard competitive micro margin behaviour in a macro model. Second, do measures of seller concentration distinguish non-competitive from competitive margin behaviour? To the extent that measures of seller concentration distinguish non-competitive from com-petitive market types, one might expect any counter-cyclical gross-margin behaviour to be positively related to concentration. Third, if counter-cyclical gross-margin behaviour is ob-served, is an annual target-rate or full-cost pricing objective respon-sible? If this is the case one would expect to observe counter-cyclical net profit margin behaviour along with the counter-cyclical gross-margin behaviour. Before proceeding to the results of the empirical investiga-86 tions I first discuss related previous studies. On the Cyclical Behaviour of Manufacturing Gross-Margins Several U.S. studies of the 30's relate to cyclical gross-margin behaviour. Ruggles examined the movement of prices relative to wage and material costs in sixteen major industry groups, 1929-31. ^  Using the Biennial Census of Manufacturers, he found that ine seven industry groups prices fell by approximately the same percentage as costs (within 1%); in five industry groups prices fell by a greater percentage than costs; in four by a smaller percentage than costs. He concluded that for the most part price changes are well explained by cost changes, the gross-margin remaining relatively constant during depressions. Ruggles confined his sample to agricultural product and mineral processing industries. In a comment following Ruggles' 2 article, K. Gordon stated that had other manufacturing industries been included the correspondence between cost and price changes would have been less striking. More important, he pointed out that gross-margin behaviour cannot be directly inferred from a knowledge of percentage change in costs and percentage change in price. Algebraically, margin behaviour and the difference between price and cost behaviour are related as follows: R. Ruggles (1955). 2 K. Gordon (1955). 87 A "rn m 0 rn _ m 0 m •"o P J - A V C J Po-AVCo AVC 1 AVCo m Q Po A V C i AVCo mo 1 m.o • P AVCo - Po- A V C 1 AVCo- A V C } 1 Wo AVCo- AP-Po- A A V C AVCo - A V C 1 1 JPo Po A V C i A P Po A A V C A V C o P o A V C o 1 m o A V C o AVC i A P Po A A V C A V C o where: .m P A V C 1 Mo 1 1 +• A A V C A V C o A P A A V C Po AVCo gross margin price average variable (i.e., direct) costs second period initial period It is clear, therefore, that the degree of margin change depends not only on the difference between price and cost behaviour . A P A A V C x , , , • •„ , r / A A V C v ( —= TT77=— ), out also on cost behaviour itself ( , — ) and on the Po AVCo , AVCo initial margin ( l mo). It is interesting to note that in a depression, with costs falling ( A AVC < 0), the change in the margin is greater than the 88 difference between the change in price and in cost. In other words, even a difference between price and cost change of less than 1% could 3 still entail a large margin change. Thus Ruggles 1 conclusion is not that strong. The constancy of the gross-margin is not well established. Nevertheless, an interest-ing result does emerge from the above formulation. Where prices fall by more than costs, i.e., where gross margins do fall, industries with smaller declines in costs and industries with a higher initial gross-margin will, cet. par., experience a smaller decline in the margin. Since smaller declines in costs and higher margins are generally associated with non-competitive industries this suggests that their gross-margins may behave less pro-cyclically than other industries. In a study prior to Ruggles', Neal also used the Biennial 4 Census data to determine if prices were well explained by cost changes. Over a large sample of manufacturing industries (accounting for 59% of total manufacturing employment for 1929-31, 57% for 1929-33), he cal-culated expected price changes assuming that price minus average variable (direct) cost remained the same. Comparing expected price changes with the actual he found high correlations (.85 for 1929-31 and . 92 for 1929-33). Neal concluded that price changes were well explained by 3 For example, if the gross margin were . 5 and costs fell by 25%, the magnifying factor would be ?fio + 1 1 &o ' f A AVC . 5 + 1 _ _ i _ 4 A V C o .5 1 - .25 4 A. C. Neal (1942). ' 89 changes in direct cost. The interesting point though is that this result does not imply a constant gross-margin, rather a gross-margin which varies inversely with direct costs. Neal's result states that, for the most part, the numerator of the gross-margin (P-AVC)remains constant. Therefore, the gross-margin will vary inversely with the denominator (AVC). To the extent that average variable costs vary pro-cyclically, which they will with sufficiently pro-cyclical wage and material prices, Neal's result implies counter-cyclical gross-margin behaviour. In fact, counter-cyclical gross-margin behaviour had been 5 observed in an earlier study by Dunlop. He examined percentage changes in direct costs relative to prices for several U.S. manufacturing industries. Generally, he found prices falling proportionately less than costs in downswings, rising less in upswings. This implies counter-cyclical gross-margin behaviour. Specifically, looking at annual changes in six industries, 1928-35, he found that steel, boot and shoe, paper, and tobacco exhibited marked counter-cyclical gross-margin behaviour. Cottons and woollens showed little variation. (Because, as in Ruggles' study, gross-margin behaviour is inferred from price behaviour in relation to costs, the 'little variation' of cottons and woollens is subject to the same criticism as discussed regarding Ruggles' result. What may appear as little variation from percentage changes in prices and costs may indeed be substantial gross-margin change). 5 John Dunlop (1939). 90 Looking at 1929-33 changes and 1933-36 changes (i.e., depression and upswing), Dunlop found that three industries (shoes, agricultural machinery and autos) exhibited counter-cyclical gross-margin behaviour. The gross-margins of the other four industries in his sample (leather, cottons, woollens and paint) did not exhibit any cyclical tendency. An aggregative study by Tsiang^ tends to support Dunlop's and Neal's results. Examining the U.S. manufacturing sector from 1919 to 1938, he found that the gross-margin varied inversely with average variable costs from 1921 to 1938 and inversely with output from 1929 to 1938. Tsiang postulated that the counter-cyclical gross-margin behaviour from 1929 to 1938 was due primarily to pro-cyclical variation in average variable costs. As he noted, this would be con-sistent with standard volume target-rate or full-cost pricing. In more recent studies price rather than gross-margin be-haviour has been the focus of attention. In these studies price equations in terms of cost and demand variables are estimated with multiple re-gression techniques. Because price movements in relation to average variable costs (i.e., holding them constant) describe at least the direction of gross-margin movements, (again, as discussed regarding Ruggles' results, the extent of gross-margin charges cannot be directly inferred) these studies reveal some evidence regarding cyclical gross-margin be-haviour. 6 Sho-Chieh Tsiang (1947). 91 Several U.S. and U.K. studies of this nature have generated seemingly contradictory results. Eckstein and Fromm's aggregative study of U.S. data, 1954-65, suggested elements of com-7 petitive and of target-rate behaviour in the price equation. In com-petitive fashion, price (and, therefore, the margin) respond positively to measures of excess demand (inventory disequilibrium and unfilled orders). On the other hand, in keeping with a standard volume target-rate model, normal in contrast to actual unit labour costs and other target-rate variables are found to be significant. Different results were obtained by Yordon. In his sample of 14 U.S. manufacturing industries, 1947 to 1958, he found that demand (capacity utilization) was insignificant in explaining price behaviour. o Different results again were found by Laden. He studied the U.S. manufacturing sector, 1954 to 1966, and found evidence of annual full-cost or target-rate pricing. He found that demand (income) has a negative impact on price (and, therefore, the margin). Furthermore, capital cost was found to have a positive impact on price. Studies of U. K. manufacturing by Neild and by Godley and N o r h a u s ^ support the standard volume target-rate model. Demand variables generally are not significant. Normal unit direct costs, on the _ O. Eckstein and G. Fromm (1968, p. 1171). 8 W. Yordon (1961). 9 B. E. Laden (1972). 10 R. Neild (1963); W. Godley and W. Nordhaus (1972). other hand, explain price behaviour well. Challenging Neild's results, Rushdy and L u n d ^ found that excess demand variables were significant in explaining price behaviour in the U.K. To a large extent the different results stem from the use of 12 different demand variables. Although generalizations are difficult from the above price studies, it appears that variables reflecting the level of demand (income, capacity utilization) have no effect or a negative effect on prices and the gross-margin. Variables reflecting excess demand (unfilled orders, inventory depletion) have a positive effect. - An examination of two recent Canadian studies for the Prices and Incomes 1 3 Commission suggests that such generalizations do hold for Canadian data. It is important to note regarding the price studies already discussed and the two Canadian studies to follow that it is demand variables as opposed to excess demand variables which are relevant when examining the cyclical behaviour of prices and margins. Cyclical be-haviour means behaviour in relation to cyclical fluctuations of demand. In his study of 57 sectors of the Canadian economy, 1961-69, Scarfe concluded that price tends to respond positively to demand (capa-city utilization). However, a close examination of his final equations reveals that in a majority of cases demand was not significant or an T l F. Rushdy and P. Lund (1967). 1 2 Of course, different sample periods and data sources, and different problems associated with the data, can also be responsible for the dif-ferent results. 1 3 B.L. Scarfe (1972); L. Taylor, S. Turnovsky and T. Wilson (1973). accumulated lag function of demand was significant (34 out of his 57 sectors). As he noted, the accumulated lag result relates the rate of price change to capacity utilization, not price change itself. Moreover, in 14 of the 23 cases where either an unlagged, one-quarter or five-quarter lag of demand was significant, trend labour costs or no labour costs entered the equation. The significant positive effect of demand in those cases could be due to pro-cyclical variation of actual unit labour costs, not a traditional demand-induced price change. While Scarfe's results indicate that in a majority of cases demand does not affect prices (and margins), Taylor, Turnovsky and Wilson, in their study of Canadian manufacturing as a whole and of 13 two-digit manufacturing sectors, found that excess demand (inventory depletion and unfilled orders) generally had a significant positive effect on prices. To summarize, both the 30 's gross-margin studies and the more recent price studies suggest that counter-cyclical or at least an absence of pro-cyclical gross-margin behaviour may be common in manufacturing industries. Moreover, some of the studies suggest that an inverse gross margin-average variable cost relation could be playing an important part in generating this non-competitive behaviour (i.e., be-haviour inconsistent with the standard competitive model). On Concentration and Gross-Margin Behaviour In Neal's study, the extent to which actual prices conformed to his expected prices was regressed against corresponding concentration 94 ratios. He found that the difference between actual and expected prices was inversely related to concentration. For the period 1929 to 1931 the correlation coefficient was -.25; for the period 1929 to 1933 the correlation coefficient was -.19. This result indicates that price minus average variable cost rose for highly concentrated industries relative to less concentrated industries. However, Neal himself noted that average variable costs in highly concentrated industries also rose relative to average variable 14 costs in less concentrated industries. Therefore, the relative behaviour of the gross-margin (price minus average variable cost divided by aver-age variable cost) is ambiguous. Neal's result, therefore, does not necessarily indicate a gross-margin behaviour-concentration relation. Studies of more recent U.S. data suggest that during down-swings, at any rate, a gross-margin behaviour-concentration relation 1 5 may exist. In these studies, a different formulation was adopted. Essentially, price changes are regressed against changes in unit material costs, unit labour costs, output and concentration. The argument is that any effect concentration has on price change reflects an adjustment of the gross-margin unrelated to cost or demand factors. A positive concentra-tion effect during a downswing would then indicate gross-margins of highly concentrated industries rising relative to unconcentrated industries because of concentration, not different market conditions. _ A.C. Neal (1942. p. 119). 