Microeconomics for Management Students [2 ed.] 0195647351

Citation preview

.

—

—

Microeconomics for Management Students eeressntatatin ater

MICROECONOMICS FOR MANAGEMENT STUDENTS Second Edition

RAVINDRA

H. DHOLAKIA

AJAY N. OZA

OXFORD UNIVERSITY

PRESS

OXFORD UNIVERSITY

PRESS

YMCA Library Building, Jai Singh Road, New Delhi 110001 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship and education by publishing worldwide in Oxford

™._

New York

Auckland Bangkok Buenos Aires Cape Town

Chennai

“Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkat a la Lumpur Madrid Melbourne Mexico City Mumbai Nairob i Sao Paulo Shanghai Taipei Tokyo Toronto Id is a registered trade mark of Oxford University Press in the UK and in certain other countries. Published in India by Oxford University Press

ase right Oxford University Press (maker)

z wn

First published 1999

Eighth impression 2004

VB

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the approp riate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Depart ment, Oxford University Press, at the address above. You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer. ISBN 0-19-564735-1

Typeset by Excellent Laser Typesetters, Delhi 110034 Printed in India by Rashtriya Printers, Delhi 110032 and published by Manzar Khan, Oxford University Press YMCA Library Building, Jai Singh Road, New Delhi 110001

To

Archana and Suhas

ae

ay + =

|

y

: x

gh cN

vee ot

.

=. e

58D Panic Shang

%

k

an iat

S & a eed = P

as

;

Sects Tins Wines k

a

(ride iark of Catind ‘ Wes ;

. “es

ao) (be UR ain! th certain

‘\

sa

=

.

aoe

e

— -Coeee

_

=|

~

=

2

Mie

=

a.

atl

-

]

——

)

>

=

.

oy,

2

4

St

re

:

=

=

ao

a

7

-

oh

,

:

a

,

= p

4

=

Oe

Siwih toed

ee

i

7a

Se

BSR

u

SS

av

C5

__

6g

Sa

-

tz

am

>

ee

~

‘

A

he

PR a

P

*

Ho

ea ow

ee

Sipe

7 Sar

ASL

te

fo

7 7

>

:

Sk

eal ges ;

tA

Aa

SOR qAyste-

7 7

ceria ;

:

u ep

a

. Ske

;

=

peti o

a

TUE.

ay.

ots et

@f te

Cathe

Yee

ee

e355) on ter

Jes ‘Sige wt,

risen

Been *

ware the mppeigyiete i

Fier ne. cag -

_

i /RpyDelin :

;

ee

OX es erat

aS Pvt th dee

4

= Uped

—

S Sees

LE—— eS

7

—

acer

ee

a

a?

‘

2079

iy =

si

)

; =

:

ire Py

7

:

¥

g

RS

:

er

we

“es

.

:

.

=>

a

a

"-

g

:

are

Caviereigy Meee ip aie)

Mi eIaies

:

Wd

=

P

-

i

,

_

rh

-

;

an

6

7

'

Rae Gh The satiety ape ey a

ere

7

~_

—

ight 4 THs

ve

o4

=

=

ofa

ae

a

,

be a4 _

—

ge,-~ % Caters Uasires sity Press weee if.

74

an

Tae

a

retaed Univetvity Press

=

:

eet

:

,


B and B=C,A must be greater than C. Similarly, if C=A and A>B,C cannot be less than or equal to B. Geometrically, the consistency

of the preference ordering implies that the indifference curves cannot intersect one another. It does not necessarily mean that the indifference curves have to be parallel. They may not be parallel but they should not be intersecting. If two indifference curves intersect, inconsistency of preference orderings arises. This can be shown through Fig. 7 which contains two indifference curves u, and u, intersecting at the point C. Now,

66

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Figure 7 pointA yields a higher level of satisfaction than pointB since point A contains more of Y and the same amount ofX in comparison to point B. Thus, A is preferred to B. But A is equivalent to C since

both of them lie on the same indifference curve (u,). However, point C is as preferred as point B since they also lie on the same indifference curve (u,). This is inconsistent because if A = C and if B = C, then A cannot be greater than B. But A is shown to be preferred to B. Therefore, either A # C or B#C. This can happen only if the two indifference curves do not intersect. If they intersect, inconsistency in the preference orderings arises.

4. Diminishing Marginal Rate of Substitution The Marginal Rate of Substitution (MRS) of X for Y is defined as

the rate at which a consumer wants to substitute good X for good Y in his consumption basket without affecting his utility. It can be seen from the definition that MRS is the absolute value of the slope of the indifference curve. It is measured along the indifference curve where the utility level does not change. Since MRS is the absolute value of the slope of the indifference curve, it is simply

a ratio of the marginal utilities of the two goods. It is assumed that the utility function of the consumer is such that the MRS of X for Y goes on falling as he increases his consumption of X.

CONSUMER DEMAND THEORY

67

Figure § Fig. 8 shows what this assumption implies in terms of the curvature of the indiffererice curve. U, is an indifference curve

yielding U, level of satisfaction at all points on the curve. If this curve is a straight line, its slope would remain constant by definition. As a result, the consumer would sacrifice a fixed number of

units of the good Yfor every additional unit of the good X without affecting his total utility. A concave curve would imply an increasing MRS of X for Y as X increases. The assumption made about MRS of X for Y, however, is that the consumer is willing to progres-

sively sacrifice less quantity of good Y for each successive additional unit of good X. This can happen only when the combinations of the goods X and Yare like points A, B, C, D, etc., as shown in the Fig. 8. The horizontal distance between points A, B, C, and D

in the figure remains the same. It may be considered to be one unit of the good X. The vertical distance between points A, B, C, and Din the figure, however, progressively declines in accordance

with the assumption of diminishing MRS of X for Y. Under these circumstances, the curve passing through all such points showing an equal level of satisfaction has to be convex from below. The assumption of diminishing MRS of X for Y implies convex (from

68

MICROECONOMICS

FOR MANAGEMENT STUDENTS

below) indifference curves.® We will later discuss a further implication of the assumption of diminishing MRS of X for Y. An interesting implication of convex indifference curve is that points lying on the line segment joining any two points on the same indifference curve, such as points A and D in Fig. 8, always show higher utility than either of the points on the given indifference curve.

The Budget Constraint So far we have restricted our discussion to the subjective world of a consumer.

The indifference map, its shape, curvature, etc.

represented the subjective valuation of the consumer of the goods available in the market. The objective realities of the market would not enter. This appears to be an artificial device, but is necessary because it is important to study the interaction of the pure subjective world of the consumer with the objective realities to understand consumer choice. The more we are able to reduce the subjective zone, the greater the success of the theory is considered to be. The objective realities of the market are captured by the concept of budget constraint in the analysis. In the present framework, it consists of the prices of the goods and the

money available with the consumer during the time period under consideration. We refer to it as the consumer’s money income by convention. When we consider such a definition of the money income of the consumer, he simply cannot overspend because it provides an absolute ceiling on the consumer’s expenditure during the given period. He can spend less, but not more on the available goods. In terms of an algebraic expression, the consumer’s budget constraint can be written as:

(10)

— M2P,X+P,.¥+...

Where M is consumer's money income; P,, P,, etc. are the prices 6 Mathematically, the convex indifference curve would require that the slope of the curve should be an increasing function of X. This does not contradict the

diminishing MRS of X for Y because MRS is defined in absolute terms or is given as the negative of the slope of the indifference curve. When the negative variable increases, its absolute value declines.

CONSUMER

DEMAND THEORY

69

of the goods X, Y, etc., and x, y, etc. are the quantities consumed by the consumer of the goods X, Y, etc.

Geometrically, the consumer’s budget constraint is represented by the line BB in Fig. 9. The line BB divides the whole space in the XOY plane into the attainable and non-attainable region. Any point lying on or below line BB is attainable by the consumer

given his money income and the prices of goods in the market. If the point lies on the line BB, the consumer spends his entire budget on the two goods in question. If the point lies below line BB, the consumer spends less than his budget on the two goods. The points above the line BB are not attainable by the consumer because they require higher expenditure than the consumer can afford at the given prices of the goods. Line BB in Fig. 9 is a downward sloping straight line because, by assumption, the objective realities of the consumer’s income and prices of the goods are considered to be given and exogenous for him. If we consider that consumer spends all his income on the consumption of the two goods, X and Y, the expression in (9) becomes an equation. This equation, moreover is linear in _X and Y, representing a straight line curve in the XOY plane. The slope and intercepts of this line on the two axes are given in terms of the parameters of the equation. The vertical and horizontal

M/P

O

M/P,.

Figure 9

70

MICROECONOMICS

FOR MANAGEMENT STUDENTS

intercepts of this equation are given by (M/ P,)and (M/P,,), respectively. Thus, the level of the budget constraint is determined by the consumer’s money income and the prices of the goods. If the money income increases, the budget constraint would shift upward, and if the money income decreases, line BB would shift downward. Again, if the prices of the goods change, the budget constraint would shift outward or inward, depending upon which price changes and in which direction. The slope of the budget constraint is given by the negative of the price ratio (—) ae le an? This is also obtainable by the two intercepts. Since both the prices are treated as parameters, the slope is constant and hence the curve is a straight line. Similarly, the negative sign of the slope indicates that the budget constraint is also downward sloping. Before considering the interaction of the subjective world of the consumer represented by his indifference map and the objective realities in the market given by his budget constraint, we may note that the concept of the budget constraint can be used in a very general way. The prices of the commodities may not be always very explicit and clearly given. Similarly, the “money income’ of the consumer may not be in money terms nor an income. Moreover, the interpretation of the attainable non-attainable region and the shape of the budget constraint also change, depending upon the nature of the commodities

even and may con-

sidered. For instance, if we consider the case of a consumer faced

with a choice of location of his residence between the place of work and the pollution, both of these are bads or discommodities. The consumer would ideally be interested in minimizing his consumption of both. He cannot however do so. There may be geographical and climatic constraints that compel him to consume the two discommodities in a certain definite relationship. Such a relationship, then, represents the consumer’s budget constraint. If we measure the two bads on the two axes, and draw the budget constraint, it is likely to be downward sloping; but the area below the budget line would not be attainable while that above or on the line would be. In this particular, example the budget constraint for all the workers of a given factory would be the same, though workers of factories located elsewhere might have very different

budget constraints. In short, budget constraint can also be viewed as a line of substitution between the goods allowed by the market

CONSUMER DEMAND THEORY

71

or the environment. Such an interpretation is further strengthened because the absolute value of the slope of the budget line given by (P,. fF) is the relative price of good X, which shows the rate at which the good X can be exchanged for good Yin the market.

Consumer’s Equilibrium Fig. 10 superimposes the budget constraint on the consumer’s indifference map. This allows us to see the interaction of the consumer’s subjective world with the objective realities. The objective of the consumer is to obtain the maximum possible satisfaction from his consumption of goods X and Y, given his tastes and preferences, the prices of the two goods, and his money income. In other words, he has to choose his consumption basket in such a way that he attains the highest possible indifference curve while remaining within his budget constraint. As we have already seen, all points lying on or below his budget line BB in Fig. 10 are attainable, in the sense that they are possible options before him. His objective realities allow him to be on line BB or below it. This is because, by assumption, we are considering the case of both the goods being positively desired. It is possible,

Figure 10

12

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

therefore, to further argue that there is always at least a point on line BB which would be preferred to a point below it. Thus, essentially, the consumer has to choose from among the points lying on his budget line. The solution obviously is that he should choose the point along BB line which lies on his highest indifference curve. In Fig. 10, point E is such a point which represents the consumer’s equilibrium. No other point on the budget line BB

lies on the u, curve which shows the highest attainable utility level since it is the only indifference curve that touches the budget line BB and does not intersect it. All other points on line BB lie on lower indifference curves and hence have the lower utility level. The condition for the consumer's equilibrium ts, thus, the tangency

between the indifference curve and the budget line. Tangency implies equality of the slopes of the two curves. Thus, the condition of consumer’s equilibrium is given by:

(11)

MU, P —~=-— MU > tsBy

or

MU, MU SSeS ies Ry

It can be seen that the condition of consumer’s equilibrium with the ordinal utility approach remains the same as that in the cardinal utility approach. The law of equi-marginal utility of money also applies here, though with much less restrictive assumptions. In Fig. 10, point £, which represents the point of tangency between the indifference curve and the budget line BB shows the highest attainable utility level because the indifference curves are assumed to be convex from below, implying the assumption of diminishing MRS of X for Y. Thus, the role played by the assumption of diminishing marginal utilities in the cardinal theory is played by the assumption of the diminishing MRS in the ordinal theory. These assumptions are required to ensure the stability of the consumer’s equilibrium. Let us see how the stability of the consumer’s equilibrium is ensured here. Let us assume that the consumer is by chance located at point

A on the budget line BB in Fig. 10. At the point A, the indifference curve u, is steeper than the budget line BB. This implies that the rate at which the consumer is willing to substitute X for Y to maintain his level of satisfaction is higher than the rate that the market allows. In other words, the marginal valuation of X in terms

CONSUMER DEMAND THEORY

i

of Y by the consumer is higher than the value of X in terms of Y in the market. The consumer would, therefore, be better off sub-

stituting more X for Y. Thus, if the consumer’s equilibrium is internally disturbed such that he is found at point A, he has a tendency to return to point E where the consumer’s subjective valuation of good X matches the market’s valuation. Very similar logic can be applied to show that if the consumer is found initially at a point to the right of point E in Fig. 10 (such as point C), where the indifference curve would be flatter than the BB line, he would

have a tendency to substitute Y for X and move towards point E. Thus, equilibrium point £ represents a stable equilibrium. This stability is basically ensured by the convexity of the indifference curves or the law of diminishing MRS of X for Y. Moreover, the

condition of the consumer’s equilibrium has an interesting implication about different consumers’ subjective valuation of goods X and Y. All the consumers consuming both the goodsX and Y would have the same subjective marginal valuation of the two goods as given by their market prices. This is because all utility maximizing consumers would be equating their subjective marginal valuation of goods X and Yto their market valuation as given by their prices.

Diminishing MRS versus Diminishing MU The MRS of X for Y can be expressed as a ratio of (MU,,/MUy). From this, it may appear that diminishing MRS of X for Y as X increases would amount to the same thing as the law of diminishing marginal utilities. This is because, as X increases in accordance with the law of diminishing MU, MU,, would fall. Similarly,

when _X increases, Y has to fall along a given indifference curve. As Y falls, MU,

would

increase

in accordance

with the law of

diminishing MU. Therefore, the ratio WWU,,/MUy) would fall as X increases along an indifference curve, which amounts to the law of diminishing MRS. This is not always true, however, because,

under the ordinal approach to utility, the interdependence of utilities can be explicitly considered. Thus, when_X increases and Y decreases along an indifference curve, their marginal utilities may be so affected that their ratio may actually rise rather than fall, though both the goods individually may be subject to the

74

MICROECONOMICS FOR MANAGEMENT STUDENTS

law of diminishing MU. Similarly, even when the goods are not subject to the law of diminishing MU, the ratio of their marginal utilities could be falling as X increases and Y decreases along an indifference curve. Thus, because of the possibility of interdependent utilities, neither does diminishing MRS necessarily imply diminishing MU nor diminishing MU necessarily imply diminishing MRS. The law of diminishing MU is neither a necessary nor a sufficient condition for the law of diminishing MRS of X for Y, and vice versa. We cannot therefore say, merely on the basis of a comparison of these two assumptions, that the theory based on cardinal utility is more or less general than that based on ordinal utility. There are, however, other grounds on which we can say that

the ordinal theory is more general than the cardinal! one. Unlike the cardinal theory, the ordinal one allows interdependent utilities and hence has the capacity to explain the phenomenon ofsubstitutes and complements. Moreover, the ordinal theory is also able to effectively tackle the case of those goods which play important role in the consumer’s budget, whereas the cardinal theory can at best explain the case of goods which are not significant in the consumer’s budget.

Corner Solution and Curvature of Indifference Curves

We frequently find that a consumer does not consume all the goods available in the market. Even when, for the sake of simplicity, only two goods are considered, it is possible that the consumer may find it most satisfying to consume only one of them. This is a case of specialization in consumption. The brand loyalty of a consumer may be considered as a special case of specialization in consumption. However, it is a common experience that, in the final analysis, brand loyalty is not absolute. It does depend on a certain level of the prices. More importantly, when brand loyalty is broken due to a very high relative price of the good, the consumer does not immediately shift wholly to another brand.

Genrerally, he experiments with other brands by combining them with his favourite one. Similarly, it is expected that at a sufficiently low relative price, any brand can eliminate other brands from the

CONSUMER DEMAND THEORY

TS

market, and at a sufficiently higher relative price, any brand may be ousted from the market. These are casually observed phenomena. The ordinal utility theory can tackle such cases by considering two brands of the same good on the two axes for the representative consumer’s indifference curves. At this stage, however, the question about the behaviour

of

MRS of X for Y as X increases becomes very critical. If the indifference curves are straight lines implying a constant MRS through out, the consumer does not consider the two goods to be different. There is a subjective scale established in his mind by which he converts good Y into good X and vice versa. If the market also allows the exchange of good Y with good X at the same rate, the consumer would be totally indifferent among all possible combinations of X and Y given by his budget line BB because the line BB would completely coincide with the straight line indifference curve. This case is depicted in Fig. 1] where the budget line

coincides with the indifference curve showing u, level of utility. Now if the price of X slightly increases, the consumer’s budget line becomes BB’ and the highest indifference curve attainable by his budget constraint is still u,, but now his equilibrium point is at B indicating a Corner Solution and specialization in consumption of Y. However, if the price of X had slightly fallen so that his budget line were to become BB”, he would have specialized in the consumption of onlyX because in such a case, his equilibrium

Figure 11

76

MICROECONOMICS

FOR MANAGEMENT STUDENTS

B’

-

B”’

Figure 12

would be at point B” on the indifference curve showing u, level

of utility. Thus, the consumer would show wild fluctuations in his consumption of X and Y in response to slight variations in the prices of the goods. This is quite contrary to the casually observable phenomena discussed above. The assumption of constant MRS, thus, does not appear to be workable or plausible. Fig. 12 shows the indifference map when we assume increasing MRS of X for Y as X increases. The indifference curves in this case would be concave from below. Here the consumer would invariably specialize in consumption of eitherX or Y, depending on the slope of the budget line and the indifference curves. For instance, if the budget line is BB’, the consumer’s equilibrium is a corner solution at point B which gives u, level of utility. But if

the budget line is BB”, he would be specializing in consumption of X at point B” which provides him u, level of utility. Moreover, when the indifference curves are concave

from below, a utility

maximizing consumer would never be found consuming both the goods X and Ytogether. Thus, the assumption of increasing MRS of X for Y is also not workable or plausible. By a simple process of elimination, therefore, only the assumption of diminishing MRS ofX for Y remains. It is possible to find corner solutions with the convex indif-

ference curves. This is shown in Fig. /3. It can be seen from the

CONSUMER DEMAND THEORY

v7

Uy uy

Up

Figure 13 figure that at very high relative price, X is not demanded by the consumer. However, as the price of X falls in relation to Y, he starts consuming some X along with Y. Thus, if point 6 showed

equilibrium with brand loyalty for Yin the initial situation, as the

relative price of X falls with the price line shifting from BB’ to BB”, and hence the brand loyalty of the consumer is broken, he does not totally abandon brand Y. He only starts mixing some X with Y, as indicated by point £. It is clear that the assumption of the convexity of the indifference curve is a more general assumption capable of encompassing a wider phenomenon casually observable. Thus, the assumption of the diminishing MRS of X for Y is more plausible and workable than alternative assumptions.

Chapter 4 |

Consumer's Equilibrium and Demand Curves

In the ordinal approach to utility, a consumer reaches equilibrium at a point where his budget line is a tangent to an indifference curve from his indifference map. Since the indifference curves

are assumed

to be convex

from below, such a

point of tangency between the budget line and the indifference curves would invariably imply maximization of utility for the consumer. The level of his purchases of good X, given all the factors like his money income, prices of goods, and his tastes and preferences, is determined by his equilibrium. At the given price of good _X, his utility maximizing consumption of X is, thus, determined and with it, we have a point located on his demand

curve for good X. Now, we must examine how his equilibrium would change if some of the underlying factors change, particularly the price of good X. It is important to recognize that when all other things remain the same and only the price of the good X changes, it leads to a change in, (i) the relative price of X given by P,. /Py, and (ii) the real income of the consumer. This is because good X may be significant in the consumer’s budget. If it is not, the effect of a change in the price of good X may be very small in terms of the consumer’s real income. Therefore, if we wish to analyse a general case, it is necessary to view the change in the price of X as leading to simultaneous changes in the real income of the consumer and in the relative prices of the goods. It is, therefore, customary to consider the effects of these two changes in isolation before combining them in order to capture the total effect of a change in the price of X, other things held constant.

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

fe,

Income Effect (IE) When we consider a change in the consumer’s money income with all other things remaining the same, it amounts to a change in the real income by the same amount and in the same direction. A change in consumer’s money income shifts the budget line upward or downward, depending upon whether the consumer’s money income has increased or decreased. Fig. 1 shows the consumer’s budget line AB in the XOY plane. It represents the consumer’s budget constraint given by

(1)

M=P,-X+P,-Y

where consumer’s money income (¥), prices of good X (P,), and good ry) are taken as parameters. From this equation, it can be seen that the vertical intercept OA is obtained as (M/P,). At the point of vertical intercept, the consumption of good X is by definition zero. Thus, point A is arrived at when the consumer spends

Figure 1

SO

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

all his income on Y. Obviously, he can buy at the most (M/P,) units of Y. Similarly, the horizontal intercept (point B) is given by (M/P..) when the consumer spends all his income only on good X. Now, when we consider an increase in his income from M to

M’, the vertical and the horizontal intercepts shift to A’ and B’ respectively. More is available to him for spending on the two goods X and Y. His budget constraint is relaxed to that extent. This is represented by an upward and parallel shift of the budget

line AB to A’B’ in Fig. 1. The shift is parallel because the slope of the budget line, which is given by (-P, /P,), does not change when the income changes. The new and the old budget lines have, therefore, the same slope.

If the consumer’s money income falls, the budget constraint becomes tighter and the budget line would shift downward and parallel. The changes in consumer’s income, with the prices of the goods remaining the same, would shift the level of the consumer’s budget line, leaving the slope unchanged. Alternatively, the farther the budget line from the origin, the higher the consumer’s money income. If, however, we are comparing two different consumers at two different locations where the prices of the goods are different, the level of their budget lines would reveal their real income, since the budget constraint considers both

money income and the prices of goods. Turning back to Fig. /, at the initial level of money income (1), the consumer’s equilibrium is obtained at point F on his budget line AB, where the indifference curve with the utility level of u, is tangent to the budget line AB. At the point of equilibrium, the consumer is consuming OM units of X and ON units of Y. Now, let

us consider an increase in the consumer’s money income from M to M’ and the consequent shift in his budget line from AB to

A’B’. It can be seen from Fig. J that the new budget line A’B’ intersects the indifference curve (u,) at two points Rk, and R,. Given the property of the utility function that two indifference curves do not intersect one another, the new equilibrium point along the new budget line A’B’ would always lie on the line segment R,R,

(or between the two points R, and &,). Moreover, given that the indifference curves are convex from below, the points, like £, and E,, would also be located between points R, and R, on A’B’. The point i, ie obtained by extending the neve euial line NE and point

CONSUMER'S EQUILIBRIUM AND DEMAND CURVES

81

E, by extending the vertical line ME to the new budget line A’B’. These points (£, and £,) represent the consumption basket in the new situation with the same quantity of good Y and good X, respectively as at the initial point of the consumer’s equilibrium. As we have argued, the new equilibrium of the consumer with the increased income has to lie between the two points A, and

R, along the new

budget line A’B’. Therefore, the consumer

moves on to higher level of utility since he would invariably be located on the higher indifference curve. However, his consumption of goods X and Y may or may not increase in response to an increase in the consumer’s real income. There are three different cases here: (i) If the new indifference curve is tangent to A’B’ at points between E£, and E,, the consumption of both X and Y would increase. This is the normal case and hence both goods are considered to be normal goods. (2) If the new equilibrium lies between the points A, and £, along the new budget line A’B’, the consumption of good X falls and the quantity of good Y rises in comparison to the initial equilibrium at point F. Here the consumption of good X declines as the consumer’s income increases. Thus good X is an inferior good. (3) If the new equilibrium lies between points £, and Rk, along the new budget

line A’B’, the consumption of good X increases but that of the good ¥Y declines in comparison to the initial equilibrium point E. Therefore, good Y is an inferior good. In all the three cases, at least one good is normal, in the sense that its consumption will increase as the consumer’s income increases. In other words, in

a consumer's basket, it is not possible to have all inferior goods, but all the goods may be normal goods. The /ncome Effect (IE) is defined as the change in the equilibrium quantity of a good in response to a change in the consumer’s money income when prices of the goods and tastes and preferences of the consumer remain the same. When the IE ts positive for a good, it is Rnown as a normal good. Positive JE means a direct relationship between income and the quantity of the good demanded. When the IE is negative for a good, it is called an inferior good. Thus, the demand for an inferior good would tend to rise when the income of a consumer falls. Since the income effect relates quantities of goods demanded with the consumer’s income, other things remain-

ing the same, we can derive the relationship in terms of a curve

82

MICROECONOMICS FOR MANAGEMENT STUDENTS

Figure 2(a)

Figure 2(b)

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

83

joining all such points of consumer’s equilibrium along the parallel budget lines. Fig. 2(a) shows such a curve which is known as the Income Consumption Curve (ICC). The JCC is the locus of the points of consumer’s equilibrium obtained from a given indifference map, as the points of tangency between higher and higher indifference curves and higher and higher budget lines which are parallel. Thus, as we move from the left to right on the JCC, the consumer’s money income, his real income, and his utility level rise. Since there is a one to one correspondence, the consumer’s real income, is considered to be

the same thing as his utility level. Fig. 2(b) gives 3 different shapes of the JCC. (A) The JCC is upward sloping when both the goods are normal goods. (B) The JCC is backward bending when good X becomes an inferior good at some critical level of the consumer’s income. Good Yin this case is a normal good throughout. (C) The ICC is forward falling when good Y becomes an inferior good at some critical level of the consumer’s income. However, good X in this case remains a normal good throughout. If the /CC is vertical, the income effect for good X is zero, which means that

the consumer spends the same amount of money on good X whether he becomes richer or poorer, and the quantity of X demanded remains constant. Similarly, if the JCC is horizontal, the same amount of good Y is consumed irrespective of the consumer’s income. His consumption of such commodities may, however, vary if the price of the good changes. From the JCC, it is possible to derive a demand

curve

for a

good, say X, with regard to the income of the consumer where the price of the good along with the tastes and preferences of the consumer are held constant. Such a curve is known as the incomedemand curve or Engel Curve. Fig. 3 shows an Engel Curve for good X. As the consumer’s money income goes. on rising (with the prices of goods held constant), the consumer’s demand

for

good X would increase upto ON units of X. When the consumer’s income rises further to OM’, his consumption of good X tends to fall from ON to ON’. Most of the goods would tend to become inferior goods when the consumer’s income is at a sufficiently

high level. The threshold level of income which converts a normal good into an inferior good for a consumer is given by the point where the Engel curve bends backward (or changes direction).

84

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Income

Figure 3 In Fig. 5, OM is such a threshold level of income. It is important to note that though most of the goods tend to become inferior goods for a consumer, the threshold level of income for different

goods is different. Moreover, the threshold level of income in the case of the same good differs from one individual to another. Thus, market demand, which represents an aggregate demand of individuals, is not likely to show such a distinct fall as the average income in the economy increases. However, if the average income in an economy continues to increase substantially we are bound to find noticeable changes in the demand patterns for goods — some declining, some rising. If the quality of the same goods starts improving in response to consumer demand, it may also corroborate the income demand relationship discussed above.

Substitution Effect (SE) The substitution effect (SE) is defined as the change in the equilibrium consumption

of a good in response to a change in the

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

85

relative prices of the goods when the tastes and preferences as well as the real income (or the utility level) of the consumer remain the same. Thus, the substitution effect is always measured through a movement of the consumer’s equilibrium point along the same indifference curve. Since the consumer’s equilibrium is obtained as a point of tangency between the budget line (or the price line) and the indifference curve, the movement along a given indifference curve necessarily implies changes in the slope of the price line, making it either steeper or flatter than it initially was. Since the slope of the budget line is determined by the negative of the relative price of good X (i.e. —P, Tey), a steeper price line implies an increase in the price of X in relation to the price of Y; and a flatter price line implies a fall in the price of X in relation to the price of Y. The substitution effect thus measures the effect of a good becoming relatively cheaper or costlier on the basis of the quantity demanded of it without making the consumer better or worse off. Since indifference curves are assumed to be downward sloping and convex from below, the substitution effect always increases the demand for the relatively cheaper good and reduces the demand for the relatively costlier one. If there are only two goods, X and Y, good X can become relatively cheaper only when the good Y becomes relatively costlier. Thus, in a two-goods world, when X becomes relatively cheaper (and hence Y relatively costlier), the demand for X increases at the cost of demand for Y if the consumer is to remain on the same level of the satisfaction. Fig. 4 shows the substitution effect as a result of a fall in the relative price of X as given by the slope of the budget line A’B’ from the initial line AB. The consumer’s equilibrium point, therefore, shifts from E to E’. At point E, the consumer is

consuming OM of X and ON of Y. At £’, the consumer is consuming OM’ of X and ON’ of Y. Thus, in response to a fall in the relative price of X, the SE leads to an increase in demand for X from OM to OM’. If we consider an increase in the quantity demanded as a favourable effect, we can say that a good becoming relatively cheaper always experiences a favourable SE. When we consider more than two goods in the consumer's basket, it is not necessary that consumption of one good increases only if consumption of all other goods falls, with the utility level

86

O

MICROECONOMICS FOR MANAGEMENT STUDENTS

M

M’

B

B’

Figure 4 remaining constant. It is possible that interdependent utilities may lead some goods to move together in consumption even when the utility level does not change. However, in order to ensure that the consumer’s utility level remains the same, it is necessary for at least one of the goods to experience an opposite movement in consumption in comparison to the commodity experiencing a change in its price, other things remaining the same. Thus, demand for complementary goods would move in the same direction although there may be a change in the price of only one of them. The substitution effect in the multiple goods case

gets modified to that extent. However, the important result that all goods in a be complements. There has to be at substitute. However, all the goods in the substitutes.

it also provides us with consumer’s basket cannot least one good that is a consumer's basket can be

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

8/

Price Effect (PE) The price effect (PE) is defined as the change in the equilibrium consumption of the good in response to a change in the price of the good, all other things like the consumer’s

money

income,

prices of other goods, and his tastes and preferences remaining the same. When only the price of X changes, it leads to a change in the relative price of X in the same direction as well as a change in the consumer’s real income in the opposite direction. Conceptually, therefore,

a change only in the price of X elicits both the

income effect and the substitution effect. Let us consider am example to clarify this point. Let there be a consumer with a money income of Rs 1,000 which he uses to consume 50 units of X and 50 units of Y. The price of X is Rs 15 per unit and that of Yis Rs 5. Now suppose the price of X falls to

Rs 10 per unit. As a result of this change in P,, the relative price of X which was 3 © Rs 15/Rs 5), now changes to 2 (= Rs 10/Rs 5). The direction of the change in the relative price of X is the same as that in P,. This is not all. The consumer's real income has now increased because, with his given money income of Rs 1,000, he

can buy not only the initial combination of 50 units of X and 50 units of Y but more of either or both goods X and Y. The fall in the price of X by Rs 5 would obviously save him Rs 250 © Rs 5 x 50 units). This represents the minimum increase in his real income on account of the fall in the price of X. Thus, when the price of X falls, the consumer’s real income increases and the relative price of X falls. Xbecomes relatively cheaper. The demand for X will increase, firstly because of the substitution effect, since

good X has become relatively cheaper; and secondly because of the income effect, since the consumer experiences a rise in his real income. The price effect can thus be viewed as a combination of the income effect and the substitution effect.