15 H. Levinson (I960); W. Yordon (1961); L. Weiss (1966). 95 Looking at price changes for the period 1947 to 1958, Levinson found a strong positive concentration effect. Looking at price change for the period 1953 to 1959, Weiss also found a positive concen-tration effect. Since both these periods compare years of high demand to low demand, the studies suggest that margins in highly concentrated industries do rise relative to margins in less concentrated industries during downswings. Weiss repeated his tests for the period 1959 to 1963, a slight upswing. Here he found an insignificant negative correla-tion. Yordon also examined the period 1947 to 1958. He used a sample of seven concentrated and seven unconcentrated industries. He regressed these two groups separately against cost and demand variables and found that he could not reject the hypothesis (90% confidence level) that the responses of the two groups were the same. Nevertheless, he did find a positive, though insignificant price-to-utilization response for the unconcentrated industries and a negative, though insignificant, price-to-utilization response for the concentrated industries. This is consist-ent with more counter-cyclical behaviour of concentrated industries. Two recent studies of Canadian manufacturing industries have found results similar to those of the recent U.S. s t u d i e s . ^ In his study, McFetridge examined three time periods, 1957-61, 1961-64 and 1964-68, periods he described as a downswing, an upswing and a period of high stable demand. For the 1957-61 downswing he found a positive concen-16 ' D. McFetridge (1972); K. Dennis (1973). 96 tration effect. For the other two periods significant correlations were not found. McFetridge interpreted his results as follows: "The oligopolistic sector used the contractionary phase of the business cycle to rebuild margins. There was no such activity during the later periods ."17 These results are similar to Weiss' findings for the U.S. K. Dennis has also examined Canadian manufacturing industries. Dennis divided his sample of 90 three- and four-digit industries into high concentration and low concentration industry groups. Arguing that firm size is another indication of market power, he also divided his sample into large and small firm size industry groups. He then regressed the groups separately in order to compare the R 's of the different groups' price response equations. Using 1962-63 as his reference downturn and 1965-66 as his reference upswing, Dennis found that prices in the highly concentrated group were more rigid during the downturn, no different during the up-swing. This is consistent with the relative margin behaviour suggested in the above studies. It is interesting to note, although not explored in this thesis, that Dennis found the exact opposite effect of firm size. The large firm group was more rigid during the upswing, no different during the downturn. Both the U.S. and Canadian studies suggest that highly con-__ D. McFetridge (1972, p. 189). 97 centrated industries do behave differently from less concentrated indust-ries. Their gross-margins tend to rise more or fall less (whatever the case may be) than the others during downswings. However, a very serious question with regard to these studies remains. The question is whether concentration is an appropriate variable to distinguish between non-competitive and competitive price and gross-margin behaviour. In his seminal studies on cyclical price behaviour, Means adopted the concepts "administered" and "market" prices in order to 18 distinguish between non-competitive and competitive prices. Market prices, as in the classical competitive model, emerge from the inter-action of many buyers and sellers. They change frequently in response to changing conditions of demand and supply. Administered prices are set by the seller, , by administrative action, and held constant for a period of time. Clearly, it is the administered price versus market price distinction which is directly related to theory comparing non-competitive with classical competitive price and margin behaviour. Moreover, the relationship between concentrated industries and administered prices is not as obvious as some might think. As Nelson wrote in a follow-up to Means' initial studies: "To say that a business man has a price policy neces-s a r i l y means he has some degree of latitude in determining the prices of the commodities he has to s e l l " , 19 G. Means (1935); idem (1939). 9 Saul Nelson (1940, p. 4). 98 and monopoly in an anti-trust sense (high concentration) is not required to have a price policy. In Means' and Nelson's studies market prices were distin-guished from administered prices on the basis of frequency of price change. Then, using correlation tests and comparisons of different fre-quency groups, it was determined that administered prices fell less during the depression and rose less during the upswings prior to and after the 29 to'33 depression. To the extent that prime costs behaved similarly (note that products with volatile input prices were eliminated from the samples), this result would hold true for gross-margins as well. Means did undertake a concentration-price behaviour test. Out of his sample of 617 wholesale prices in the 1939 study, he found 37 for which appropriate concentration data were available. Here he did find a rough correlation, the higher concentrated behaving more like the administered prices (less pro-cyclical). However, it should be noted that his sample included agricultural as well as manufacturing industries and agriculture did have low concentrated industries with market deter-mined prices. A study of only the manufacturing sector might not include such industries which must underlie any significant correlation. In sum, especially in a study of manufacturing industries, it is important to recognize that concentration is not necessarily the most appropriate variable in distinguishing between non-competitive and com-petitive behaviour. The key distinction is between administered and market prices. The results with concentration, therefore, depend in 99 large part on the extent to which concentration distinguishes between these two types of prices. It is interesting to note that Means' and Nelson's administ-ered price result (administered prices being less pro-cyclical) has stood up well in that light of recent studies. In particular, a study by Stigler and Kindahl provides much evidence of not only less pro-cyclical be-2 0 haviour of administered prices, but even counter-cyclical behaviour. The evidence is particularly strong when price behaviour is examined relative to industry specific cycles. Articles by Means and M. Moore 21 point out the striking results. It is also interesting to note that administered price pheno-mena have been evidenced in Canada. In the Royal Commission on Price  Spreads, 1937, it was noted that administered prices and profit margins fell much less than the market-determined prices during the 30's depres-sion. For example, it was noted that from 1929 to 1933 agricultural implement prices fell only 7% while agricultural prices fell 50%. Also, in the meat packing industry "the dominating packer has been able substantially to protect his profit margin, while other branches of the industry, more especially the livestock pro-ducer, have had to bear a disproportionate share of the burden of depression". 22 G. Stigler and J. Kindahl (1970). Gardiner Means (1972); M. Moore (1972). Royal Commission on Price Spreads (1937, p. 60). 100 On Cyclical Net Profit Margin Behaviour In contrast to the predictions of an annual target-rate pricing model, studies of net profit margins do not indicate counter-cyclical net profit behaviour. Rather, in studies of U. S. and U. K. data, pro-cyclical behaviour is observed. In his study of British manufacturing industries, Neild found 23 pro-cyclical net profit margin behaviour. This, he argued, resulted from pro-cyclical output per man behaviour. Edwin Kuh found similar 24 results with U.S. data. Kuh noted that overhead labour could be res-ponsible for the pro-cyclical output per man behaviour and thus important regarding pro-cyclical net profit margins. 25 T. Hultgren, in his study of U.S. manufacturing industries, noted that reversals late in upswings often occurred. In other words, net profit margins often started to fall well before cyclical peaks. This phenomena, also observed by Kuh, was attributed to falling output per man as the industry approached its capacity. Of course, some sort of 2 b limit price policy could also be at work. In sum, these previous studies do not support an annual target-rate pricing model. Nevertheless, an examination of Canadian data will be pursued. 23 R. Neild (1963). 24 E. Kuh (I960). 25 T. Hultgren (1965). 26 This phenomena could also be due to reversals in the behaviour of inventories over sales. See Archibald (1955, esp. p. 263). 101 C H A P T E R VI T H E EMPIRICAL STUDIES: T H E D A T A AND THEIR RELATION TO T H E THEORY In the empirical tests I examine the behaviour of gross-margins and of net profit margins. The behaviour is examined in rela-tion to reference and industry specific cycles and in relation to concen-tration. The sample studied is all three and four digit Canadian manu-facturing industries, over the period 1954-1969, for which data are available. In i960, changes in the Standard Industrial Classification (S. I. C. ) were introduced which present irreconcilable alterations in the definitions of industries. In addition, a new establishment definition and an extension to cover non-manufacturing activities of manufacturing e stablishments were introduced. Because of this, the years studied are divided into two periods: data over the period 1954-1959 based on the 1948 S. I. C. , and data over the period 1957-1969 based on the I960 S. I. C. The tests are divided accordingly. That is, all tests are performed on the data available on the basis of the 1948 S.I.C. and then on the data available on the basis of the I960 S.I.C. Data Requirements and Sources (1) Gross-Margin Data Micro gross-margins are defined as value of shipments (of own produced manufactured goods) minus costs of fuel, electricity, 102 m a t e r i a l s and s u p p l i e s m i n u s wage c o s t s a l l d i v i d e d b y c o s t s of f u e l , V S i - M i - w L i e l e c t r i c i t y , m a t e r i a l s and s u p p l i e s p lu s wage c o s t s ( m i = M i 4- w L i — F o r the g r o s s - m a r g i n tes t s these s e r i e s a r e t aken f r o m the p r i n c i p a l s t a t i s t i c s of the annua l C e n s u s of M a n u f a c t u r i n g . ^ The d e f i n i t i o n s o f the s e r i e s u s e d a n d t h e r e f o r e the p r e c i s e 2 d e f i n i t i o n o f m y g r o s s - m a r g i n a r e those e m p l o y e d b y the C e n s u s . V a l u e of s h i p m e n t s r e f e r s to the f. o . b . s e l l i n g p r i c e t i m e s the amoun t o f o w n m a n u f a c t u r e s o l d . G o o d s a r e c o n s i d e r e d s o l d when c o n t r o l i s r e l i n q u i s h e d ( e . g . , o r d e r a c c e p t e d ) . In four i n d u s t r i e s w h e r e the re a r e long g e s t a t i o n p e r i o d s ( F a b r i c a t e d M e t a l s , A i r c r a f t , R o l l i n g S t o c k , S h i p b u i l d i n g ) va lue of p r o d u c t i o n i s g i v e n i n the p l a c e of v a l u e o f s h i p m e n t s . C o s t of fue l and e l e c t r i c i t y r e f e r s to the l a i d - d o w n cos t of the amoun t u s e d . It i n c l u d e s the r e l a t i v e l y s m a l l amoun t u s e d i n n o n -m a n u f a c t u r i n g a c t i v i t i e s . T h e exac t s o u r c e s a r e : (a) b a s e d on the 48 S. I . C . , 1954 -59 , M a n u f a c t u r i n g I n d u s t r i e s R e v i e w ( S t a t i s t i c s C a n a d a 31 -201 ) , T a b l e II, i s s u e d a n n u a l l y ; (b) b a s e d on the 60 S . I . C , (i) 1965 -69 , The G e n e r a l R e v i e w of M a n u - f a c t u r i n g I n d u s t r i e s of C a n a d a . V o l u m e I, p r i o r to 1969 c a l l e d M a n u f a c t u r i n g I n d u s t r i e s of C a n a d a , S e c t i o n A , S u m m a r y for  C a n a d a (31-203) , T a b l e 8, i s s u e d a n n u a l l y ; ( i i ) 1961 -64 , M a n u - f a c t u r i n g I n d u s t r i e s of C a n a d a , T a b l e 8, 1964 i s s u e ; ( i i i ) 1 9 5 7 - 6 1 , G e n e r a l R e v i e w of M a n u f a c t u r i n g I n d u s t r i e s (31-201) , T a b l e 10, 1961 i s s u e . (The 1957-61 da ta r e f l e c t the I960 r e v i s i o n of the S . I . C . and the new e s t a b l i s h m e n t concep t , but not the r e v i s i o n s c a u s e d b y the r e p o r t i n g of n o n - m a n u f a c t u r i n g a c t i v i t y . • S l i g h t d i f f e r e n c e s b e c a u s e o f th i s n e c e s s i t a t e l i n k i n g the 57-61 da ta w i t h the 61 -69 da ta at 1961) . F o r m o r e d e t a i l on concep t s and d e f i n i t i o n s see M a n u f a c t u r i n g Indus t -r i e s o f C a n a d a (31-203) , 1965 i s s u e , p p . 3 6 - 4 5 . 