Fig. 5shows the movement of the consumer’s equilibrium from the initial point £, which is the point of tangency between the initial budget line AB and the indifference curve showing u, level of utility, to the new point £, in response to a fall in the price of X. The new budget line AB” is flatter than AB with the same vertical intercept because the consumer’s money income and the price of the good Y have not changed. The total change in

88

MICROECONOMICS FOR MANAGEMENT STUDENTS

M, M,B

B

B”

Figure 5 the consumption of X given by MM, is called the price effect (PE). This total change of MM, in the consumption of X in response to a fall in P, can be broken up into two components: movement of the consumer’s equilibrium from point £ to point £, along the

same indifference curve showing u, utility level and the movement of the consumer’s equilibrium point from E, to £, on to the higher indifference curve showing u, level of utility. In terms of

the consumption of X, we can write the total PE of MM, as being made up of the SE of MM, and JE of M.M,. Thus,

(2)

MM, = MM, + MM, i.e. PE=SE+IE.

This is a very important result yielded by the ordinal approach to utility. Such a neat division of the price effect into the income and substitution effect is achieved by introducing an artificial or abstract price line A’B’ which is parallel to the new price line. Thus, the key to the division of the PE into JE and SE lies in considering the consumer’s purchases at the new relative prices without allowing him to become better or worse off than before.

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

89

Although such a division at the first sight might appear to be artificial and a purely theoretical construct, it has very important practical uses and applications which we will discuss later. Let us discuss here some relevant concepts which would be required to operationalize this division. As has been clearly stated above, the most critical element in

such a division is to identify a budget line A’B’ which would leave the consumer at the same level of the satisfaction as before with the new relative prices. This would require us to withdraw some purchasing power from the consumer in response to a fall in the price ofX. From the Fig. 5, this amount should be equivalent to AA’ in terms of Y or B’B” in terms of X. This is the level of purchasing power that needs to be withdrawn from the consumer in order to leave him neither better nor worse off than before in response to a fall in the price of X and is termed Compensating Variation in income. If we are considering a rise in the price of X, we have to provide additional purchasing power to the consumer in order to compensate him for the loss in his real income. Thus, the compensating variation in income is negative for afall in the price of a good and positive for a rise in the price of a good. In theory, it is possible to define the precise amount of compensation required to bring the consumer back to the same level of utility with the help of his indifference curve showing that particular level of the utility. In practice, however, it becomes virtually impossible to estimate such an amount with precision, and therefore, the compensating variation can only be approximated. One of the most popularly used methods of such approximation is termed the Cost-Difference Method. In this method, the initial point of the consumer’s equilibrium is used as a point of reference. As shown in the Fig. 6, point F is the initial equi-

librium along the initial budget line AB. When the price of X falls, the price line becomes AB,. The ideal compensating variation in income is given by the price line A,B, which is tangent to the same indifference curve (u,) at E,. However, the compensating variation of AA, is not possible to estimate with precision if the utility function of the consumer is not known. As an approximation, therefore, the cost-difference method is used in which a new price line A,B,, parallel to the new budget line AB, but passing through

90

O

MICROECONOMICS

M

FOR MANAGEMENT

M, M, BM),

|

STUDENTS

B,

Bs

|

By

Figure 6 the initial equilibrium point £, is considered. The compensating

variation is estimated as AA,. As it can be seen from the Fig. 6, the cost-difference method of estimating the compensating variation consists of determining the cost of the initial equilibrium combination at the new as well as the old prices of the goods and taking the difference between the two. It considers that the consumer would be approximately as well off as before the price change if he has just enough money income to buy his initial combination of the goods at the new prices. Thus, in our earlier example, the cost-difference would work out to Rs 250 (= Rs 5 of the fall in P, X 50 units of X). It may be noted that this amount always represents an underestimate of the true compensating variation in income required to leave the consumer as well off as before. This is because, if the old equilibrium point £ is available to the consumer after the fall in P, and the cost difference

being withdrawn

from him, his budget line

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

9]

A,B, would intersect the initial indifference curve (u,) at point E£. At £ he would now find that his subjective valuation of X given by the MRS of X for Y or the slope of his indifference curve (u,) was higher than the market valuation given by the new relative price or the slope of the new price line. Therefore, he would be better off substitutingX for Y in his consumption basket. In Fig. 6, it may be seen that the consumer can attain a higher indifference curve (u,) along the budget line A,B, at point £,. Such an approximation of the true compensating variation by the cost-difference

method,

therefore,

introduces

a systematic

bias in the

estimates of the income and substitution effects of a price change.

(3)

MM, = MM, + M,M, based on True Compensating Variation. PE = SE

(4)

IE

MM, = MM, + M,M, based on Cost-Difference method

Pie St. alk, (5)

Since, MM, = MM, + M,M, ets = SE, MM, eo

(6)

and M,M,=M,M, —M,M, bee SIE SM M7,

ee

eae

Thus, the method of cost-difference overestimates the substitution

effect and underestimates the income effect of a fall in the price of a good. What happens if we consider an increase in the price of X ? We leave this as an exercise for our readers.

Giffen Good As an important implication of the price effect being expressed as a sum of the income effect and the substitution effect, we may consider the case of those goods where both the effects W/E and

92

MICROECONOMICS FOR MANAGEMENT STUDENTS

SE) do not work in the same direction. For normal goods, the /E

is positive. Therefore, when a fall in the price of X is considered, the real income of the consumer rises and hence the positive income effect would lead to an increase in the quantity demanded of X. The substitution effect also works in favour of good X which has now become relatively cheaper. However, if X is an inferior good, the income

effect is negative for it. But the SE is always

favourable to a cheaper good. Therefore, the two effects would work in opposite directions. The net result could therefore be a favourable, zero, or unfavourable price effect. If the negative in-

come effect is overwhelmingly important, the favourable SE cannot help the quantity demanded of the good to increase in response to a fall in its price. We may, therefore, get a genuine exception to the law of demand. Such a commodity is known as a Giffen Good. For a giffen good, the price effect and the substitution effect have opposite signs. Fig. 7 shows the case of a giffen good whose demand decreases

Y

PE = SE +1E MM, = MM, + (-M.M,)

Figure /

CONSUMER'S EQUILIBRIUM AND DEMAND CURVES

93

as its price falls. It is obvious that a giffen good has to be an inferior good because the substitution effect always favours a cheaper good. The price effect can therefore be adverse only when the income effect is negative. However, it is not necessary that whenever the income effect is negative, the price effect should also be adverse. Indeed, in most cases of inferior goods, the income effect

is not strong enough to outweigh the favourable substitution effect and as a result the inferior good does not become a giffen good. For a good to become a giffen good, not only has it to be an inferior good but should have a very important place in the consumer’s budget. Besides, it should not have many close substitutes such that its substitution effect becomes weak. We can think of a poor consumer barely living above the poverty line who spends over 70 to 80 per cent of his budget on coarse grain. If the price of this item increases, he might no longer find it possible to afford the luxury of mixing it with other varieties of foods. He might therefore cut the consumption of the latter and increase his consumption of the coarse grain! Such an extreme reaction is relatively uncommon. Even when individual consumers find some good to be a giffen good, it is not likely that several consumers would consider the same good to be a giffen good in the given range of the price of the good. Actually, giffen good behaviour may be found, if at all, over

a very narrow price range. The price of the good cannot be very high nor can it be very low because otherwise it would not play a major role in the consumer’s budget. Such a relevant price range

is likely to vary from consumer to consumer. Thus, the giffen good is unlikely to be encountered in the market, and the market demand

curve for a good is not likely to behave abnormally.

Price Consumption Curve (PCC) The price consumption curve for good X (PCC,) is the locus of the points of the consumer’s equilibrium on his given indifference map when the price of the good X goes on falling with his income, and prices of other goods remain the same. It represents the path of the points of tangency between higher and higher indifference curves and a progressively flatter price line rotating in an anticlockwise direction from a given point of the vertical intercept.

94

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Figure &

Fig. S shows the PCC, derived from different points of consumer's equilibrium such as E,, E,, E,, etc. It can be seen from Fig. & that PCC,, always starts from the intercept of the budget line on the other axis. This intercept remains the same when we consider a

fall in the price of X because we keep the consumer’s money income and the price of good Y constant. Secondly, as we move from the left to the right along the PCC,,, the price ofX keeps falling; and the consumer’s utility level goes on rising. The PCC, normally has a U-shape. In other words, it falls in the beginning when the price of X starts falling from a very high level; it becomes almost horizontal or attains a minimum at medium prices; and then, rises when the P, falls to very low levels. The PCC, may be considered as a parent curve for an individual’s demand curve. The demand curve can be derived from the PCC, by associating the price of X with the quantity demanded.

This is: shown, .in, Fig, 9. The prices F,, F,, and 7, and te

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

95

Figure 9

corresponding quantities M,, M,, and M,, representing the consumer’s equilibria at the respective prices as shown in Fig. S, are connected to one another, yielding the familiar demand curve of the consumer for good X. This process of deriving the demand curve from the PCC, clearly shows that different points on the demand curve show the consumer’s equilibrium at different prices, other things remaining the same. Thus, at every point on the demand

curve,

the consumer

gets the maximum

satisfaction. Moreover, the points on the demand

possible

curve show a

rising level of utility as the price of X continues to fall. This happens because the demand curve is based on the assumption of constant money income. As the price of X falls, the consumer’s real income continues to rise, and with it the consumer’s utility.

Compensated Demand Curve It is possible to derive a different type of demand curve from the consumer’s demand theory discussed so far. The ordinary demand curve as derived in Fig. 9 is based on the assumption of

96

MICROECONOMICS

FOR MANAGEMENT STUDENTS

constant money income on the part of the consumer. Alternatively, it is possible to derive a demand curve based on the assumption that the consumer’s real income (or utility level) remains constant.

In order to derive such a demand

curve, the

division of the price effect into the substitution effect and the income effect is very crucial. The income effect of a price change arises only because the consumer’s real income changes as a result of the price change in relation to his constant money income. If on the other hand, we assume that the consumer is always promptly compensated in order that he maintains a constant real income in response to a price change, the component of the income effect in the price effect would prove to be zero. The price effect would then coincide only with the substitution effect. Thus, the demand curve based only on the substitution effect is termed the compensated demand curve. This type of demand curve is also known as the Real Income constant Demand Curve or the Hicksian Demand Curve sincé it was popularized by J.R. Hicks. As against this, the ordinary demand curve is Rnown as the Marshallian Demand Curve since it is based on the demand theory popularized by Alfred Marshall. Fig. 10 shows both these types of demand curves and their relationship. The DM curve is the ordinary or the Marshallian demand curve and the DH is the compensated or Hickian demand curve. Let us consider a price OP, of good X. At this price, the consumer demands OX, units of the good X. Now, when the price falls to OP, other things remaining the same, the substitution effect (SE) and the income effect (JE) occur and, as a result, the quantity demanded of good X expands. Let the SE be given by X,X, and

the JE by X,X,. Fig. 10(a) represents the case of a normal good where the JE is positive. Fig. /0(b), on the other hand, shows the case of an inferior good where the JE is negative. The Marshallian demand curve (DM) incorporates both the SE and JE since it assumes constant money income on the part of the consumer. The Hicksian demand curve (DH) is based only on the SE because itassumes constant real income on the part of the consumer. Thus, the Hicksian demand curve would be steeper than the Marshallian counterpart for a normal good, but the reverse is true in the case of an inferior good. Another important point regarding these curves is that a distinct

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

P

Normal Good

O

WM, Bosco, Figure 10(a) Inferior Good

0

X,

X,

Figure 10(b)

X)

o¢

98

MICROECONOMICS FOR MANAGEMENT STUDENTS

Hicksian demand curve would pass through every point on the Marshallian demand

curve. This is because, on the Marshallian

demand curve the consumer’s real income goes on rising as the price continues to fall. Since the Hicksian demand curve is based on the assumption of constant real income, for every level of the consumer’s real income, there would be a distinct Hicksian demand curve. Similarly, through every point on the Hicksian demand curve, a distinct Marshallian demand would pass. Again, this happens because the Hicksian demand curve would show a constant level of the consumer’s real income, but when the price of X falls, the consumer’s

real income cannot remain the same

unless some money income is withdrawn from him. Thus, the Marshallian demand curve would be shifting downward in the case of anormal good and upward in the case of an inferior good.

Consumer’s Surplus The demand curve can be used to estimate the amount a consumer is willing to spend for the given level of the consumption of a commodity rather than go without it. The excess of such an amount over what he actually pays for the good is a surplus that accrues to the consumer for consuming the given quantity of the good. This notional surplus is known as Consumer’s Surplus. Let - us see how we can obtain it from the demand curve of an individual. | Fig. 11 presents an individual’s ordinary demand curve for good X. Ata price of Rs 6 or more, the individual does not demand any amount of X. If the price is Rs 5, he would demand 1 unit of X. Similarly, when the price is Rs 4 and Rs 3, he would demand 2 and 3 units of X respectively. The same data can be interpreted to mean that the consumer would be willing to pay Rs 5 for the first unit of X rather than go without it. Similarly, he would be willing to pay Rs 4 for the second unit and Rs 3 for the third unit of X rather than

go without it. Thus, the total amount that the consumer would be willing to pay for two units of X is Rs 9 (= Rs 5 + Rs 4). However,

he actually pays only Rs 8 © price of Rs 4 x 2 units of X). Thus, his notional gain is Rs 1 when he consumes 2 units of X. For the consumption of the third unit of X, he is willing to pay Rs 3. He is,

CONSUMER'’S EQUILIBRIUM AND DEMAND CURVES

99

Figure 1] therefore, willing to pay altogether Rs 12 (= Rs 5 + Rs 4+ Rs 3) for the 3 units of consumption of X rather than go without it. But he is required to pay only Rs 9 (= Price of Rs 3 x 3 units of X). Thus, his notional gain is equivalent to Rs 3 which represents the consumer’s surplus. In more general terms, the consumer’s surplus is measured as

the excess of the area under the demand curve up to the point of his consumption and the area of the rectangle formed by the price and quantity demanded of the good. In other words, the area of the triangle formed by the demand curve above the horizontal line from the given price measures the consumer’s surplus. If the demand curve is nota straight line, the consumer’s surplus would be given by the area of the figure formed by the demand curve, vertical axis, and the horizontal line through the given price of X. It may be observed from Fig. // and the definition of consumer’s surplus that it goes on increasing as the price of the good falls. Earlier we had noted that the consumer’s total satisfaction would go on increasing as the price of X falls. It is, therefore, obvious that consumers would always prefer a lower rather than a higher

100

MICROECONOMICS

FOR MANAGEMENT STUDENTS

price for the good. The consumer’s surplus as a concept provides an estimate of the potential benefits accruing to consumers. Alternatively, it also shows the potential revenue a seller or the government can raise from the consumers of a good under very special conditions.

Market Demand

Curve

The market demand for a good is an aggregation of individual demands for the good. Since demand for any good is always at a price, the market demand for a good as a concept would make sense only when the individual demands for the same good is aggregated at the same price. Geometrically, the market demand curve is derived as a horizontal aggregation of all the individual! demand curves. Fig. /2 illustrates this point. From Fig. /2 it can be seen that one of the important determinants of the level of the market demand curve is the number of individuals or the population in the market. It should be noted that the population includes not only those who are actually consuming or demanding good X at a particular price but also those likely to enter the market as consumers at lower prices. It

IEA

ind

ne. Figure 12

Market

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

101

is, therefore, customary to consider the whole population in an economy since they all could be potential consumers of the good. Depending on the nature of the good, however, one can consider appropriate sections of the population as the determinant of the demand for the good. Another thing which can be readily observed from the Fig. 12 is that the curvature of the market demand curve is not necessarily the same as the individual demand curve. In the Fig. 12, we have

considered all the three individual demand curves to be linear or straight lines. However, because their vertical intercepts differed, the market demand curve has kinks at two points. If there are several individuals, each of them entering the market sequentially at lower and lower prices, the market demand curve would have far too many kinks which would make it appear to be virtually a smooth curve. It is also possible to argue that the market demand curve is not affected by the peculiarities of a few consumers. If there are consumers insensitive to small price changes, their demand curves are likely to be vertical in the given price ranges. Again, the very similar logic of horizontal summation ensures that the market demand curve may not have any such vertical patches. This is because not all consumers in the market would be price insensitive. Those who are, would also not be price insensitive in the same range of prices. Moreover, as discussed earlier, market behaviour in terms of quantity response to the price change is largely dictated by some individuals who are price sensitive and are active in the market. /t is this price-sensitive group of marginal buyers that make the market demand curve a downward sloping one. Similar logic applied to the horizontal aggregation of individual demand curves also rules out the possibility of a giffen good being observed in the market. This is because the giffen good case arises only in some specific range of prices of the commodity for different individuals. Several other conditions, as discussed above, have also to be met to turn an inferior good into a giffen good. All individuals are not likely to experience such things simultaneously and to the same extent. The market demand curve is, thus, more likely to iron out abnormal individual behaviour.

An important assumption for deriving the market demand curve simply as a horizontal summation of the individual demand

102

MICROECONOMICS

FOR MANAGEMENT STUDENTS

curves is that they are all independent. One individual’s demand curve does not affect another’s in any way. However, there are certain types of goods for which such an assumption does not hold. They are peculiar goods, some of which may appear to pose a problem for the law of demand. Let us consider them briefly. One category is of those where individual’s utility derived from consumption of the good depends on how many persons consume the good or how much of the good is consumed by society as a whole. There are some goods, such as fashion goods, intoxicants, drugs, tobacco,

alcohol, etc., whose

utility to an individual

in-

creases with greater consumption of the good in the society at large. Such goods are subject to the bandwagon effect. Fig. 15 depicts the case of individual and market demand for a bandwagon good. It is assumed that as the price falls, the demand for the good expands not only because the existing buyers demand more but also because new buyers enter the market. As a result, the individual’s demand curves (D, and D,) shift upward leading to an upward shift in the market demand curve (D,, to D_) for a bandwagon good. Thus, the quantity demanded associated with a lower price is not given by the movement along the old market demand curve (D,,). The effective market demand curve (DD) of a bandwagon good is likely to be flatter than the curve derived

Ind. 1

Ind. 2

\ Market

Figure 13

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

103

as a simple horizontal summation of the individual demand curves (D,,,). Such commodities are most likely to be highly price sensitive in the aggregate unless by their very nature they are habit forming. Some of the bandwagon goods belong to this category of the habit forming goods which further complicates the analysis since their response is not symmetrical with price. In other words, when

the price falls their demand

does increase, but when the

price rises, their demand may not fall to the same extent. This happens because, as consumption increases, the consumer forms the habit which shifts his demand curve upward so that when the price again rises, his demand does not fall. Another category of goods are by nature quite the reverse of bandwagon goods. Some goods have greater attraction because they appear to be rare and exclusive. People consuming such goods derive greater satisfaction from their consumption if they are consumed by fewer individuals because of the snob appeal this imbues them with. Demand for exclusive clothing, custommade suits, paintings, antiques, art objects, holiday resorts, etc.,

may be considered such goods would them. As shown in D,, etc.) would shift

Ind. 1

to be snob goods. The individual’s demand for decrease if more people began to consume Fig. 14, the individual’s demand curves (D,, downward as the number of buyers increased

Ind. 2

D

Figure 14

Market

104

MICROECONOMICS

FOR MANAGEMENT STUDENTS

in response to a fall in price. As a result, the market demand curve also shifts downward with a fall in price. The effective demand curve (DD) may, therefore,be steeper than the market demand curve (D,,) derived through a simple horizontal summation. In an extreme case, the effective market demand curve may prove to be either vertical or even upward sloping in some cases. This is because the new entrants may not be heavy consumers while the existing consumers might be. However, this may occur only for a certain middle range of prices. At low prices, the good may no longer remain a snob good. There are also some peculiar types of goods meant usually for conspicuous consumption. Diamonds, necklace, wedding rings, earrings, celebrations, dresses for special occasions, etc. are among

the commodities where high prices are a source of special satisfaction to the consumer. The higher the price he pays, the greater the satisfaction he derives. Such commodities are subject to what is known as the Veblen effect. Higher price leads to an upward shift in the consumer’s demand for such goods. Fig. 15 shows this

case. The individual’s demand curves (D,, D,, etc.) shift upward in response to higher price in the market. If the price is lower, the individual

demand

curves

would

shift downward.

In this way,

such goods are very similar to the snob goods. As is shown in the

Py

Py Ind. 1

:

Py Ind. 2

Market ,

or Ys

.

D XN \ \

\

\ Me

\D,,

D Dy

Figure 15

CONSUMER’S EQUILIBRIUM AND DEMAND CURVES

105

Fig. 15, the effective market demand curve (DD) in the case of goods subject to the Veblen effect may slope upward. This, again, is aphenomenon restricted to a certain range of prices of the good. Outside the range, the good may not be subject to the Veblen effect because it may not be considered worthy for purposes of conspicuous consumption! However, the Veblen effect can produce a market demand curve of a good which ts upward sloping at least in a certain price range. The giffen good effect, which can

produce an upward sloping demand curve at the level of the individual, may not produce an upward sloping market demand curve for any price range. The Veblen effect applies to super luxury goods whereas the giffen effect applies to a special type of inferior goods. That is why we expect only a small number of buyers of the goods subject to the Veblen effect and a large number of buyers of the giffen good. Ultimately, it is the sheer number of buyers that irons out the irregular behaviour. When the price of a Veblen good also falls sufficiently, there are a large number of buyers and the demand for the good would again behave as predicted by the law of demand.

Chapter 5 Some Applications of the Theory of Demand

There can be several real life situations where the demand theory can be directly or indirectly applied. Here, we consider only some selected applications which have become very popular because of the special insights they provide of the demand theory. The situations selected here are in the fields of subsidies, commodity taxation versus income taxation, labour supply, supply of marketable surplus of agricultural goods, voluntary exchange, and use of cost of living index numbers.

Subsidies in Cash or Kind

The poor in various countries receive special consideration from society as a whole and politicians in particular. Specific programmes subsidizing the poor in one way or another are announced to alleviate their poverty. If we closely examine the contents of such special poverty alleviation programmes, we realize that they are based either on direct income subsidy or direct food, clothing, or shelter subsidy. The consumer demand theory can be used to analyse which of these two alternative ways of subsidizing the poor is more

efficient. In other words, if the stated objective of

the policy is to improve the welfare level of poor consumers, consumer demand theory shows us whether a direct income subsidy or an equivalent amount of a commodity (say, food) subsidy, is the better way of achieving the goal. A commodity

subsidy can be given in two different ways: either in the form of the commodity itself (e.g. food stamps in the West, or the midday meal and food-for-work schemes here); or in the form of subsidized price of the commodity (e.g. the popular election promise of making wheat/rice available for Rs 2 per kg). The direct income

SOME APPLICATIONS OF THE THEORY OF DEMAND

107

subsidy may be given in the form of tax concessions or direct

income transfers (e.g. unemployment allowance, welfare payments, etc.). Although these types of choices arise in the field of public policy, the same types of situations can also arise for a business enterprise. The nomenclature may differ but the nature of the problem from the analytical angle would remain the same. For instance, workers in a particular category need to be given some incentive. Should it be in the form of cash or kind? Let us consider a more concrete case of the managerial paypackage in a company. The company accepts in principle that its executives need to have access to health clubs and that it should facilitate this. It considers 10 visits per executive per month to be adequate. The market price per visit is Rs 120. The options before the company are: (a) It reimburses its executives for 10 visits per month; (b) it subsidizes each of the executives to the extent of Rs 100 per visit to clubs of his or her choice; and (c) it pays each of the executives a sum of Rs 1,200 per month whether or not the Money

O

B

Figure 1

D

F

108

MICROECONOMICS

FOR MANAGEMENT STUDENTS

executive visits the health club. We can consider all these three options on a single diagram if we measure the number of visits to the health club per month by an executive on the horizontal axis and denote it by X, and if the executive’s money balances are measured along the Y-axis. Fig. 1 shows the budget constraint to be AB without any intervention from the company. The slope of the budget line (AB) shows the market price of a visit to the health club. Different executives would be visiting the health club with differing frequency, depending on their preferences. For an executive with what may be considered to be the normal preference, the point of equilibrium would lie at £, along the budget line AB. For an executive with a stronger preference for visits to the health club, the equilibrium point would be £,. Similarly, for an executive with strong preference against health club visits, the equilibrium point would be E£.,. Now under option (a) the executive’s budget constraint would shift rightward by 10 visits per month and would become ACD as shown in the Fig. 7. This is because, providing 10 free visits to the

health club amounts to raising the real income of the executive

to that extent. It can be readily seen that all the three types of consumers are encouraged to visit the health club with greater frequency now than before the scheme was instituted. Moreover,

all the three types of consumers experience an increase in their utility levels. When we consider option (b), the budget constraint shifts from AB to AF. This is because under option (b), the company gives a subsidy (or an incentive!) of Rs 100 to its executives for every visit to the health club. This amounts to reducing the price of a

visit by Rs 100 for its company’s executives. As a result, the budget line AB becomes flatter and rotates in an anticlockwise direction to the position of AF with the centre at A. If the average number of visits per executive is 12, the total amount of subsidy involved would remain the same. In this case, as is evident from Fig. /, all the three types of consumers are moving on to a higher level of satisfaction than they would have without any scheme. But the consumers with a strong preference for health club visits! are | The second type of consumers whose initial equilibrium point is at £, in FI;

SOME APPLICATIONS OF THE THEORY OF DEMAND

109

benefiting to a far greater degree from option (b) than from option (a); whereas the consumers with a strong preference against health club visits’ benefit much less from option (b) than from option (a). The first type of consumers whose preferences for health club visits are normal or average (initial equilibrium point E’,) are more or less indifferent to the two options. Now, let us consider the option (c) in which the company pays each of the executives a sum of Rs 1,200 per month or encashes the unutilized number of visits to the health club subject to a maximum of 10 visits per month. Under this option, the budget constraint of the consumer would shift from AB to GD. It is a parallel shift since the market price does not change and real income of the consumer increases by Rs 1,200 per month. In this case, the consumers of the first and second type (corresponding to the initial equilibrium points £, and E, respectively) are neither better nor worse off in comparison to option (a). But the third type of consumers (with the initial equilibrium at £,) are better off than they would have been under the option (a). Their consumption of X would be higher than the initial equilibrium point E,, but would be less than what it would be under option (a). Actually, such executives would prefer the encashment option and yet would end up consuming more of X than they would had there been no scheme. The second type of consumers who have a stronger preference for visits to the health club (with initial equilibrium at £,) are, however, worse off under the option (c) than under option (b). Thus, the company manager should assess the type of consumers his executives are. If most of them have an average or normal preference, they would be more or less indifferent to the

three options in terms of the scheme for encouraging visits to the health club. However, if most of the executives have very strong preference

for visiting the health

club, they would

prefer the

option (b) of price incentive over the other two options. On the other hand, if most of the executives have a strong preference against visiting the health club, they would prefer option (c) of cash incentive over the other options. With a large number of executives in a company, everybody cannot be satisfied. Some 2 The third type of consumers whose initial equilibrium point is/at £, in Fig. 1.

110

MICROECONOMICS

FOR MANAGEMENT STUDENTS

group or other is bound to be grudging or complaining about the wrong choice of option. However, the manager’s skill lies in assessing the consumption preference of the majority. Moreover, the principal goal of the company in operating a particular scheme must also be borne in mind.

Commodity versus Income Taxation Another important application of the consumer demand theory is in the field of taxation. Several alternatives are available to the government. The government can levy direct taxes such as income tax, wealth tax, etc., or indirect taxes like excise duty, tariff,

sales tax, etc. Indirect taxes are. basically commodity taxes. By imposing such taxes, the government interferes with market functioning and influences the relative prices of commodities. When imposing a tax burden on consumers, the government has several objectives. Raising tax revenue is only one of these. The government may also want to encourage the consumption of some goods and discourage the consumption of others. When we analyse the options available to the government in choosing a particular tax policy, its objectives have to be borne in mind. For our purpose let us assume that the government is interested only in raising revenue through taxation. The basic choice is between commocdity tax and the income tax. This will also help in an appreciation of the basic drift in the economic policy reforms undertaken in a number of developing countries in recent times. For the sake of simplicity, let us consider only two goods X and Y. The government considers imposing a duty on good X as option (a). As option (b), the government considers imposing an income tax that garners an equal amount of revenue. The consumer’s equilibrium is shown in Fig. 2. AB is the consumer's budget constraint without any intervention from the government.

E is his equilibrium point with u, level of satisfaction. Now, the government imposes a duty on good X which results in an increase in the price of X in absolute and relative terms.’ This 3 This is because price of Y would remain the same. If the government also imposes a duty on good Y, but at a rate lower than on good X, the relative price

SOME APPLICATIONS OF THE THEORY OF DEMAND

111

Figure 2 would shift the budget line AB downward in a clockwise direction to AB’. The consumer’s real income is reduced and he is obviously worse off when the government imposes a tax on X. He shifts to the lower level of the indifference curve with u, utility. His consumption of good X is reduced. Whether his consumption of good Y is reduced or not depends on the nature of the price consumption curve for X. At the new point of equilibrium £, under the option (a) of the commodity tax, the government would get by way of tax an amount equal to the vertical distance £, k between the new point of good X would still increase. Only when the government imposes an ad valorem duty on all commodities at the same rate would their relative prices remain unchanged.

112

MICROECONOMICS FOR MANAGEMENT STUDENTS

of equilibrium £, and the old budget line AB. If the government collects the same revenue by way of the income tax (i.e. option (b)), the new budget line becomes A’B”, as shown in Fig. 2. The

new budget line A’B” is parallel to the old budget line AB since the relative prices of the goods do not change when the government levies the income tax. The new budget line A’B”’, however, intersects the indifference curve u, at point £,. Actually, at £,, the consumer’s subjective evaluation of good X in terms of good Y is higher than the relative price of X in the market. He can, therefore, gain more satisfaction by substituting X for Y in his consumption. As shown in the Fig. 2, the consumer is better off

with the equilibrium at point £, along A’B” when the government imposes

the equivalent income

tax as compared

to his equi-

librium at £, along AB’ when the government imposes the commodity tax. With the commodity tax, the consumer can get a maximum utility of u); whereas with the income tax, he can reach a higher utility level of u,. Under both the options, the government’s tax revenue remains the same. However, as can be seen from Fig. 2, his consumption of X is higher under the income

tax than under the commodity tax. It may however be noted that even

with income

tax, the consumption

of good X is less than

was the case without any tax, assuming of course that X is a normal good. The efficiency of the alternative taxes can be measured by the amount of revenue collected, leaving the consumer at the same

level of satisfaction after the tax. Again referring to the Fig. 2, we can see that the government can collect more revenue through the income tax leaving the consumer at the same level of utility (u,) as in the case of the commodity tax. The commodity tax of AA’ (in terms of good Y ) imposes the same burden on the consumer as the income tax of AA”. Alternatively, the government can collect an amount of A’A” more from the consumer without affecting his level of satisfaction if it decides to shift from the indirect taxation to the direct taxation. The income

tax, thus,

appears to be more efficient than the commodity tax. Why then does the government persist with commodity taxation? The possible explanation could be that developing countries are poor with a large majority of their population hovering around the poverty line. The base for income tax is therefore likely to be

SOME APPLICATIONS OF THE THEORY OF DEMAND

113

very narrow. Moreover, most of the concerned countries believed

in interventionist policies, and direct intervention with market forces to channelize the resource allocation in the ‘desired’ direction was considered very important. Discouraging the consumption cf a particular commodity was the specifically stated objective in several cases of the commodity taxation. Our analysis here shows that this was not without costs. The society was paying the

cost of real income loss of A’A” to achieve a reduction in the consumption of the good X from point £, to E,. This was the society's explicit preference. With active consideration of economic policy reforms these societies have started questioning the desirability of such costs. In order to avoid these types of costs, policy reforms in taxation emphasize three basic directions: (i) shift from the specific duty structure to the ad valorem duty structure; (ii) reduction in the number of duty slabs in order to reduce variation in duty rates; and (iii) shift from the indirect taxes to the direct taxes as a source of tax revenue.