1 0 3 Cost of materials and supplies refers to the laid-down cost of physical commodity inputs (including repair materials) used in manufacturing activity. It excludes service costs such as advertising, insurance, repair work contracted out, etc. It should be noted that some of the service costs excluded from costs of materials and supplies are not of an overhead nature. There-fore, this Census-defined gross-margin will overstate the theoretical gross-margin because not all variable costs will have been subtracted out. However, on the assumption that the cyclical behaviour of variable ser-vice costs parallels the cyclical behaviour of all other variable costs, the cyclical behaviour of the gross-margin will not be distorted. The overstatement of the theoretical gross-margin will be proportionately the same in the boom as in the slump, leaving cyclical behaviour unaffected. Wages refer to all wages, bonuses, profits shared, etc., paid to production and related employees (including assembly, warehouse, inspector, maintenance and repair workers, and line supervisors). Wages are gross earnings before taxes and all social service contributions. Value added refers to value of shipments plus the change in the book value of inventories of finished goods and work in progress minus the cost of materials and of fuel and electricity. 3 Although not 3 Although adequate for a model of the manufacturing sector, this definition would involve some double counting in a model of the entire economy. This is because service inputs (e. g. , insurance) are in-cluded in value added. 104 required in the calculation of the gross-margin as defined above, this series is required in special tests using change of inventories and tests using value of output instead of value of shipments. The use of book value of inventories necessitates an inven-tory valuation adjustment for a truly correct value added series. This adjustment has not been undertaken and as a result the tests using this series, in particular the tests involving a value of output gross-margin, are biased. As noted in footnote 16, the direction of the bias will be the same as the direction of cost movements. When costs are rising the observed value of output gross-margin will overstate the theoreti-cal margin; the opposite will hold when costs are falling. (Z) Reference Cycle Turning Points The NBER method of identifying reference turning points entails locating a consensus of peaks and troughs of various specific series. The turning points I use in this study are the consensus of three studies of Canadian business cycle or manufacturing sector reference turning points - a "consensus of consensuses" approach. The three 4 5 studies are by White, by Waterman and by Taylor, Turnovsky and Wilson. ^ 4 D. White (1967). White reports the reference troughs and peaks deter-min ' S d when the NBER consensus method is applied to Canadian data. For the period in which I am interested, some of White's figures are taken from a study by Chambers (1958). 5 A. M.C. Waterman (1972). Waterman calculated reference peaks and troughs on the basis of 45 indicators published in the Canadian Statis- tical Review. He used the same methods as in his Australian study, Waterman (1967). 105 The peaks and trough s determined in these studies are: WHITE WATERMAN TAYLOR, et al. Trough Peak •Trrough Peak Trough Peak May 53 Jul. 53 June 54 Apr i l 57 Sept. 54 Dec. 56 Jul. 56 A p r i l 58 Jan. 60 May 58 June 62 A p r i l 58 A p r i l 60 Mar. 61 Jan. 63 Jul. 64 Apr i l 61 Apr i l 66 Mar. 65 Nov. 66 Nov. 67 Jan. 69 Apr i l 68 A p r i l 69 For purposes of this study, ambiguities concerning the exact timing of a turning point pose no problem. In the micro theory, a change in the margin is seen as an adjustment in response to a different level of demand. A l l that need be distinguished are years of relatively low demand (slumps) from years of relatively high demand (booms). With this in mind, the following generally agreed upon 'upswings' and 'down-swings* were selected for my tests. ^  UPSWINGS DOWNSWINGS 1954 - 57 1957 - 58 1958 - 60 1960 - 61 1961 - 66 1966 - 68 1968 - 69 Graphs of output and average variable cost indices for 6 L. Taylor, S. Turnovsky and T. Wilson (-1971). In this study a manu-facturing capacity utilization chart is produced for 1956-69. It is on the basis of this study that McFetridge, in his margin behaviour-concentration study, described the cyclical characteristics of the time periods he examined. 7 W. Daly (1969) suggested a similar breakdown. The only difference is that 1958-60 is described more as a period of little change than as an upswing. 106 the manufacturing sector as a whole, over the period 1954-69, are presented on the following page (for the exact sources, see footnote 1 and footnote >9). The output graph indicates that my reference cycle breakdown does tend to distinguish boom years from slump years. The most dubious periods in light of this series are the 58-60 upswing and 60-61 downswing. The average variable cost graph indicates its parallel behaviour to output over this period. (3) Industry Specific Cycles A method of identifying specific cycles is suggested by Stigler and Kindahl. A cyclical movement is defined as "a change in seasonally corrected monthly output data having the following characteris-tics: 1. The expansion (contraction) must last at least eight months. 2. The level of output rises (falls) by at least 20 percent. 3. The expansion succeeds a period of contraction or at most stability of output, and similarly for contractions. These criteria are intended to identify periods in which one would expect a substantial force to have been exerted on the market price by changing demand". 8 The industry output data I use are seasonally adjusted Q monthly indexes of real domestic product. These indexes provide con-8 G. Stigler and J. Kindahl (1970, p. 491). 9 The exact sources are: (a) based on the 48 S. I. C. , 1954-59, Indexes of Real Domestic 108 stant dollar measures of the contribution of each industry to Gross Domestic Product at factor cost. A Laspeyres base-weight formula is used and a double deflation process employed. (First value of output is deflated; then intermediate goods are deflated and subtracted).^ Unfortunately, attempts to apply Stigler andKindahl's method to these data did not prove fruitful. . Very few cycles were iden-tified. Because of this the monthly data are summed to annual indexes and annual changes "in output are used for the tests where margin be-haviour is examined relative to industry specific output behaviour. (4) Concentration Data A variety of measures (concentration ratio, inverse measure, Herfindahl, etc.) over a variety of variables (sales, employ-ment, assets, etc. ) can be used to indicate levels of industry concentra-tion. Because of the generally observed high rank correlation between the different measures and variables only one measure and variable is used for each sample period. For the industries and sample period based on the 48 S.I.C. the inverse measure of firm concentration of 9 Product by Industry (1961 Base) (Statistics Canada 61-506), Table 8; (b) based on the 60 S. I. C. , (i) 1968-69, Indexes of Real Domestic  Product by Industry (61-005), Table 2, June 1972 issue annual supplement; (ii) 1961-67, Indexes of Real Domestic Product by  Industry, 1961-69, 1961-100 (61-510), Table 2; (iii) 1959-60, Indexes of Real Domestic Product by Industry (1961 Base) (61 -506), Table 5. 10 For a detailed discussion of concepts and methods see Indexes of Real Domestic Product by Industry of Origin 1935-61 (61-505), pp. 35-64. 109 employment (i.e. , the number of firms accounting for 80% of industry employment) is used. For the industries and sample period based on the 60 S.I.C. , the Herfindahl index of f i r m concentration of employment (the sum of the squares of each firm's share of industry employment) is used. These data are available in studies by Rosenbluth and by the De-partment of Corporate and Consumer Affairs. ^ It should be noted that while these concentration figures are based on Census data for a single year, they are applied throughout the entire sample periods. Rosenbluth's figures are calculated with 1948 data; the Department of Corporate and Consumer Affairs figures are cal-culated with 1965 data. However, evidence suggests that these figures are roughly accurate for their entire respective sample periods. Drama-tic changes in concentration have not taken place over the post-war period? It should also be noted that Rosenbluth's industry clas s i -fication is based on the Chief Component System used by the Census prior to the application of the 1948 S.I.C. Nevertheless, with a conversion table I received from the Industry and Merchandising Division, Statistics Canada, I determined that 93 of his industries correspond to 1948 S.I.C. three- and four-digit industries. (5) Net Profit Margin Data Unfortunately, the Manufacturing Census data do not provide any data on net profit margins. Therefore, for the tests examin-^ G. Rosenbluth (1957, pp. 111-13). Department of Corporate and Consumer Affairs (1971, pp. 56-71). 12 Department of Corporate and Consumer Affairs (1971, p. 43). 110 ing net profit margin behaviour corporation financial statistics data are used. ^  3 The corporation statistics are compiled on the basis of a sample of corporate tax returns. Unlike the Census data, unincor-porated businesses are not included. Also unlike the Census data and more important, the reporting unit for the corporation statistics is the firm, not the establishment. This causes classification differences bet-ween the Census and the corporation statistics data even where the same S.I.C. framework is being used. Nevertheless, I assume that concen-tration data and output data which are calculated on the Census industry classification apply well to the corporation statistics classifications. In this study net profit margins are defined as net profits after taxes less non-recurring items (e.g., capital gains and losses) all divided by total revenues. Where reported profits have been adjusted to exclude certain non-manufacturing sources of income, sales replaces total revenues. A second net profit margin definition adds depreciation and depletion to after tax profits. This second definition (a cash flow margin) is also used for the net profit margin tests because of the dubious economic meaning of reported depreciation and depletion allow-13 ' The exact sources are: (a) based on the 48 S. I. C. , 1953-59, Department of National Revenue, Taxation Statistics Part II, Table 4, 1955-62 issue s; (b) based on the 60 S. I. C. (i) 1960-64; Department of National Revenue, Taxation Statistics Part II, Table 4, 1963-66 issues; (ii) 1965-70, Corporation Financial Statistics (Statistics Canada 61-207), Table 2, issued annually. I l l ances. These allowances relate as much to tax laws as to real economic costs. A good part of them, therefore, may reflect hidden profits. The corporation statistics also provide asset data (a reported book value). As well as net profit margins, capital-output ratios are also collected and used in the examination of net profit margin behaviour. Included in capital are inventories, land, buildings and equipment. The precise definitions of these variables as well as the net profit margin variables are given in each issue. It should be noted that the 60 S. I. C. period is examined in two parts. The 60-64 data cannot be linked with the 65-70 data because of widespread regrouping of firms among the industry classifications. Relating the Data to Theoretical Propositions and Concepts Strictly, tests based on these data relate only to observed gross and net profit margin behaviour as measured by their average annual accounting values. To relate these tests to theoretical propositions and concepts several factors must be considered and several assumptions made. (1) The Use of Annual Data In using average annual data I am assuming that a basic annual plan exists and is pursued. If in fact planning periods generally are less than one year (e.g., quarterly), then I am assuming that the annual aggregation of the plans should bear roughly the same relation to the annual aggregation of changing output as the single period plan bears to the single period change in output. If planning periods are longer than 112 one year, e.g., because of multi-year labour contracts, I am assuming that a basic annual plan remains. In other words, I am assuming that the longer period considerations do not dramatically influence and distort planned annual relations. Finally, if firms do have annual planning periods, but not over the same twelve months as reported in the Census or corporation statistics, I am assuming that the overlap period is suffi-ciently great and the non-coincident months sufficiently similar that no major distortions arise. (2) The Use of Industry Data In using industry data to examine micro behaviour I am assuming that intra-industry weight shifts do not seriously obscure the relation between planned firm behaviour and observed industry behaviour. With respect to unconcentrated industries, I am assuming that the firms' weights and behaviour are sufficiently similar to preclude any intra-industry aggregation problems. With respect to concentrated industries, I am assuming that'the weights of the leading firms are sufficiently great and stable, and their behaviour sufficiently similar to preclude such problems. (3) Realization of Expectations and Intentions In examining reported data, one only observes, for