Income Tax and Labour Supply From

the above

discussion, we

need

not jump to a conclusion

that all is well with income tax, that the incidence of the income tax falls equally on all commodities and hence is neutral; or that income tax does not distort any relative prices in the society. Actually, income tax is a tax on the consumer’s income. It, therefore, has implications for the use of resources such as labour, capital, land, and enterprise, at his command. For instance, if the

consumer’s income largely comes from wages he earns, the income tax acts as acommodity tax on his labour. In order to analyse this situation, it is necessary to consider the consumer’s choice to supply his labour in the market for wages or income.

Fig. 3 shows an indifference curve diagram between the consumer’s income (Y-axis) and his labour supply (X-axis). Since work or labour is a discommodity, the consumer needs compen-

sation if he has to work. His indifference curves are, therefore,

upward sloping as shown in Fig. 3. ON is the maximum number of hours that a consumer can work. This is determined by physical constraints. The budget constraint in this case is denoted by an

114

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Income

IN

Labour

Figure 3 upward sloping line like OB. Its slope is determined by the wage rate per hour. The higher the wage rate, the steeper the curve with the centre at the origin (O). The consumer reaches equilibrium at point E along the budget line OB, reaching the utility level u,. At £, he is in equilibrium earning EL of the wage income and working for OL hours. He consumes LN hours of leisure. Now, let us consider the case when the government imposes an income tax whereby his effective wage rate goes down and his

budget line shifts to OB’ in a clockwise direction. This happens because the income tax becomes a specific tax on the consumer’s

labour. Like all commodities that are taxed, the consumer’s supply of labour is also likely to be reduced under the normal circumstances when it is taxed. Thus, the new equilibrium point is likely to be at E, where he supplies only OL, hours of labour and earns EL, of wage income, actually increasing his consumption of leisure. This is very consistent because the wage rate is a price

SOME APPLICATIONS OF THE THEORY OF DEMAND

115

or cost of enjoying leisure. When the income tax is imposed and hence the wage rate available to the consumer falls, the leisure becomes cheaper. It is quite understandable if its consumption, therefore, increases.

It is not however necessary that the consumer will always respond as we have shown in Fig. 3. This is because the consumer’s response can be considered as a price effect of a fall in the effective price of labour as a result of the income tax. We can break up this price effect into its income effect and substitution effect by first imposing the amount levied as head-tax (which is a flat rate of tax not related to the level of the consumer’s income)

equal to BB” which would leave the consumer at u, level of uiility. The budget constraint OB would shift downwards but parallel to the new level, AB”. This is because the head tax is neutral with

regard to all commodities including labour and hence does not distort the relative price of labour. The new equilibrium point

E, along the budget line A’B” at the utility level of u, helps to separate the income effect from the observed price effect of the movement of E to E,. In this case, the movement from E to E£, in

Fig. 3 is the income effect (JE) and E, to E, is the substitution effect (SE). In terms of labour, the price effect (PE) of LL, can be divided into LL, of JE and L£,L, of SE. The consumer’s demand theory clearly enables us to see that the income effect of a proportional income tax, by making the consumer poorer, compels him to reduce his consumption of leisure from LN to L,N, and hence compels him to work more from OL to OL, hours. However, the substitution effect of the income

tax which reduces the relative price of labour (and hence leisure), induces him to consume more leisure from NL, to NL, and work less from OL, to OL,. In such cases, the income and the substitution effects work in opposite directions even when the good (namely leisure) is a normal good. This is because the fall in the price of labour does not result in increased but reduced real income. Under such circumstances, the response of the consumer to the increased income tax could be anything in terms of his labour supply. It all depends on the relative strengths of the income and substitution effects. By the same token, it can be argued that a reduction in the income tax rate may also evoke any type of response from the consumer in terms of his labour supply. It is

116

MICROECONOMICS

FOR MANAGEMENT STUDENTS

not necessary that it must always increase in response to a cut in the income tax rate. However, if the objective of the government is to increase the work effort or labour supply in the economy through taxation, the most efficient policy seems to be to impose head tax which would compel consumers to increase their labour supply and cut their consumption of leisure without distorting the relative price of labour.

Marketable Surplus Another application of the consumer’s demand theory lies in analysing the behaviour of producers who are also consumers of their own products. Farmers in developing and poor countries represent an excellent illustration. However, the same logic would apply to the case of a housewife’s allocation of time between the outside work and household responsibilities. Her time for outside

Money

Figure 4

SOME APPLICATIONS OF THE THEORY OF DEMAND

117

work may be considered to be her marketable surplus. Let us consider the case of supply of the marketable surplus of foodgrains. Fig. 4 shows a farmer’s equilibrium as a consumer of the foodgrains which he himself produces. On the X-axis, we measure units of foodgrain denoted by X. On the Y-axis, the farmer’s money income is measured. It is assumed that the farmer derives his income only through the sale of his produce in the market. His budget constraint is given by the line AB whose slope measures the price of the foodgrains. Point B on the X-axis shows the total foodgrains produced by the farmer during a given period. His equilibrium is at point £ where his self-consumption is ON units of foodgrains and his marketable surplus is VB units. He attains u, level of utility with his money income being EN. Now when the price of foodgrain increases, his budget line rotates in a clockwise direction from BA to BA’. This is because, when the price of foodgrain increases, his potential real income increases rather than falls. His new equilibrium is at point £, along the new budget line BA’ on the higher indifference curve u,. At the new equilibrium level, his self-consumption of foodgrains increases from ON to ON’, and hence his marketable surplus falls from NB to N’B. Thus, at a higher price, the farmer’s supply of the foodgrain to the market falls instead of rising. This is not a very abnormal response. It can be explained in terms of income and substitution effects. As shown in the Fig. 4, B’A” is the artificial budget line drawn parallel to the new price line BA” so as to ensure that the consumer is compensated for the rise in his real income. With the budget constraint B’A”’, the consumer is neither left worse or better off than before the price rise. His level of satisfaction remains the same as u,. But his equilibrium would have been at point £,. Thus, the movement from point£ to E, along the same indifference curve (u,) measures the substitution effect (SE). Then, the movement from point E, to E, represents the income effect (JE). The total price effect (PE) on the foodgrain axis is measured by NN’. It can be divided into the

SE of NN” and JE of N’’N’. The substitution effect works against his self-consumption since the item of his consumption, namely, the foodgrains, have become costlier than before. But the income effect is highly favourable and more than compensates for the

118

MICROECONOMICS FOR MANAGEMENT STUDENTS

negative effect of the SE, so that the final outcome is an increase

in the farmer’s self-consumption. This happens because the farmer feels rich enough to consume a larger quantity of his own product though it has become costlier. He does not consider his own product inferior. It is important to note that the price rise is so large that even after his increased self-consumption and consequent fall in his marketable surplus, he realizes a higher income. This is not likely to happen with any rise in the price of foodgrains, but only when the prices have risen to a very high level. Moreover, all the farmers are not likely to simultaneously experience such a backward bending supply curve of their foodgrains. The threshold price would be different for different farmers. The aggregate supply of foodgrains is, therefore, not likely to be backward bending, though it may prove to be much less price sensitive at very high prices, particularly in poor, developing countries with a large subsistence farming sector.

Voluntary Exchange With the aid of the indifference curve technique, show that voluntary exchange of goods is always cial to both the parties involved in the exchange. two individuals, A and B, with two goods X and has X, and Y, amounts of the two goods X and

it is possible to mutually benefiLet us consider Y. Individual A Y, respectively,

and individuals B has Xp, and Y, amounts. If they consume these quantities themselves, both would derive some satisfaction, as indicated by their respective indifference maps shown in Fig. 5. For the sake of simplicity we have drawn only three indifference curves for each individual. Both the individuals’ initial endowments are shown on their respective middle indifference curves.

It may be observed from Fig. 5 that the two individuals may or may not have different preferences or utility functions, as reflected by the nature of their indifference curves. Now if these two individuals happen to come into contact with one another they will soon discover that their marginal subjective valuation ofX in terms of Y (i.e. the slope of the indifference curves at points E, and E,) are very different. Individual A has a relatively

higher subjective value of X in terms of Y than individual B. It,

SOME APPLICATIONS OF THE THEORY OF DEMAND

Individual A

Y

119

Individual B

Figure 5 therefore, implies that individual & has a relatively high subjective value of Y in terms of X than individual A. Both would, therefore,

find it useful and beneficial to exchange the commodities that they value relatively less for those that they value more. In order to show the possibility of voluntary exchange leading to mutual gain, a tool frequently used, is known as the Edgeworth Box. It uses the type of information we have depicted in Fig. 5. It is a simple geometrical device to develop some important concepts relevant in the theory of exchange. Fig. 6 shows the Edgeworth Box created with the help of the indifference maps of the two individuals A and B. As Fig. 6 shows, the initial endowments of the two individuals in terms of each of the goods is summed up to arrive at the size of the Box. Thus, the height of the box is given by the total Y (= YA + YB) and its length by the total X (= XA + XB). The Box is then created by superimposing individual 6’s indifference map by turning it upside-down on individual A’s indifference map, as shown in the Fig. 6.

The initial endowment points £, and £, coincide in the Box at point £. As Fig. 6 shows,

this point lies on

individual

A’s

indifference curve Uu,, and on individual 5’s inverted indifference curve Up. Since the indifference map of individual B is inverted, a higher level of his utility is obtained on the indifference curve

120

MICROECONOMICS

FOR MANAGEMENT STUDENTS

(Yala)

77 O

x,

Bess Figure 6

farther away from his origin (O’) or as we move in the south-west direction. Individual A’s utility increases when we move in the usual north-east direction. The most interesting feature of this Edgeworth Box is that the slope of the tangents to the curves measures the Marginal Rate of Substitution (MRS) of X for Y in the case of the respective individuals. It does not have to be adjusted or inverted. At point F, which shows the initial endowment point for both the individuals simultaneously from their respective origins, the slopes of the indifference curves u,, and Up. passing through £ are totally different. Indifference curve Uy) Of individual A is much steeper than the ug, curve of individual B at point £. As a result, individual A can exchange Y for X with individual B. If the exchange takes place along the

indifference curve Ug, individual A can go on exchanging Y for X until he reaches a point like C’ where he attains the utility level u,3. At point C’ the exchange would stop because the slopes of the two indifference curves are equalized. In other words, at C’, the marginal subjective valuation of X in terms of Y of both

SOME APPLICATIONS OF THE THEORY OF DEMAND

121

the individuals is the same and hence there is no incentive for further exchange. This point C’, however, is by no means a unique solution. It actually represents an extreme case where one of the parties does not gain anything from the voluntary exchange, the other party

cornering all the gains. The other extreme of similar nature is given by the point C in Fig. 6, where individual A remains on the same

indifference curve as U,5, and all the gains of the exchange accrue to individual

B who moves on to his higher indifference curve

Up. In between these two points C and C”, there would be an infinite number of points at which the MRS of X for Y of both the individuals would be equalized. Of course, as we move along the locus of such points from C to C’, individual A’s utility level would increase and individual B’s would decrease. However, so long as

the final point of exchange lies between C and C’, both the individuals gain through the voluntary exchange when compared to the initial endowment point. The line defined by the locus of the points of tangency of two sets of indifference curve, such as line CC’, is known

as the Contract Curve. This is a very useful

concept in the negotiations.

Price Index and Welfare

Dearness allowance (DA) is a very important component in the pay-package of most employees in the organized sector. The quantum of such DA is invariably linked to some price index which is mutually agreed upon by the employer and the employee. Such agreements are often made by the unions on behalf of the workers and employees. Pay commissions may themselves recommend the precise formula and the type of price index to be used for the purpose. DA is basically intended to protect employees from the adverse effect of price rises, and it illustrates the direct application of the consumers’ demand theory. It is interesting to see how different types of price indices can be formulated enabling inferences on consumers’ welfare to be made over time. For this purpose, we have to make some simplifying assumptions. For instance, we assume that the tastes and preferences of the consumers remain unchanged over time. The preferences of

122

MICROECONOMICS FOR MANAGEMENT STUDENTS

all individuals can be averaged out, and an inference drawn about the representative consumer can be applied to the society as a whole. Moreover, for expositional purpose here, we further assume that there are only two goods available in the economy over time. This assumption can be easily dropped once we derive a usable formula for price indices, enabling us to examine changes in the welfare. Let us consider a hypothetical example of an economy where only two goods, X and Y, are consumed

during two time period

denoted by t= 0 and t= 1. As we have seen earlier, the observable points can be considered as consumers’ equilibrium. Let Table 1 represent the quantities of goods X and Y consumed at the prices prevailing in an economy during the two time periods. The table also gives the total expenditure of the society on these goods. Let us assume that the society as a whole can be represented by an individual who has a well-behaved utility function which is unknown. The problem is to compare the utility level of the individual at the two points in time. Table 1 Prices and Quantities of Goods consumed

Time

| eee «

a

Y

Jud pe

3 eal

e

- ee

[=o

10

10

20

15

400

400

555

=]

18

rf

25

12

606

410

606

It can be seen from the table that over time the prices of both the goods have increased. The quantity consumed ofX has increased but that of Y has fallen. Had the consumption of Y also increased, it would have been obvious that the utility level would also have increased from t= 0 to t=1. However, the case given in Table 1 is not so obvious. The total expenditure in period 1 is much higher than in period 0. This could be due to increased prices. It is necessary to adjust the expenditures relating to the price rise in order to arrive at real income or expenditure of consumers. It is for this purpose that we require price indices.

We can adjust expenditures caused by the price change by evaluating the goods consumed in the two period by keeping the

SOME APPLICATIONS OF THE THEORY OF DEMAND

123

prices constant. The prices can be kept constant either at the initial level (t=0) or at the terminal level (t= 1). Table J] also provides these calculations. Now let us try to interpret these figures from Table 1. The cost of the quantities consumed in the base period at the base period prices is Rs 400 (=2P,Q,). The cost of the quantities consumed in the terminal period at the terminal period prices is Rs 606 (= P,Q,). The cost of the quantities consumed in the terminal period at the base period prices is Rs 410

(x P,Q,). Finally, the cost of the quantities consumed in the base period at the terminal period prices is Rs 555 (2 P,Q,). A comparison of these figures provide very useful insights.

Now Rs 606 (= =P,0,)> Rs 555 (= ZP,Q,).

(1)

This implies that the cost of the quantities consumed in the terminal period (Q,) at the terminal prices (P,) is higher than the cost of the quantities consumed in the base period (Q,) at the terminal prices (P,). Thus, Q, is available to the consumer in period 1 and yet the consumer chose Q, in period 1. The consumer has therefore to be better off in the period | than in period 0. This is because consumer demand theory tells us that the consumer chooses that combination of goods which maximizes his satisfaction. Thus, the combination chosen by him has to be giving more utility than any combination of goods available or affordable by him. From this, we can derive the following generalization:

DP

(2)

O22 Ps

implies

u, > Up.

In order to derive index numbers, we divide both the sides by X

PoQo:

EFOiis ogysec bs up.

where (2 P,Q,

72 P,;Qy)) x 100 = Paasche’s Index of Real Income.

The Paasche’s index of real income is also known as the index of real income at the constant terminal prices. In Table 1, we considered a situation where the welfare level improved over time 0 to 1. Now let us consider another situation in Table 2.

Table 2 Prices and Quantities of Goods consumed

f{=9

10

10

20

15

400

400

595

Bi

18

14

25

12

a4

380

oe

As it can be readily seen, we have retained all the basic figures

unaltered in Table 2 as for Table 1, except the quantity of X in period 1 which is now taken to be 14 units as against 17 units in Table J]. As a result, the costs of the goods consumed in the period 1 change. Now, we can see that

(5)

Rs 552 (=EP,Q,) < Rs 555 (= EP,Q,).

This shows that in the second situation as given in the Table 2,

when the consumer chooses Q,, Q, is not available to him because he cannot afford to buy it any longer. However, on this basis alone no conclusion can be drawn regarding the consumer becoming better or worse off in t= 1 than in t=0. However, we can make use of the other two types of costs which are given in the Table 2.

(6)

Rs 380 (=x P,Q,) < Rs 400 (= 2 PpQy).

SOME APPLICATIONS OF THE THEORY OF DEMAND

125

This implies that Q, was available to the consumer in period t= (0 when he actually chose Q,. Thus, the utility level associated with Q, must be higher than that associated with Q,. This is particularly so because Q, was not available to the consumer

when he chose Q,. Thus, the consumer of Table 2 must have become worse off during period 1 when compared to period 0. We can generalize this as follows:

(7) i.e.

Dy) Ser oOr

EPO, = SPs.

This implies that Q, was not available when Q, was chosen. In the case considered in Table J, this is precisely what happens.

Rs 410 (== P,Q,) > Rs 400 (= ZP,Q,). Thus, applying Paasche’s price index, we cannot conclude anything about the consumer’s utility in the two period. Laspeyre’s price index however gave us conclusive evidence about improvement in the consumer’s utility over time. On the other hand, in the case of Table 2, Laspeyre’s price index did not give us any clues about the consumer’s utility, but Paasche’s price index provided conclusive evidence about the deterioration in the consumer’s utility over time. Two things can be noted from this discussion: (1) Both Laspeyre’s and Paasche’s price indices should be used together, and can serve also to provide a consistency check; and (2) Laspeyre’s price index when compared with the Jncome /ndex can show welfare improvements over time; whereas Paasche’s price index compared with the income index can show welfare deterioration over time. Laspeyre’s Price Index cannot conclusively determine deterioration in welfare, and Paasche’s Price Index cannot conclusively detect improvements in welfare. More formally, we can state the necessary and sufficient conditions for the welfare improvement and deterioration over time,

as follows:

A. Welfare Improvement Over Time (t = 0 to t= 1): (i) Necessary condition: Income Index > Paasche’s Price Index.

(ii) Sufficient condition: Income Index 2 Laspeyre’s Price Index.

SOME APPLICATIONS OF THE THEORY OF DEMAND

ea

B. Welfare Deterioration Over Time (t = 0 to t= 1): (i) Necessary condition: Income Index < Laspeyre’s Price Index.

(ii) Sufficient condition: Income Index < Paasche’s Price Index. There are two further possibilities regarding the interrelationships among the Income Index (II), Laspeyre’s Price Index (LPI), and Paasche’s Price Index (PPI). These possibilities are as follows:

(1)

PPI> I > LPI

This shows inconsistency of preferences (if preferences remain unchanged) because when Q, is chosen, Q, is available to the consumers; and when Q, is chosen Q, is available to them.’ Therefore, if Q, and Q, are not the identical combinations of goods, we have to conclude that the consumer’s preferences would have significantly changed during the time period under consideration.

(2)

LPI > al > PPL

This is a truly indeterminate case, in the sense that no conclusive statement based on these indices can be made about either improvement or deterioration in the consumer’s utility over time. Table 3 provides two sets of hypothetical data to illustrate these two possibilities. Readers may make the necessary calculations and inferences.

ee Time

pa Pl eihed (b)t=1 NOTE:

Table 3 Prices and Quantities of Goods Consumed

se fee rake i ye

ee Yaueiel,

Vala Siggesinge belt atars, 18 Mau wien

sigs a. Ah:

olsagg 80 acvBGd

ge acide PyQy

400 390 410

-cXP,Q,

555 570 551

The two cases would be obtained when the row of t= 0 is consid ered

with the row (a) f= 1; and when row f = 0 is considered with the row CD) f= 1.

4 Readers may verify this statement by considering the formulae of these indices and applying the logic discussed in this section.

Chapter 6 Demand

Elasticities

The law of demand is a statement about the nature of the relationship of the quantity demanded of a good with its price. A demand function is a more precise statement of such a relationship of the quantity demanded of a good not only with its own price but also with prices of other commodities,

income,

and other variables

influencing the tastes and preferences of consumers. The demand functions can be estimated with the aid of appropriate statistical techniques applied to the data collected for the different variables. The desired properties of such demand functions are very important. If such properties are overlooked when selecting the functional form or defining variables, interpretation of the estimated function becomes problematic. Sometimes, sufficient data are not readily available for estimation of the demand function. However, some selected properties of such demand functions may be adequate for the purpose at hand. We may, therefore, discuss a few concepts related to the demand functions that measure the extent of the relationship of the quantity demanded with independent variables.

Slope versus Elasticity The extent of the relationship is generally measured by the rate of change in the dependent variable with regard to the independent variable. In geometric terms, it is measured as the slope of the

curve. The rate of change is measured as a ratio of the changes in the dependent and the independent variables. The slope is, therefore, always expressed in terms of the units of the dependent variable per unit change of the independent variable. The rate of change in the quantity demanded of a good with respect to its price would also be measured in the same units as the good. Different goods are expressed in different units of measurement.

DEMAND ELASTICITIES

129

For instance, cloth is measured in metres, milk in litres, and wheat

in kilograms. The extent of the relationship of the quantity demanded with the price as measured by the slope of the demand curve cannot be compared across commodities. Moreover, the slope of a curve is also dependent on the scale of measurements on the two axes. Thus, if we adjust the scales, we can convert a very flat curve into a very steep one and vice versa. In order to make the rate of change (or the slope) independent of the units of measurement and the scale of measurement, the changes in the variables can be expressed in relative or percentage terms rather than in the absolute terms. The percentage rate of change is called elasticity. It measures the degree of responsiveness of the dependent variable with respect to the independent variable. Elasticity is defined as the ratio of the percentage changes in the two variables. It is always expressed as a unitless or absolute number. Since it is a modified version of the slope, it can be related to the slope of the curve. The elasticity of a dependent variable (say, Y) with regardto an independent variable (say, X) is measured as:

(1)

Hiei

Pekceninge Change tml

EX

PereentageChangeiniX /4.4

ia

AY wo Xp ¥isXe

GAYx 5A

y

It can be seen from this definition that the slope is a component of the elasticity. The slope can also be interpreted as a marginal value of Y with respect to X at the given point. Similarly, the ratio (Y/X ) can be considered as the average value of ¥ with respect to X at the same point. The elasticity of Y with respect to X can, therefore, be expressed as the ratio of the marginal value to the

average value. Alternatively, it can also be considered as the slope

deflated by the average value at the given point.! [t can be seen from the definition that the elasticity, would vary from point to point on a curve unless the curve is of a special type where changes in the slope are exactly offset by changes in the

average value. The slope remains constant on a straight line curve, | The average value (Y/X) is obtained as the slope of a line joining the origin

and the point on the curve. Such a line is known

as the radius vector. Thus,

geometrically, elasticity can be defined as the ratio of the slope of the curve to the slope of the radius vector.

130

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

but unless the average value also remains constant, the elasticity would not be constant on a straight line. We will return to this point when discussing the price elasticity of demand.

Point Elasticity and Arc Elasticity From the above discussion, it can be inferred that elasticity, like slope, is a concept specific to a point. Therefore, it is based on the concept of instantaneous rate of change rather than average observed rate of change. However, in practice, it is very difficult

to observe such minute and insignificantly small changes in variables. The theoretical concept of true elasticity has to be approximated by some formula. There are two ways of approximating true elasticity: (i) Point elasticity that measures the instantaneous rate of change between two observed points by assuming that they lie on the same straight line curve; and (ii) Arc elasticity that assumes that true elasticity remains constant over the relevant range of the values of the variables actually observed. The point elasticity formula is the same as that given in equation (1) above. It can be readily seen that point elasticity would differ from one point to the other because the slope component remaining the same, the average value component would differ for the two points. The Arc elasticity formula should, however,

ensure

that it yields the same estimate of the elasticity at both the points in question. There are several such formulae of the Arc elasticity. The most popularly used one is:

(2)

an aldeue! e stdlsat pe t =colee eianicity iscerc slitKiast

Where (X,, Y,) and (X,, Y,) are the two points observed.

On a closer examination of the formula in (2) above, we find that the first component is the same as the slope component in the point elasticity formula (1). Thus, even the Arc elasticity formula (2) is based on the assumption that a linear curve passes through the two points. The second component in the Arc elasticity formula differs from the point elasticity formula. In point

DEMAND ELASTICITIES

|

131

elasticity, the values of X and Y are those given by the coordinates

of the point in question. In the Arc elasticity formula, on the other hand, the ratio of the aggregate X and Yis taken. It can be viewed as the ratio of the coordinates of the mid-point joining the two points in question. Thus, Arc elasticity can be considered as the point elasticity at the mid-point of the line joining the two points. This is why the same estimate of elasticity is obtained for both the points (and hence, by assumption, for any point'within the range) by the Arc elasticity formula. Let us now discuss various demand elasticities with respect to different independent variables. In particular, we examine the formula, critical values, the determinants, and related properties of the demand function.

Price Elasticity Price elasticity of demand is the measure of the degree of responsiveness of the quantity demanded of a good to the changes in its price, other things remaining the same. It is thus a partial percentage rate of change in quantity demanded with respect to the price of the good. The price elasticity is thus measured only through the movement along a demand curve. If the demand curve is known to have shifted on account of changes in some of those ‘other factors’, appropriate adjustments have first to be made in order to estimate the partial elasticity. The formula to calculate the price elasticity of demand is:

(3)

inQ % ange e,= FQ _ Ch % Change in P ‘EP

e VENQ2 FOSSA Vigd AP

Q> Slope

P Q

It is presumed that these movements in the quantity demanded (Q) and price (P) are along a given demand curve where all other factors are held constant. It may be noted that the slope of the demand curve, as we usually draw it, is the reciprocal of the slope coefficient of the usual demand function. This is because we

132

MICROECONOMICS

FOR MANAGEMENT STUDENTS

measure the dependent variable (Q) along the horizontal axis and the independent variable (P) along the vertical axis in a demand curve. Thus, our usual demand curve represents an inverse demand function. Another thing to note is that the price elasticity of demand would be invariably negative since the law of demand postulates an inverse relationship between the quantity demanded and the price of a commodity. Let us take a hypothetical illustration to calculate the price elasticity of demand. Table ] provides the price-quantity combinations along a demand curve and the calculations of the elasticity of demand at different points. Table 1 Calculation of Point Elasticity of Demand for Good X

Point

Price

Quantity

Point Elasticity

A

12

20

aU

B

10

30

%

g

40

40-30 10 Ss Meininate aan earner tae

ii?

20

apie

The price and quantity figures in 7able / clearly indicate that in this range, the demand curve is linear (i.e. a straight line). The slope of the curve, therefore, remains constant at (—)5 for all three points. Under such circumstances, the point elasticity formula yields the same estimate of price elasticity at a given point irrespective of the point intervals we consider. Thus, even when we consider points A and C to calculate the elasticity at point A, it proves to be (-)3 only. Readers may verify for themselves that the point elasticity at point B would remain the same if it is calculated by considering points B and A instead of points B and C The price and quantity figures in the Table J can also be used 2 This would not, however, happen if all the three points were not lying on the same straight line. This is because, in such a situation, the slope component would differ depending upon the points chosen.

DEMAND ELASTICITIES

133

to illustrate the calculation of the Arc Elasticity. Let us consider the Arc Elasticity between pointsA and Cin Table /. Itis calculated as follows:

Arc elasticity

dp

(between pointtnA&C)

See

Or Ad

8-12

40420

= Oy

It can be seen from this calculation that the Arc elasticity between pointsA and C proves to be exactly the same as the Point elasticity at point B which is a mid-point of the line joining points A and C. Moreover, it can also be seen that numerically, the Arc elasticity

(-1.67) turns out to be lower than the Point elasticity (-3) at point A, which shows higher price and lower quantity; and is higher than the Point elasticity (-1) at point C which shows lower price and higher quantity. It may also be noted that the Arc elasticity would be different if the price range alters such that the mid-point changes. Thus, the Arc elasticity between points A and B and between the points B and C would be:

and

Arc elasticity (between pointtA&B)

a= Ce 10-12 30420

Arc elasticity (between points B & C)

Re 8-10

LUI

(-)2.2 ° j

a ae = (-)1.3 40+ 30

Geometric Formula

It is important to note that price elasticity of demand ts always calculated at a given price or quantity demanded. In other words, it is a point-specific concept that is valid only for a movement along a demand curve. If we use the arc elasticity formula, it can still be interpreted as an elasticity at a point, though it amounts to assuming the same elasticity at different points within the given range of prices and quantities. We may, therefore, derive the geometric formula for the point elasticity of demand. To begin with we assume that the demand curve of the good is a straight line as shown by the line DC in the Fig. /. We are interested in finding the formula for the price elasticity of demand at point T which shows price OP and the quantity demanded OQ.

|

134

MICROECONOMICS

|

FOR MANAGEMENT STUDENTS

ae Figure 1

(4)

BOR EP

OO, PP,

Or OQ

Fee (By definition)

Where (QQ, /PP,) is the reciprocal of the slope of the demand curve. Since the slope of the demand curve is given by (DP/PT)

and (TQ/QC), (5)

QQ, /PP, = PT/DP = QC/TQ

We may, therefore, rewrite equation (4) as:

(6) or

EQ°

PF OP

::OQ ;OP:

OP

EP DP OQ DP. OQ DP

DEMAND ELASTICITIES

Ba we JF EP TO OO

(7)

185

OC. OF OC OP OQ OO

Equations (6) and (7) provide the formulae for measuring the price elasticity of demand at any point 7 on a demand curve, as shown in the Fig. 7. Even when the demand curve is not linear, the same formula can be utilized. When the demand curve is non-linear, as

given by dd in Fig. 1, we can find out the price elasticity of demand at a point on the curve like point T by drawing a tangent (DC) to the demand curve at point 7. Then, by applying the same formulae as given by equations (6) and (7), we can obtain the price elasticity of demand at point 7. This is possible because, of two components of price elasticity, slope of the curve can be measured by the slope of the tangent to the curve at the given point. The geometric formulae given in equations (6) and (7) are extremely useful gaining some insights into the concepts. For instance, the way we measure the price elasticity on the price axis or the quantity axis, immediately makes it clear that it does not remain constant on a linear demand curve. Indeed, it would go on falling numerically on a downward sloping straight line demand curve as we move from higher to lower prices. It would vary from an infinitely large value at point D to zero at point Cin Fig. 1]. At the mid-point, the price elasticity would be (-)I. Secondly, the geometric formula of price elasticity also makes it very clear that the elasticity depends on the level of the demand curve and hence on all those ‘other factors’ which are assumed to remain constant. Higher income on the part of the consumer, for instance, would make the demand less elastic at a given price. This is because the demand curve would shift upward with an increase

in the

consumer’s

income.

As

a result,

the vertical

intercept of the demand curve, as given by point D in the Fig. /, would shift upward. At the given price OP, therefore, the price elasticity would tend to fall numerically. Similarly, when the price of a substitute commodity rises, the price elasticity tends to fall numerically at the given price. Even when the marketing or sales promotion effort or an advertisement campaign for a product is successfully launched, the demand for the product tends to become less price elastic.

136

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

Py

Dy

Ds

D,

ae

RE, Ce rin

Be,Tatreaioietet ret:

he

(3) (2)

(1)

Ox

O Figure 2 It also becomes clear from the geometric formula that the elasticity at a given price is numerically higher on a flatter demand curve than on a steeper one provided such demand curves intersect at a positive quantity demanded. Fig. 2 shows 3 demand curves

|, 2, and 3. The demand

curves

2 and 3 intersect at a

positive quantity. At price OP, the price elasticity is numerically higher on 3 than on 2 because 3 is flatter than 2. However, curve | is steeper than both 2 and 3, and yet the price elasticity at price OP on curve 1 is numerically higher than on the curves 2 and 3. Moreover, if different linear demand curves have the same vertical intercept, they are all equally elastic at the given price.

The geometric formula also implies that the minimum information needed to estimate the elasticity of demand at a given price is an estimate of the minimum price at which nothing is demanded of the good in question. Such a price is given by the vertical

137

DEMAND ELASTICITIES

intercept of the demand curve. Once we know the vertical intercept, the price elasticity can be estimated at the given price by assuming a linear demand curve. This becomes possible because the slope of such a linear curve, then, becomes

irrelevant for

calculating the price elasticity at a given price. Similarly, the price elasticity of demand at a given quantity can be estimated if we can somehow estimate the maximum quantity demanded at a zero priceof the good. Such a quantity is basically the satiating level of the good for a consumer. Any further consumption would lead to negative marginal utility and hence a fall in the total satisfaction of the consumer. Such a quantity is, therefore, given by the horizontal intercept of the demand curve. The geometric formula given in equation (7) enables us to estimate the elasticity of demand at a given quantity if we know the horizontal intercept of the demand curve.

Determinants of Price Elasticity It is clear from the above discussion that the factors governing the level of the demand curve determine the value of the elasticity of demand ata given price. Thus, the level of the consumer’s income, prices of related commodities,

his tastes and preferences,

and

expectations regarding future income and prices also determine the degree of price responsiveness of the quantity demanded of a good. Moreover, those factors that influence the slope of the demand curve, such as ease of substitution of the good by other goods, also determine the elasticity of demand. The higher the number of close substitutes available for a good, the more price elastic the good becomes.” Similarly, the numerical value of elas-

ticity depends directly on the number of different uses to which the good can be put. This is because an increase in the price of the good leads to a reduction in the quantity demanded of the 3 An important implication is that demand for an aggregative commodity such as toothpaste or soap would be less elastic than the demand for a particular brand

of the same product. This is because there would be very few substitutes to soap as a good but there may be several close substitutes for a particular brand of soap.

138

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

good for each of its uses. Hence, the greater is the number of uses, the greater is the aggregate reduction in demand and the larger the elasticity. In certain cases, the proportion of income spent on the good becomes a determinant of the price elasticity of demand. If the consumer is spending a very large proportion of his income ona good, he is likely to be more sensitive to its price. On the other hand, if he is spending a very small or negligible proportion of his income on a good, he may not be very sensitive to its price. There are however several exceptic::s A poor consumer generally spends a large proportion of his :::come on food. However, since

there are very few substitutes to food, his demand for food is not likely to be very price sensitive. Similarly, a consumer is likely to be spending a small proportion of his income on a specific brand of soap. However, since there are many close substitutes available, his demand for the specific brand is likely to be highly price elastic. However, given the nature of the good and all other factors, the proportion of income spent on the good by the consumer is likely to determine the price-sensitivity of his demand. This may be an important basis for comparison of the price elasticity of demand for the same good in different market segments.

Price Elasticity and Total Revenue A demand curve shows the price-quantity relationships the consumers would be willing to accept. For a seller the demand curve shows the possibility of his revenue if he decides to charge different prices. Thus, ifa seller correctly estimates the market demand curve for his product, he can interpret it as his average revenue curve. This is because the area of the rectangle formed by the coordinates of any point on the demand curve shows the total expenditure by consumers on the good. However, the expenditure of the consumers is the revenue of the seller. Therefore,

(8)

Total expenditure = PQ = Total revenue.

(9)

Total Average revenue = ae

ied

P 7 = P.

DEMAND ELASTICITIES

(10)

139

R _ PAQ+QAP Marginal revenue = AT AQ AQ a Pi 1D pers OF = A : i.e. Mia P+ AO =PLles “a

(11)

ie. MR=P(1+1/e).

Where e is the price elasticity of demand as defined in quation (3) above.

It may be noted here that e would always be negative and hence the marginal revenue (MR) would always be lower than the price or the average revenue (AR) at any given quantity. Marginal revenue is defined as a rate of change in the total revenue with respect to quantity.’ In other words, MR is the addition to total revenue when the quantity increases by one additional unit. Equation (11) shows the definitional equality among the values of the price elasticity of demand, price of the product, and marginal revenue. This is a very useful relationship because it clearly defines the relationship between the price and the total revenue a seller can obtain, given the demand curve of the good in the market. The relationship among the price (P), marginal revenue (MR), and price elasticity of demand (e) as given in equation (11) implies the following:

(a)

When e = (—) © (i.e. infinity), MR = P.

(b)

When e=

(c)

When e lies between 0 and (-)1, MR is negative.

(d)

When e lies between (-)1 and (—) ~, MR is positive.

(—)1, MR=0.

When MR is positive, an increase in quantity as a result of a fall in the price of the good would lead to an increase in total revenue 4 For a seller, his total revenue is a function of the quantity of the good he is able to sell. Therefore, marginal and the average revenue are also defined with respect to quantity. However, average revenue is just the price. Thus, if the whole framework is viewed from a seller’s viewpoint, this can justify measuring quantity on the horizontal axis and price on the vertical axis.

140

MICROECONOMICS

FOR MANAGEMENT STUDENTS

or alternatively an increase in the consumers’ expenditure on the good. As it is stated above, this happens when the price elasticity of demand is numerically greater than one. The demand for the good is, therefore, said to be elastic in this case because it leads to a more than proportionate change in the quantity demanded in response to a given percentage change in the price. It should be noted that elastic demand would also imply a reduction in the total revenue when the price increases and, as a result, the quantity demanded falls. In other words, when the demand is price elastic, the total revenue (TR) ts directly related to the quantity and inversely related to the price of the good. The complete opposite is true when the price elasticity of demand lies between 0 and (-)1. As stated above in (c), under such conditions, the marginal revenue is negative. An increase in the quantity demanded as a result of a fall in the price of the good leads to a reduction in the total revenue. When the price elasticity of the demand for a good is numerically less than one, it is Rnown as a price inelastic demand. This is because the quantity demanded does respond to a change in the price of the good, but the response is less than proportionate. If the quantity demanded simply does not respond to changes in the price of a good, the demand for the good is known to be perfectly inelastic. A perfectly price inelastic demand means that the price elasticity is zero. The dividing line between the elastic and inelastic demand is given by the value of price elasticity (-)1. When the demand ts price inelastic, the total revenue is inversely related to the quantity and directly related to the price of the good. Generally a demand curve shows very high elasticity at higher prices (or lower quantities) and very low elasticity at lower prices

(or very high quantities). Therefore, the relationship between the total revenue and price, and the total revenue and quantity can be represented as shown in the Figs. 3(a) and 3(b), respectively. A linear demand curve which can also be viewed as a linear AR curve displays the property of the falling elasticity of demand (numerically) as the price falls. Fig. 4 shows the relationship among AR, MR, and e, as given in equation (11) with a linear demand function. From this equation, if we know any two variables, the third can be derived. Since the geometric formula makes it clear that for any demand curve we can derive the elasticity at

DEMAND ELASTICITIES

Figure 3(a)

Figure 3(b)

14]

142

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Figure 4 a given price, for any given demand curve (and hence AR curve), the corresponding MR curve is derivable. If the AX curve is linear, the corresponding MR curve is also linear and falls twice as rapidly as the AR when quanity increases. In any case, so long as the AR curve is downward sloping, the MR curve lies below the AR curve. Moreover, when the MR curve intersects the quantity axis and becomes negative from the positive values, the total revenue reaches its maximum value. Thus, 7R is at its maximum

when MR

= 0 or when e = (-)1. This result can be directly observed from Figs. 5 and 4.

Constant Elasticity Curves On a linear curve, the slope remains constant. Similarly, there is

a family of curves on which the elasticity remains constant at different points. In general, such a family of the curves is given by the following equation:

DEMAND

fle)

ELASTICITIES

143

Pe Are);

Where eé is price elasticity and A is a positive constant. For the demand curves, e is negative and hence Q and P are inversely related. On all possible combinations of Q and P on this curve, the elasticity works out to e which remains constant.” The demand curves showing constant price elasticity at different points are all hyperbolic. They are downward sloping, convex from below over the entire range of the positive values of Pand Q. When the critical value of the price elasticity, (-)1, remains constant throughout the range, the demand function becomes: (13)

Oe Ayr OF= Zz.

It shows that the expenditure on such a commodity remains invariant to price or quantity fluctuations. Under such circumstances, the quantity demanded and the price of the good are not only inversely but also proportionately related. The curve that shows the constant price elasticity to be (-)1 is known as Rectangular Hyperbola or Equilateral Hyperbola. The property of such a curve is that the area of the rectangle formed by the coordinates of different points on the curve remains the same. The constant elasticity curves are very popular for empirical estimation of elasticities because we get a unique estimate of the elasticity in question. Moreover, it is very easy and convenient to fit such a function empirically. Consider equation (12) which gives the demand function showing constant price elasticity. If we take

logarithms on both sides, the following equation is obtained: (14)

iO = iA +e. IiP.

Thus, the price elasticity is the slope coefficient of the linear

equation if the variables are measured in logarithms. With easy availability of computing facilities and software, this function is almost indiscriminately used in the empirical research. Apart from imposing several unwarranted restrictions on the utility and

144

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

expenditure functions,® the constant elasticity demand curves violate the generally well accepted condition for normal demand behaviour. The demand behaviour is considered normal if the demand curve shows price to be inversely related to the quantity demanded and directly related to the numerical value of the price elasticity of demand. Under normal conditions, we would expect

that there would be some very high prices at which nothing of a good would be demanded. Similarly, even when the good is available free (i.e. P= 0), the consumers would reach a saturation point and stop consuming it at some finite quantity level. These are very plausible propositions and have intuitive appeal. The constant elasticity demand curves, however, do not show such demand behaviour. They, on the contrary, imply that some quantity would invariably be bought so long as the price was finite; and that the consumers would not be satiated with the commodity at finite levels of consumption.

Cross-Elasticity of Demand A demand function relates the quantity demanded of a good to those independent variables that are likely to play a significant role in influencing the consumer’s choice of buying the good in question. Generally, the demand function is not as simple as given by equations (12) or (14) above, but invariably involves other variables too. Since elasticity is the degree of the responsiveness of the dependent variable to the variations in the independent variable, there can be as many demand elasticities as there are independent variables in the demand functions. Thus, we can have population elasticity of demand and expectational elasticity of demand. We may even have a time elasticity of demand if we explicitly introduce the time dimension in the demand function. However, the most popular and frequently used concepts of the other demand elasticities are cross-elasticities and income elasticity of demand. We briefly discuss them here. Cross elasticity measures the degree of responsiveness of the 6 See, for details, L. Evans, ‘On the Restrictive Nature of Constant Elasticity Demand

Functions’, /nternational Economic Review, vol. 35, no. 4, Nov.

pp. 1015-18.

1994,

DEMAND ELASTICITIES

145

quantity demanded of a good to the variations in the price of a related good, all other things remaining the same. Like price elasticity, it is also a partial measure. It is defined as:

(15)

EQx AQxPy Cross-Elasticity = EP ey Os

There can be as many cross-elasticities of demand for good X as there are related goods considered in the demand function of X. The sign of the cross-elasticity determines the nature of the related good. If it is positive, it implies that an increase in the price of good Y leads to an increase in the quantity demanded of good X. Good X and good Y must, therefore, be substitutes. On the other hand, if the cross elasticity is negative, the two goods must be complements.

In order to make

such statements,

however,

it is

necessary to have the cross elasticity of demand for good Ywith regard to the price of good X also of the same sign. If these signs are not the same, the nature of goods X and Y cannot be unequivocally determined.’ This is because we consider the price effect

PCC, > BEE, => As Py t, Q, + PCC, > AEE, => As P, 1, QyT

O

B

B,

Figure 5 7 This is possible because EQ,/EP,, is measured along the Price Consumption

146

MICROECONOMICS

FOR MANAGEMENT STUDENTS

which includes the income effect as well as the substitution effect of a price change. As a result, such elasticities are called ‘gross’ cross-elasticities and the subsequent identification of the nature of the commodities are also known as Gross Substitutes and Gross Complements. If on the other hand, we consider only the substitution effect of a price change keeping the real income constant rather than the money income constant, the corresponding elasticities are known as ‘net’ cross-elasticities and the nature of the commodities as Net Substitutes and Net Complements. It is important to note that anomalies can arise in the signs of only the ‘gross’ cross-elasticities and not in the ‘net’ cross-elasticities. Thus, if the demand functions forX and Yare estimated incorporating the real income variable, the signs of the cross-elasticities FQ,/EPy and EQ,/EP, should remain the same in both the demand functions.

The magnitude of the cross elasticity reveals the closeness of the substitutes or complements, as the case may be. If the cross elasticity is positive and very large, goods X and Yare considered to be close or nearly perfect substitutes. If the cross elasticity is close to zero, the two goods are considered to be distantly related or practically unrelated. Therefore, the concept of the cross elas-

ticity is found to be very useful in delineating ‘an industry’ or in defining ‘a product’ for empirical and analytical investigation.

Income Elasticity of Demand Another very popular and useful concept of demand elasticity is income elasticity of demand. It is defined as the degree of responsiveness of the quantity demanded of a good with respect to the variations in the consumer’s

income, all other things remaining

the same. The price of all goods should remain the same income elasticity of demand to be measured, i.e.

(16)

FQ_ AQ Income Elasticity = EMZ. AM.

Curve for good Y, and Ear,

for

MO

is measured along the Price Consumption Curve

for good X. Fig. 5 shows the possibility when FQ,/EP,, is positive (movement from point £ to E,), but EQ,,/EP, is negative (movement from E to E,).

147

DEMAND ELASTICITIES

Where M is the money income.

Another related concept is expenditure elasticity which is defined as: Crt)

nowin

Expenditure Elasticity=

i cM RO»

codOAM dO:

cold

PO

If all prices, including P, do not change, the expenditure elasticity and the income elasticity would be identical. However, if the prices do not remain constant, expenditure elasticity would differ from true income elasticity. In order to ensure that prices remain the same as far as possible, expenditure elasticities are generally estimated on the basis of the cross-section data on the family budget or consumption surveys. There may still be the problems of aggregation where the product composition may not have remained the same. The expenditure elasticities are, therefore, con-

sidered only as an approximation of the true income elasticities. Income elasticity can take any value, positive or negative, depending upon whether the good in question is a normal or inferior good. As we have noied earlier, the income effect for an inferior good is negative; and hence its income elasticity would

also be negative. However, for normal goods, the income elasticity of demand is positive. It is possible to imagine a situation in which the real income of the consumers grows so substantially that they start reducing consumption of the good when their income further increases. For an individual there is a possibility of practically every good becoming an inferior one at some level of income. However, the aggregative behaviour is not likely to show such a pattern, at least in the relevant income range. At the aggregate level, therefore, all the three possibilities of increasing, constant,

or decreasing income elasticity as income increases are acceptable. The principal use of the concept of the income elasticity of demand is to distinguish between the necessities and the luxuries.

If the income elasticity of demand for a good ts less than (+)1, it can be considered to be a necessity. On the other hand, if the income elasticity is greater than one, the good may be considered to be a

luxury since the income elasticity of unity implies that the proportion of the income spent on the good by the consumer remains

148

MICROECONOMICS

FOR MANAGEMENT STUDENTS

the same at different levels of income. This is because unit income elasticity means that percentage change in the quantity demanded of the good is the same as the percentage change in income when all other variables, including prices, remain constant. If the price of the good remains constant, a given percentage change in the quantity of the good leads to the same percentage change in expenditure on the good. Therefore, when the income elasticity of demand is (+)1, the percentage change in expenditure on the good is the same as the percentage change in the income which makes the proportion of income spent on the good constant as income changes. When the income elasticity of demand is greater than (+)1, the proportion of income spent on the good rises as the income rises and falls as it declines. Such goods are, therefore, called luxuries. On the other hand, when the income elasticity of demand is less

than (+)1, the proportion of income spent on the good falls as the income rises, and rises as it falls. Such goods are the necessities

because the consumers have to consume them in certain quantitative range. They would not want to consume such goods in large quantitities nor would they like to cut their consumption below a certain quantity level. Food, shelter, clothing are some of the standard examples of necessities. Since the income elasticity of demand is an important component of the price elasticity of demand, a very low income elasticity tends to be associated with inelastic goods. This also happens because necessities by definition do not have many close substitutes. Thus, demand for them also tends to be price inelastic. The demand for luxuries, on the

other hand, tends to be price elastic both because they are likely to have many close substitutes and the income elasticity of their consumption is higher. It may be pointed out that all the goods in a consumer’s basket cannot be necessities with an income elasticity less than unity. This is because, if all goods in a consumer’s basket were necessities, then, how would he be able to increase the proportion of income spent on all those goods simultaneously when his income

dropped? There would at least be one luxury good in his basket. The same logic would convince us that all the commodities in a consumer’s basket cannot be luxuries. At least one good has to be a necessity. This logic can, indeed, be formalized in terms of

DEMAND ELASTICITIES

149

the relationship among the income elasticities of different goods,” as follows: (18)

Re PE ee

ae

where K,, K,, K, etc. are the proportions of income spent on goods X, Y, Z, etc. respectively; and e,,, Gi, Gp, etc. are the income elas-

ticities of demand for respectively the goodsX, Y, Z, etc. Moreover,

(19)

Buco

tots aye b

Equations (18 and (19) state that the weighted average of all income elasticities is unity, where the weights are given by the proportion of income spent on the respective goods. Equation (18) provides a very useful restriction on the system of demand equations. If there are only two goods, e.g. food and non-food, an estimate of the income elasticity of one of them would imply an estimate of the income elasticity for the other good through equation (18). Such restrictions are very helpful in choosing the most appropriate equation demand functions.

and the estimates

of parameters

of the

8 Assuming that there are only two goods X and Y, and that their prices P, and »

P, are given, the budget constraint is:

M=Px

X+Py.>¥

. AM=PxyAX+PyAY ”

1=Px(A X/AM) + Py(A Y/A M)

AY M AX M,PyY _PxX Bt a NI MRM CF “.

1=Kyeix + Kyeiy.

This proof can easily be extended to include many goods.

Appendix A

Are Cigarettes Overtaxed in India?

Cigarettes are consumed all over the world and are considered to be the most sophisticated form of tobacco consumption. In most parts of the world, they constitute an overwhelming proportion of about 80 to 90 per cent of the total tobacco consumption. In India, however, they account for barely a sixth of total tobacco consumption. Itis not that smoking is not popular in India. Indeed, the proportion of tobacco smoked is constantly rising and currently stands at 71 per cent of the tobacco consumed in the country. Within the smoking segment, however, it is bidis which have been growing at the expense of cigarettes both relatively and absolutely. Table I shows the trends in the tobacco consumption in India over the period 1971-94.

Table 1 Tobacco Consumption in India (in million kg) Tobacco Used in ]

1971-2

1991-2

1993-4

Ps

mi

4

Cigarettes

71

82

66

Bidis

I

202

Zo

Smoking

162

284

2d

Total

309

404

417

SOURCE:

Tobacco Excise Tariff Committee and Indian Tobacco Statistics.

Such a trend in tobacco consumption has several implications for the government’s excise policy in this sector. The principal objective of the government’s excise policy for tobacco appears to be revenue generation. This is because it is one of those products where the governmentis neither interested in increasing domestic consumption nor appears to be seriously interested in curbing

APPENDIX A

151

consumption. As a matter of fact, tobacco used for smoking has been increasing in India at the annual compound rate of 2.8 per cent over the period 1971-2 to 1993-4 (See Table /). After the withdrawal of duty on leaf tobacco in 1979, the government had little option but to raise duty on the final products to bridge the revenue gap. The nature of the final products in the Indian tobacco sector further restricts the government's options. Cigarettes in India are produced by only a few companies in the large-scale registered manufacturing sector which make it very cost effective to collect excise. Bidis, on the contrary, are produced in the informal sector on a small scale by a very large number of units widely dispersed geographically. Moreover, since they are handmade, they are not uniform. All this makes bidis extremely cost inefficient to collect excise. It does not mean that the government cannot levy excise on bidis. Rather, it is difficult to raise net revenue by doing so. Most of the tobacco products in the nonsmoking segment share such features with bidis. Thus, cigarettes were considered the most relevant tax base in the tobacco sector. Table 2 shows that cigarettes are virtually the only tax-base in the Indian tobacco sector.

Taple’Z Tax-Base of the Indian Tobacco Sector

Year

Excise from Tobacco

Cigarettes as a percentage in Tobacco

(Rs crores)

Volume%

Excise%

]

Z

3

4

1971-2

Jeo

23

70

1981-2

834.8

2]

82

1993-4

3140.0

16

87

SOURCE:

Tobacco Excise Tariff Committee and Budget Documents.

The table clearly shows that the tax-base in the tobacco sector is shrinking rapidly over time. The diminishing tax-base is being burdened with larger and larger contribution of excise revenue in both absolute and relative terms. This implies a very high incidence of taxation which gives rise to the equity issue. The

152

MICROECONOMICS

FOR MANAGEMENT STUDENTS

inequality of tax incidence in the tobacco sector is significantly higher than all possible measures of income inequality in the country. Only 12 per cent of the tobacco users who are cigarette smokers contribute as much as 87 per cent of the excise revenue from the sector, the remaining 88 per cent of the tobacco users contributing only 13 per cent. In terms of the excise revenue per kilogram of tobacco used, cigarettes yield Rs 415 whereas bidis yield only Rs 10. Such a high incidence and inequality is likely to encourage tax evasion and smuggling that would go against the interest of the government, industry, and the consumers alike. It

needs correction sooner than later considering the potential for growth and exports by the Indian cigarette industry. At present when the Finance Ministry is putting emphasis on expansion of tax-base rather than increasing tax rates in order to achieve growth in tax revenue, the issue of a rapidly shrinking tax-base in the tobacco sector cannot be ignored. This situation has primarily arisen because the Indian government chose the easier option of almost continuously raising excise duty in order to garner increased excise revenue from cigarettes. Currently, the excise on cigarettes has crossed the level of 150 per cent of the price net of excise. It is time to question whether cigarettes in India are overtaxed.

If they are, any further increase

in the

excise rate would lead to a decline rather than a rise in excise revenue. The basic assumption behind following a policy of steep and continuous increase in the excise duty on cigarettes is that the demand for cigarette is price inelastic. This implies that in response to an increase in excise duty and hence in price, the demand for cigarettes would not fall proportionately and therefore yield higher excise revenue. However, whether cigarettes in India are price inelastic or not is an empirical question which can be scientifically addressed. There are two ways of estimating the elasticity of demand for cigarettes in India: (1) By taking the annual changes in the price and quantity demanded of cigarettes over the recent period and

apply the Arc Elasticity formula to calculate the elasticity of demand relevant for the current period and, (2) Estimating the demand function for cigarettes using long-term time series data and obtaining an estimate of the elasticity of demand

for the current

APPENDIX A

153

period based on the regression equation. Both the methods are used

here so that the most conservative estimates

of the price

elasticity of demand for cigarettes in India can be obtained.

Arc Elasticity The advantages of the Arc Elasticity method are: 1.

2.

dh

It requires annual averages over short periods of time and therefore effectively uses the most recent data, capturing the current trends in the price responsiveness of the commodity. Since estimation is over a short period of time, the long-term influences such as changes in consumer taste and preferences in favour of or against the commodity in question is not likely to play significant role. In comparison to the alternative Point Elasticity measure, the Arc Elasticity measure, always gives a lower (and hence more conservative )estimate at higher price and lower quantity consumed. /n the /ndian cigarette industry, volumes have

been consistently declining since 1991-2. The principal limitation of the Arc Elasticity measure is that it is a direct method, not capable of separating the influence of some important factors such as price of a close substitute (bidis) or the real income of the consumers. However, the influence of some other variables such as seasonal variations, population growth,

and general inflation on the demand for the commodity (i.e. cigarettes) can be neutralized by measuring the volume of cigarettes as a monthly average on a per capita basis, and the price of cigarettes as the real price which is obtained by deflating the nominal price by the consumer price index. The Arc Elasticity formula utilizes the concept of the average percentage changes in the demand and price over the interval, ie, Ago

Ay

i

om oe

——_—" Arc Elasticity Elasticity = X, +X, x -—___— P,P,

where, (P,, X,) and (P,, X,) are the observed values of price (P) and volume (X) for the year 2 and 1 respectively. Using this

154

MICROECONOMICS

FOR MANAGEMENT STUDENTS

formula, and the data from Table 3, estimates of the price elasticity

of demand for cigarettes are obtained for consecutive periods from 1991-2. Table 3 Monthly Average Consumption and Price/1000 Cigarettes

Year

Total Volume (million cigs.)

Per Capita Average Volume _ _ Price in Rs (in sticks) i

4

Real Price in Rs

]

D4

1991-2

1144

8.346

905.95

276.26

1992-3

6681

7.653

988.90

291.53

1993-4

6514

Volk

634.65

293.82

1994-5

6322

6.955

667.70

284.13

*

Excluding minors.

**

Based on the period March to Aug. 1994.

SOURCE:

bs

i. For volume and consumer price of cigarettes: Cigarette industry sources in India

ii. CPI is for Urban non-manual workers, obtained from CMIE, Aug. 1994.

Period

Arc Elasticity of Demand for Cigarettes

1991-2 to 1992-3

(-)1.61

1991-2 to 1993-4

—

1991-2 to 1994-5

()2.15 (-)6.47

It appears that the price elasticity of demand for cigarettes in India has been sharply rising over recent years. It may be noted that these estimates are conservative when compared to Point Elasticity estimates.

Regression Model The estimate of the price elasticity of demand for cigarettes for the current period can also be obtained from the regression model

APPENDIX A

155

for cigarette demand in India. The data considered for the purpose

pertain to the period 1973-4 to 1994-5. The principal objective of this exercise is to obtain an overall demand function for cigarettes capturing as many possible effects as permitted by the data set. This would also enable us to obtain a reliable estimate of the price elasticity of demand relevant for the recent period. The dependant variable is, therefore, taken as the monthly average volumes

of

cigarettes measured in million cigarettes (MNC) consumed during a year. This would neutralize the seasonal variations in consumption and yield a more stable measure of cigarette consumption in India. The independent variables included in the demand functions are in accordance with standard consumer theory. The basic data set on cigarettes in India for the regression model is given in Table 4. It may be noted here that a consumption commodity like cigarette is likely to be subjected to some significant shifts in consumer tastes and preferences in favour or against over a period of time like a decade or two. If such gradual but clear shifts are occurring, there is no way of capturing it unless the conventional assumption about the lack of money illusion! is made and the model is estimated in its non-restrictive form. This implies considering nominal prices along with real income and the general price level rather than real prices and real income as independent variables. Moreover, the choice of functional form is also clear once the

well accepted Marshallian conditions of ‘Normal Demand’ are considered. According to Marshall, the normal demand curve should be the one that fulfils the following two conditions: I.

2.

Price and the quantity demanded of a commodity should be inversely related, other things remaining the same, and The price elasticity of demand should be higher at higher prices and lower quantities; and lower at lower prices and higher quantities.

' Lack of money illusion implies that the consumers would make no changes in their consumption if all monetary aggregates such as prices and money income change by the same proportion. Thus, if all prices and money income rise by 10 per cent, the budget constraint would remain unchanged and so also the consumption ofX and all other commodities.

156

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Table 4

Basic Data for the Regression Models

Year

1973-4 1974-5 1975-6 1976-7 1977-8 1978-9

income Rs in Index Real of vol. cigs. in MNC Price capita perdisposable Rs/1,000° Price of biris Rs/1,000°CPI Population in Avg. Monthly of cigs. cneas=)—_— ~]

peNO

9052.4 oa meg af 979 6238.1 6670.1

1980-1

6551.0

1981-2

P2020

1982-3

7423.1

1983-4

(272.4

1984-5

8006.9

1985-6

6867.8

1986-7

6754.5

1987-8

6404.1

1988-9

6687.5

1989-90

6920

1990-1

7084

1991-2

1144

1992-3

6681

1993-4

6514

1994-5

ptqn

9044.0

1979-80

a eee

D on

6322

90

lll

Estimates as available from industry sources in September 1994.

APPENDIX A

157

Given these two conditions of normal demand, a double-log form of the function which shows a constant elasticity of demand throughout the whole range of price and quantities is ruled out. On the other hand, the simple linear demand curve, which can subsequently be put to further analytical uses, becomes not only handy but the most appropriate functional form, satisfying both the conditions of normal demand if it is found to be downward sloping, i.e. the coefficient of P. proves to be negative and statistically significant. The estimates of the demand functions for cigarettes in India are obtained for two periods: (i) 1973-4 to 1994-5 and (ii) 1982-3 to 1994-5, the latter being more recent and relevant since priceincreases of cigarettes on account of increased excise rates became more pronounced since 1982-3. (See Table 4). During 1973-4 to 1982-3 the price of cigarettes barely doubled but it more than quadrupled during 1982-3 to 1994-5. This sub-period is, therefore, considered separately for estimation of price elasticity, though the functional form and measurements of the variables remain the same. The results are presented below in the form of a table. It can be observed from these results that the regression model for the recent sub-period, 1982-3 to 1994-5 fits the data better than that covering the experience of the past two decades. Although there are some significant structural shifts in the parameter estimates between the two regressions, the slope parameters of the price of cigarettes in both the functions are remarkably similar. Even the estimates of the price elasticity of demand for cigarettes around 1994-5 prove to be almost the same in both the equations. This enhances the confidence in the results for further use. The overall goodness of the fit, as given by the R-square, is very satisfactory in both the models. Considering the values, and their statistical significance,

however,

fhe second model seems

to be

preferable. In any case, all the estimates of the price elasticity of demand, whether through the direct Arc Elasticity method or through regression methods using the time series data prove to be well in excess of unity. Considering this evidence collectively, the most conservative estimate of the price elasticity of demand for cigarettes in India relevant for the current period seems to be (-)2.

158

MICROECONOMICS

Time Period

3

FOR MANAGEMENT

Model 1

1973-4 to 1994-5 Variable

Coefficient

t-value

]

2 -20.78 68.705 58.825 31.049 0.174 749.50 0.8014

3 Lah 1.10 0.85 1.35 0.12 0.25 (12.91)

PC PB Pop. CPI Income Intercept R-Square

STUDENTS

Model 2

1982-3 to 1994-5 Coefficient

t-value

4 20.239 138.390 -347.61 39.712 1.703 23782 0.8727

2 ZBI" 3.66. aA. 2.78 1.65 5.61. (9.60)

Prive elisttity’" (adeeb vidted estistie:: (32Bite odpeeraot 5 at 1994-5 levels NOTE:

_ Figures in parenthesis are F-values. Statistically significant at 5 per cent level of significance.

These estimates imply that a one per cent reduction in the price of cigarette would result in a minimum increase of two per cent in the total consumption of cigarettes. In other words, the demand for cigarettes in India is price elastic (perhaps highly elastic). The statistical evidence does not support the belief that the demand for cigarettes is price inelastic in India. This estimate for India may be compared with that obtained for the US by the Nobel laureate Gary Becker and his colleagues (American Economic Review, June

1994) who found that the price elasticity of demand for cigarettes in the US is (-)0.4 in the short run and (-)0.75 in the long-run. Elastic demand in india and inelastic demand in the US is understandable because, as noted earlier, cigarettes form only 16 per cent of the total consumption of tobacco in India whereas elsewhere in the world, particularly in the US they are virtually the

only form of tobacco consumption. Availability of the perceived close substitutes is an important factor raising the elasticity of demand for a commodity. This is a well-accepted proposition in Economics. It may be noted here that in spite of the fairly inelastic demand

APPENDIX A

159

for cigarettes, in the US both in the short- and long-run, the US Federal rate of excise on cigarettes is less than 12 per cent (total tax being 35 per cent) of consumer pack price as compared to an average of about 60 per cent in India. There was a proposal by some Congressmen in the US that the Federal tax rate be raised substantially from the present rate of 24 cents to $ 1.24 per pack of cigarettes. Professor Gary Becker had warned against such a move, arguing that it would amount to unnecessarily high taxation. According to his estimates, the tax revenue from cigarettes would be maximized when the tax rate increased from 24 to 95 cents. Thereafter, the tax revenue

would

fall as the tax rate

was increased further. (See Business Week, 15 Aug. 1994, p. 8.) It is learnt that the US Government has since deferred the proposal, taking cognisance of Professor Becker’s warning. Professor Gary Becker’s argument is based on a well known proposition in price theory that price elasticity increases as the price rises. Thus, ceteris paribus, there is always an optimal rate of excise duty that maximizes the excise revenue from the commodity (See Fig. /). Any higher rate of tax would make it an overtaxed commodity hurting the interest of everybody involved.