example, ex-post values of the gross-margin in relation to ex-post changes in output. These observed relations relate to theoretical planned relations only insofar as expectations and intentions are fulfilled. I 113 assume that a year is a sufficiently long period for information-up-dating and revision of plans to bring about approximate realization of expectations and intentions. With regard to intentions, however, one particular problem should be noted. If unintended inventories exhibit a consistent cyclical pattern, the cyclical behaviour of the gross-margin and net profit margin will be affected. If unintended inventory change is positive, then the observed margin will be less than the planned margin because of the extra cost of production to which no sales correspond. If unintended inventory change is negative, then the observed margin will be higher 14 than the planned because of the extra sales to which no sales correspond. At a theoretical level, not much can be said about the behaviour of unintended inventory change. It is a question of whether the firm underestimate or overestimate sales. However, many would argue 14 This problem also applies to the value of output gross-margin (where output replaces shipments). In this case, a positive unin-tended inventory change, for example, causes the observed margin to understate the planned margin because of the extra cost of pro-duction to which no gross profits correspond (assuming that inven-tories are values at cost of production, not market value. The Census inventory data may be valued at either cost or market value, but likely at some intermediate value). See G.C. Archibald (1955, pp. 258-60). It should be noted that the value of output gross-margin has another problem regarding inventories. Since F.I.F.O. method is generally used to value inventories, reported inventory change is inflated when costs are rising; deflated when costs are falling. This will bias the measurement of the value of output gross-margin in the same direction as cost movement. See Archibald (1955, pp. 265-66). 114 that firms tend to underestimate fluctuations in demand and, therefore, unintended inventory change behaves counter-cyclically. ^  This would imply that observed counter-cyclical behaviour was evidence of an even stronger planned counter-cyclical relation, a convenient circumstance when testing for counter-cyclical behaviour.^ In sum, it should be emphasized that while data are available for the different tests, they do have problems. As always, assumptions must be made to infer from observed phenomena to theore-tical relations. Thus, while the following tests are informative and interesting, a certain degree of skepticism regarding the results should be maintained. C.C. Archibald (1955, p. 263). With regard to the inventory valuation adjustment discussed in the last half of footnote 14, it is interesting to note that with pro-cyclical cost behaviour, the observed value of output gross-margin will have a pro-cyclical bias on this count as well. In other words, observed counter-cyclical behaviour would be evidence of even stronger counter-cyclical gross-margin behaviour, after value of inventories had been adjusted appropriately. 115 C H A P T E R VII T H E EMPIRICAL STUDIES: T H E TESTS AND RESULTS In this chapter the empirical tests and the results of these tests are discussed. As noted at the outset of Chapter V, three issues are examined: (a) Do manufacturing industries for the most part exhibit the pro-cyclical gross-margin behaviour predicted in the standard competitive model? (b) Is any observed counter-cyclical gross-margin be-haviour positively related to concentration? (c) Can any observed counter-cyclical gross-margin be-haviour be explained by annual target-jate pricing with its counter-cyclical net profit margin behaviour also? The tests relating to these issues follow. On the Cyclical Behaviour of Gross Margins Proportionate changes of the gross-margin ( — -) were m o calculated for eighty-one 1948 S.I.C. industries over the period 1954 to 1959; one hundred thirty-three I960 S.I.C. industries over the period 1957 to 1961; and one hundred thirty- six 1960 S.I.C. industrie s over the period 1961 to 1969. These changes were calculated over reference upswings and downswings within the periods to be examined in relation to the reference cycles, and were calculated annually to be examined 116 in relation to annual changes in industry specific output. In the reference cycle tests I find that for a majority of industries the gross-margin behaves counter-cylically; on average the gross-margins behave counter-cyclically; and as a weighted average (weighted to discern the contribution of changing micro margins to the change of the macro margin) the gross-margins behave counter - cyclically. These results are supported by non-parametric tests. Also, the observed counter-cyclical behaviour is not dependent on the exact dates of the reference turning points. Moreover, the counter-cyclical result is not altered by re-defining the micro gross-margin whereby value of output would replace value of shipments. The 1948 S.I.C. period contains one upswing, 1954-57, and one downswing, 1957-58. The proportionate changes of the gross margin m 5 7 - m54 a n d m 5 8 - H 1 5 7 indicate that in a majority of industries m K i r n --54 57 the gross-margin fell during the 1954-57 upswing (49 out of the 81) and in a majority the gross-margin rose during the 57-58 downswing (50 out of the 81). Also over the entire cycle 1954-58, a majority of industries (53 out of the 81) exhibited absolute counter-cyclical (falling in the up-swing; rising in the downswing) or relative counter-cyclical behaviour (falling more or rising less in the upswing than the downswing). The 1960 S.I.C. period contains three downswings, 57-58, 60-61, and 66-68, and three upswings, 58-60, 61-66, and 68-69. Except for the 58-60 upswing (perhaps the most dubious of all the upswings), ^  See p. 105 above. 117 again a majority of gross-margins fell during upswings and rose during downswings. Again over entire cycles (adjacent downswings and up-swings) a majority of industries exhibited absolute counter-cyclical or relative counter - cyclical behaviour. These results are summarized in the following tables. Certainly, a predominance of counter-cyclical gross-margin behaviour is suggested by the behaviour of the majority of industries. In this regard it is interesting to calculate the probability of arriving at these results given the null hypothesis that gross margins do not tend to behave counter-cyclically, i.e., that gross-margins exhibit no consistent cyclical behaviour or tend to vary pro-cyclically. On the assumption of no consistent cyclical behaviour, during any of the five cycles there is an equal likelihood of observing pro-cycli-cal as counter-cyclical behaviour in each industry. The probability of a counter-cyclical observation would be one-half. Assuming each industry observation is independent, the probability of as many as r counter-cyclical observations out of n industries would be given by the binomial distribution value B(r, n; 1/2). On the assumption of no consistent cyclical behaviour or a tendency to pro-cyclical behaviour the probability of a counter-cyclical observation would be less than or equal to one-half. Therefore, the probability of as many as r counter-cyclical observations out of n industries would be less than or equal to the binomial distribution value i'B(r, n; 1/2). 118 T A B L E 7. 1 INDUSTRY GROSS MARGIN BEHAVIOUR DURING UPSWINGS AND DOWNSWINGS No. of Upswing Period  Industries: 54-57 58-60 61-66 68-69 With rising gross-margin 32(39.5%) 73(54.8%) 53 (39.0%) 51(37.5%) With falling gross-margin 49(60.5%) 60(45.2%) 83(61.0%) 85(62.5%) Downswing Period No. of 57-58 (48 57-58 (60 Industries: S. I. C. ) S. I. C. ) 60-61 66-68 With rising gross-margin 50(61.7%) 91(68.4%) 67(50.4%) 91(66.9%) With falling gross-margin 31(38.3%) 42(31.6%) 66 (49.6%) 45(33.1%) 119 T A B L E 7. 2 INDUSTRY GROSS MARGIN BEHAVIOUR OVER ENTIRE C Y C L E S No. of Cycle Period  Industries: 54-58 57-60 58-61 61-68 66-69 With absolute counter-cyclical behaviour 36 41 33 62 61 Relative counter-cyclical 17 41 35 32 36 Total counter-cyclical 53(65.4%) 82(61.6%) 68(51.1%) 94(69.1%) 97(71.3%) With absolute pro-cyclical 18 25 41 25 19 Relative pro-cyclical 10 26 24 17 20 Total pro-cyclical 28(34.6%) 51(38.4%) 65(48.9%) 42(30.9%) 39(28.7%) 120 Looking back at Table 7. 2, it is now possible to calculate the probability of these results on the null hypothesis that gross-margins do not tend to behave counter-cyclically. A normal distribution approxi-mation is used in the calculations. Quite clearly, on the basis of data shown in Table 7. 3, the null hypothesis would have to be rejected. Gross-margins do tend to behave counter-cyclically. Not only the behaviour of the majority of industries, but also the average and weighted average gross-margin behaviour indicate a predominance of counter-cylical behaviour over the reference cycles. As shown in Table 7.4, the average gross-margin (the sum of the changes in the gross margins divided by the number of industries) varies counter-cyclically in all cases except for the I960 S.I.C. 58-60 upswing. However, more important than the behaviour of the simple average is the behaviour of the weighted average margin, weighted to discern the contribution of changing micro margins to the change of the macro margin. In Chapter III, the relationship between micro margins and the macro margin was determined as: Jm = ^ 1 Mi. o<i w L i .. (M+wL)i where: cKi = M+w L M+w- L w L 121 T A B L E 7. 3 PROBABILITY OF T A B L E 7.2 RESULTS ON N U L L HYPOTHESIS  OF NO TENDENCY TO COUNTER-CYCLICAL 7BEHAVIOUR r = No. Industries Normal Probability Counter- n = Total No. Approximate on Null Cycle Period Cyclical Industries Value Hypothesis 54-58 53 81 2.66 .004 57- 60 82 133 2.59 .005 58- 61 68 133 .17 .433 61-68 94 136 4. 36 < .001 66-69 97 136 4.88 < . 001 T A B L E 7.4 AVERAGE GROSS MARGIN BEHAVIOUR Average of Upswing Gross Margin Period Changes 54-57 58-60 61-66 68-69 .02709 .03175 -.04513 -.03119 Average of Downswing Margin Period 57-58 (48 S.I. C. ) 57-58 (60S.I.C. ) 60-61 66-68 Changes .06160 .06329 .01904 .05724 122 It was also determined that a change in the macro margin could be broken up into the sum of weighted averages of changes of the component factors. In the breakdown formulae a weighted average of the changes of the micro margins was derived (though not uniquely) which would indicate the contribution of changing micro margins to the change of the macro margin. Table 7. 5 shows that minimum and maximum values of the weighted average of the changes of micro gross-margins were calculated over the reference upswings and downswings. Except for the 58-60 up-swing and 60-61 downswing, these weighted averages indicate counter-cyclical behaviour. These results are very important. They indicate that the changing micro margins have a counter-cyclical influence on the change of the macro margin. Moreover, although the macro margin does not always vary counter-cyclically with the weighted average of the changes of the micro margins because of the behaviour of the other determining factors, as demonstrated in Chapter III, the micro margin behaviour is sufficiently independent of the other three factors to have an independent effect on the macro margin. In other words, the counter-cyclical influence of the micro margins does indeed cause the macro margin to vary more counter-cyclically than otherwise. Thus, the majority of industries, the average and the weighted average over industries all indicate that micro gross-margins tend to vary counter-cyclically over reference cycle periods and that this 123 T A B L E 7. 5 CHANGE IN MICRO MARGIN AND WEIGHTED AVERAGE  O FEGHANGES IN MICRO MARGINS OVER R E F E R E N C E PERIODS-H Weighted Change of Change Micro Up- in Margins swing Macro (Min. -Max. Period Margin Values) 54-57 .0362 -.0310 to -.0250 58-60 . 0367 0340 to 0416 61-66. -.0223 -.0751 to -.0694 68-69 -.0042 -.0224 to -.0210 Weighted Change of Change Micro Down- in Margins swing Macro (Min.-Max. Period Margin Values) 57-58 . 0398 to (48 .0446 S.I.C.) .0404 57-58 .0411 to (60 S.I.C.) .0590* .0476 60-61 .0147 -.0175 to -.0139 66-68 -.0155 . 0189 to . 0222 + A l l values (changes)are written as a proportion of initial year macro margin. * This does not include the change in inventories over wages. Inventory data were not available on the basis of the I960 S. I. C. for 1957. 124 contributes to a more counter-cyclical macro gross-margin than other-wise. It should be noted that this counter-cyclical behaviour is not very sensitive to the exact dates of the reference periods. Annual weighted averages of changes in the micro margins indicate that altering the turning points by a year would not affect the results. These annual weighted averages, portrayed graphically in the appendix of Chapter III, are listed in Table 7.6. For both the 48 and 60 S.I.C. periods they are expressed as a proportion of the average macro margin over the period. Therefore, changes between any two years within a given period can be seen as the sum of the annual changes between the two years. Thus, during the 48 S.I.C. period for example, the weighted average of the changes of micro margins was negative not only when com-paring 1957 to 1954 values (my reference upswing), but also when com-paring 56 to 55, 56 to 54, 57 to 56 and 57 to 55. Also, the weighted average of the changes was positive not only when comparing 1958 to 1957 values (my reference downswing), but also when comparing 58 to 56, 59 to 57 and 59 to 56. In other words, counter-cyclical behaviour would be observed not only with my turning points, 54, 57 and 58, but also with turning points 1955, 56 and 59 or any combination of these and my turning points. One might argue that the observed predominance of counter-cyclical behaviour is due to the specific definition of the micro gross-margin adopted here. In particular, it might be argued that a definition 125 T A B L E 7.6 ANNUAL WEIGHTED AVERAGES OF CHANGES OF MICRO MARGINSf 48 S. I. C. PERIOD Year s 55- 54 56- 55 57- 56 58- 57 59- 58 Weighted Change in Micro Margins 0432 -.0071 to 0687 to 0391 to 0477 0658 to -.co;o62 0438 0375 to .0393 60 S. I. C. PERIOD Years 58- 57 59- 58 60- 59 6 1 - 60 62- 61 63- 62 64- 63 65- 64 66- 65 67- 66 68- 67 69- 68 Weighted Change in Micro Margins 0387 0095 0230 0169 0162 0104 0099 0197 .0329 0163 0012 0243 to to to to to to to to to to to to . 0450 . 0133 . 0258 -.0134 -."0151 -.0062 -.0077 -.0161 -.0306 . 0182 . 0062 -.0228 +• A l l values (changes) are written as a proportion of the average macro gross margin over the respective period. 126 in which value of output replaces value of shipments is more appropriate and would lead to different results. Whether such a definition is more appropriate, i.e., whether it is more appropriate to examine potential gross profits from output than realized gross profits from sales is a matter of conjecture. However, a quick examination of the 48 S.I.C. period indicates that using such a definition does not lead to different results. During the 54 to 57 upswing in a majority of the industries the margins so defined fell (48 out of 81), the average change of the margins was negative (-.01319), and the weighted change in the margins was negative (-.0051 to -.0093). During the 57 to 58 downswing in a majority of the industries the margin rose (51 out of 81), the average change was positive (.03874) and the weighted average was positive (.0242 to .0264). Again, a predominance of counter-cyclical behaviour is indicated. The results above concern the behaviour of gross-margins in relation to reference cycles. It is this micro behaviour in relation to reference cycles which is incorporated into the macro model. However, in terms of micro theory micro gross-margin behaviour is most directly related to industry specific changes in demand. Any reference cycle-gross margin relation would have to be based on an underlying specific cycle - gross-margin relation plus sufficient coincidence of specific and reference cycles. In this regard it is interesting to examine briefly gross-margin behaviour in relation to industry specific changes of output. As 127 noted in the last chapter, annual proportionate changes in output are used instead of the seemingly non-existent specific cycles as defined by Stigler and Kindahl. First, for both the 48 S. I. C. period and 60 S. I. C. period, the data are examined in three groups. The first group includes the proportionate change of the margin for all industries for all years. The second group includes only those industries and years where the pro-portionate change in output is positive. The third group includes only those industries and years where the proportionate change in output is negative. The group averages of the changes of the gross-margins (Table 7. 7) suggest counter-cyclical behaviour in relation to industry specific changes of output. Over the 48 S.I. C. period, when all indust-ries and years are considered, the average of the margin changes is . 01759. With only industries and years where changes in output are positive, the average of the margin changes is smaller, equal to .01129. With industries and years where changes in output are negative, the average of the margin changes is greater, equal to .04151. Similar results are found over the 60 S.I.C. period. With all years and industries the average of the margin changes is .01234. With only positive changes in output the average of the margin changes is smaller, equal to .00114. With only negative changes in output the average of the margin changes is larger, equal to .03951. The counter-cyclical behaviour of the group averages of 128 T A B L E 7. 7 GROUP AVERAGE CHANGES OF T H E MARGINS 48 S.I. C. 60 S. I. C. Group Average Average A l l industries and years .01759 .01234 Industries and years with rising output .01129 .00114 Industries and years with falling output .04151 .03951 T A B L E 7. 8 GROSS MARGIN-INDUSTRY SPECIFIC OUTPUT AND AVC REGRESSION RESULTS A A , 2 Sample Period b * c * R 48 S.I.C. -.1876 -.3799 (. 0810) ( < . 0001) 60 S. I. C. -.2610 -. 5625 (.0002) ( < . 0001) 046 . 093 Probability of the coefficient equal to zero is given in the parentheses. Note that to accept the null hypothesis one-half the pro-bability-must be less than one minus the confidence level. The probability is multi-plied by one-half because the hypothesis es-tablishes one-tailed tests. 129 the micro gross-margin changes suggests counter-cyclical gross margin behaviour in relation to industry specific output fluctuations. A multiple regression test tends to confirm this result. Annual changes of micro gross margins were regressed against annual changes of the industry's output. Because some non-competitive models suggest that average variable cost behaviour may affect gross-margins, annual changes of average variable costs were included as another independent variable in the regression. The regression test (Table 7.8) was run over all industries and years, first for the 48 S.I.C. period, then for the 60 S.I.C. period. The results for both periods indicate significant negative effects (95%. confidence level for one-tailed tests) of output and average variable costs on the gross-margin. H o : (A™, = a f b ( A j L ) f c ( A A V C ) m ij q ij AVC ij i = l,n (# industries) j = 1, m (# years) where: b < 0 ; c < 0 The reference cycle and industry specific results are of paramount significance. Gross-margins do not behave as predicted in the standard competitive model. When incorporating gross-margin behaviour into a macro model this must be recognized and applied. The importance of this for predicting and understanding macro phenomena has been suggested in Chapters II and III and is discussed further in the 130 next chapter. On the Role of Concentration The hypothesis that counter-cyclical gross-margin behaviour is based on a concentration-counter-cyclical gross-margin relation and a predominance of highly concentrated markets is now con-sidered. Insofar as gross-margins in highly concentrated markets behave more counter-cyclically over reference cycles than in less con-centrated markets one would expect at least one of the two following hypotheses to hold in most of the reference upswings and downswings: (a) Assuming an incremental effect of concentration Ho : ( A ™ L ) = a + b C i m i where: b > 0 for downswings b < 0 for upswings Ci - is a measure of concentration (b) Assuming a threshold effect of concentration Ho : (-£-2L) 4 ( AZS) M HC m LC where: HC - refers to a high concentration group LC - refers to a low concentration group where the inequality hypothesized is greater than in upswings; lower than in downswings. 131 Tests of these hypotheses were undertaken for all the 1948 S.I.C. and I960 S.I.C. upswings and downswings. Over the 1948 S.I.C. period the negative of the log of Rosenbluth's inverse measure was used to measure concentration. For the I960 S.I.C. period the un-transformed Herfindahl index was used. A simple correlation test (Table 7. 9) was used for hypo-thesis (a) and a difference of means T-test (Table 7. 10) for hypothesis (b). The results of both tests indicate that concentration has an insigni-ficant effect on gross-margin behaviour. Only in one time period, the 60-61 downswing, is a significant effect of concentration observed. During that period a positive effect of concentration was observed in both the correlation and T-tests (at the 95% confidence level). In the one downswing he examined, the period 1957-61, McFetridge found a significant positive effect of concentration. Repeating my correlation test over his period, I also found a significant positive concentration effect at the 95% confidence level (for a one-tailed test). The beta coefficient equalled .0055 with a probability of .076 of being equal to zero. The R was .0229. Looking at this period alone one would be tempted to infer some counter-cyclical gross margin effect of concentration, at least in downswings. This, in fact, is what McFetridge inferred. However, the results of the four different downswings I examined demonstrate that McFetridge's result is unique to his time period and my 60-61 period. In general, concentration incrementally or in terms of a threshold does 132 T A B L E 7. 9 T H E CORRELATION TESTS Ho : (-^ -—-) = a f b C i m i Upswing Period 54-57 58-60 61-66 68-69 No. Observations (Industries) 81 133 136 136 A b 01111 00261 00022 00042 Prob (b=0)* .4290 . 2954 . 8434 . 8061 0081 0084 0002 0004 Down-swing Period 57-58 (48 S.I.C. ) 57-58 (60S. I.C. ) 60-61 66-68 No. Observations 81 133 133 136 A b 00110 -.00025 00356 00179 Prob (b=0)* 8857 8374 . 0375 . 2429 , 0001 , 0002 0319 0101 Note that the hypothesis establishes one-tailed tests. 133 T A B L E 7. 10 Upswings 54 - 57 58 - 60 61-66 68-69 , A m , A M T H E T-TE5TS Ho :( —)„^4 ( ,^ ) m HC Top 40 vs. Bottom 40 Concentrated Industries Prob. Difference T-Value = 0* + .495 + .482 -.800 -.092 628 637 431 . 888 Im 'LC Top 20 vs. Bottom 20 Concentrated Industries Prov. Difference T-Value = 0* +1.266 + . 137 - .066 + . 762 . 211 . 862 .904 .461 Downswings 57 - 58 (48 S. I. C. ) 57 - 58 (60 S.I. C. ) 60 - 61 66-68 -1.016 - .235 +1.811 + .945 . 315 . 801 . 073 . 352 + .317 + . 347 +1.862 + .607 . 749 . 728 . 070 . 557 * The hypothesis establishes one-tailed tests. 134 not appear to be related to cyclical gross-margin behaviour. One might argue that these negative results are due to some dubious concentration figures. In particular, concentration figures for industries with regionally separated markets, for export industries and for industries facing much import competition do not reflect the true state of competition. Concentration is understated where regional markets exist; overstated in export industries and industries facing import competition. However, further tests indicate that the dubious concentra-tion figures are not responsible for the negative results. The 1948 S.I.C. period upswing and downswing were retested with regional market, high export and high import market industries omitted. In other words, only national market industries were included in the sample. Forty-seven of the eighty-one industries were left for the re-testing. For both the 54-57 upswing and the 57-58 downswing the corre-lation and T-tests again indicated an insignificant concentration effect, (Table 7. 11). One might also argue that these negative results are due to the non-coincidence of industry specific changes in output and reference 2 cycles. The consistently low R especially suggest that industry specific factors should be considered. In this regard a different formulation of the test of the role of concentration was postulated. Here the effect of Rosenbluth's concentration data also indicates which industries have national markets. 135 T A B L E 7. 11 1948 S.I. C. NATIONAL MARKET TESTS (The Correlation Tests (48 S.I.C. National Markets) A 2 Period No. Industries b Prob (Jo ~ 0 ) R 54 - 57 47 .00003 .8745 .0003 5T-S5B 47 .00008 .6906 .0036 The T-Tests (48 S.I.C. National Markets) Top Z0 vs. Bottom 20 Period T-Value Prob (Diff.= 0) 54 - 57 .897 . 379 57 - 58 . 288 . 767 136 concentration on cyclical margin behaviour in relation to industry specific changes in output is examined. The test combines a times series estimation of the elas-ticity of margin behaviour with respect to industry output behaviour and a cross sectional test of the role of concentration. In this test, where the change of the margin in relation to the change in industry output is accounted for, it is hoped that the role of concentration is better isolated. The algebraic formulation of the test is as follows: A m A n for each industry i , ( ) = a i f bi ( a -) j = 1, m (# years) m ij q ij with the coefficients hypothesized to be functions of concentration. a j L = <X f A C i b. = t f S c . 