Excise Revenue

Excise Rate

160

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Optimal Tax on Cigarettes Let a linear demand curve for a commodity like cigarette be considered with the price (P) measured on the vertical axis and the quantity demanded (X) on the horizontal axis as in Fig. 2. A is the initial point with OX of quantity demanded and OP of price. Let 7 be the rate of specific excise duty in the initial situation.

OP - OP, = T= PP,

(1)

Let the price elasticity of demand relevant in the initial situation be given by e. By definition,

e = OP/CP

CP =:QP7e

(2)

OC = OP + PC = OP [(l+e)/e] Again by definition

(3)

3

Price

C

A P

B

-f;

y

O

a

X,

Figure 2

|

Quantity

APPENDIX A

161

erredOP S DERMSNOe

OX XX, abit = Gp eF)

4 (4)

OX, = OX+ XX,=OX(1+eT/OP) Tax revenue = PP,

O4=T

|

(5)

*OX

If 7 = 0, then OP coincides with OP, and the Tax Revenue is zero. Similarly, if T=P,C, then OP = OC, OX falls to zero, and Tax Revenue = 0. There is, therefore, maximum.

an

optimal

7 where

Tax

Revenue

is

This is equivalent to finding that value of T where the area of the rectangle PP, - PA is maximised within the right-angled triangle P,CB. According to a theorem, the rectangle formed by joining the mid-points of the lines of the right-angled triangle has the maximum

area. Thus,

The Optimal Excise Rate “=f = (1/237€ Ke)

Fo ey

2(0C+ Or]

=i lg) Or CL +e ye

| (Or ir

using (1) and (3)

= (172) (Or (te) 4 i4 Po = @/2)(OP £exD ie

(6)

Similarly, at 7* the quantity demanded (X%*) is given by

X” = (1/2) OX, using (5), =(OX/2): [1+ e-T/ OP]

(7)

The optimal tax revenue (7h*) is obtained, then by using (7) and (6) as:

(os he

a

162

MICROECONOMICS

Les

FOR MANAGEMENT STUDENTS

TR’ = (OX/4) [OP/e + 2T + eT’/OP]

(8)

Moreover, whenever the government imposes a tax, the consumers have to pay a higher price and, as a result, suffer the loss of consumer’s surplus. In Fig. 2, when the excise of T per unit of X is imposed, consumers lose satisfaction to the extent measured

by the area P,PAB, i.e.

Loss of Consumer’s Surplus = Area of P,PAB

=T-OX+ (1/2) T- XX, =T- OX [1+ (1/2)eT/OP]

(9)

Since out of this loss, the government gets its excise revenue of (7- OX), it represents a transfer from the consumers to the government.

The net loss of consumer’s

surplus to the society

=(1/ ie. 8 “Qa OF

(10)

However, the society also loses on the production side. Because of the excise of T per unit of the commodity, the production

declines from its potential at OX, to OX. When this decline leads to the capacity under-utilization and net unemployment of labour and other resources which could have been employed productively by the industry if it were to operate at OX, instead of OX, all this would represent social loss on account of loss of production. It would also include the net loss of farmers’ or raw material suppliers’ income when they shift their resources to the second best occupation. While generally all such costs do not occur ina perfectly competitive, ideal market economy, in relatively less developed economies like India with considerable specificity and immobility of resources, these costs are not likely to be small or

negligible. Indeed, the monetary equivalent to such opportunity costs may be considered to be half the net revenue loss to the industry, i.e. (1/2) XX, - OP,,.

The total net social loss

= (1/2) XX, - OP, + (1/2)e- T* - OX/OP = (1/2) e- T- OX/OP -(OP-T) + (1/2) e- T° - OX/OP

163

APPENDIX A

= (1/2) (e- T- OX/OP) [OP-T+ T] = (1/2) e- T- OX

C1

It may be noted that all formulae derived here are in terms of the current level of demand, nominal price, excise duty, and the price elasticity of demand for the commodity currently prevalent.

These basic data are given in Table 5.

Table 5 Basic Data for Cigarettes in India (monthly average excluding minors, March to Aug. 1994) Price (in Rs/1000 cigs.)

667.70

Excise (in Rs/1000 cigs.)

402.36

Volume (in million cigs.)

6322.00

Price elasticity of demand

(-)2.00

Excise revenue (lakh rupee)

25438

Questions for Discussion 1.

3. 4.

5.

When is a commodity considered overtaxed? What is conservatism? Why do we need the most conservative estimates of the elasticity of demand for cigarettes? | What is the income elasticity of demand for cigarettes relevant for the year 1994-5 based on the regression model 2? Using the formulae (1) to (11) and the data given in Table 5, estimate all the entries in 7able 6. Using the estimates of the Regression Model 2, and making appropriate assumptions about the values of the independent variables, calculate the optimal rate of excise on

6.

cigarettes in India for the year 1995-6. Inthe budget for 1995-6, the excise duty on cigarette is raised by 7 per cent (i.e. Rs 430 from Rs 402 per 1,000 cigarettes). Critically evaluate this step.

164

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Table 6 Optimal Excise Rates and Related Aggregates for Cigarettes (excluding minors) Aggregates

Units

Optimal excise

Rs/1000 cigs.

Optimal excise revenue

Rs lakhs/month

Demand at optimal excise rate

Million cigs./month

4

Incremental demand if excise = 0

Million cigs./month

9

Present loss of consumer surplus

Rs lakhs/month

6

Present net social loss on consumption side

Rs lakhs/month

¢

Present total net social loss

Rs lakhs/month

8

Loss of consumer surplus at optimal excise

Rs lakhs/month

9

Net social loss on consumption side at optimal excise

Rs lakhs/month

Total net social loss at optimal excise

Rs lakhs/month

|

10

Estimates for 1994-5

Appendix B Elasticity of Demand for Food in India

Tewari and Pandey (1993) attempted to estimate the Social Time Preference Rate (S7PR) for the Indian economy over the period 1951-85. The STPR is also known as the consumption rate of interest because it denotes the rate at which present consumption is foregone for additional future consumption. Thus, in an intertemporal context, it is measured through the slope of the social indifference curve between the present and future consumptions. More precisely, let_X= present consumption; Y = future consumption after one period; and r=

social time preference rate; then the

present value of consumption, Z, is given by:

is

Le > o l+r

le a

Nea os = 0 if the present value of consumption remains constant.

pe BAL sia| hy

per

res se =/11a ¥ Le. (-) Vy SRS MES = Sips

FSV

1.

They considered the following utility function for an average Indian: i=

Act l-e

AC A 8 CA a l-e

where U is utility, A is constant, mt is the probability of survival from period 1 to period 2, C, and C, are real consumption in

166

MICROECONOMICS

FOR MANAGEMENT STUDENTS

periods | and 2, respectively, and (—)e is the consumption elasticity of the marginal utility. Then,

MRS = A.C,/7A CG, = (C,/C,)* (/n) = (1 + g)* (1/2), where g is the growth rate in real consumption.

“. STPR= MRS — T= (14+ 9)" (17x) = 1. It is obvious from this formula that STPR requires estimates of 3 parameters, namely g, m,ande. Tewari and Pandey (1993) estimated g = 1.076 per cent or 0.01076, and m = 0.9795. In order to estimate the consumption elasticity of marginal utility, (—)e, they used the formula suggested by Fellner (1967): (—)e =e, 7ey, where e, and e,;, are respectively the income elasticity and the compensated price elasticity of demand for food. The income elasticity (e,), compensated price elasticity (e,,), and uncompensated price elasticity (e,) for any commodity (like food) is related by the following formula:

Cy = ep + Ke,, where K is the proportion of income spent on the good.

Tewari and Pandey (1993) estimated a food demand function using the time-series data for the period 1960-1 to 1984-5. Their estimated model is as follows:

(1)

InQ= 4.74 — 0.0724 In (P,/P,) + 0.2044 In (Y/CPN) — 0.0902 DUM (t-values) (18.3) (1.66) (5.31) (5.99) Adjusted R-Square = 0.72; DW Stat = 1.6609;

F= 21.6.

where Q is the per adult equivalent real food expenditure in 1970-1 prices, P, and P, are aggregate price indices formulated for

food and non-food products in the consumers’ basket (1970-1 = 100); Y is the per adult equivalent disposable income, CPI is consumer price index (1970-1 = 100), dummy variable, which takes a value of | in the years when foodgrain production dipped suddenly from the previous level due to drought (1965-6, 66-7,

APPENDIX B

167

(6-7, 79-80), and otherwise takes zero values for normal years.

(See Tewari and Pandey, 1993: 65 and data given in App.: 75). Their estimate of the uncompensated price elasticity of demand for food is (-)0.0724, and of income elasticity of demand for food 0.2044. The average value of the proportion of food expenditure in the total consumption for the period under consideration is 54 per cent.

Alternative models of the food demand functions have also been estimated by us using the same set of data as follows: Z in ie

4.65 — 0.1903 In (P, /P,) + 0.2201 In (Y/P,) — 0.0899 DUM

(t-values)

(19.09) (4.25)

(6.02)

Adjusted R-Square = 0.7611;

(6.45)

DW Stat. = 1.891;

F= 26.48.

CP) — 0.1995 In

CE)

(3) In O=

ao

OU2197 iy

— 0.0379 In (P, 7CPD — 0.0898 DUM (t-values) (18.6) (6.29) (2.53) (0.53) Adjusted R-Square = 0.7494; DW Stat. = 1.90;

(6.29) F = 18.94.

(4) in d=

A636 — 01882 In VP. AP.)

(t-values)

(11.4) (2.49)

0.2134 Ind ¥7P3) (3.46)

Adjusted R-Square = 0.3203;

DW Stat. = 2.20;

F = 6.65.

(5) nO =

4.09 + 0.2126 In (¥/ CP) = 0.2087 In ./ CPI) —0.043 lne,/7 GED

(t-values) (11.1) (3.36) Adjusted R-Square = 0.2891;

(1.57) DW Stat. = 2.20;

(0.35) F = 4.25.

References 1.

W. Fellner (1967): ‘Operational Utility: The Theoretical Background and Measurement’, in W. Fellner (ed.), Ten Economic Studies in the

Traditions of Irwing Fisher (New York: J. Wiley).

168

2.

MICROECONOMICS FOR MANAGEMENT STUDENTS

D.D. Tewari and I.M. Pandey (1993): ‘An Estimation of Social Time Preference Rate for India and its Public Policy Implications’, /ndian Journal of Economics, vol. Lxxiv, pt. 1, no. 292, July, pp. 61-75.

Questions for Discussion 1. 2. oe

4.

What is the estimate of the compensated price elasticity of demand for food in India as per equations (1) to (5)? Which is the most appropriate model for estimation of the elasticity of demand for food in India? Why? Which is the most appropriate estimate of STPR in India? | What can be said about the aggregative demand function for ‘non-food commodity’ in India?

Chapter 7

Methods of Demand Forecasting

Every business enterprise interested in planning its activities must have clear idea about the demand for its product. Important business planning decisions, including the strategy to be followed, the amount of capital to be raised, the extent of working capital that is likely to be necessary, labour requirement and skills, the necessary distribution and after-sale service networks, sales incentives, sourcing of the raw materials, etc. are all critically dependent on the perception of the demand for its product. If this perception is substantially faulty, most of these decisions of the enterprise are likely to prove to be erroneous and lead to avoidable losses. A reasonably correct estimate of demand,

on the other hand, can

prove to be the key for a successful venture. The consumer’s demand theory provides very useful insights into forecasting the demand for a product. However, very frequently business enterprises employ ad hoc and rudimentary methods of forecasting, often depending on surveys which have their own limitations. In some

cases, broad industry level forecasts are used, exercizing

intuition and judgement. Here we discuss the various methods of demand forecasting, but the discussion is by no means exhaustive. It only provides some idea of the wide variety of methods available. The decision to choose a particular method should depend upon the objective for generating the forecast, the cost involved in doing so, and any specific constraints, such as the time, resources available, etc.

The specific steps involved in generating a forecast can be listed as follows:

(a) (b) (c) (d) (e)

Specifying the objectives for generating the forecast; Identifying relevant data; Selecting an appropriate forecasting technique; Applying the technique to the relevant set of data; and Evaluating the forecast.

170

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

Basically, it involves gathering the necessary information, analysing it, and forming a judgement. All these steps are very important, and at every stage well-balanced decisions have to be taken.

Level of Forecasting Forecasts may be required at various levels. We can divide them into three broad levels: (i) micro-level, (ii) meso-level, and (iii) macro-level. At the micro-level, we may include the demand forecasts of a particular brand of the product or the firm or company level forecasts. The macro-level demand

forecasts, on the other

extreme, consist of an aggregative level commodity or the industry or the sectoral level forecasts. The meso-level forecasts would include all the in-between levels of demand forecasts such as those for the product categories or product groups. To give a concrete example, we may say that the demand forecasts for tobacco products may be considered to be macro-level forecasts. But demand for cigarettes, bidis, snuff, chewing tobacco, etc., may

be regarded as meso-level forecasts. Even demand for the filtered and non-filtered cigarettes; or even within the filtered category, King-size or regular-size filtered cigarettes may be regarded as meso-level forecasts. Micro-level demand forecasts are for specific brands like Classic, Gold-flake, or Four-Square filtered cigarettes, Cow-brand zarda, etc. The logic for generating different levels of forecasts is that the higher level demand forecasts necessarily represent an aggregation of forecasts at the lower levels. Most of the time, the assumption of ‘other things remaining the same’ — particularly with regard to consumer’s tastes and preferences — are more sustainable and workable at aggregate level rather than at micro-level. The generally followed practice is, therefore, to obtain the demand forecasts

at the aggregative level, whether at the macro-level or the mesolevel, depending on which can be relied upon to represent a reasonably stable situation. Then to use this forecast, together with further information about the market share of the product of interest in the product-category, to obtain the desired forecast. The advantages of using this methodology for forecasting demand at various levels are twofold: (i) it ensures internal consistency of

METHODS OF DEMAND FORECASTING

171

the forecasts which basically amounts to recognizing the overall constraints on the plausible range of the quantity demanded; and (ii) it explicitly recognizes that the objective realities not within the control of a business enterprise are likely to be the same for all competitors in the industry. Their reactions to such stimuli are also likely to be similar and hence particular care needs to be exercised in formulating an appropriate business strategy if the aim is to garner a major change in the market share of the product in question. The next question may arise as to who is responsible for conducting the actual exercise of industry-level forecasting. For very big players in the market, this exercise is not likely to be irrelevant or very costly. They may, therefore, conduct it themselves, entrusting it to their market research wing. If, however, the industry

market is shared among many players of more or less equal size, the concerned industry association may be interested in either conducting this exercise itself or hiring experts/consultants to undertake it. The Market Research Information Bureau (MRIB) also engages itself in producing demand forecasts at various levels for different products and product groups. However, it needs to be recognized that business enterprises or associations are not the only entities interested in generating the demand forecasts. Government

and its various

Ministries — particularly,

Finance,

Trade, Industry, and Planning — are likely to be equally interested in the demand forecasts, though not necessarily at a very microlevel.

Indeed,

their interest is likely to be only in macro

and

meso-level forecasts. This is because virtually all their policies, such as taxation, export-import policy, developmental and planning strategy, etc. consider the demand for industrial or sectoral products representing a high degree of aggregation. Such policies cannot be addressed to specific product brands. The Plan documents and numerous policy papers prepared by government Ministries — sometimes with the help of external experts and consultants — represent a rich source of useful information on macro- and meso-level demand forecasts. These are valuable because they are supposed to be consistent with the numerous policy changes which have taken place, are taking place, or are contemplated. Independent forecasts at micro-level not using such information are likely to be misleading and inaccurate.

172

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Methods of Information Gathering There are essentially three different ways of gathering the necessary information for forecasting the demand for a product. Data Surveys Consumers|

Experiments Experts

Sales Force

Secondary Source Indicators}

|

Parameters

Variables

Test Marketing

Laboratories Experiments

Figure 1 Consumer Surveys are very costly propositions, particularly when they have to be conducted only for a specific product. If such surveys are commonly conducted for many products, for each one of them the information collected on the relevant aspects and factors would be limited. Thus, there is a trade-off between the extent and depth of information gathered and the cost of doing so. Moreover, the reliability of the data also poses questions, because the survey, when conducted on a sample basis, has to ensure that the choice of the sample is appropriate to the need and that the questions, their sequence, etc. do not distort the response. Even the concepts, methodology,’ and measurements used in the sample surveys have to be uniform across all field investigators. Assistance from organizations specializing in conducting such surveys may be sought. However, there is a danger. If repetitive exercises are conducted by the field staff, they may

! For example, how a non-response is to be dealt with, or the absence of the

respondent, or how an entry is to be cross-checked, etc. need to be uniform and consistently followed by all field staff.

METHODS OF DEMAND FORECASTING

173

tend to substitute facts by guess work and presumptions. At the national level, there are periodical consumer expenditure surveys conducted by specialized organizations like the National Sample Survey Organization (NSSO). These surveys are conducted with a fair degree of uniformity and reliability on various macro-level product groupings. Their findings form the basis for several policyoriented research and official forecasts. The MRIB also periodically conducts consumer surveys for numerous products and

product groupings, sometimes obtaining the information on the basis of even specific brands. Consumer surveys may be resorted to once in a while, bearing

in view the cost in terms of time and resources. For annual or shorter duration forecasts, a company can hardly depend on elaborate consumer surveys unless the product is a raw material or intermediate product with a very limited number of buyers. If the number of buyers is large and are geographically widely dispersed, consumer surveys would prove to be a very inferior option as a means of generating a demand forecast. Under such circumstances, firms frequently depend upon their sales forces to provide some estimates. A market survey based on responses from the company’s sales forces is relatively a cost-efficient way of gathering the necessary information, particularly since the sales force is directly in touch with the market and the consumers. Their judgement is therefore likely to be relatively more accurate and reliable than that of those not specifically concerned with the product and its market. However, the principal danger is that the employees in the sales department may consider such information as ultimately being related to their performance evaluation. Unwarranted optimism or pessimism is, therefore, likely to colour their estimates. Depending on their perception of the top management’s views, they may consciously overstate or understate the likely demand for the company’s product. In order to overcome these limitations, there are various methods of linking the provision of such crucial information and feedback with incentives and disincentives that foster accuracy. It is possible to maintain an information system enabling each salesman’s past performance in relation to his estimates to be readily verified. Such feedback may also be given to him before he is asked to estimate the demand from his territory. This would, by itself,

174

MICROECONOMICS

FOR MANAGEMENT STUDENTS

substantially discourage systematic under- or over-estimation of demand by the sales force. The principal disadvantage of data gathered through the sales force is that they represent perceptions or judgement of individuals with very diverse backgrounds and intelligence. It is difficult to comprehensively consider the overall effect of ‘other’ factors influencing the demand for a product. The prices of related goods, incomes of the consumers, and, above all, the price of the

product itself tend not to be clearly reflected as important influences on the demand for a product. In short, the survey through the sales forces, would not provide sufficient information to generate a demand

schedule

and may, at best, generate a point estimate

which may not prove useful in considering different options or taking major decisions about a change in strategy. Thus, a survey conducted by the sales force is capable of generating a routine, short-term demand forecast, but cannot be very useful in providing important inputs in long range business planning decisions. An alternative to a survey conducted by the sales force is to contact experts and knowledgeable people in the field and conducting an opinion poll. It is also possible to combine the experts’ opinion with the sales force survey. Basically, the perception of experts such as consultants, agents, academicians, industry leaders, etc. would be very valuable in determining the future prospects and problems likely to be faced by the product. These inputs would typically be available in terms of several possible scenarios. In-depth discussions with them would help in working out probabilities from such possibilities. Since they would typically consider several factors in conjunction, their inputs might prove to be very useful for long-range business planning and strategy formulation. Specially designed experiments can be conducted to collect the necessary information for demand forecasting by a company. This method is generally employed when the company seeks to introduce some innovation. It could be in terms of the design, content, or quality of the product, or the development of a new market segment or new distribution channel. The most popular means of conducting such experiments is Test Marketing in which

any innovation is tested out on experimental basis. The data collected over the test marketing period is then analysed and

METHODS OF DEMAND FORECASTING

175

used for large-scale introduction of the innovation incorporating necessary improvements in the design, scope, or delivery system, whichever is relevant. The test marketing method is thus very useful in generating some estimates about a few relevant parameters useful for forecasting demand for a new product or new market. It is also possible to use this method for testing some hypotheses about the consumers’ response to several factors considered relevant by the sales department of a company. Another experimental method is that of conducting laboratory experiments. Under ideal conditions created specifically to test some relevant hypotheses about the consumers’ behaviour, selected individuals are experimented upon. Their cases are studied intensively by presenting them with several options and alternative conditions. The principal purpose here would be to estimate some useful parameters and gain insights into key, specific factors in determining the level of the demand for the product coicerned. This is not a very popular method. The utmost care and caution needs to be exercised to conduct the right experiments and then to draw correct inferences from them. Since the participants are intelligent human beings, the very knowledge that their behaviour is being studied can lead them to act differently. If their responses are not spontaneous but artificial, the whole experiment may mislead the investigator into drawing incorrect conclusions, and the resulting forecasts, therefore, prov-

ing to be very misleading. However, if properly conducted, this technique can provide very useful insights both for long range business planning decisions and regarding the most effective marketing strategy for the product. Secondary sources of data are perhaps the least costly options for gathering useful information. However, there is always the problem. of the data being dated, and sometimes so out of date that they are useless and irrelevant. However,

exercising some

care

and caution, such secondary sources as the Census Reports, NSS Reports, Plan documents, National Accounts Statistics, studies by

individual scholars, articles in journals, news papers, etc., can be used to a great advantage in arriving at a demand forecast for a product or product-group. Secondary sources do provide data on several indicators at the national and regional level. Of these, Leading Indicators are useful for demand forecasting. Leading indicators

176

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

generally precede the event and hence are very useful in making reasonably accurate forecasts. For instance, the birth-rate in a region can serve as an indicator of the future demand for many children related goods such as childwear, games, toys, etc. Secon-

dary sources also provide very useful information on macro-level variables like population, urbanization, the age-sex composition of the population, literacy level, income level, inflation rate, investment, exports, imports, sectoral output, etc. These variables play the level of the demand curve a very important role in determining for several products and product groups. Secondary sources provide time series data on such variables as well as some idea about their future values based on the most current assumptions made about them in the relevant quarters, such as the Finance Ministry or Planning Commission, etc. After all, most of these variables are outside the control of an individual company or an industry. They are ‘givens’ in a genuine sense of the term. Only the government with several policy instruments at its command can hope to marginally influence the future values of such variables, yet they are so crucial that they affect everybody and hence the demand for most products. Another important aspect of information available from secondary sources is that on relevant parameters, such as the demand elasticities of specific products or product-groups. There are individual studies or studies carried out by some organizations which may in the past have estimated such demand elasticities for the product in question or for a very similar product. Such estimates of parameters can be fruitfully applied to estimate the demand for a product or to forecast the demand for a product group. They may also be very useful in generating interesting policy options. Box J presents one such problem that serves as an illustration. Readers may attempt to seek a solution. The problem uses approximate figures based on targets over the next five years in India.

Methods of Demand Forecasting Having collected the information and data on the basis of a survey or experiments or the secondary sources, it is necessary to apply

METHODS OF DEMAND FORECASTING

Igy

Box I An Illustrative Problem in Demand Forecasting

During the next five years, the government has worked out the following targets: (i) Gi) (ii) (iv)

growth growth overall growth

of total real income of population inflation rate of foodgrain production

6% Pp.a. 2% p.a. 6% p.a. 2.5% p.a.

Moreover, it is estimated that the income elasticity of demand for foodgrains is 0.21 and the price elasticity of demand for foodgrains is (-)0.19. Cross-elasticity is negligible and can be ignored. (a)

If the government wants to maintain the relative price of foodgrains constant, how much foodgrain per annum should it export or import? If the government does not want to export or import foodgrains, by how much would the prices of foodgrains change per annum in absolute terms and relative terms?

some statistical techniques and methods to obtain the required demand forecast. The informal method of historical comparisons and using some selected estimates of the parameters to arrive at some qualitative demand forecast would constitute a part of gathering relevant information through secondary sources. Here we will discuss more formal methods yielding quantitative estimates of demand forecasts. Basically, there are three methods for this: (1) time Series Models which use the time-series data on the demand variable,

treating it as an exclusive function of time alone; (2) Econometric Models which consider the most relevant determinants of the demand for the product as dictated by the economic theory and, (3) End-Use Method which uses survey based data from customers of an intermediate product who produce the final products

178

MICROECONOMICS FOR MANAGEMENT STUDENTS

for consumers. Here we briefly outline these methods, providing specific illustrations from a recent study on Forecast of Domestic Demand for Cement in Indian Economy (VIII and IX Plan Period) carried out at IIM, Ahmedabad during 1993-5 by Professors Bakul H. Dholakia and K.V. Ramani (henceforth referred to as ‘BHD-KVR Study’). In this study, the authors have used a number of different methods to generate alternative forecasts of the domestic demand for cement in India. The annual domestic consumption of cement is measured in the study as the total annual despatches of cement made by the large and mini cement plants less exports. These data were made available by the: Cement Manufacturers’ Association (CMA) for the period 1980-1 to 1992-3 on annual basis. It is argued that trends in cement consumption prior to its partial decontrol in the early eighties is not comparable to trends thereafter. In order to avoid biased estimates and focus on the more relevant recent period, the sample period chosen by the authors was 1980-1 onwards.

Time Series Models

Time series models require data only on the dependent variable. It can, therefore, address the need of a forecast for even a very short period like a week, a month, or a quarter, for which the data on other variables may not be available. 7ime series models

generally imply an assumption of a continuation of past trends in the future. This is because they generally obtain the demand forecast by extrapolating the time relation established with the dependent variable through a formal model. Such formal models can be specified in many different ways. Some of them are called naive models which do not require any elaborate estimation and their data requirement is also not very high. Some of the naive models are: (1)

X41

= Xp + (Xp — Xy_ 3)

where X is the demand for the good and fis the time period. This ‘model’ is based on the simple additive assumption that equalizes the absolute difference between two consecutive quantities. Another naive model is:

METHODS OF DEMAND FORECASTING

(2)

#be

X41 = (Xp 7X; _ XY

This is based on the simple assumption that the percentage change between two consecutive terms remains the same. There may be several such simple ‘models’. If the number of observations are limited, such ‘models’ are very handy. However, if longer time series data is available, it is preferable to use other models.

Trend Models These are basically regression models with the demand as the dependent variable and time as the independent variable. Again several functional forms are possible here. The most popular are:

(3)

Linear Trend Model: X,= a + bt.

(4)

Exponential Trend Model: In X,=a + bt.

Both these trend equations are estimated for the cement demand in the BHD-KVR study. Their estimates are:

(5)

Linear: X,= 14556 + 3088.9*t

(R* = 0.9946)

(6)

Exponential: In X,= 9.8034 + 0.09094 *t

(R*=0.9794)

where X, is domestic cement demand in India in year tf, and t= 1 for 1980-1 to 13 for 1992-3, and 14 for 1993-4 to 22 for 2001-2

(sample) (forecast period)

The trend models can be further modified and used to identify several components ofa time series. If our interest is in a quarterly or monthly forecast for a product which is subject to some seasonal fluctuations in demand, we can use the monthly or quarterly data to track the seasonality and prepare a seasonality index on the basis of standard statistical methods such as moving averages. After eliminating the effect of seasonal variations, we can obtain a deseasonalized time series and estimate its trend. There are standard statistical techniques for identifying the four components of a time series, namely (i) Trend, (ii) Seasonal Variation, (iii) Cyclical Fluctuations, and (iv) Random Fluctuations. Detailed discussion on these methods is beyond the purview of this book.

180

MICROECONOMICS FOR MANAGEMENT STUDENTS

Autoregressive Models Another category of time series models, again more sophisticated in terms of the statistical techniques used, recognize serial cor-

relation. The models using the serial correlation between X, and X,_1,X;_9, etc. in a series may be considered to be a class of autoregressive models and may also explicitly consider the correlations with the residuals in the previous periods, i.e. X, with

€,_ 1, &;_», etc. The most sophisticated version of these is called the Auto Regressive Integrated Moving Average (ARIMA) model, also known as Box-Jenkin’s method. All these models, with what-

ever sophisticated statistical techniques they may be using, are ultimately based on the analysis of a single time series. They do not consider any causal relationships or any behavioural influences on the dependent variable. In the ultimate analysis, they amount to assuming some from of continuation of past behaviour in the future. However, with greater and greater sophistication achieved in time series analysis, the additional insights provided by the behavioural sciences like economics, sociology, etc. can be more rigorously evaluated. The BHD-KVR study also estimated the most appropriate ARIMA model and on the basis of it derived their demand forecast.

Econometric Models

These are again regression models which are specified on the basis of the determinants suggested by economic theory. They thus represent causal and behavioural relationships among different variables. This method requires data not only on the dependent variable but also on all identified independent variables. The regressions can be estimated by using either a time series data set or a cross-sectional data set obtained from surveys or experiments. The regression models too could also be of two types: (a) The single

equation model where only uni-directional causality is considered; and (b) The simultaneous equations model where two-way Causality is also considered. The simultaneous equations model requires estimation techniques dealing with the estimation of a system of interrelated equations with several constraints imposed by economic theory. This technique of simultaneous equations is

METHODS OF DEMAND FORECASTING

181

particularly appropriate if both the demand and supply of a product can be considered well-defined functions freely determined by behavioural variables. If any one of them is constrained or determined exogenously, it is possible to treat the single equation estimation to be workable if not appropriate. The principal advantage of the econometric model is its ability to identify the causes for a particular forecast. It is actually able to decompose, if necessary, the sources of the change predicted in demand. It also becomes possible to consider various alternative scenarios and their impact on the demand forecasts. Such forecasts are most useful in policy formulation or strategy evaluation or business planning. The forecasts based only on the time series models lack this feature and to that extent are less useful. However,

it should be ciear from the discussion so far that the

quality of the forecasts based on the econometric model critically depends upon:

(i)

the correct identification of the most relevant explanatory variables;

(ii)

the degree of confidence that can be placed on the estimated relationship or the explanatory power as well as overall fit of the model; and

(iii) the forecast of the explanatory variables, which have to be very carefully and meticulously derived by selecting the most expected values of such variables using all possible information rather than deriving them mechanistically by applying some standard methods such as trend or equal percentage growth. While demand forecasters generally consider and devote a lot of attention to the first two aspects stated above, they usually tend to ignore the third. As a result, they are not able to make optimum

use of their estimated econometric model. The distinguishing feature of the BHD-KVR study is that it considered all these three aspects particularly the third very carefully. The causal or explanatory variables considered in the econometric model of the cement demand function in the BHD-KVR study are: the aggregate purchasing power of the buyers, the relative price of cement, the structure of the domestic economy, the extent of capital formation in the economy, the extent of

182

MICROECONOMICS FOR MANAGEMENT STUDENTS

infrastructural development and the relative growth of development oriented public expenditure. The aggregate purchasing power is measured as the real gross domestic product (GDP). The relative price of cement is measured as a ratio of the wholesale price index for cement to the GDP deflator. The structure of the domestic economy which would affect the demand for cement is measured by the share of the non-agricultural sector in GDP (VAS), and also by ratio of urban population to total population. Over and above these factors, the level of cement consumption in the immediately preceding period is also considered a determinant of the demand for cement in the current period. The BHD-KVR study considered numerous alternative specifications of the model, incorporating different combinations of the explanatory variables. Of these alternative models, the one that fitted the data best and also performed best at the prediction tests was selected for the generation of demand forecasts for the future. It may be mentioned here that the study carried out the forecasting exercise not only at the national level but also at the state level. For different state economies, different specifications of the econometric model performed better than the others. Besides, all these factors could not have been considered simul-

taneously in a single equation because, as there were only 13 data points, this imposed a genuine constraint on the degrees of freedom available. The most acceptable econometric model reported by BHDKVR for all India domestic demand for cement is:

(7)

X, = (—)82526 + 0.2683 GDP, + 1109.3 NAS, R* = 0.9939

The regression shows an excellent fit given by a very high value of R2 which is also known as the coefficient of determination.” It is interesting to compare the forecasting errors of the Linear and Exponential Trend models and the one of the Econometric models for the year 1993-4. The provisional estimate for actual cement consumption in India during 1993-4 is 55,509 thousand 2 The maximum possible value of R’ is 1.0. It basically shows the proportion of the total variation in the dependent variable explained by the independent variables considered in the model.

183

METHODS OF DEMAND FORECASTING

tonnes. The forecasting error for the year 1993-4 turns out to be (+)4.13 per cent in the case of the Linear Trend model and (+)16.45 per cent in the Exponential Trend model. However, the margin of error is quite small at (+)1.32 per cent in the case of the Econometric model. The Econometric model, thus, appearseto be more dependable than Trend models in forecasting cement demand in India.

End-Use Method

ae

This is the most widely used method for demand forecasting of industrial products used for further production, or intermediate

products. They would also include the capital goods or investment goods. All such products are basically in demand not because they satisfy some consumer needs directly, but because they help to produce final goods and services that satisfy the consumers’ wants. The demand for the intermediate products is, therefore, a derived demand. Moreover, since such products are used for further production of specified goods, the users are also the producers, interested in producing their product most efficiently using the latest available technology. The physical requirements of the intermediate products to produce one unit of the final product is likely to remain stable over time. In other words, if such

a requirement is not constant, it can at least be known or predicted fairly accurately given the technology used by the producers of the final products. The end-use method of demand forecasting consists of the following steps:

i)

ii)

iii)

Identification and listing of all possible end-uses of the product. Let there be n different uses of the product (X) given by Keel ot eerie Estimating the requirement of the product per unit of the final product in each end-use; i.e. estimate the ratios given by X,/Y,, where X, represents the amount of the product X used in the production of the :-th final product Y.. Estimating through survey or any secondary sources the desired level of the production of the final outputs in

different uses; i.e. estimating future levels of Y,’s. Let us denote them as Y;, ,, 1.

184

(iv)

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Applying the estimated requirements to the estimated future output levels of the final outputs in the different end-uses of the product_X; i.e. obtain n

fig sy Raytele Lk ghUe,t+l i=]

The principal advantages of the end-use method are as follows: (i) It yields the demand forecast by direction of use, specifications of the product, qualitative precision, etc. Thus, only aggregate forecast is not produced. The essential information on product composition and the consequent demand pattern are also provided. (ii) The end-use method helps to incorporate critical inputs on changes in technology or breaks in the pattern of demand for the final products that are likely to be relevant to the demand for industrial goods. Such information is not likely to be captured | by other methods. (iii) When there are major deviations between the forecast and

actual

demand,

since

the forecast using this

method provides details of uses and direction, it helps to locate the sources of the deviations and to take corrective measures or revise future forecasts. The principal limitations of the end-use method are: (i) It can be applied only to industrial products used for further production, and is not applicable to most consumer goods. (ii) The data requirement for this method is enormous. Few organizations would be in a position to afford it on a regular basis. (iii) Since customer surveys are necessary to obtain very specific data, all the limitations of the survey method are also applicable to this one. In order to circumvent

these limitations, survey-based

data collection

is

attempted only occasionally, generally, available macroeconomic data from the well-known secondary sources being used. (iv) Finally, it is difficult to estimate the product requirements per unit of the final product because possibilities of substitution exist. The price of the product may be an important determinant of it. The BHD-KVR study also attempted to forecast the demand for cement in India through the end-use method. The principal prob-

lem faced by them was the complete lack of basic data and relevant source material at the industry and company levels. The pattern of cement demand according to the end-uses is simply not

METHODS

OF DEMAND

185

FORECASTING

looked into by the industry sources nor is it a concern for their internal information

system.

It, therefore, became

necessary

to

resort to a sample survey, albeit with much reservation about the possibilities of its success. The survey was conducted with ‘commendable support and assistance from the CMA’, and yet it failed to produce the desired results. As the authors point out, this happened because of a very high proportion of non-response and incomplete or inadequate response received from the companies, highlighting an important limitation of the method. The authors, however, did not give up. When the end-use method based on the micro-level data did not prove to be feasible, the use of macrolevel data was resorted to. The actual end-users of cement are units engaged in construction activities of various types. They may be building residential non-residential buildings, bridges, dams, canals, roads, powerhouses, railway tracks, factories, infrastructure, etc. These users can, therefore, be classified into the standard economic houses,

sectors as defined in the National Accounts Statistics. Accordingly, the following seven sector — categories were identified as the major cement consuming sectors in the Indian economy:

Agriculture, Forestry & Fishery (Canals, Dams & Land Development) Gi) Mining & Quarrying (iii) Manufacturing (Factories) (iv) Electricity, Gas & Water Supply (Power) (v) Railways (vi) Transport by other Means & Communication (Roads, Bridges & Communication Infrastructure) (vii) Residential & Non-residential Buildings. (i)

To estimate the cement requirement, cement consumption was

related to the Gross Fixed Capital Formation (GFCF). By using numerous secondary sources from the Ministry of Planning, they estimated the proportion of construction in GFCF for each of the specified sectors and obtained estimates of expenditures on cement by those sectors. By using the price data, cement consumption in the physical units was derived for each of the end-use sectors. The study goes into details of estimation of the technical norm by sectors on a time series basis for the period 1980-1 to

186

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

1992-3. It is the projected value of such technical norms for the future date by each sector of the end-use which is considered for deriving the forecast. Table 1 Comparison of Cement Demand Forecasts In India Obtained by Alternative Methods (million tonnes) Method

Scenario-[

Scenario-I!

Scenario-Ill

For 1996-7 1. Trend Analysis

(i) Linear

-

67.07

~-

(i) Exponential

~

84.92

~

2.

ARIMA

66.80

70.20

73.60

3.

Econometric Model

65.06

69.19

(3.42

4.

End-use

(i) Direct

64.45

68.18

72.60

(ii) Simulation

67.34

69.91

76.28

For 2001-2 1. Trend Analysis

(i) Linear

~

O2ia1

-

(ii) Exponential

~

133:61

~

2.

ARIMA

91:20

9730

103.40

3.

Econometric Model

52.00

Leis

109.64

4.

End-use

(i) Direct

81.78

95.03

i215

(ii) Simulation

89.46

92.06

115.48

SOURCE:

Bakul H. Dholakia and K.V. Ramani, Forecast of Domestic Demand For Cement in Indian Economy (VIII & IX Plan Period), part! (Feb. 1994); and part I! (May, 1995), IIM, Ahmedabad.

Moreover,

the demand

forecasts

are derived

under

three

alternative scenarios: (i) Scenario I (pessimistic), representing a

METHODS OF DEMAND

FORECASTING

187

continuation of the actual performance of the Indian economy during the years 1990-3 over the following 10 years. This scenario implies: Real GDP growth of 4 per cent p.a., Non-agricultural sector growth of 4.5 per cent p.a., general Inflation rate of 8.5 per cent p.a., and Investment Rate of less than 22 per cent. (ii) Scenario I/ (most likely) representing actual realization of the Eighth Plan projections. It assumes a real GDP growth of 5.6 per cent p.a., Non-agricultural growth of 6.5 per cent p.a., an inflation rate of less than 6 per cent p.a., and an Investment rate of over 23 per cent.

(iii) Scenario III (optimistic) represents the scenario of a high degree of buoyancy over the next 10 years. It assumes a real GDP growth of 7 per cent p.a., Non-agricultural growth of 8.5 per cent p.a., an Inflation rate of less than 5 per cent p.a., and an Investment rate of over 25 per cent. Table I summarizes the BHD-KVR study’s forecasts for the domestic demand for cement in India by alternative methods for the period ending the VIII Plan and IX Plan under the three alternative scenarios. The authors’ comment on these forecasts are presented below in their own words: It is interesting to examine the forecasts under the optimistic scenario (Scenario II[) obtained through alternative methods, especially for the year 2001-2. It is evident from Table / that the ARIMA Models (based on monthly data from April 1982 to March 1992) yield relatively lower forecasts, while the end-use method yields relatively higher forecasts as compared to the econometric models. The main reason behind the relatively lower forecasts generated by the ARIMA model seems to be the comparatively high weightage attached by this forecasting method to the latest available data, which in this case happen to be the monthly data for the year 1992-3 — by now clearly identified as an abnormal year. In fact, if the ARIMA Model forecasts are obtained after taking into account the more recent data for the year 1993-4, they would be more or less in

agreement with the corresponding forecasts obtained from the econometric models. Between the end-use method and econometric models, we would like to place greater weightage on the latter mainly on account of two reasons: (a) Reasonably satisfactory state of data availability as well as the wellestablished estimation procedures and validation methods applicable in the case of econometric models; and (b) Relatively less satisfactory state of data availability and the lack of well-established estimation procedures

applicable in the case of the end-use method. Thus, based on a detailed

188

MICROECONOMICS

FOR MANAGEMENT STUDENTS

comparison of the forecasts obtained by alternative methods, we have come to the conclusion that the forecasts of domestic demand for cement

in Indian economy obtained by applying econometric models are to be regarded as more reliable as compared to the corresponding forecasts obtained through other methods. [BHD-KVR Study, part tl, p. 54].

Chapter 8 Theory of Production

The supply of a good represents an offer from the sellers backed by an ability to produce the good. The supply response is, therefore, primarily determined by the production conditions of the seller. The production condition is an objective reality that is likely to remain the same for different sellers. The methods whereby a production unit may use various factors of production and inputs to produce different output levels of the good in question are provided by the current fechnology. It essentially represents the knowledge about the state of the art, comprising various techniques of production defined as combinations of the factors of production. Thus, production requires the factors of production to be combined in specific proportions at various stages. The production of any good or service has to be visualized, organized, and

monitored. The entrepreneur performs all these tasks either himself or by hiring the services of other factors of production, himself bearing the risk involved in the venture.

Production and Firm

Production is associated with process of undertaking any alteration or transformation of a good or a service into another good or service. It involves a change in the specification of the commodity in terms of its size, shape, quality, content, form, location, etc. In short, production involves any use of the scarce resources to produce commodities, i.e. goods or services that are able to satisfy some consumer needs. A firm is defined as the unit where production takes place. It is an organization that generates output. A firm internalizes a number of activities and processes leading to the production of a commodity. Conceptually, it is possible to consider the outcome of every single process or activity carried out as an output of some commodity for use in the production of the

190

MICROECONOMICS

FOR MANAGEMENT STUDENTS

final output of a good or a service produced by the firm. When we treat all such distinct intermediate products as potential outputs, the existence of a firm has to be justified on some socioeconomic or political grounds. The existence of a firm on economic grounds can be justified under the following two conditions:

(i)

(ii)

The aggregate output of all such intermediate processes and activities internalized in a firm should be greater than their output when they are independently carried out externally; and The additional cost of supervision and monitoring of such activities internalized in a firm over the transaction cost involved in independently carrying out the same set of activities externally should be less than the output gain obtained in (i) above.

Let us take an illustration of a factory producing ready-made garments. We may consider four distinct stages of production which might be internalized: (a) Designing, (b) Cutting, (c) Stitching, and (d) Packing. All these four activities can be carried out independently and externally. Under these circumstances, at every stage a formal exchange of goods should be taking place, involving the cost of entering into a contract, insurance, excise,

sales tax, etc. All these are transaction costs, arising as they do from transactions. They would also include the cost of enforcing the contracts, such as legal expense, delays involved, etc. Several of these transaction costs are saved when these activities are internalized. However, some additional costs arise such as those of supervision, monitoring and organization of activities and resources to ensure efficiency. Some of these costs are also affected by the politico-legal environment in the country. There are several Acts of the Parliament and regulations governing the

organization of the business and production, e.g. the Factory Act, the MRTP Act, FERA, etc. They also-impose some direct and indirect costs on internalizing such activities in a firm. If all such costs taken together are less than the transaction costs involved in carrying out these activities externally, there is an economic justification for a firm to exist. Even if these costs are

equal or adverse, a firm can still have an economic justification

THEORY OF PRODUCTION

191

for its existence provided the output gains can compensate for such additional costs. It is important to recognize that the basic conditions necessary to justify the existence of a firm can and do change over time.

In consequence new firms in new lines are being set up or existing firms expanding horizontally or integrating vertically, or firms being split with the passage of time. It should also be recognized that firms can have non-economic justifications for their existence. Many public sector companies have been set up for socio-political considerations. Profits or any economic criteria may either be irrelevant or at best of secondary importance. If the national goal is industrial development and if private enterprise is not well-developed in the country, the government may decide to set up public enterprises to produce and provide goods and services considered necessary. Whether such an alternative Is economically viable or not is then politically irrelevant. With changed environment under the economic policy reforms of the 1990s, the basic conditions for the existence of a firm or business have substantially changed. Substantial organizational changes in production and the nature of the firm may also, therefore, be expected.

Firm’s Objective Business activity is generally defined as any activity undertaken with a predominant profit motive. However, not all businesses are organized with the sole objective of making profits. There may be several objectives other than earning a profit to undertake production. The psychology and motivation of entrepreneurs could be very complex. For private entrepreneurs, profits could be one of the motives but not the sole motive. Moreover, we should also

recognize that the motive of the entrepreneur at the time of launching an activity could be very different from the motives after the business is established and starts production. Earning some minimum profit may become essential because it may represent the opportunity cost of the entrepreneur to remain in the production of the commodity. For a running private business, it is extremely unlikely that profit is not an important motive.

192

MICROECONOMICS

FOR MANAGEMENT STUDENTS

The modern business has a far more extensive scale and scope of operation, spread and diversity than can be effectively handled by one or two individuals alone without substantial assistance in managing the business. There are still several small business enterprises that exist and also flourish over time. They are basically run by an individual proprietor or partnership as a form of business organization. The principal advantage of such individualcentered forms of business organizations is that ownership and control of the business are in the same hands. Business decisions are, therefore, likely to be based on the basic objectives of the business, and objectives and goals likely to be clearly defined (albeit not formally on paper). Large-scale business organizations based on corporate or cooperative structures are likely to follow practices where the ownership and control of the business are invariably separated. The owners in such cases are expected to exercise overall control over the managers and executives by evaluating their periformance and changing them if this is found to be unsatisfactory. For this, the owners must specify formally the principal objectives of the business, its goals, and prescribe definite criteria for performance evaluation. However, control of this type is only indirect,

with a substantial amount of discretion delegated to the managers. Besides, formal statements of objectives cannot be very precise and some ambiguity is bound to remain. For instance, earning the maximum possible profits could be the stated objective, but it is difficult to evaluate in practice. In order to exercise effective control over the managers, the owners of a business have to take some

active interest by participating in important meetings. Its

shareholders, who are not only numerous but also geographically dispersed, may not individually have sufficiently high stakes and may not, therefore, participate very actively in several relevant and important matters. All this gives rise to an expectation that the managers who control the day-to-day operations and also several medium-term business decisions would try to impose their own objectives on the business. They are thus likely to act ina way that maximizes their own utility rather than the utility of the owners of the business. Managers generally derive special satisfaction from power, authority, position, and status they enjoy in society. Such matters

THEORY OF PRODUCTION

193

directly or indirectly depend upon some well-defined and well-accepted criteria, such as the total sales turnover of their company,

number of employees, the growth of the business over time, etc. The firms may, therefore, effectively have such objectives although their stated objectives may remain profits, social service, etc. Another very powerful driving force of the behaviour of managers may be a preference for fewer tensions associated with slow but steady business over high tensions in a fast moving business. Some popular alternative objectives of a firm may be: (a) (b) (c) (d) (e)

Profit maximization Sales maximization Growth maximization over a given period Employment generation Satisficing behaviour

Firms are likely to have several objectives, but the theory of firm is developed by considering only one objective at a time. This is because it is only then that the implications of the given objective on the behaviour of the firm can be fully worked out. Even here,

the most popular postulate considered is the maximization of the profits. We will consider some of the above-mentioned objectives later. At this stage, it may be noted that objectives other than that of

profit maximization also require that some minimum profits are treated as an effective constraint on a firm’s management. For private enterprise, profits or the net asset value becomes an important parameter. Even when effective control is delinked or separated from the ownership of a business, it is not necessary

that the managers are entrusted with total freedom to set any goals of their choice in its running. They are answerable to the general body of the shareholders. If they do not perform, they will be changed. This, however, raises the question of the availability of trained and competent managers in the economy. If their supply is increased, this acts as an effective check on their behaviour by

curbing their bargaining power. An effective and ever growing market for managers and senior executives is likely to go a long way in ensuring greater adherence of the management to the

stated objectives of the business. Besides, if there is a thriving narket for managers, the better performing managers can aspire

194

MICROECONOMICS

FOR MANAGEMENT STUDENTS

to prosper because free market forces always reward talent and competence when this is coupled with performance. Even when managers are in short supply, they may not be in a position to move a business away from its stated or desired objectives if there is a threat of takeover. Thus, if the legal and political climate in a country facilitates easy takeovers, a company not performing properly always runs the risk of being taken over by another. This in itself creates a competitive environment, in which the management of every company has to remain constantly on their toes. The existence of a free market for company takeovers ensures certain minimum level of the financial performance by companies. Moreover, it also provides incentives for the better performers to earn more and expand their business. The process of economic liberalization thus would imply greater protection of the investors’ interests than under a regulated regime. The basic assumption of profit maximization being the principal objective of a firm is thus defensible under the ideal free-market conditions. Capitalism as a system of organizing economic activity places great emphasis on voluntary and free interplay of the market forces of demand and supply. What we would like to examine is given such interplay is freely allowed, how resources are allocated in the economy. Profit maximization, therefore, appears to be the most acceptable and workable assumption as a firm’s principal objective.

Production Function The production function is a formal statement of the linkage between the inputs and outputs of a firm or a production unit. It essentially describes the state of technology available to a firm for the production of various alternative quantities of output with different quantities of inputs. Thus, a production function can be looked upon as providing information on numerous alternative techniques of production permitted by different combinations of the resources or inputs. The production function would differ from commodity to commodity. Although it is basically a technical relation, it can be empirically estimated by various statistical techniques if sufficient time series or cross-section data exist. At

195

THEORY OF PRODUCTION

an empirical level, we can estimate the production function not only for a product but also for a group of products. It is possible to conceptualize even an aggregate production function for the economy as a whole. As aconcept, the production function is very useful in describing several production- and technology-related aspects of a firm. A production function is based on the notion of efficiency in resource use. Since it is a technical relationship between inputs and output, it represents the most efficient methods of producing the output with different combinations of resources. Thus, a production function does not relate any arbitrary level of output to the given combination of the inputs. /t represents the maximum possible output obtainable under the present state of Rnowledge from a given combination

of inputs. In other words,

it ensures

efficiency in resource use. If any circumstances make a firm decide to underutilize some of its resources, they should be considered separately as a constraint or as exogenous forces. The basic production relation as set out by the production function should describe only the most efficient methods of producing different levels of output. The production function can be written as:

(1)

X=f(K,L)

where X is the level of output,

K is capital, and LZ is labour. We

may also consider other factors of production. For simplicity, however, we assume here that there are only two factors of production: Labour and Capital. It is important to point out at the outset that output, labour, and capital are all assumed to be measured in their respective homogeneous units. In other words, when we talk about labour being increased from 10 to 20 units, we mean

the labour of an identical nature; of the same

quality.

Similarly, capital also consists of homogeneous units. In reality, we observe several categories of labour and several types of machinery, equipment, buildings, etc. Ideally we should incorporate all of these different qualities and categories of labour and capital as separate factors or inputs in the production function. Our assumption of only two relevant factors of production is essentially a simplifying assumption which is adequate for broad analytical purposes.

196

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Another important thing to note about the production function is that it is a relationship among flows and not stocks. Production of a commodity occurs over a period of time. The inputs that are used in the production process, either fully or partially, should also represent the corresponding flows. If the inputs are not measured as flows over the same period as the output, a correct technical relationship between the inputs and outputs cannot be defined. The production function relates the output that can be produced with the inputs which are used for producing it. Like a utility function, a production function also represents technological options in terms of various alternative input combinations to produce the output. We may now discuss important features of the production technology available to a firm by considering its production function.

Total, Average and Marginal Product A formal functional relationship between the inputs and output, as denoted in equation (1), can be used to relate one input with the output, the other inputs remaining constant at a given level. Thus, we may rewrite equation (1) as:

(2)

X =f (K,, L) or

(3)

X =f (K, L,)

In equation (2), capital is held constant at K, level. It represents a functional relationship between labour and output. It is called a total product function with respect to labour. It is defined for a given value of capital. If the value of capital increases from K, to K,, the total product of the labour function would change in such a way that for every value of labour, the corresponding output increases. This is because we are assuming that capital contributes positively to production. Similarly, we can interpret the equation (3), which represents the total product of capital as a function of

capital with given amount of labour at L, level. When the labour increases from L, to L,, the total product of capital would also increase at all values of capital. This is illustrated with the help of a hypothetical example in Table J.

197

THEORY OF PRODUCTION

Table 1 Total Product in Output Units

Total Product Units

K=10

| ee

fox I2

K=13

ke=5

100

106

it

IS

Loe 6

115

ted

128

[33

les 7

128

136

143

149

b=

140

149

boy

164

It can be seen from Table / that both capital and labour contribute positively to production. All the columns in 7able / represent the total product of the labour schedule at the respective constant levels of capital. Similarly, all the rows show the total product of the capital schedule at the respective constant levels of labour. These schedules can be converted into the corresponding curves as given in Figs. I(a) and /(b). As it can be readily

Figure 1(a)

198

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

TPx

TP AL,) TPL.) TPAL,) TP (Ly)

Figure 1(b) observed, both these 7otal Product (TP) curves are upward sloping, implying that given all other things constant, including the amount of the other factors used, the total product would rise as we increase the employment of the factor in question. The slope of the total product curve represents the marginal product (MP) of the factor. The marginal product of labour (MP,) is the partial rate at which total product would change with respect to labour when all other things are held constant. Thus, MP, represents additions made to the total product when labour is increased by one unit, all other things remaining the same. Similarly, the marginal product of capital can be measured as the additions made to the total product when capital is increased by one unit, all other things remaining the same. As it can be seen from Table J, the marginal products of labour and capital can be calculated along the columns and the rows, respectively. Table 2 presents the marginal product of labour and the marginal product of capital calculated from the data given in the Table 1.

199

THEORY OF PRODUCTION

Table 2 Marginal Product of Labour and Capital

Marginal Product of Labour When Units

A=

7G

' Gaoe wi

aoe bg

Kes

in

is

16

Vs

|

18

Lo

ie

14

15

16

N= 3

12

3

14

Lb

_ Marginal Product of Capital When

cai Kile #2 K =13 SOURCE:

L=5 6 : 4

pe 7 6 5

Pay 8 7 :

L=8 3 8 7

Table 1 above.

The marginal product of the factor is calculated as the ratio of the change in the total product and in the amount of the factor employed, keeping all other things the same.’ Thus,

(4)

MP, =A TP,/AL=(TP1)y—(TPi)n -1

(5)

Vi ea

Iie / ih

= (IP) mila t

where rn is the number of units of the variable factor.

It can be seen

from the J7able 2 that MP, and MP, are both

positive, which essentially means that both the factors contribute

positively to the production of the commodity. Moreover, the table also shows that the marginal product of one factor depends not only on the employment of that factor but also on the employment of the other factor. Thus, when capital is held constant at K = 10

units, the MP, changes as the employment of labour increases. This is true for all different levels of capital. Similarly, the MP, of | It may be noted that MP, being the rate of change in the 7P, is measured in terms of the same unit as output per unit change in labour. In order to emphasize this aspect, it is often referred to as the Marginal Physical Product (MPP) of Labour. Similarly we may also define MPP...

200

MICROECONOMICS

FOR MANAGEMENT STUDENTS

6 units of labour also changes when capital changes from K = 10 units to K = 11 units or K = 12 units. This is again true for the different levels of labour. The same holds in the case of MP,. It may further be observed from the Table 2 that on margin, both labour and capital contribute positively to the marginal productivity of the other factor. Thus when K changes from 10 to 11 units, the MP, increases at all levels of the labour employment.

Similarly, when L changes from 6 to 7 units, the MP, increases for

all levels of capital employment. This implies that the level of the MP, schedule depends on the level of the capital employment;

and that the level of the MP, schedule depends on the level of the labour employment. Figs. 2(a) and 2(b) show the MP, and MP, curves. These curves are always drawn with respect to the factor whose marginal product is being measured. The behaviour of the marginal product curve is generally governed by what may be termed the Law of Diminishing Marginal Returns. According to this law, the marginal product of a factor would eventually start declining as greater and greater quantity of the factor is employed, all other things remaining the same. This is basically an empirical assumption about production technology that is so popular as to be referred to as a law. It is intuitively appealing too. For this ‘law’ to hold it is required that all units of the factor (say, labour) are identical in all respects. Moreover, it is also necessary that the technology, as given by the production function, remains the same. The employment of other factors (such as capital, etc.) should also remain the same. Under such circumstances, if we employ more and more labour for the same

unit of time, total production will increase but at a

diminishing rate. Production will increase because additional labour does make a positive contribution. In other words, the MP, is positive. But these consecutive additions to the total product cannot be sustained at the same rate as before because now each worker has to work with less capital than earlier. The additional output obtainable keeps falling as we go on increasing labour with the same amount of capital employed in production. The output produced by a firm is an objective reality and hence can be measured in terms of well defined units like a litre, a metre, a kilogramme, etc. In fact, most outputs can be cardinally meas-

ured. This property makes it meaningful to consider the concept

201

THEORY OF PRODUCTION MP,

MP, (K))

Figure 2(a)

Figure 2(b)

202

MICROECONOMICS

FOR MANAGEMENT STUDENTS

of the average productivity of a factor. The average product (AP) of a factor, say labour, can be defined as the ratio of the total

product and the number of units of labour employed (6)

AP, =

TP, /jrand

AP

TP es K.

The average product usually differs from the marginal product because the former (i.e. AP) distributes the total product equally among all the factor units employed to produce the output, whereas the latter (i.e. MP) considers the contribution of the additional or marginal factor units in the production. If the MP > AP, the average product would have a tendency to increase as the employment of the factor increases. This is because the marginal factor unit contributes on an average more to production than that contributed by the existing factor units. By the same logic, when MP < AP, the average product tends to rise as we withdraw a factor unit from employment. The average product would not change only when AP = MP. It is important to note that all these interrelationships between the marginal and average productivity of a factor hold only when all other things remain the same. If, for instance, some

other factor, say capital, changes from a K, to K, level, all the three aggregates of 7P,, AP,, and MP, are likely to change without any change in the labour units employed. Fig. 5 clearly illustrates the relationship between the marginal and average product of a factor (say, labour) when another factor (say, capital) remains constant at the given level (say, K=K,). The figure shows that the average product of labour rises initially,

reaches a maximum point when L = L, and falls thereafter as the labour units in employment increase. The marginal product of the labour curve lies above the AP, curve when AP, is rising. It lies

below the AP, curve when AP, is falling. And the MP, curve intersects the AP, curve at the point where AP, is at its maximum (i.e. when L=L,). This nature of the interrelationship between AP, and MP, remains the same when we consider the relationship between MP, and AP, or, for that matter, between any marginal

and average values. The behaviour of AP, with regard to labour units, as shown in

Fig. 3 is consistent with the Law of Diminishing Returns in average terms. The law, when stated in terms of average quantities, would

203

THEORY OF PRODUCTION

MP, & AP;

AP, MP,

Figure 3 imply that all other things given, as more and more units of a factor are employed, the average product would rise initially but would eventually begin declining. Thus, when we have 100 acres of land with four ploughs, the average product of labour is most likely to increase initially as we employ more and more workers. However, as we go on increasing the employment of workers on the same piece of land with the given equipment, the average product of labour would ultimately start falling.

Iso-Quants and Factor Substitution

The production function, as given by the equation (1), can be rewritten as the function of only one factor by holding the other factor (or factors) constant. Equations (2) and (3) above represent such partial total product functions. It is also possible to generate another function from equation (1) by holding the total production

at a given level like X= X, or X=X,, etc. Thus, we may write the function as:

204

(7)

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Ag = £ 1K, 1)

This function shows various combinations of the two factors of production — K and L — to produce the given level of output X,. In geometric terms, the function generates a curve in the KOL space which can be termed the equal product curve or an /soQuant curve. Fig. 4 gives such Iso-Quant curves, e.g. curves X,, X,, X,, showing different levels of output of a good. Different points (like A, and B,) on a given iso-quant (X,) shows different combinations of capital and labour, and hence different techniques of producing the output with existing technology. The technique of production is defined in terms of the factor-proportion which is given by the slope of the straight lines like OA or OB. Such straight lines passing through the origin are known as the radius vectors. An important property of the radius vector is that different

points on it like A), A,, A, along OA; or B,, B,, B, along OB would show the same factor proportion (or capital-labour ratio). Thus a technique of production ts geometrically represented by a radius vector. Different radius vectors like OA and OB would represent different techniques of production. The one-to-one correspondence between the radius vectors and techniques of production makes it possible to distinguish between the labour intensive and the capital intensive techniques of production. A capital intensive technique of production is one that uses relatively more capital per unit of labour than the others. In the Fig. 4, OA represents a capital intensive technique when compared to OB. Similarly, a labour intensive technique of production is one that uses. more labour per unit of capital than the others. OB represents a labour

intensive technique when compared to OA in the Fig. 4. Any radius vector steeper than OA would represent a more capital intensive technique than OA and any radius vector flatter than OB would show a more labour intensive technique vis-a-vis OB. It may also be observed from Fig. 4 that, using the same technique of production as OA or OB, the farther away we go along the ray, we get higher and higher outputs. This is because A,

represents more of both the factors K and L than A,, which, in turn, shows more K and L, than A,, and so on. Since we have assumed that both the factors positively contribute to the production, a higher level of the iso-quant shows higher level of production.

THEORY OF PRODUCTION

205

Figure 4 Moreover, since both the factors — L and K — have positive marginal productivities, the iso-quant has to be downward sloping. It is not possible to maintain the output rate constant without reducing labour when capital is increased, and vice versa. More formally, we can derive the slope of the iso-quant as follows:

(8)

AX=MP,-AL+MP,:-AK=0

Since X = X, on the iso-quant.

i.e. A K/A L = (—) (MP, /MPx) = slope of the iso-quant with

X =X)

Since we assume that both the factors have positive marginal oroductivities, the slope of the iso-quant is negative, i.e. is downward sloping. The slope of the iso-quant is given by the negative yf the ratio of the marginal productivities of the two factors. The

atio of MP, to MP, is known as the Marginal Rate of Technical

substitution (MRTS). The iso-quant, as noted earlier, shows posibilities of factor substitution in the production of a commodity.

206

MICROECONOMICS

FOR MANAGEMENT STUDENTS

The rate at which one factor (say, labour) can be substituted for another factor (say, capital) in production is an important feature of the technology available for the production of a commodity. When the technology embodies infinite techniques of production so that the factors of production can be smoothly substituted for one another, the iso-quants are continuous as well as differenti-

able. However, it is not necessary that the production technology provides infinite techniques of production. When the technology provides only a finite number of techniques of production, the iso-quants are either discontinuous or with sharp kinks along the rays through the origin, representing the available techniques of production. Fig. 5 shows an iso-quant map of this type. In Fig. 5, the production technology has only three possible techniques of production, namely those given by the OA, OB, and OC radius vectors. Points A, along OA, B, along OB, and C, along OC yield the same level of output. These points, therefore, lie on the iso-quant X,. Similarly, points A,, B,, C, lie on the higher iso-

quant X,. Although, at the first sight, it may appear that the isoquants are discontinuous, rather than being discontinuous they

may only have kinks at points A,, B,, and C, or A,, B,, and C,. This K

Figure 5

THEORY OF PRODUCTION

207

is because, it may be possible to have a combination of the two

techniques of production such that we get the same level of output. Geometrically, it is possible to combine any two techniques of production linearly to get the same output along the straight line

joining points like A, and B,, or B, and C,, or A, and C,.” In Fig. 5, it is interesting to observe that at point B, we get the same level of output X, by combining the two techniques A and C linearly. Therefore, if technique B is efficient, it must use less of both the

inputs — K and L — to get the same level of X, output. Thus, production technique B is efficient if, and only if, the point B, producing X, level of output using technique B lies below point B, which is obtained along the straight line joining points A, and C, representing a linear combination of the two production techniques A and C. This concept of Technical Efficiency is very useful in determining the curvature of the iso-quant. It is clear from the Fig. 5 that the iso-quant joining the points on the radius vectors representing the efficient techniques must have the shape

of a convex curve to the origin” An iso-quant with infinite efficient techniques would imply a smooth convex curve to the origin. It implies a diminishing marginal rate of technical substitution of factors in production. Since MRTS is only a ratio of the marginal products of L and K, in several cases, diminishing MRTS and diminishing marginal returns would go together. However, there may be cases of the production technology where the two may not go together.4 Under such circumstances, it is the assumption of diminishing MRTS which is more workable than the assumption of diminishing marginal returns. Diminishing MRTS implies that the amount of K substituted by an additional unit of Z goes on falling as we keep increasing L so as to obtain the same level of output. This again is only a 2 The proof of this statement uses the principle of vector addition which is obtained by the diagonal of a parallelogram. Moreover, it also requires the assumption that the output changes proportionately with the quantity of each factor used in any technique of production. 3 In Fig. 5, the exact shape of the iso-quant X, would be a vertical line upto ed

a

and a horizontal line thereafter. Thus, there would be kinks at A, 5. and Cy Similarly, at Ay; Bb and C, for iso-quant Ky, 4 This is largely on account of the nature of the cross effects of the factors on each others’ marginal products.