1 i substituting: (~^-) = A C i f X ( " ^ ) + £ ( - ^ ) • C i m ij q ij q ij The hypotheses are: A n (a) With output falling ( -t. <_ 0 ) t n e coefficient A will q be greater than zero (i. e. , given the elasticity of margin response, with higher concentration the margin will move higher). (b) With output rising ( ^  9- y n) the coefficient A will be q less than zero. (c) In both instances of rising and falling output, the coeffi-cient £, maybe different from zero (i.e., the elasti-137 city of margin response to change in output may be affected by concentration). This multiple correlation test was run over the 1948 S.I.C. and I960 S.I.C. periods (1954 to 1959 and 1957 to 1969, respectively). Both periods were divided into samples of rising and falling output to be tested separately. The results follow in Table 7.1Z. Again, the coeffi-cients are not significant and the R 's are low. Thus industry specific tests agree with the reference cycle tests in demonstrating that concentration does not relate to the behaviour of gross-margins. This is true despite the fact gross-margins do not behave as predicted by the standard competitive model. Because this tells what does not explain rather than what does, many possible explanations remain. My explanation is based on the fact that it is the administration of prices, not monopoly in the anti-trust sense, which is responsible for the non-competitive behaviour. As discussed in Chapter V, administered prices may emerge from very low concentrated industries. On Target-Rate Pricing Although not explained by concentration, a predominance of counter-cyclical gross-margin behaviour has been observed. In the theoretical micro chapter one of the models predicting such behaviour was the annual target-rate model. The target-rate model, however, predicts not only counter-cyclical gross-margin behaviour; but also counter-cyclical net profit margin behaviour. In the target-rate model, 138 T A B L E 7. 12 INDUSTRY SPECIFIC OUTPUT CHANGES: CONCENTRATION " E F F E C T REGRESSIONS A A A Sample Period X * y * fc * 48 S. I. C. -.01357 .39125 .03285 where A q < 0 (.3861) (. 5929) (.8714) 60 S. I. C. .00151 .28695 .02357 where A q < 0 ( . 3835) (.5946) (.4431) R' . 0192 0279 * Probability of the coefficient equal to zero is given in the parentheses. The hypothe-sis regarding A establishes a one-tailed test. Sample Period A A * A jr * A 6, * R2 48 S. I. C. where A q > 0 60 S. I. C. where A q > 0 -.00647 (.4639) .00074 ( .4107) .32315 (. 2835) -.19524 (. 2113) .12322 (. 1918) .00599 (. 5365) 0062 0739 Probability of the coefficient equal to zero in the parantheses. The hypothesis regarding A establishes a one-tailed test. 139 net profits assume the characteristic of overhead. As output and sales fluctuate in one direction, the net profit margin (net profits over sales) would vary in the opposite direction. Thus, the validity of the annual target-rate model can be tested by examining net profit margin behaviour. Proportionate changes in the net profit margin were cal-culated for nineteen 1948 S.I.C. industries over the period 1954-59; twenty-five I960 S.I.C. industries over the period 1960-64; and forty-eight I960 S.I.C. industries over the period 1965-69. Paralleling the study of gross-margins, these net profit margin changes were calculated over reference upswings and downswings to be examined in relation to the reference cycles and were calculated annually to be examined in relation to annual changes in industry specific output. Two points must be noted before proceeding to the results. F i r s t , there are much fewer industries in these samples than in the gross-margin samples, with the consequence of less reliable results. There are fewer industries because the corporation statistics data more often than the Census data were aggregated beyond three- and four-digit industries for which concentration data were available. The existence of corresponding concentration data was used as a criterion for inclusion in the sample not only because of planned concentration-net profit margin tests, but also because it signified that the industry grouping defined in the corporation statistics corresponded to a well-defined industry in the economic sense. Second, as discussed in the last chapter, two net profit margins are calculated for each industry: a reported after-tax net 140 profit margin and a reported after tax profits plus depreciation and depletion allowance net profit margin (a cash flow margin). With respect to the cyclical behaviour of net profit margins over reference cycles, the same consistent counter-cyclical behaviour as with the gross-margins was not observed. With two ex-ceptions, both measures of the net profit margin indicate that only a minority of industries had rising net profit margins in any of the three downswings or had falling net profit margins in either of the two up-swings (Table 7. 13). Over entire cycles, both net profit margins measures indicate pro-cyclical behaviour im a majority of industries. (Because of breaks in the data at I960 and 1965, only two cycle periods could be examined: 1954 to 1958 and 1966 to 1969). As noted in Table 7. 14, in a majority of industries, therefore, the net profit margins do not behave counter-cyclically. Clearly, the b i -nomial test as described in the analysis of gross margins would not reject the hypothesis that the net profit margins do not behave counter-cyclically. Not only the net profit margins of the majority of industries, but also the average of changes in the net profit margin behaves pro-cyclically (Table 7. 15). This is true for all periods except for the 66-68 and 6 8-69 periods when the net profit margin is defined to include depre-ciation and depletion. Thus, net profit margin behaviour in relation to reference cycles is not consistent with a target-rate pricing model. The same 141 T A B L E 7. 13 INDUSTRY NET PROFIT MARGIN BEHAVIOUR DURING  UPSWINGS AND DOWNSWINGS Upswing Period No. of Industries 54-57 68-'69- . With falling net 7 9 profit margin (out of 19) (out of 48) With falling cash 9 27 flow margin (out of 19) (out of 48) Downswing Period 57-58 60-61 66-68 No. of Industries (48 S.I.C.) With rising net 9 13 9 profit margin (out of 19) (out of 25) (out of 48) With rising cash flow margin 7 (out of 19) 12 (out of .25) 23 (out of 48) 142 T A B L E 7.14 INDUSTRY NET PROFIT MARGIN BEHAVIOUR Cycle Period: No. of Industries With absolute counter-cyclical Relative counter-cyclical Total counter-cyclical With absolute pro-cyclical Relative pro- cyclical Total pro- cyclical 1954 - 1958 1966 - 1969 Net Profit Cash Flow Margin Margin 8 (42. 1%) 7 (36. 8%) Net Profit Cash Flow Margin Margin 33 17 9 (18. 8%) 23 (47. 9%) 14 11 11(57.9%) 12(63.2%) 39(81.2%) 25(52.1%) T A B L E 7. 15 AVERAGE OF CHANGES OF NET PROFIT MARGINS Upswing Period 54-57 68-69 Average Change Net Profit Margin  .10664 .19425 Average Change Cash Flow Margin  . 16190 -.10055 Downswing Period 57-58 (48 S.I.C.) 60-61 66-68 •.00915 •.04798 •.26223 -.01138 -.09206 .01493 143 holds true for net profit margin behaviour in relation to industry specific changes in output. For industries and years when output is falling, the average of the changes in the net margins is negative; for industries and years when output is rising the average is positive. The examination of net profit margin in relation to both reference cycle and industry specific output behaviour demonstrates that annual target-rate pricing is not what underlies the counter-cyclical gross margin behaviour. The counter-cyclical gross margin behaviour is due to the overhead but not the net profit margin component of the gross margin. Further tests with the net profit margins had been planned to help explain any gross margin-concentration relation. Since no such relation was found these further tests are not required. However, the corporation statistics data available do allow for an expanded model where other industry characteristics can be considered. A more complicated relationship between concentration and the behaviour of net profit margins could yet be found despite the lack of a simple concentration-gross margin behaviour relation. In the expanded model net profit margin behaviour is pos-tulated as a function of concentration, the capital-output ratio, and the level of barriers to entry. Capital-output data are available in the cor-poration statistics data, but only a proxy formulation is possible to cap-ture the level of entry barriers. In the proxy formulation a generally observed profit rate 1 4 4 relation is inverted and substituted into the model. It has been observed by Bain and others that average profit rates are positively related to concentration and the level of entry barriers. Thus average net profit margins would be related to concentration, the level of entry barriers, and the capital-output ratio. Assuming this relation is monotonic (inver-tible), concentration, the capital-output ratio and the average net profit margin should determine the level of entry b a r r i e r s . The model, where net profit margin behaviour is postulated as a function of concentration, the capital-output ratio and the level of barriers, can be written as net profit margin behaviour as a function of concentration, the capital-output ratio and the average level of the net profit margin, with the last variable capturing the direction of the effect of barriers. Algebraically: Ho : — - r - H = F(C, . B) mN Y But, H . = f(C,B) K ... k i N = _ f i _ = K - TT Y Y K K . = g(C, B, — ) K Inverting, B = h (C, — , rn;N) Substituting into F ^Jfk = F (C, h (C, *L( M N ) ) m N * = H<C ."IL. M N ) 145 K_ Y where: m N " ne^ Vro^ margin (in a given year) C - concentration capital-output ratio B - level of entry barriers - average net profit margin (over the period) average profit rate TT K A m ^ A m Note that = is of the same sign as F 0 = This is because H 3 = F 3 • ^ _ B and 3 B _ 1 2 m N ^ m N. m.N B is greater than zero. (Higher barriers afford higher average profit rates, and, given the capital-output ratio, higher average profit margins. Note also that in the final formulation H, the variables C and — include an effect of barriers. Assuming positive correlations among these variables, the estimated coefficients exaggerate the true IS isolated effects of C and _ . To the extent they work in opposite directions, the estimated coefficients underestimate the true effects. This-model was tested with net profit margin behaviour during the reference upswings and downswings. Because these two variables are designed to capture structural features, average net profit margin and the capital-output variables were calculated as the averages over the time periods for which consistent data were available, i.e., averaged over 1965 to 1969 for the upswing and downswing in that period, over I960 to 1964 for the downswing in that period, and over 1954 to 1959 146 for the upswing and downswing in that period. There are two reference upswings with corporation statis-tics data and it is interesting to note that in both (the 68 to 69 and the 54 to 57 upswings) the concentration variable has a significant negative effect. The other two variables are consistently insignificant, (Table 7.16). Afo N With a one-tailed test (Ho : _ < 0), these estimated concentration coefficients are all significant at the 90% confidence level, three are significant at the 95% confidence level. Thus the negative results in the gross-margin-concentration tests must be modified here. Concentration, perhaps in conjunction with barriers (since the concentra-tion variable defined in this test can include the effect of barriers) does have a counter-cyclical effect on the net profit margins during upswings. Unfortunately, whether such an expanded model would also demonstrate a counter-cyclical effect on gross-margins cannot be tested. The net profit margin and capital-output data are not available for the Census-defined industries from which the gross-margins are calculated. There are three reference downturns and significant concen-tEatiojnieffe.cfscare .notobbserv ceid. In fact, for the 66-68 and the 57-58 downswings no significant effects are found. Nonetheless the 60 to 61 downswing does provide an interesting result. It will be recalled that in the gross-margin tests, concentration was found to have a significant positive effect on the gross-margins for the 60-61 downswing. In the expanded model test over this period the concentration effect is insigni-147 T A B L E 7. 16 CONCENTRATION COEFFICIENTS IN THE REGRESSION , A J m \ j K (— ). = a H C i t d ( T U e M N I Concentration Coefficient* Upswing Period 1954-57 1968-69 With net profit margin as dependent variable -.00027 (. 1216) -.01419 (.0842) With cash flow margin as dependent variable -.00082 ( . 063Z) -.02902 (.0968) Probability of the coefficient equal to zero is given in the parenthesis. T A B L E 7. 17 CAPITAL-OUTPUT RATIO COEFFICIENT IN T H E REGRESSION: A m N K (— ). = a f b C i H ( T ) i + e (m N). FOR T H E PERIOD m^ i x Y 1 l 1960-61 Coefficient * With netpprofit margin -.99779 as dependent variable ( . 0148) With cash flow margin -1.8012 dependent variable (.0052) Probability given in the of coefficient parentheses. equal to zero is 148 ficant, but the capital-output ratio has a strong significant negative effect. Thus, assuming a positive capital-output concentration relation, it very well could be that the concentration effect observed on the gross-margins would have been even stronger had capital-output ratios been included in the gross mar gin-concentration regression. Summary of the Empirical Results At the outset of this chapter it was determined that gross-margins in Canadian manufacturing industries tend to behave counter-cyclically. They do not tend to behave in the pro-cyclical or constant manner predicted by standard competitive micro models. In this it was observed that the counter-cyclical micro margins contribute to counter-cyclical or at least less pro-cyclical macro margin behaviour. This is an important result. In terms of this thesis it is the key result. It means that non-competitive gross-margin behaviour must be incorporated into macro models. It is an error to assume standard competitive behaviour. The remainder of this chapter served to demonstrate that this result has not been well-explained. In examining gross-margin be-haviour it was found that concentration was not responsible for the counter-cyclical behaviour. Concentration did not distinguish between competitive and non-competitive behaviour. The tests at the end of the chapter on net profit margins suggest that an expanded model, where the capital-output ratio and the level of barriers are considered may be re-quired in any attempt to distinguish between non-competitive and compe-149 titive gross-margin behaviour. True as this may be, I also believe that a key point is that non-competitive behaviour may emanate from unconcentrated industries. A l l that is required is administered instead of market determined prices. It was also found that annual target-rate pricing did not underlie the counter-cyclical gross margin behaviour. This was demon-strated by a predominance of pro-cyclical net profit margin behaviour, behaviour inconsistent with an annual target-rate model. Perhaps this leaves much unexplained. Iedo not apologize for these results. Economic activity undoubtedly is much more com-plicated than we would like to believe and theorize. Moreover, despite the fact that I do not explain it well, non-competitive gross-margin has been clearly observed. Whatever the reasons for it (and it is important to pursue them), in the meantime the non-competitive behaviour must be recognized and considered. The importance of this for theory and policy will be discussed in the final chapter. 150 C H A P T E R VIII CONCLUSION It is quite common to ignore non-competitive markets in the analysis of macro behaviour. Even where the existence of non-com-petitive markets is acknowledged, it is often argued that resource allo-cation, not the behaviour of macro variables would be affected. ^  In this vein, the Price s and Incomes Commission wrote with respect to inflation analysis: "The existence of a certain degree of monopoly power can explain high prices at a given level of demand, but not rising prices". ^  In this thesis I have argued that such a view is a mistake. One cannot ignore non-competitive markets in a macro model because a hypothetical non-competitive economy does generate different macro behaviour from a hypothetical competitive economy and, more importantly, because an examination of Canadian manufacturing data suggests that actual behaviour conforms better to a model incorporating non-competi-tive than standard competitive markets. In other words, the theory sug-gests that non-competitive micro behaviour is hypothetically important in macro analysis and the empirical work suggests that it is in fact important. In my analysis, it is through the macro gross-margin that 1 See, e.g., R. T. Seldon (1959) and M. Bailey (1959). Prices and Incomes Commission (1972, p. 4). 151 non-competitive micro behaviour can enter and affect the macro system. I argued that non-competitive micro behaviour generates micro gross-margin behaviour different from competitive gross-margin behaviour (Chapter IV) and that this, in turn, causes the macro gross margin to behave differently (Chapter III). The different behaviour of the macro gross-margin entails different behaviour of other macro variables (Chapter II). My macro model, i t w i l l be recalled, simplified to the following two equation system: (1) * f((l+m)-L, m, Y/Ytr, I) = (1+m)-L /1 \ + _ 1 + m ( 2 ) P = g T L T • where: m - gross margin Y — - income over trend income Y t r I - exogeneous investment L - employment p - price w - money wages g(L) - output per man Differentiating, i t was seen that different gross-margin behaviour (because of underlying non-competitive markets) entails d i f -ferent behaviour of employment and the price level. It also entails different relative shares behaviour because the gross-margin equals the 152 r a t i o of non-wage to wage income. dL dm r2 -(1 + m) < 0 2^- = — ^ - r - + —.W . •! • q 1 (L)(L--^-^2) ^ o (assuming dm g(L) g(L)* . 1 - f i ^ g - ( L ) > 0 ) P'O-wL m = _ w- L In the empirical work non-competitive behaviour was ob-served. The micro gross margins tended to vary c o u n t e r - c y c l i c a l l y , opposite to the p r o - c y c l i c a l v a r i a t i o n predicted i n the standard competi-t i v e model. Thus, consistent with a model of a non-competitive economy, the actual macro gross-margin varies more c o u n t e r - c y c l i c a l l y (or at le a s t l e s s p r o - c y c l i c a l l y ) than would be the case with underlying com-p e t i t i v e behaviour. Thus, i n a downswing for example, wages share f a l l s more (or r i s e s l e s s ) , p r i c e s r i s e more (or f a l l less) and employ-ment and output f a l l more because of the underlying non-competitive behaviour. Non-competitive behaviour i s important i n a macro model. It does have important e f f e c t s regarding the behaviour of d i f f e r e n t macro var i a b l e s . I t i s i n t e r e s t i n g to note that these e f f e c t s have been observed by others, with d i f f e r e n t writers emphasizing d i f f e r e n t e f f e c t s . In h i s study, Tsiang emphasized the importance of non-competitive p r i c i n g on d i s t r i b u t i o n . He wrote: " I t i s . o f great t h e o r e t i c a l i n t e r e s t how t h i s percentage margin [my gross-margin] i s decided upon i n the mind of the business executive; for t h i s i s i n f a c t the funda-153 mental problem of price determination and the theory of distribution".3 Of course, Kalecki, too, emphasized the role of non-com-petitive margin determination on distribution. Regarding the effects on employment and output, two d i f -ferent aspects have been emphasized. Means was concerned with what I would c a l l the aggregate supply effect. Higher margins (and prices) in a depression entail lower employment and output because of the 4 L upward shift of aggregate supply. This is the _ ^ ^ + m ^ component of . Neal, on the other hand, emphasized what I would c a l l the aggre-gate demand effect (the ^_£^/- (1+m) component of . He wrote: "Insofar as concentration leads to maintenance of unit margins...concentration tends to accentuate depres-sion. This result follows from the fact that margins constitute the source, directly or indirectly, of much of the economy's saving. The maintenance of saving in the fact of a depression f a l l in investment is likely to be depressing to income".5 Regarding the effect on prices, many have noted the obvious, yet important, point that a more counter-cyclical gross-margin because of non-competitive pricing inhibits price declines in depressions and slows price increases in upswings. In a recent work, Blair has summarized a l l these effects.^ In a certain sense, my analysis can be considered as formalizing his dis-cussion in a Neo-Keynesian macro model. 3 Sho-Chieh-Tsiang (1947, p. 4). 4 G. Means (1935); idem (1939). 5 A.C. Neal (1942, p. 166). 6 John Blair (1972, pp. 523-550) 154 On the basis of my work and the other studies I have referred to, I believe it is essential to incorporate non-competitive pricing into macro models. As Keynes himself feared, his competitive assumption was an important one and could be responsible for inaccurate predictions. The problem is of critical importance with regard to policy. Ignoring non-competitive pricing can lead to under-estimating the em-ployment impact of a government-induced upswing or downswing and to 7 overestimating the price impact. These implications are particularly important regarding the policy of fighting inflation by creating unemployment. A given decline in government spending will cause a sharper cyclical downturn than would be expected with competitive prices. The burden imposed on workers in terms of wages' share will be greater than expected. And the price level will fall less than would be expected. In fact, the macro gross-margin could even rise in the depression (in the 57-58 downswing it rose almost 6%) contributing to a price increase. Indeed, the only hope for such a policy would be that the very sharp downswing would cause a considerable reduction of money wages. With non-competitive product markets, this would be the only reliable source of price declines. However, as many have noted, 8 sub-_ In Appendix B the employment multiplier and the price effect of a change in exogeneous demand are calculated for competitive and non-competitive economies. A precise formulation of the overestimation of the employment impact and underestimation of the price impact is found there. See, e.g., J. Garbarino (1950). 155 stantial reductions in wages in the face of rising (or very slowly falling) gross-margins and share of non-wage income is very unlikely. Only in a competitive economy where there would be substantial declines in the gross-margins and share of non-wage income can wage reductions be expected. In sum, non-competitive pricing makes fighting inflation by creating unemployment more inefficacious and inequitable than it would be otherwise. If for no other reason, non-competitive pricing should be incorporated into macro analysis simply to recognize this point. 156 BIBLIOGRAPHY Abromovitch, M. (1938), "Monopolistic Selling in a Changing Economy", Quarterly Journal of Economics LII, pp. 191-214. Ackley, Gardiner (1961), Macroeconomic Theory, Macmillan, New York. Andrews, P.W.S. (1949), Manufacturing Business, Macmillan, London. Archibald, G.C. (1955), "Inventory Investment and the Share of Wages", Economic Journal, LXV, pp. 257-270. Arrow, K. J. , Chenery, H.B., Minhas, B.S. and Solow, R. M. , (1961), "Capital-Labour Substitution and Economic Efficiency", Review of Economics and Statistics XLIII, pp. 225-48. Asimakopolous, A. (1969), "A Robinsonian Growth Model in One-Sector Notation", Australian Economic Papers VIII, pp. 41-58. (1970), "A Robinsonian Growth Model in One-Sector Notation - An Amendment", Australian Economic Papers IX, pp. 171-76. (1972), "Aggregate Supply and Aggregate Demand Curves", unpublished. Bailey, M. (1959), "Administered Prices Reconsidered (Discussion)", American Economic Association Papers and Proceedings XLIX, pp. 459-61. Bain, Joe (1956), B a r r i e r s to New Competition, Harvard University Press, Cambridge, Mass. Baron, David (1972), "Limit Pricing and Models of Potential Entry", Western Economic Journal X, pp. 298-307. Baumol, W.J. (1958), "On the Theory of Oligopoly", Economica XXV, pp. 187-98. Berle, A. and Means, G. (1932), The Modern Corporation and Private Property, Macmillan, New York. Blair, John (1972), Economic Concentration, Harcourt Brace Jovanovitch, New York. 157 Burmeister, E. and Taubman, P. (1969), "Labour and Non-Labour Income Saving Propensities", Canadian Journal of Economics II, pp. 78-89. Chamberlain, Neil (1962), The F i r m : Micro-Economic Planning and  Action, McGraw-Hill, New York. Chambers, (1958), "Canadian Business Cycles Since 1919", Canadian Journal of Economics and Political Science XXIV, pp. 166-89. Cohen, H.A. (1971), "Effects of Demand and Supply on the 'Limit Price'", Mississippi Valley Journal of Business and Economics VII, pp. 47-55. Cyert, R.M. (1955), "Oligopoly Price Behaviour and the Business Cycle", Journal of Political Economy LXIII, pp. 41-51. and March, J. (1963), A Behaviourial Theory of the Firm, Prentice-Hall, Englewood Cliffs, New Jersey. Daly, D.J. (1969), "Business Cycles in Canada: Their Post-War Persistence", in M. Bronfenbrenner, Is the Business Cycle Obsolete, J. Wiley and Sons, New York, pp. 45-66. Dean, Joel (1952), "Methods and Potentialities of Break-Even Analysis", reprinted in D. Solomons (ed. ), Studies in Costing, Sweet and Maxwell, London, pp. 227-66. Dennis, K. (1973), "Market Power and the Behaviour of Industrial Prices", in Prices and Incomes Commission, Essays on Price Changes, Ottawa. Department of Corporate and Consumer Affairs (1971), Concentration in the Manufacturing Industries of Canada, Ottawa. Dunlop, John (1938), "The Movement of Real and Money Wage Rates", Economic Journal XLVIII, pp. 413-34. (1939), "Price Flexibility and the 'Degree of Monopoly'", Quarterly Journal of Economics LIII, pp. 522-33. (1950), Wage Determination Under Trade Unions, Oxford University Press, London. Eckstein, O. and Fromm, G. (1968), "The Price Equation", American Economic Review LVIII, pp. 1159-83. 