208

MICROECONOMICS

FOR MANAGEMENT STUDENTS

desirable feature of production technology, though it is very likely to form part of most technologies permitting factor substitution. The concept of technical efficiency also rules out the intersection of any two iso-quants on the same iso-quant map. This is simply because consistency would require that any one factor combination cannot be credited with two different levels of output. The concept of technical efficiency and production function ensures that for the given combination

of factors, the maximum

possible output is considered; and that for the given output, the least quantities of the factors required along a given technique of production are considered. Finally, we may consider a special case of a production technology which has only one possible production technique. Since no other technique is possible, factor substitution is ruled out in the production of the commodity. The iso-quants in this extreme case take the form of the vertical and horizontal lines emanating from different points on a ray through the origin. Fig. 6 shows the

Figure 6

THEORY OF PRODUCTION

209

iso-quant map representing the production function with only one possible technique of production (i.e. OA). The vertical line through the A, point shows that if labour remains the same, more and more capital does not yield any additional output because no alternative techniques exist. Similarly, the horizontal line through

point A,, indicating more and more units of labour combined with the same level of capital, does not add any output. Both labour and capital have to be combined in the same proportion given by the radius vector OA. This, therefore, represents the case of the zero substitution between factors. If any other efficient technique of production is possible, factor substitution can take place.

Returns to Scale So far we have discussed various aspects of the production technology when either one factor or the output level remained constant. In order to measure the returns to a factor, we have to assume that the other factor remains constant. Similarly, to meas-

ure the degree of factor substitutability, we have to assume that the output remains the same ata given level. If we want to examine the aspect of the production technology that relates changes in the scale of production to the average resource productivities, we have to consider changes in both the factors and the output. When we consider simultaneous changes in both L and K, there are two possibilities: (i) the factor proportion (i.e. K/L), and hence the technique of production, remains the same; or (ii) the factor proportion, and hence the technique of production, changes. If ZL and K change in the same direction by the same proportion, then the K/L proportion does not change. On the other hand, when L and K change by different proportions, the K/L proportion, and hence the technique of production, changes. In order to examine the returns to the scale of production, it is necessary to keep expanding the inputs maintaining the same technique of production. It implies constancy of the K/L proportion that requires both Z and K to change in the same direction by the same proportion. In geometrical terms, it amounts to moving along a ray through origin (or a radius vector). Fig. 7

represents two iso-quants showing X, and X, levels of output

210

MICROECONOMICS

FOR MANAGEMENT STUDENTS

om 2 a Om

K

A

dg Eekgoth

sal

Pept Pl a

OA,


X,/Xp, there are Diminishing Returns to Scale (DRS)

We have so far defined the returns to scale only at one point like A, along one technique of production. If the technology is neutral to different techniques of production, the same property would hold also for those, e.g. along the ray OB in Fig. 7. Thus,

under Ji’, Ob, 705,
2/7 XgeH

the same type of returns to scale are shown at different points with different techniques of production, the production function may be interpreted to show the same type of returns to scale. However, at different points and different techniques, the degree

of the returns to scale may be different. If the degree has to remain the same at different points and different techniques, it is necessary that the production function given in equation (1) satisfies the following condition:

(9)

One

EEX

where A (pronounced lambda) represents a multiplier which can assume different values, and r is a constant showing the degree of returns to scale.° If r=1, it shows constant returns to scale because when all the

inputs change by a given percentage, the output also changes by the same percentage. If r> 1, it shows increasing returns to scale because a given percentage change in all inputs leads to a more than proportionate change in the output. Similarly, when r< 1, it shows diminishing returns to scale since a given proportionate change in all inputs leads to less than proportionate change in the output. Moreover, equation (9) would ensure that the same returns to scale are obtained for all possible techniques of ° Equation (9) is a standard way of defining a homogeneous function of the rth degree. Any homogeneous function can be used to depict a given type of return to scale. However, in equation (9), if we treat ‘r’ as a variable, we can get a production function showing changing returns to scale.

212

MICROECONOMICS

FOR MANAGERS

production and all possible points. This is because ‘r’ is treated as a constant.

From the very definition of the returns to scale, it becomes clear that production technology showing the increasing returns to scale implies increasing average productivities of both the factors — L and K — as the scale of production expands. This is because, under /RS, as K and L increase, say by 10 per cent, the output

expands

by

more

than

10

per

cent.

Therefore,

AP, and AP, would tend to increase as the scale of operation expands along the same technique of production. Similarly, under

DRS, AP, and AP, has a tendency to fall as the scale of production expands along the same technique of production. With CRS, AP, and AP, remain the same as output expands along the same technique of production. Thus, returns to scale and factor productivities have a close relationship, particularly when the technique of production does not change. The technology is said to be scale-neutral when there are constant returns to scale, leaving the factor productivities unchanged as the scale of production changes, other things, including the technique of production, remaining the same.

Euler’s Theorem and Law of Variable Proportion An important characteristic of a production function showing uniform returns to scale at all points as given in equation (9) is Euler’s Theorem, which is as follows: (10)

MP

Lop Mig

KARA,

where r is the constant representing the uniform degree of the returns to scale. This theorem has several interesting implications.® For instance, if all factors are paid according to their respective marginal products, the total product (X) would be

(i)

exhausted under CAS, i.e. when r= 1;

6 This theorem can be derived from the equations (1) and (9) by using simple partial differentiation. Those readers who are well-versed with partial differentiation may like to derive the equation (10) from equations (1), (4), ©), and (9).

THEORY OF PRODUCTION

213

(ii) more than exhausted under JRS, i.e. when r> 1; and (iii) less than exhausted under DRS, i.e. when r< 1. Because of such implications, the case of the technology snowing constant returns to scale has generated more interest. It is also because it represents an excellent point of reference since it shows the scale-neutrality of the technology. Under CRS, equation (10) becomes: (11)

MP,

-L+MP,-K=X

This would hold for all possible values of L and K when the technology uniformly shows ,CRS. Equation (11) has interesting implications particularly when it is combined with equations (9) and (10). In equation (9), we may take A= 1/L. Under CRS, r= 1. Hence we get

(12)

X/L = f (K/L). Similarly, X/K = f (L/K).

This implies that under CRS, the average productivities of the factors depend only on the factor proportions and not on the absolute values of the factors. By partially differentiating equation (9) or equation (12), we also find that the marginal productivities of the factors also depend exclusively on the factor proportions under the CRS. Thus, so long as the technique of production does not change, the marginal rate of technical substitution (MRTS) in production would not change. This implies that the iso-quants showing CRS technology are parallel along all rays through origin. Moreover, they are spaced along a given ray through origin in proportion to the outputs they show. As a result, under CRS, any one iso-quant contains all the information about the entire production function because the rest of the iso-quants can be derived merely as the radial projection of any one iso-quant. Another interesting result follows from equation (11) when we divide both the sides by X:

(13)

(MP, - L/X)+(MPx- K/X)=1 ie., EX/EL + EX/EK= 1.

In equation (13), the term (MP, - L/X) represents the partial elasticity of output with respect to labour, which is also known as labour elasticity of output. Similarly, (MP, -K/X) is the capital elasticity of output. Under CRS, the sum of all such factor elasticities of

214

MICROECONOMICS FOR MANAGERS

output is unity. Moreover, if the factors of production are paid according to their marginal product, the factor elasticity of output becomes the relative share of the factor in the output. Thus, under CRS, the factor elasticities of output can be estimated through the relative factor shares. Equation (11) also results in the following equation when we divide both the sides by L:

(14)

X/L = MP, + MPx (K/L). i.e.

(AP, = MP;,) = MPx (K/L).

As discussed above, under CRS, the average and marginal productivities of factors do not change so long as the factor proportion does not change. We therefore, get a very interesting interrelationship between the average and marginal productivities of the two factors at the given factor proportion (K/L). In accordance with equation (14), at a given proportion (K/L), if Gi)

AP, MP, i.e., when AP, is decreasing, MP, > 0. These results are symmetric for L and K. Therefore, if we replace

L by K and K by L, these results would still hold. Geometrically, these results can be shown in Fig. &. These results which are shown to be the implications of the assumption of the constant returns to scale technology are put together in the law of variable proportion. According to this law, as the labour per unit of capital increases, the average product of labour eventually starts declining and the average product of capital starts rising. Fig. & illustrates the symmetry of the law of variable proportion. On the horizontal axis, we measure L/K in the rightward direction

and hence K/L in the leftward direction. The whole range is divided into three distinct stages. Stage I represents the case when the average product of labour is increasing so that MP, > AP,, which is associated with the negative marginal product of capital.

This stage continues until AP, reaches its maximum where AP, = MP, and MP, =0. In the second stage of production, the AP, jis is downward sloping, implying that AP, > MP, and MP, > 0. The third stage of production is symmetrical to the first stage of

215

THEORY OF PRODUCTION

Stage Il

Stage |

AP &

AP

MP

a

|

Stage III

- MPx

:

|

| APx AP,

AP,

APx O

SS MP«

ig

MP,

Figure & the production because here AP, is rising as K/L increases and

MP, is negative. The second stage of production is the most relevant stage, because

it is only here that we

find that all the

average and marginal productivities are positive. In the first stage, labour is relatively so limited or capital relatively in such a large quantity that the average productivity of labour tends to rise as we increase its quantity relatively. At such a point, since capital is already in a relatively large quantity, any further increase in capital would lead to a reduction in the output, implying a negative MP,,.. The third stage is symmetrical to the first stage.

Choice of Technique by Firm Having discussed the nature of the technology available to a firm, we may now consider the question as to how, from various options, the firm decides about the technique of production to adopt.

The firm’s objective is taken to be maximization of profits. The cost of production is determined by the factor prices and the

216

MICROECONOMICS FOR MANAGERS

amount of factors used. Maximum profits can be achieved by maximizing the output given the costs; or by minimizing the costs for a given level of output. The choice of the optimum technique of production is based on all these considerations. If we assume that the firm is one of the many firms competing in the factor markets to hire the factor-services, we may infer that the firm by its own action would not be able to influence the factor-prices. It has to accept the factor-prices as given. Under such a situation, the firm’s budget constraint would be: (15)

pet,

Le

ae

where B is the firm’s overall budget or the total cost of production; and P, and P, are the prices (or rentals) per unit of labour and capital respectively which are taken as parameters. Fig. 9 shows the budget line of the firm which is also known as the J/so-Cost Curve since along the curve, the cost of production remains the same given the factor-prices. The intercepts of the line on the two axes are given by the ratio of the budget and the respective factor-price, as shown in the Fig. 9. It can also be seen from Fig. 9 that the iso-cost curve or the Iso-cline is a downward sloping straight line with the slope given K

O

By

B,

Pr

Py

Figure 9

THEORY OF PRODUCTION

217

by (—) P, /P,. Moreover, as the budgetary allocation available to the firm increases, the budget line shifts upward and parallel, as shown in the figure. Any point on the budget line shows the same total cost to the firm; whereas any point below the budget line shows lower cost to the firm. A point lying above the budget line has a higher cost at the given factor-prices. Thus, when the firm is interested in minimizing the cost of production for a given level of output, it should essentially choose the lowest iso-cost line on

a given iso-quant. This requires superimposing an iso-quant showing the desired quantity on the iso-cost lines. Fig. /0 depicts this. It can be seen from Fig. 10 that the same output (X,) can be produced with different techniques of production, such as A, B, C, D, E, etc. The corresponding points along the iso-quant for X, are given by A,, B,, C,, D,, and E,. Itis possible to evaluate the total cost of production at every point on the iso-quant curve once the factor-prices are given. Factor-prices determine the slope of the iso-cost lines. From every point indicating a factor combination, one or other of the iso-cost lines would pass which would give the total cost associated

with the factor combination.

Thus, for

instance, in Fig. 10, the iso-cost line showing TC, of total cost passes

Figure 10

218

MICROECONOMICS FOR MANAGERS

through the points A, and £, on the iso-quant X,. Similarly, points B, and D, on the iso-quant X, show the total cost of TC,. Point C, along the X, iso-quant shows the least cost combination of factors. At C,, the iso-quant X, and the iso-cost line TC, are tangential to one another. TC, is, therefore, the lowest possible cost of produc-

ing X,, given the technology and the factor prices. At C,, the slope of the iso-quant (i.e. MRTS) and the slope of the iso-cost line (i.e. relative factor prices) are the same. i.e.

(16)

MRTS = Relative factor prices

1.6/0

= Pp Pe MPPEMP

10,0:

MP.AP;=MPp/Py

This is the condition for the firm’s equilibrium in choosing the technique to produce the given level of output. It is known as the principle of Equi-Marginal Productivity of resources. Fig. 10 clearly shows that the same level of output (X,) can be produced by different techniques at different costs given the factor-prices. The firm will choose the technique which would minimize the total cost of producing the given level of output. Thus, in this case, technique C proves to be economically the most efficient technique of all the techniques. An economically efficient technique is that which produces the given level of output at the lowest cost when compared to all alternative techniques to produce the same level of output, ceteris paribus. The economically efficient technique must satisfy the law of equi-marginal productivity, as given in equation (16). The factor-prices are given in monetary terms, say rupees. Therefore, MP, /P, can be interpreted as the marginal product per rupee spent to hire labour; and MP, /P, as the marginal product per rupee spent to hire capital. If these two are not equal, the firm can gain by reallocating its scarce resources to hire greater quantity of the factor whose marginal product per rupee is higher. If the output remains the same at X,, it would imply shifting to a technique that is more intensive in the use of that particular factor vis-a-vis the other factor to produce the output. The equilibrium is reached only when the last rupee spent on each resource yields the same increase in the output. The reciprocal of the marginal

THEORY OF PRODUCTION

product of a equilibrium. unit increase output to the (17)

219

rupee worth of resource can be interpreted in the It would show additions made to the total cost per in the output. This is the Marginal Cost (MC) of the firm. Thus, we have: P,

/MP,, =

Px /MPx

=

MC.

So far we discussed the choice of technique by the firm within the framework of cost minimization for a given level of output. Now, we consider the question of the choice of technique under output maximization for a given level of cost. Fig. ]] shows line AB as the given iso-cost line showing the total cost of say, Rs 100. This iso-cost line is based on given factor-prices that determine its slope. It intersects various iso-quants showing different levels of output. In Fig. 17, we have shown one such iso-quant with the

output level at 5X units which is intersected by the iso-cost line AB at points C and D. At points C and D, the output is 5X and total cost is Rs 100. Moreover, at the point C, we find that the MRTS,

which shows the rate at which the technology allows the firm to substitute L for K to get the same level of output (i.e. 5X), is much higher than the relative price of ZL in terms of K in the market.

Figure 1]

220

MICROECONOMICS FOR MANAGERS

Therefore, the firm would find it profitable to substitute ZL for K in the production of X. It would move rightward along the iso-cost line AB by reallocating some of its expenditures from K to L. This will happen until the firm reaches the point E when it finds that the MRTS is just equal to the relative factor-prices. The point £ represents the firm’s equilibrium, because it cannot hope to produce anything more than 10X given its total cost of Rs 100, the technology, and the factor-prices. Thus, the condition for the firm’s

choice of technique of production remains the same as given in equation (16). Indeed, the two problems are not different,’ in the sense that Rs 100 of 7C can achieve the maximum output of 10X, and that the lowest cost of producing 10X is Rs 100.

Factor Prices and Choice of Technique The choice of the production technique by the firm is dependent on the factor-prices because the condition of equilibrium is that MRTS is equal to the relative factor-prices. If, for some reason, the

relative price of labour falls, labour becomes relatively cheaper and capital relatively costlier. Under such circumstances, we would expect that the firm would find it more profitable to shift to the more labour intensive and less capital intensive techniques of production. This can be shown, as in Fig. 12. In Fig. 12, RS is the total cost line with given factor-prices. The firm chooses technique A to produce X, output. Now the price of labour falls so that the new factor-price line becomes AT. At these factor-prices, the firm finds it most profitable to shift to technique B if the resource cost in nominal terms is kept constant so that the firm can change its output level from X, to X,. However, if the firm now also wants to produce the same old level of output X,, it may find it most profitable to produce it by using technique C. This is because, with new factor-price ratio, as given by the slope of RT, the lowest iso-cost line, which is just a tangent to the

iso-quant X,, is the line R’T”. It gives the optimal technique as C which is more labour intensive than technique B. This happens because of the scale effect. 7 In programming, they are known as the primal and the dual problems.

THEORY OF PRODUCTION

221

Figure 12 The movement from point C on the iso-quant X, to point B on the iso-quant X,, shows

the effect of expansion

of the scale of

production. Here we define the scale of production in terms of the output or the total cost of production.’® As the total cost of production increases with the factor-prices remaining constant, the way in which the firm’s equilibrium level of output changes is traced out by what is known as the expansion path. It represents a locus of the points of tangency between higher and higher parallel 8 It may be compared to the definition of the scale in the concept of the returns to scale. There the scale was defined in terms of output expansion along a given technique of production, which amounts to increasing all inputs in the same proportion. Thus, in returns to scale, scale is defined in terms of inputs. In the

present case, which is closely associated with the concept of the economies of scale, the scale is defined in terms of output.

a2n

MICROECONOMICS FOR MANAGERS

factor-price lines and higher and higher iso-quants. The expansion-path shows the long-term options before the firm given the factor-prices and technology. It, therefore, represents a planning curve for the firm. The expansion-path may be linear or non-linear. If the expansion-path is linear throughout from the origin, the scale of production is said to be neutral to the techniques and hence to the factors of production. However, the technology, as given by the production function, may be such that the scale of production may not be neutral to the factors of production or the techniques of production. Fig. 12 shows a case where the technology is such that higher scale of production favours more capital intensive techniques. This is because, in Fig. 72, when the equilibrium

shifted from point C to point B, with an increased output level

from X, to X,, the technique of production became more capital intensive from C to B with factor-prices, as given by the slope of RT and R’T’, remaining the same. We may also find a case where a higher scale of production at given factor-prices favours labour intensive techniques of production.

Chapter 9 Theory of Costs

In economic literature, the concept of costs of production has several meanings. These are explained below. X

Money Costs The most commonly used concept of costs is the money costs of production. It means the minimum amount of money that is required to be spent to produce a given level of output of a particular commodity. It is the expenditure, in terms of money, that must be

incurred to pay for the services of the factors of production and raw materials required to produce a given level of output. By factor services we mean the services of four productive factors broadly classified as land, labour, capital, and entrepreneur. All free gifts of nature,

such as soil, forests, rivers, mineral

deposits, etc., are described as land in economics. The payment made for the use of land as a factor is called rent. Human effort of all kinds, whether mental or manual, skilled or

unskilled, is broadly classified as labour. The payment made for the use of labour as a factor of production is called wages. (Salaries

are included in wages). All ‘produced means of production’ are described as capital in economics. Things like machines, equipment, tools, buildings, trucks, etc. are capital goods, in the sense that they are first pro-

duced with the help of land and labour and then used to facilitate production of consumption goods. In this sense, not only machinery but even the plough, the fishing net, or the hammer are also capital goods. In economics, capital means physical capital goods and not finance. The price paid for the use of capital as a factor of production is called interest. (Since capital goods can be created only out of savings, people must be paid ‘interest’ to induce them to save by postponing current consumption).

224

MICROECONOMICS FOR MANAGEMENT STUDENTS

As a factor of production, the entrepreneuris an economic agent who hires/purchases land, labour, and capital, promotes a business organization to combine their services in a chosen productive process, and takes the risk involved in producing and marketing the commodities produced by the other three factors of production. For performing these functions the entrepreneur is paid a reward which is called profit. As a factor payment, profit differs from the other three factor payments (namely, rent, wages, and interest) which are predetermined and constitute contractual obligations. Irrespective of how the enterprise which employs them performs, rent on land, wages of labour, and interest on

capital must be paid as a prior charge on revenue. In contrast, profit is a residual payment and therefore the entrepreneur’s profit can be positive, zero, or even negative (i.e. loss). Essentially, any production activity is concerned with transforming one thing into another. Something (say, cotton) which has no ‘utility’ in itself acquires utility by being transformed into another thing (say, cloth) through production activity. Similarly, a thing with less utility (say, cloth) acquires greater utility when transformed into another thing (say, a readymade shirt). Just as production activity adds ‘form’ utility to things, the activity of distribution (transport, storage, trade, etc.) adds ‘place’ and (time) utility to things that have already been produced. Cars and scooters are often produced far away from the cities, in the rural countryside where they have no or little utility. By transporting and stocking them in the cities the dealers add ‘place’ utility. Similarly, by purchasing blankets in summer, stocking them, and selling them

in winter, wholesale

and retail traders add ‘time’

utility. In other words, the four factors of production employed in production and distribution activities add utility or add value to raw and intermediate materials. The value added by land, labour,

capital, and entrepreneurship is therefore paid out to these factors in form of rent, wages, interest and profit. (Value added in production and/or distribution of any commodity is equal to value of output minus value of material inputs). The money costs of production depend upon: (a) factor prices, (b) the proportions in which the factors are combined (technique) (c) the quantities in which they are employed, (d) the quality of

THEORY OF COSTS

225

the factors and the efficiency with which they are used, and (e) scale of output.

Real Costs

The real costs of producing a particular commodity mean

the

physical quantities of land, labour, capital, entrepreneurship and

raw materials that are required to produce a given level of output of that commodity. The real costs of production are also described as ‘engineering costs’. Real costs depend upon the engineering or technical conditions of production and are not affected by factor prices. When economists and technocrats talk about efficiency and cost-effectiveness (cost-reduction effect) of a particular method of production, what they imply is the cost-saving in terms of real costs (i.e. reduction in the physica! quantity of one or more inputs required to produce a given level of output). On the other hand, when businessmen talk about efficiency or reduction in costs what they imply is the reduction in money costs or pecuniary costs attainable through a particular production method. Reduction in pecuniary costs can be achieved either by reduction in real (engineering) costs or by reduction in prices paid for one or more factors and/or material inputs. In economic analysis we have to distinguish between real and pecuniary economies of scale. Expansion in the scale of output by a particular firm may lead to a decline in its real costs as a consequence of increased specialization of labour and/or improved machinery. On the other hand, as the scale of output increases, the expanding firm, by emerging as a large buyer of one or more inputs, may be in a position to obtain one or more inputs at lower prices. As a result, the firm’s pecuniary costs decrease even though its real costs may have remained unchanged. From the viewpoint of private businessmen and the financial profitability of firms, what is most relevant is the pecuniary economies of scale. Whereas, from the point of the economy and growth of real national income, what is most relevant is the real economies of scale.

226

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Opportunity Cost The opportunity cost of a factor of production is the reward (or the value) that factor could have earned in the next best alternative occupation. For example, if a productive factor F can be used in industry A and in industry B, then the opportunity cost of employing F in industryA is the reward that F would have earned in industry B (i.e. the value that F would have contributed in

industry B). The concept of opportunity cost is derived from the fact that in any economy the available productive resources are scarce in relation to their demand, and so employing a resource in one particular use involves foregoing (sacrificing) the opportunity of employing that resource for an alternative use. Economists

distinguish

between

accounting and opportunity

costs because, from the point of view of accountants, the cost of

a productive factor means only the money that is actually paid out for using that factor. However, so far as the making of economic

decisions is concerned, what is relevant is the opportunity cost of a factor. Most economic decisions are concerned with allocating scarce resources among many alternative uses; if more of a given available resource is used for one purpose, the less of it will be available for another use. In deciding the value or contribution of a factor in one particular use the decision-maker must take into account the opportunity foregone in terms of what that resource could have earned or the value it could have produced in the

next best alternative use.

|

For instance, take the case of a farmer who owns a plot of land which he is using to produce cotton. According to the accountant, the cost of that land in cotton production is zero because the farmer, being the owner of that land, pays no rent. However, since the same land can also be used for growing wheat, the landowning cotton farmer could have earned rent on his land by leasing it out to a tenant-farmer for wheat production. The rent which the farmer could have earned for allowing his land to be used in wheat production is the opportunity cost of land in cotton production. When deciding whether cotton growing was economically viable

and profitable, the landowning farmer should take into consideration the opportunity cost of using land in cotton growing even

THOURY UF UOoTS

eek

though no rent is actually paid out to use the land for cotton production. The concept of opportunity cost plays a significant role in the theory of the firm when it is applied to the supply of entrepreneurship as a factor of production. As mentioned earlier, profit as a reward for entrepreneurship is a residual payment and not a pre-determined

contractual

payment like rent, wages, and inter-

est. In this sense, profit is not an item of cost and it can be even zero or negative. However, since the entrepreneur supplies an essential productive service, he must receive at least some minimum

positive reward; otherwise he will abandon his own busi-

ness and hire out his skills as an employee in someone else’s. This minimum positive profit that an entrepreneur must receive in order to remain in a particular business is called normal profit. Normal profit is the opportunity cost of an entrepreneur (in a particular line of business). It is the reward that he would have earned by taking up a job and offering his skills and expertise on payment of a (contractual) salary. Like rent, wages, and interest, normal profits must also be included in the total cost of production because it is the minimum price that must be paid for maintaining the supply of entrepreneurship. Suppose Mr X, who possesses a degree in engineering, starts and runs a precision casting business. He finds out that if he were to take up a job in a similar industry he would earn an annual salary of at least Rs 50,000. Rs 50,000 therefore represents

the opportunity cost of Mr X as an entrepreneur and hence the ‘normal profit’ he must earn from his business if he is to continue with it. As a rational economic agent, Mr X would close or sell his business and take up a job if the surplus of revenue after payment

of costs of land, labour, and capital in his business

is

less than Rs 50,000, for his business can be considered economically profitable only if this surplus exceeds the normal profit of Rs 50,000.

In American textbooks on microeconomics the concept of economic profit is used to signify profits that are in excess of normal profits. That is, actual profits minus normal profits is equal to eco-

nomic profits. British textbooks use the term ‘super-normal profits to convey the same meaning. Thus super-normal profits mean positive ‘economic’ profits, normal profits mean zero ‘economic’

228

MICROECONOMICS

FOR MANAGEMENT STUDENTS

profits, and sub-normal (less than normal) profits mean negative ‘economic’ profits. The concept of opportunity cost is also applicable to emphasize the fact that ‘owned’ or equity (financial) capital invested in a business is not cost-free. Although in the accounting sense the cost of ‘owned’ or equity capital is zero because no interest is actually paid out on such capital, the opportunity cost of owned or equity capital is positive. It is equal to the interest income that the owners of capital could have earned if they had lent out this capital against interest payment. While assessing the profitability of any enterprise, accountants take into consideration only the interest paid on debt-capital. But economists take into account even the opportunity cost of equity capital, i.e. imputed interest on equity or ‘owned’ capital. With regard to the opportunity cost of labour, it needs to be noted that the opportunity cost of a factor is zero if the supply of that factor is abundant in relation to its demand and, as a result,

alternative avenues to employ that resource are not available. It is in this sense that in general the opportunity cost of labour is zero in a labour-surplus, underdeveloped economy like India.

Private and Social Costs

Private costs are the opportunity costs of resources that are borne by the owners of an enterprise. Thus private costs have to be borne by only those persons/legal entities who are party to a particular economic decision. In contrast, social costs are the opportunity costs borne by a whole society or community. Thus social costs include not only the costs borne by the owners of a business but also the costs inflicted upon those who are not party to a particular economic decision. For instance, when a large irrigation dam is built by the government in a river valley, the private costs are only those costs incurred by the government agency for building the dam, canals, power station, etc. The social costs of such a dam, however, include even the costs of displacement and migration of the inhabitants of the submerged land. Similarly, when a fertilizer

THEORY OF COSTS

229

factory is located in the vicinity ofa city, the social costs of fertilizer production would include the costs of medical bills borne by the residents in the neighbourhood whose health is adversely affected by the air pollution.

Sunk Costs

Sunk costs signify the costs of those highly specialized resources/inputs which are such that after they are employed in a particular enterprise/organization they cannot be put to any alternative use. For instance, the cost of a specially designed machine installed by a firm is a sunk cost in the sense that the firm can neither sell nor lease it to any other party, or even use it for any alternative purpose. The firm has to bear the cost of that machine irrespective of whether it yields any return. In other words, sunk costs mean the costs of those inputs or resources whose opportunity costs are zero.

In economic theory costs of production are analysed in relation to changes in the output of a firm. The manner in which a firm can change its output depends upon whether the period under consideration is short or long. Certain factors of production, such as land and buildings, capital equipment

management,

administrative staff, and entrepreneur-

ship are difficult to change in a short period (i.e. one or two years). Hence employment of these factors remains unchanged when the output of a firm changes in the short period due to seasonal and temporary fluctuations in demand. These factors are called fixed factors, in the sense that their employment cannot be changed in response to variations in output over a short period. On the other hand, certain other productive factors like labour, power and fuel, and raw materials are such that a firm can increase

or decrease their use or purchase even during a short period. These factors are called variable factors, in the sense that their employment can be varied in response to short period variations in output.

Costs of fixed factors are called fixed costs and that of variable factors called variable costs. Fixed costs are also known as overhead costs.

230

MICROECONOMICS

FOR MANAGEMENT STUDENTS

The long period allows a firm enough time to change its output not only by changing the variable factors but also the fixed factors. Thus, in the long run all factors become variable. The distinction between fixed and variable factors is applicable only to short run costs because in the long run all costs can be varied to adjust to changes in output.

Costs of Production in the Short Run In the short run, total costs consist of total fixed costs and total

variable costs. The fixed factors, such as land, buildings, and capital equipment last for a long period and are used continuously in the production process. The costs of such fixed factors are therefore measured in terms of the annual allowance made for their depreciation (or wear and tear) and the expenditure on their maintenance. Thus total fixed costs of a firm include the following: (i) Salaries of managerial and administrative staff; (ii) maintenance and depreciation of machinery and equipment; (iii) maintenance and depreciation of land and buildings; (iv) the normal profit of the entrepreneur. Total variable costs include the following: (i) Costs of raw and intermediate materials; (ii) wages of labour directly involved in production; (iii) expenses on power, water supply, fuel, etc. Denoting total costs by 7C, total fixed cost by 7FC, and total variable cost by 7VC, we find that in the short run

TC =eiPC + IVC In Fig. 1, Total Fixed Cost (TFC) is graphically shown

by a

straight line parallel to the output axis.

Fig. 2 shows that the Total Variable Cost (TVC) has an inverse-S shape, which reflects the law of variable proportions of laws of returns to a variable factor. According to this law, when output is being increased by em-

ploying more of a variable factor like labour which is used in combination with given fixed factors like land and capital then in the initial stages of production the addition made to total output is proportionately higher than the increase in the variable factor.

THEORY OF COSTS

aol

it.