158 Evans, Michael (1969), Macroeconomic Activity, Harper and Row, New York. Galbraith, J. K. (1936), "Monopoly Prices and Price Rigidities", Quarterly Journal of Economics L, pp. 456-75. (1957), "Market Structure and Stabilization Policy", Review of Economic and Statistics XXXIX, pp. 124-33. Garbarino, J. (1950), "A Theory of Inter-Industry Wage Structure Variations", Quarterly Journal of Economics LXIV, pp. 282-305. Gaskins, D.W. (1971), "Dynamic Limit Pricing: Optimal Pricing Under Threat of Entry", Journal of Economic Theory III, pp. 306-22. Godley, W. and Nordhaus, W. (1972), "Pricing in the Trade Cycle", Economic Journal LXXXII, pp. 853-82. Gordon, K. (1955), in NBER, Business Concentration and Price Policy, Princeton University Press, Princeton, pp. 495-500. Hall, R. and Hitch, C. (1951), "Price Theory and Business Behaviour", in T. Wilson and P. W. S. Andrews (eds. ), Oxford Studies  in the Price Mechanism, Oxford University Press, London. Harcourt, G. (1972), Some Cambridge Controversies in the Theory of  Capital, Cambridge University Press, Cambridge. Harris, D.J. (1972), "The Price Policy of Firms, the Level of Employ-ment and Distribution of Income in the Short Run", unpub-lished. Harrod, R.F. (1936), The Trade Cycle, Clarendon Press, Oxford. (1952), "Theory of Imperfect Competition Revised", in Economic Essays, Macmillan, London, pp. 139-87. Heflebower, R.B. (1955), " F u l l Costs, Cost Changes and Prices", in NBER, Business Concentration and Price Policy, Princeton University Press, Princeton, pp. 361-92. Houthhakker, H. and Taylor, L. (1970), Consumer Demand in the U.S., Harvard University Press, Cambridge, Mass. 159 Hultgren, T. (1965), Costs, Prices and Profits: Their Cyclical Rela- tions, Columbia University Press, New York. Jacquemin, A. P. (1972), "Market Structure and the Firm's Market Power", Journal of Industrial Economics XX, pp. 122-34. Johnston, J. (I960), Statistical Cost Analysis, McGraw-Hill, Toronto. Kaldor, N. (1959), "Economic Growth and the Problem of Inflation, Part I", Economics XXVI, pp. 212-225. (I960), "A Model of the Trade Cycle", in Essays on Economic Stability and Growth, G. Duckworth, London, pp. 177-92. Kalecki, M. (1939), Essays in the Theory of Economic Fluctuations, Allen and Unwin, London. (1964), Theory of Economic Dynamics, Unwin, London. Kamien, M.I. and Schwartz, N. L. (1971), "Limit Pricing and Uncertain Entry", Econometrica XXXIX, pp. 441-54. Keynes, J.M. (1936), The General Theory of Employment, Interest and Money, Macmillan, London. (1939), "Relative Movements of Real Wages and Output", Economic Journal XLIX, pp. 34-51. Kuh, Edwin (I960), "Profits, Profit Mark-Ups and Productivity: An Examination of Corporate Behaviour Since 1947", Study Paper #14, Study of Employment, Growth and Price Levels, Joint Economic Committee of the U.S. Congress, Washington. Laden, B.E. (1972), "Perfect Competition, Average Cost Pricing and the Price Equation", Review of Economics and Statistics LIV, pp. 84-88. Kotowitz, Y. (1968), "Capital-Labour Substitution in Canadian Manufact-uring 1926-39 and 1946-61", Canadian Journal of Economics I, pp. 619-32. Leijonhufvud, Axel (1968), On Keynesian Economics and The Economics of Keynes, Oxford University Press, New York. Levinson, H. (I960), "Postwar Movement of Prices and Wages in Manu-facturing Industries", Study Paper #21, Study of Employ-ment, Growth and Price Levels, Joint Economic Committee of the U.S. Congress, Washington. 160 Lintner, John (1956), "Distribution of Income of Corporations Among Dividends, Retained Earnings and Taxes", American Economic Review XLVI, pp. 97-113. Marris, R. (1964), The Economic Theory of Managerial Capitalism, Free Press of Glencoe, New York. McFetridge, D. (1972), Market Structure and Price Behaviour: Empiri- cal Studies of the Canadian Manufacturing Sector, unpub-lished Ph. D. dissertation. University of Toronto. Means, Gardiner (1935), Industrial Prices and Their Relative Inflexibi-lity, U.S. Senate Document #13, 74th Congress, 1st Session, Washington. (1939), The Structure of the American Economy, Part I, National Resources Committee, Washington. (1959), "Administered Prices Reconsidered (Discus-sion)", American Economic Association Papers and Proceed-ings XLIX, pp. 451-54. (1972), "The Administered Price Thesis Reconfirmed", American Economic Review LXII, pp. 292-306. Modigliani, F. (1958), "New Developments on the Oligopoly Front", Journal of Political Economy LXVI, pp. 215-32. and Miller, M. (1958), "The Cost of Capital, Corporation "Finance and the Theory of Investment", American Economic Review XLVIII, pp. 261-97. Moore, John and Levy, L. (1955), "Price Flexibility and Industrial Con-centration", Southern Economic Journal XXI, pp. 435-440. Moore, M. (1967), "A Reformulation of the Kaldor Effect". Economic Journal LXXVII, pp. 84-99. (1970), How Much Price Competition, McGill-Queen's University Press, Montreal. (1972), "Stigler on Inflexible Prices", Economics V, pp. 486-93. Canadian Journal of Neal, A. C. (1942), Industrial Concentration and Price Inflexibility, American Council on Public Affairs, Washington. 161 Neild, R. (1963), Pricing and Employment in the Trade Cycle, Cambridge University Press, Cambridge. Nelson, Saul (1940), Price Behaviour and Business Policy, Temporary National Economic Committee, Washington. Nourse, E.G. (1957), In Administered Prices, Hearings before the Sub-committee on Antitrust and Monopoly of the Committee of the Judiciary of the U.S. Senate, Part I-Opening Phase, Washington. Oliver, F.R. (1968), "A Cross-section Study of Marginal Cost", re-printed in D. Solomons(ed. ), Studies in Cost Analysis, Sweet and Maxwell, London, pp. 251-61. Prices and Incomes Commission (1972), Summary Report, Ottawa. Pyatt, F. G. (1971), "Profit Maximization and the Threat of New Entry", Economic Journal LXXXI, pp. 242-55. Robinson, Joan (1934), "What is Perfect Competition", Quarterly Journal of Economics XLIX, pp. 104-20. (1936), "Review of Harrod's Trade Cycle", Economic Journal XLVI, pp. 691-93. (1962), "A Model of Accumulation", in Essays in the  Theory of Economic Growth, Macmillan, London, pp. 22-87. Rosenbluth, G. (1957), Concentration in Canadian Manufacturing Indust-ries, Princeton University Press, Princeton. (1961), "Concentration and Monopoly in the Canadian Economy", in M. OHver(ed. ), Social Purpose for Canada, University of Toronto Press, Toronto. Royal Commission on Price Spreads (1937), Report of the Royal Com- mission on Price Spreads, Ottawa. Ruggles, R. (1955), "The Nature of Price Flexibility and the Determinants of Relative Price Changes in the Economy", in NBER; Busi-ness Concentration and Price Policy, Princeton University Press, Princeton, pp. 441-95. Rushdy, F. and Lund, P. (1967), "The Effect of Demand on Prices in British Manufacturing", Review of Economic Studies XXXIV, pp. 361-74. 162 Scarfe, B.L. (1972), Price Determination and the Process of Inflation  in Canada, Prices and Incomes Commission, Ottawa. Seldon, R. (1959), "Administered Prices Reconsidered (Discussion)", American Economic Association Papers and Proceedings XLIX, pp. 454-57. Simon, H. A. (1959), "Theory of Decision-Making in Economics and Be-havioural Science", American Economic Review XLIX, pp. 253-83. Sloan Jr., A. P. (1964), My Years with General Motors, Doubleday, Garden City, New York. Solow, R. (1958), "A Skeptical Note on the Constancy of Relative Shares", American Economic Review XLVIII, pp. 618-31. Stigler, G. and Kindahl, J. (1970), The Behaviour of Industrial Prices, NBER, New York. Stykolt, S. and Eastman, H. (1967), The Tariff and Competition in  Canada, Macmillan, Toronto. Sweezy, P. (1939), "Demand Under Conditions of Oligopoly", Journal of Political Economy XLVII, pp. 568-73. Sylos-Labini, P. (1969), Oligopoly and Technical Progress, revised edition, Harvard University Press, Cambridge, Mass. Tarshis, L. (1938), "Real Wages in the United States and Great Britain", Canadian Journal of Economics and Political Science IV, pp. 253-83. Taylor, L. , Turnovsky, S. and Wilson, T. (1971), "Wage, Price and Productivity Behaviour in the Canadian Manufacturing Sector", presented at the Canadian Economic Association meetings, St. John's. (1973), The Inflationary Process in North American Manufacturing, Prices and Incomes Commission, Ottawa. Tsiang, Sho-Chieh. (1947), The Variations of Real Wages and Profit Margins in Relation to the Trade Cycle, Sir Isaac Pitman and Sons, London. 163 Wan Jr. , Henry. (1966), "Intertempor al Optimization with Systematically Shifting Cost and Revenue Functions", International Econo-mic Review VII, pp. Z04-Z5. Waterman, A. M.C. (1967), "Timing of Economic Fluctuations in Australia", Australian Economic Papers VI, pp. 77-10Z. (1972), "Measurement of Economic Fluctuations in Canada, 1947 to 1969", a preliminary unpublished report. Weintraub, S. (1961),. Classical Keynesianism, Monetary Theory and  The1- Price Level, Chilton Co., Philadelphia. (1969), "A Macro-Theory of Pricing, Income Distribution and Employment", Weltwirtschafliches Archiv CH, pp. 11-25. Weiss, L. (1966), "Business Pricing Policies and Inflation Reconsidered", Journal of Political Economy LXXIV, pp. 177-87. Weston, J. (1972), "Pricing Behaviour of Large Firms", Western Econo-mic Journal X, pp. 1-18. White, D. (1967), Business Cycles in Canada, Staff Study No. 17, Econo-mic Council of Canada, Ottawa. Williamson, J.H. (1966), "Profit, Growth and Sales Maximization", Economica XXXIII, pp. 1-16. Williamson, O.E. (1963), "A Model of Rational Managerial Behaviour", in R. Cyert and J. March (1963, chapter nine). Yordon, W.J. (1961), "Industrial Concentration and Price Flexibility in Inflation: Price Response Rates in 14 Industries 1947-1958", Review of Economics and Statistics XLIII, pp. Z87-94. (1970), "The Short Run Cost Function in Manufacturing", Quarterly Review of Economics and Business X, pp. 55-67. 164 APPENDIX A ELASTICITY OF DEMAND AND TARGET R A T E MARGIN BEHAVIOUR Consider, for example, a downswing. The immediate effect of a fall in demand is. lower output and gross profits. The firm will want to adjust the margin to restore profits back to their former, the target level. The following algebra demonstrates how the margin will behave given different elasticities of demand. p. q - (Mfw- L)  m = M+w-L T T Q = T l f F - gross profits TT + F q = q(p) - demand p. q- ( TT + F) curve AVC - (constant) 1 average variable P- q _ l TT+F costs T T + F = m • P • q 1+m i.e., TTV = . p . q(p) u 1+m Also, p = x (1+m) AVC /. d TT, = q(p) • p • - — T . dm + (q'(p)« P + q(p)) dp G (1+mr 1+m = q(p) • p • 1 . dm + - H - . (ed + 1) q (p) dp (1+m) 2 1 + m - where ed = q'(p)* p/q (p) is the elasticity of demand /.dp = AVC dm 165 S u b s t i t u t i n g : d T T G q(p) .p 1 4- ——(ed + l ) - q ( p ) - A V C ( l J+ im) 2 l f m d m d m _ 1 d T T G 0 ( e d ) w h e r e : 0 (ed) = q (p) • - 4- _ ™ _ (ed + 1) q(p) A V C N o w : (1+m) 2 1+m (1) i f ed = - 1 (unit e l a s t i c ) 0(ed) = q(p) -p . - > 0 (1+m) 2 d r r L > 0, i . e. , w i t h a un i t e l a s t i c d e m a n d c u r v e the d T r G m a r g i n w i l l r i s e to i n c r e a s e g r o s s p r o f i t s (and t h e r e f o r e r i s e to m a i n t a i n t a rge t p r o f i t s i n a d o w n s w i n g ) . (2) i f - 1 < ed < 0 ( i n e l a s t i c ) 0(ed) = q (p ) ' p - 1—— + (1 + ed) . q(p) • A V C (1+m) 2 l f m 0 (ed) > 0 (-1) > 0 s i n c e 0 < 1 + ed < 1 d m d m , . , ^ 0, i . e . , w i t h an i n e l a s t i c d e m a n d e a = l ^ l - l<ed<0 c u r v e the m a r g i n w i l l r i s e , but not as m u c h as w i t h a un i t e l a s t i c d e m a n d c u r v e . (3) i f ed < -1 ( e l a s t i c ) 0 (ed) = q (p) .p + m (1+ed). q(p). A V C (1+m) 2 l f m then, i f q(p) -P > - m (1 + ed)• q(p). A V C (1+m) 2 l f m ( w h i c h e n t a i l s that I ed I < 1 + — ) 1 1 m 166 0 (ed) > 0 and if q(p). p-0(ed) < 0 and dm arrG  l (1+m)2 dm ed -1 < > d Tre- ed -1 dm dTT, G > 0 ed=l m 1+m < 0 (1+ed). q(p). AVC i.e., with an elastic demand curve if |ed|<l +_L then the m margin will rise by more than with a unit elastic curve; if ) ed| > 1 + — , then to maintain m target profits in a downswing, the margin will fall. 167 APPENDIX B EMPLOYMENT MULTIPLIER AND PRICE EFFECT OF A CHANGE IN EXOGENEOUS DEMAND The two equation system describing the macro model i n Chapter II can be closed with respect to the margin with a function r e l a t i n g the margin to the l e v e l of employment.1 This y i e l d s the following three equation system: f ( ( l+m) -L , m, Y / Y t r , I) = (l+m)-L 1+m p = g U T ' w m = m(L) T o t a l l y d i f f e r e n t i a t i n g and solving f o r dL and dp i n dL = £ r d l f 2 ) - 2 terms of d l : (1+m) + (L - yr^j -) m' (L) _ -w(l+m) _ „ , , T , ^ . „ , , T X dL d l f4 d p = g ( L ) ^ * g ( L ) + g l L f ^ ( L ) — ' d I w 1 - f l = -^y • ( 1 + m ) ^ f ? ) - d l (assuming g ' (L)=0) m^TU + ( L T=fi) I have argued that m1 (L) would be le s s i n a non-competitive economy than i n a competitive economy. This lowerm 1 (L) e n t a i l s a 1 The exposition here i s very s i m i l a r to a model by D.J. Harris (1972). In his model, however, the margin was a function of investment. 2 See the discussion on p. 27 above. If g 1 (L) were less than zero, the p r i c e e f f e c t r e s u l t might not hold. 168 higher value of —3 (the employment impact of an increase in exogeneous d i demand) and a lower value of —— (the price effect of an increase in d i exogeneous demand). This can be seen by taking the partial derivative of ^ 3 and of with respect to m1 (L) . 3 d i d i d L dp ( 1 + m ) Because of a discontinuity in and —3 at m 1 (L) = _ f 2 , i t could d i d i L IZf-j-appear that these results would not hold i f m 1 (L) were less than -(1+m) L - . However, i f such an inequality held, then m 1 (L) " g ^ ^ If d L T . - f 2 since — 1-t]_ . This would mean that the ela s t i c i t y of employment - ( 1 - m ) with respect to the margin times the e l a s t i c i t y of^jthe margin with respect to employment would be greater than one j^ /^^ " /^m ^ ^) * This implies a highly explosive model in which these comparative statics would not be appropriate. A dynamic model expli c i t l y incorporating the ratio of current to trend income would have to be examined to consider the effects of a change in exogeneous demand. 

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.831.1-0093527/manifest

Comment

Related Items