TEC

output

Figure |

TVC

Lve

output

Figure 2 Consequently, total variable costs increase proportionately less than the increase in total output. This situation continues till a point is reached when the proportionate increase in total output is the same as the addition to the variable factor. That is, the increase in total variable costs is of the same proportion as the increase in total output. Beyond this point, the increase in total output is proportionately less than the increase in variable factor.

232

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Hence total variable costs increase proportionately more than the increase in output. The following table illustrates the operation of the laws of returns to a variable factor. Table 1 Operation of Laws of Returns to a Variable Factor Assumptions: Fixed Factor:

10 machines; Wage: Rs 10 per

worker (fractions are rounded off)

:

sf

32 &8i Se 88

Coe ofte

>

ee

or ReeseConk a

a

Se PES

10

100

_

1000

_

1000

-

10

110

10

1100

10

1200

20

10

121

10

1210

10

1500

mae

10

133

10

1330

10

Lliz5

15

10

146

10

1460

10

1898

10

10

161

10

1610

10

2050

10

vei

10

1770

10

pAbe:

10

195

10

1950

10

2218

By adding the 7FC curve to 7VC curve we obtain the 7C curve of the firm in Fig. 3. Here at any output level the vertical distance between the 7C and 7VC curves is equal to the 7FC which is constant at all output levels. The Average Fixed Cost (AFC) is found by dividing the TFC by the quantity of output. AFC = TFC/Q where Q is the quantity of output.

Graphically, the AFC is a curve called a rectangular hyperbola which shows that the area of the rectangle corresponding to any point on the curve is the same at all the points. This area is equal to total fixed cost which is the same at all output levels.

oe

THEORY OF ‘COSTS TFC TVC

Te 7c TVC

TFC

tput outpu

O

Figure 5

AFC \

AFC output

O

Figure 4

Fig. 4 shows that as output increases, the AFC declines continuously, tending towards zero but never becoming zero because the 7FC is positive.

234

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

TVC

> output

Figure 5 The Average Variable Cost (AVC) is obtained by dividing TVC by the quantity of output. AVC = TVC/Q, where Q is the quantity of output. Graphically, the AVC at any particular level of output is derived from the slope of the line drawn from the origin to that point on the 7VC curve which corresponds to that level of output. For instance, in Fig. 5, the AVC at Q, is the slope of ray Oa, at Q, it is the slope of ray Ob, and so on. It is clear from Fig. 5 that the slope of a ray through the origin declines continuously until the ray becomes a tangent to the 7VC curve at the point corresponding to output Q,. To the right of this point the slope of rays through the origin starts increasing. Thus, the AVC curve falls initially as the employment of a variable factor like labour increases. It reaches a minimum when the fixed factors are operated optimally (i.e. when the combination of fixed and variable factors is at its optimum). The AVC curve rises beyond the optimum level of output. Fig. 6 shows that the AVC curve is U shaped. The Average Total Cost (ATC) is obtained by dividing the TC by the corresponding level of output.

235

THEORY OF COSTS

AVC AVC

- output

O

Q»

Q3

Figure 6 ATC = TC/Q = (TFC + TVC)/Q = AFC + AVC, where Q is the quantity of output. Graphically, the ATC curve is derived in the same way as the AVC curve. The ATC at any level of output is the slope of the

output Q,

Q,

Figure 7

236

MICROECONOMICS

FOR MANAGEMENT STUDENTS

ATC

ATC

O

Q\

Q»

Q3

output

Figure & straight line from the origin to the point on the 7C curve corresponding to the particular level of output. See Fig. 7. The shape of the AJC curve is also U-shaped. Initially the A7C declines and reaches its minimum point at output Q, where the fixed factors are optimally used. Subsequently, the A7C curve rises as output increases. See Fig. &. The U-shape of both the AVC and ATC curves reflects the operation of the law of variable proportions which states that eventually returns to a variable factor used in combination with one or more fixed factors tend to diminish.

Marginal Cost Marginal Cost (MC) is defined as the increment to total cost made by one additional unit of output. It measures that change in total cost that occurs when output changes by one unit. Thus MC is

given by the ratio of the change in 7C to the change in output. Mathematically, the marginal cost is the first derivative of the TC function with regard to output.

ot

THEORY OF COSTS

When rn units of output are being produced, the MC of the n-th unit is given by the difference between total costs of n and total cost of (n-1) units. MC (n) = TC (n) — TC (n-1) To be precise, if ATC and AQ show changes in total cost and output, then MC = ATC/AQ. Using simple calculus,

Graphically, the MC is the slope of the 7C curve at any given level of output. That is, MC is the slope of the tangent to the iG curve at the point corresponding to a particular output level. Since the TC curve has an inverse S-shape, the MC curve will be U-shaped. From Fig. 9 it can be seen that the slope of the tangent to the TC curve declines gradually until it becomes parallel to the X axis. Its slope is equal to minimum at this point. After this point the MC curve starts rising. Fig. 10 shows that the MC curve is U-shaped. It can now be concluded that according to accepted microeconomic theory, the three short run cost curves showing the LC

Output O

Q)

Figure 9

Q»

Q3

Qs

238

MICROECONOMICS FOR MANAGEMENT STUDENTS

MC

MC

output

O

Q

Q»

Oy. Figure 10

average variable cost (AVC), the short run average total cost (ATC), and the marginal cost (MC) are U-shaped and their shape reflects the operation of the laws of returns to a variable factor. In the short run, with a fixed plant size determined by the size of fixed factors such as machinery and building, there is a phase

of increasing returns followed by a phase of decreasing returns. The minimum point of each of these three cost curves is reached at different levels of output.

Relationship between SATC and AVC We have already seen that, in the short run, total cost is the sum

of total fixed cost and total variable cost. Hence, at any given output SATC = AFC + AVC. The minimum point of the SATC curve occurs to the right of the minimum point of the AVC curve. This is because SATC includes AFC, and the latter falls continuously with increase in output. After the AVC starts rising, the rise in AVC is offset by a

fall in the AFC over a certain range of output. SATC therefore

Paty

THEORY OF COSTS SATC AVC AFC

output Figure 11 continues to fall over that range of output despite the increase in AVC. When the rise in AVC eventually becomes greater than the fall in the AFC, the SATC also starts increasing. The AVC curve approaches the SATC curve asymptotically as output increases. That is, the vertical distance between the AVC and SATC tends to

decrease as output increases because the fall in AFC becomes smaller and smaller. However, since AFC never becomes zero, the SATC and AVC curves never merge.

In Fig.

1] the minimum AVC is reached at output Q, while the

SATC is at its minimum at higher output Q,. Between Q, and Q, the fall in AFC is more than the rise in AVC so that the SATC

continues to fall. Beyond Q, the increases in AVC exceeds the fall in AFC so that SATC rises.

Relationship between Marginal Cost (MC) and Average Cost (AC) We have already seen that by definition if the changes in total cost and output are indicated by ATC and AQ then:

240

MICROECONOMICS

FOR MANAGEMENT STUDENTS

MC = ATC/AQ Now

T7Cis=ACxQ

“. MC = A(AC x Q)/AQ “, MC = (AC xAQ + Q AAC)/AQ’

-. MC =(ACxA AQ+ (Q x AAC)/AQ “, MC =AC+(Q x AAC)/AQ’ *. (MC - AC) = (Q x AAC)/AQ. Here AAC/AQ is = Slope of the AC curve. “. (MC — AC) = Qx (Slope of AC curve). (See footnote to understand the above derivation.) Here Q being output is always positive, but the slope of the AC curve can be negative, zero, or positive because it is U-shaped.

When AC is falling the slope is negative, when it is at its minimum the slope is zero, and when it is rising the slope is positive. The relationship between MC and AC can also be shown by using simple calculus. We have seen that by definition, MC = dTC/dQ But TC = (AC) (Q). Hence MC =d [(AC)(Q)|/dQ. Thus we get,

MC = dTC/dQ =d [(AC)(Q)]/dQ = Q(dAC/dQ) + AC(dQ/dQ) ». MC = AC + Q(dAC/dQ). “ The change in the product of two variables, namely AC and Q, can be explained as follows: Let AC =X, O= Y and (AC) x (Q) =Z= (C. Then. Z=v"7

Now the change in Z from Z to (Z + A Z) is made up of the X~i16 (X+AX) and changein Y {trom--¥ to (Y+AT) ae AX)(Y+AY) (Zt ALZ)=(XVY+XAV+VYAX+AXAY). But XY = Zand Soth sides, we getA Z=XAY+YAX+AXAY. sipeced irda AX AYcan be treated as negligible. ‘-AZS=KA eA

change in X from Hence, 2+i2= deducting Z from ga A Y are small,

THEORY OF COSTS

241

Here dAC/dQ is the slope of the U-shaped AC curve which is negative zero or positive.

“. (MC — AC) = Q (Slope of AC curve). Here

Qis>0OQand AC is 0.

Hence, MC-—AC< 0, i.e. MC < AC when slope of AC is negative MC-AC> 0, i.e. MC > AC when slope of AC is positive MC—AC= 0, i.e. MC = AC when slope of AC is zero. From

the above

relationship

between

MC

and AC, we

can

conclude that the relationship between the U-shaped AC curve and U-shaped MC curve can be stated as follows:

(a) (b) (c)

When < AC. When When MC =

AC is falling (slope of AC is negative), then MC is AC is rising (slope is positive), then MC is > AC. slope of AC is zero and AC is at its minimum then AG.

The MC curve should be situated below the AC curve up to the minimum point of the AC curve because AC is falling. But the MC curve should be situated above the AC curve after the minimum point of the AC curve because AC is rising. And MC is equal to AC at the minimum point of the AC curve. Hence the MC curve should come from below to cut the AC curve at the minimum point of the latter so that MC is below AC before the point of intersection and MC is above AC after this point. The U-shaped MC curve intersects the U-shaped AC curve from below at the minimum point of the AC curve. Hence the minimum point of MC curve is situated to the left of the minimum point of AC curve. That is, the MC curve reaches its minimum point at an output level that is lower than the output at which the AC curve is at its minimum. Note that when AC is constant for all levels of output, MC is also constant and equal to AC at all output levels. That is, the MC curve coincides with AC curve when the latter is horizontal and parallel to the X-axis. The relationship between marginal cost (MC) and average cost (AC) explained above is equally applicable to average variable

cost (AVC), short run average total cost (SATC), and long run

242

MICROECONOMICS

FOR MANAGEMENT STUDENTS

average total cost (LATC). (Since by definition the total fixed cost remains fixed and unchanged as output changes, the concept of marginal cost is not applicable at all in relation to total fixed cost). Note that the output at which MC is at its minimum is lower than the output at which AVC is at its minimum. Fig. 12 shows the relationship between MC, AVC and SATC curves.

SATC AVC MC AFC

output

O

Q;

Q2

Q3

Figure 12 It needs to be noted here that the relationship which exists between average and marginal cost holds good for the average and marginal values of any variable like utility, productivity, etc.

Relationship between Factor Productivity and Costs The Average (Physical) Productivity of any factor of production measures the number of units of output produced per unit of that factor. That is, average productivity of any factor is equal to the ratio of total output to total number of units of that factor used in producing that quantity of output.

243

THEORY. QF) COSTS

If Q is total output and L is the number of units of factor L employed to produce Q, then average productivity (AP) of L is equal to Q/L. The Marginal (Physical) Productivity of any factor of production is the addition to total output made by one additional unit of that factor. That is, marginal productivity (MP) of a factor is equal to the ratio of the change in output to the change in employment of that factor. If A shows change in Q and L then AQ/AL = MP of L. Due to the operation of the laws of returns, when the employment of a variable factor like labour increases and factors such as machinery (or land in farming) remain fixed, then the marginal and average productivity of the variable factor increase initially, reach the maximum level and then decrease. Thus both the AP and MP curves are inverse U-shaped and MP curve intersects AP curve at the maximum AP.

Relationship between Cost and Productivity Let

W =Price of factor L L = Units of factor L employed TVC = Total Variabie Cost AVC = Average Variable Cost MC = Marginal Cost

We can then explain the relation between productivity of factor L and costs as follows:

TVC=WxL;

AVC = TVC/Q and MC = ATVC/AQ

MC = (Wx AL)/A OQ = Wx (AL/AQ) But AL/AQ is = 1/(A Q/AL) = 1/MP “MC = W/MP And AVC=(WxL)/Q = Wx (L/Q)

Bat

£70=1/(Q0/L) = V/AP

“, AVC = W/AP (Note that Wis assumed to remain constant) The relationships is at its minimum

AVC = W/AP and MC = W/MP show that MC

when MP is at its maximum,

and AVC is at its

244

MICROECONOMICS

O

bj

FOR MANAGEMENT STUDENTS

Lo

Labour Employed

Figure 13(a)

output

Figure 13(b) minimum when AP is at its maximum. maximum, AP = MP.

Also, when AP is at its

Figs. 13(a) and 13(b) show that the optimum output Q, at which AVC is at its minimum and MC = AVC is arrived at by employing L, units of factor L where its AP is at its maximum and MP = AP.

THEORY OF COSTS

245

Costs of Production in the Long Run In the long run, a firm has enough

time to change

output by

changing all the factors of production and so all factors are considered variable in the long run.

We assume that firms being rational producers, any output that they produce is produced most efficiently under given conditions. While in the long run the given conditions are technology and factor prices, in the short run, the firm has to face the additional constraint of the given size of fixed factors. The long run average cost curve (LAC) is derived from several short run average total cost curves (SACs). Each point on the LAC curve corresponds to a point on a SAC curve which is at a tangent to the LAC curve at that point. The SAC curve cannot be located below the LAC at any output level because any firm faces many more constraints in the short than in the long run. Whatever it can do to reduce costs in the short run it can always do in the long run because in the long period a firm can change every factor. However, all the the changes that a firm can make to lower the LAC are not possible to make in the short run, e.g. it cannot change the fixed factors. Hence for any rational producer the SAC for any output is bound to be higher than the LAC for that output. But since the long run and short run are not independent, there is at least one point which is common to both: the point at which SAC is at a tangent to and equal to LAC. Derivation of LAC from SAC Curves

Assume that the existing technology is such that the firm can produce with plants of three sizes, namely a small plant, a medium plant, and a large plant. Let the short run average (total) cost curves of the small, medium, and large plant be denoted by

SAC,, SAC, and SAC,. This is shown in Fig. 14. Suppose the firm starts production with the small plant. As demand increases, it will produce at lower average cost up to level Q, on SAC,. Beyond that point average cost start increasing

on SAC,. When the firm’s demand reaches level Q,, the firm can either continue to produce with the small plant at average costs = A or it can install a medium size plant and produce Q, at same average cost. If the firm expects demand to expand permanently

246

MICROECONOMICS

FOR MANAGEMENT STUDENTS

SAC

O

Qe oc Qe

Qs

Q4 Qs

output

Figure 14 beyond output Q, it will install the medium size plant. With a medium size plant the average cost will be lower if the demand increases to Q,. Output Q, can be produced even with the smaller plant but only at average cost = B, while with the medium size plant, the same output can be produced with lower average cost = D which is lower than B. Similarly, when the firm reaches the output level of Q, the firm has to decide whether to install a large plant or continue with the medium size plant. Both give same average cost E for Q,. If the demand is expected in the long run to expand beyond Q, the large plant is profitable because the output level Q. is produced at lower average cost = G, in comparison to the medium size plant. For the medium size plant, output Q, has average cost = F which is higher than G. Thus, considering that the firm in the long run will increase its scale of output or size of plant as demand for its product increases

247

THEORY OF COSTS

and output rises from Q, to Q, to Q, to Q, to Q,, its LAC will change along the curve shown by points abADeEGH in Fig. 14. The previous assumption of 3 discrete sizes of plants can be

relaxed by assuming that the available technology includes many plant sizes, each suitable for certain level of output.

The limiting case of an infinite number of plant sizes will lead to a continuous LAC curve, the ‘planning curve’, of the firm. Each point of this curve shows the best or optimal method for producing

the corresponding level of output. Hence, the ZAC curve is the locus of points denoting the least cost of producing the corresponding output. The firm chooses that short run plant size which allows it to produce the anticipated long run output at the least possible cost. The LAC curve is U-shaped and as it envelops the SAC curves it is often called the ‘envelope curve . In Fig. 15 each point on the LAC curve is a point of tangency with the corresponding SAC curve. The point of tangency must occur at the falling part of the SAC curves for points lying to the left of the minimum point of the LAC. Since LAC is falling, the slope of the LAC is negative up to the optimum scale of output Q,. Hence the slope of SAC curves for outputs below Q, must

SAC ta

GAL aes.

SAC} SAC)

| O

Q,

output Qo

Figure 15

Qs

248

MICROECONOMICS FOR MANAGEMENT STUDENTS

also be negative because at the tangency points the SAC and LAC curves have the same or negative slope. Thus, on the falling part of the LAC, each of the plant sizes is not worked to optimum or ‘full’ capacity. That is, the level of output does not correspond to minimum SAC (or SATC). (Note that ‘Full’ capacity, according to the economists, means optimum capacity, whereas for the business people, full capacity often means capacity of maximum possible output). On the rising part of the LAC the points of tangency between SACs and LAC are located on the rising parts of each SAC. At output levels higher than the optimum level OQ, the LAC is rising and its slope is positive and so at points of tangency with SACs the slope of SAC is also positive. In other words, on the rising part of LAC with output above the optimum scale, each plant size is over-worked. Itis only at the minimum point of LAC that the SACis at a tangent to LAC at the minimum SAC. Hence at the optimum scale or optimum plant size, the plant is optimally used.

Returns to Scale and Economies of Scale

In order to understand the significance of LAC, it is necessary to understand the distinction between returns to a variable factor and returns to scale. Economies of scale means reduction in long run average costs (LAC) brought about by increase in the scale of operations, i.e. increase in the size of the plant. Diseconomies of scale mean rise in LAC brought about by increase in plant size. The U-shape of SAC shows laws of returns to a variable factor in the sense that it shows how AVC and SATC fall initially, reach the minimum point, and then ultimately rise as the /evel of output increases on account of an Increase in the employment of a variable factor used in combination with the given fixed factors. That is, changes in average and marginal costs occur as output varies within the same size of a plant. The U-shape of LAC, on the other hand, reflects the laws of

returns to scale according to which the average costs in the long run decrease as employment of all factors varies so that the plant

249

THEORY OF COSTS

size increases. Returns to scale mean variations in costs arising

from variations in the scale of operations, i.e. changes in plant size or productive capacity arising from variations in all the factors of production. The traditional theory of the firm assumes that economies of scale, i.e. decreasing LAC, A exist only upto a certain size of plant, which is known as the optimum pliant size, because with such a

O

Q;

Qo

Q3

Figure 16 SAC? EAC

SAC]

SAC3

SAC)

SACy LAC

output

O

Q;

Qs

Q3

Figure 17

Q%4

250

MICROECONOMICS

FOR MANAGEMENT STUDENTS

size the LAC is at its minimum and economies of scale are fully exploited. If the plant size or scale of operations increases beyond the optimum size there are diseconomies of scale arising from managerial inefficiency. A very large plant leads to difficulties in coordination, overworked managers, bureaucracy, and inefficien-

cy in the decision-making process. This in turn leads to upward turning of the LAC curve, i.e. diminishing returns to scale. The LAC is a straight line parallel to the X-axis when returns to scale are constant. Here, each SAC is a tangent to the ZAC at its minimum. The ZAC can also be L-shaped, falling up to a certain scale and then remaining constant. See Figs. 16 and 17.

Long-run Marginal Cost (LMC) The LMC is derived from several SMC curves. But it does not envelop them. The LMC is formed from points of intersection of the SMC curves with vertical lines drawn to the X-axis from points of tangency SAC curves and the LAC curve (see Fig. 18). The LMC must be equal to SMC for output Q, at which the

corresponding SAC, is at a tangent to the LAC at point A. This fact can be explained as follows in Fig. 18: For levels of output to the right of the tangency point A, the SAC > LAC. But at the tangency point, SAC = LAC. The movement

from pointA to the point B on SAC, by raising output above Q, is a movement from a position of equality of SAC and LAC to a position of inequality such that SAC, is > LAC. Hence the change (rise) in total cost, i.e. MC must be larger for the short run curve than the change (rise) for long run curve. Thus SMC must be

greater than LMC for an output higher than Q,. For a decrease in output below Q, (say, Q7) the SAC > LAC, ie. the movement from equality (at Q,) to inequality between SAC and LAC brought about by lowering the output must be such that the change in the total cost or MC must be smaller for the short run curve than change in the total cost for the long run curve. Thus SMC is smaller than LMC for output lower than Q.. If a vertical

line is drawn from point A to Q, on the X-axis, then we find that for output greater than Q,, LMC is less than SMC, and for output

below Q,, LMC is greater than SMC. Hence SMC and LMC must be

251

THEORY OF COSTS

SAC f SMC LAC LMC LAC

mate

SMC, .

LMC

oaee

Sac, l

Suc: ys

LAC

|

SAC%

cilia

O

Qi

Qn

Q;

output

Figure 18

equal at Q, output. The point at which this vertical line from A intersects the SMC curve must also be a point on the LMC curve. This shows that SMC = LMC at output Q,, which corresponds to

the tangency point If this procedure and LAC curves to section of the LMC

A between SAC and LAC. is repeated for all points of tangency of SAC the /eft of the minimum point of the LAC, the that lies below the LAC is obtained. At the

minimum point of LAC, the LMC intersects the LAC. To the right of output Q,,, the LMC curve lies above the LAC. At output Q, we have LMC = LAC. Also, as noted earlier, at optimum scale, minimum SAC = minimum LAC. But when SAC is at its minimum, SAC = SMC. Hence, SAC = SMC = LAC = LMC at optimum scale, Q_..

Chapter 10

Market Structure: Concepts and Determinants

Plant and Firm

A plant is a unit of production. It is a technical unit in the sense that its size is determined by technical or engineering considerations relating to production. The choice of size depends upon the technical economies of scale, i.e. reduction in the average technical cost by expanding the scale of production. A firmis a unit of organization; a managerial or decision-making unit. Its size is determined by managerial capacity or managerial economies of scale. A firm can be ‘a single plant’ or a ‘multi plant’ organization. When the size of firm exceeds the size of a single plant, it tends to become a multi plant firm. This happens when there are economies of scale in the managerial costs, i.e. expansion in the scale of output leads to reduction in the average cost of managing production, finance, and marketing.

Industry and Market Industry means a group of firms producing the same product or similar products, e.g. the textile or paper industry. Industry may also mean a group of firms using the same process or raw material,

e.g. the chemical industry, electronic industry, etc. Thus an industry is defined in terms of supply or production. On the other hand, a market means a group of firms supplying products that buyers consider to be substitutes. Thus, a market is defined from the point of view of buyers or demand. The terms ‘industry’ and ‘market’ do not always mean the same thing. For instance, the footwear market consists of products that are supplied by more than one industry, namely, leather, rubber, and

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

293

plastic. The packaging material market is supplied by firms from industries that are, in the industrial production data, classified as paper, glass, plastic, and aluminium. Similarly, an industry may be so described that its products cater to more than one market. The aluminium industry caters to markets for several products like utensils, packaging material, building material, etc. The same

applies to the Chemical industry.

|

Structure-Conduct-Performance Hypothesis The theory of the firm is to a large extent based on the ‘StructureConduct-—Performance’ hypothesis. Broadly speaking, the concept of market structure means the manner and the extent to which firms constituting a particular market are functionally interrelated with each other. Since firms operating in the same market are related to each other as competitors or rivals, the term structure of a market refers to the degree of competition prevailing in that market. The degree of competition is indicated by the power of an individual firm to influence or control the (market) price by changing its own output. This power varies inversely with the degree of competition. The greater the degree of competition, the less the power of an individual firm to control price by its own actions. When the market structure is such that competition is weak or non-existent, an individual firm wields considerable power to influence the price.

By ‘conduct of a firm’ we mean its behaviour as revealed by its policies and practices in terms of price, output, product-mix, sales promotion, etc. The performance of a firm means the outcomes or results of its conduct (behaviour) as indicated by profits,

costs (efficiency), capacity utilization, product quality, growth, innovations, etc.

In microeconomic theory there is an implicit hypothesis that the performance of a firm is determined by its conduct which, in turn, is determined by the structure of the market in which it is

operating. The performance and conduct of a firm situated in a keenly competitive market are qualitatively different from those of a firm operating in a market with weak or no competition.

254

MICROECONOMICS

FOR MANAGEMENT STUDENTS

The factors that determine the structure of a market are called structural variables. These are discussed below.

Number of Independent Sellers and Buyers When the number of independent sellers is large, the proportion of total supply controlled by any single seller is quite small and insignificant. The effect of a change in an individual firm’s output on the total market supply is thus negligible and it cannot therefore influence the price by its own actions. Thus the larger the number of independent firms in a market, the greater is the degree of competition and the degree of competition decreases as the number of firms becomes smaller. Similarly, competition among buyers and an individual buyer’s power to influence price depends upon the number of independent buyers. We can classify the number of firms into following four categories:

(i) i) (iii) (iv)

Large Few Two One

Degree of Seller-Concentration An individual firm’s power to influence market price depends not only upon the number of firms but also upon the proportion of total output of a product that the individual firm controls. When the total supply of a product is concentrated in the hands of a few big sellers, each of these sellers has power to influence price even

though the total number of firms may be large. On the other hand, no single firm has the power to influence price when the market shares of all firms are nearly equal and insignificant in relation to total supply — that is, not only is the number large but the degree of concentration in supply is very low. Thus, in addition to the number of firms, the degree of seller concentration also determines the market structure or degree of competition. For instance,

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

259

if the market for productX consists of twenty firms but the largest three control 75 per cent of total supply and if the market for product Yconsists of ten firms each controlling between 9 and 11 per cent of total production, then, other things being equal, the market for product X is less competitive than that for product Y. We classify the degree of seller concentration into the following four categories: (i) (ii) (ii) (iv)

Non-existent Low Medium High

For instance, under the Indian MRTP Act, a firm controlling 25 per cent or more of total supply of a specified product is considered to be dominant’ in the sense of possessing effective market power to control price. The same is the case under the alti-monopoly law in the UK. The Monopolies Inquiry Commission, which submitted its report in 1965 on concentration in Indian industry, considered the combined share of the largest three firms and delimited concentration as High, Medium, Low, and Non-existent if their combined percentage share was 75 or more, between 60 and 75, between 60 and 50 and below 50. Thus, the limits on the basis of which seller concentration is classified, i.e. as High, Medium, Low, or Non-existent, are rather arbitrary. The concentration ratio is the simplest and most widely used indicator of market (or seller) concentration. It is obtained by the sum of the percentage (relative) shares of the largest R number of firms in the total output or sales of an industry. The same can be also obtained by dividing the sum of output or sales of the largest R firms divided by total output or sales of all N firms in the industry. Here FR can be 1, 2, 3, 4, 5, 8, or 20. For example if R is = 4, then

four firm concentration is 80 per cent if the shares of the top 4 firms are 30, 25, 15 and 10 per cent, respectively.

Product Differentiation

Products competing in a market are considered differentiated if the buyers do not regard them as identical or homogeneous or

256

MICROECONOMICS FOR MANAGEMENT STUDENTS

perfectly substitutable. An individual firm has some, though very limited, power to influence (or hike) the price when the buyers consider its competing products as only slightly different, i.e. close though not perfect substitutes. For instance, the various brands of toilet soap like Lux, Liril, Cynthol, Rexona, are not considered

as perfectly substitutable by their users and hence if the seller of Lux raises its price slightly, some but not all the buyers will switch over to the one or other of competing brands. In the case of products like eggs or potatoes, on the other hand, the products sold by the competing sellers are homogeneous. Hence, even a slight increase in the price of such a product by an individual firm would reduce its demand to zero. Thus the power of an individual firm to control the price or the degree of competition also depends upon the substitutability of the products of competing firms. That is, it depends upon whether the buyers regard the products of competing sellers as perfect, close, or remote substitutes. The degree of substitutability or product differentiation is measured by the cross-elasticity of demand between two competing

products.! We classify the degree of product differentiation into the following four categories: (i) (ii)

Perfect substitutes or homogeneous products. Close substitutes or slight differentiation in form of different brands (e.g. brands of tooth-paste such as Colgate, Close-Up, Cibaca, Promise, etc.) (iii) Remote substitutes (e.g. radio and 7 aoe

(iv) No substitutes (e.g. salt, milk, etc.)

Condition of Entry By condition of entry we mean the difficulty or ease with which a new firm can enter an industry (or market). In the long run the number of firms and the degree of seller concentration in an | The cross-elasticity of demand between two products X and Y is shown by the ratio of the proportionate change in demand for productX to the proportionate change in price of product Y.

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

re

industry would depend upon whether the condition of entry is difficult or easy, i.e. whether the barriers to new entry are high or low. The number of sellers will tend to become large and the degree of concentration low if there is free entry. Moreover, the threat of competition from the potential new entrants will limit the power of an existing firm to control price. Thus, in the long run, the degree of competition in any market depends upon the condition of entry. The more difficult the condition of new entry, the

weaker the threat of potential competition from new entrants and the greater the power of existing firms to influence price. We classify the condition of entry into following three categories: (i) Free or easy entry Gi) Difficult entry (iii) Entry barred or impossible.

Factors Determining Condition of Entry Whether the condition of entry into an industry will be easy or difficult depends upon the following factors.

(a) Legal Barriers In many industries new entry is barred by legal restrictions. Almost all countries have a patent law to promote and protect the interests of inventors and innovators. The patent law provides that in the case of a patented product or process, no firm other than the

patent-holder or the firm holding the licence from the patent-holder is allowed to produce that product or make use of that process. This restricts entry of new firms in markets for patented products or those requiring the use of patented processes. Similarly, in most countries, for the supply of public utility services such as railways, telecommunication, water supply, electricity supply, etc., the law permits only one private or government enterprise. In these industries, the capital costs are very high but the operational (marginal) cost for one extra customer is very small. Hence there would be underutilization of fixed capital if many firms were allowed to serve the same market. This is the reason why public utilities are called natural monopolies.

258

MICROECONOMICS

In India, legal by laws such as tion of products industries or by

FOR MANAGEMENT STUDENTS

barriers to new entry in many markets are created the Industrial Licensing Regulation and Reservafor exclusive manufacturing by the small-scale the public sector.

(b) Initial Capital Cost In industries like steel, petrochemicals, cars, and trucks, etc., new

entry is quite difficult because the initial capital investment required is very large, which only a few entrepreneurs are capable of raising. Generally, industries producing basic inputs (coal, steel, power, etc.), heavy capital goods, and consumer durables require

a relatively substantial amount of initial capital. On the other hand, in industries producing light engineering goods, non-durable consumer goods, and personal services, the amount of initial investment involved is relatively small and hence new entry into such industries is easy.

(c) Vertical Integration When the existing producer of a final product is also producing a raw material or an intermediate good, it is called vertical integra-

tion. New entry becomes difficult in the case of such a product because the new entrant will either have to find an alternative (domestic or imported) source of supply of that input or he will have to depend upon his competitor for that input. For instance, if the existing producers of detergent also control the production of a chemical essential in its production, then a new detergent producer would have to buy that chemical at a high price from his competing detergent producers or would have to arrange for an alternative source of supply of that chemical.

(d) Optimum Scale of Production By optimum scale we mean that scale of output at which the long-run average cost of production is the minimum. In the long run a new entrant must aspire to produce at the optimum scale in order to face effective price competition. Now, if the optimum

scale of output for any product is quite large and if the total market

for itis relatively limited, then there is room only for a few existing producers to produce that product at the optimum scale. In such industries a new entrant is discouraged because he will have to

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

FS ba

be satisfied with sub-optimum production. For example, if the total potential demand for cars in India is 10 million, the optimum scale is 3 million cars, and the three established producers are already supplying 9 million cars, then a new entry into the Indian car industry would be difficult because the new entrant will have to produce a sub-optimum output of one million cars. The average costs of the fourth new entrant firm will be higher since his scale of output will be less than optimum.

(e) Product Differentiation New entry is difficult in an industry where products are highly differentiated from the buyer’s point of view and they are compulsively attached to the existing brands principally as a result of the high advertising expenditure of the existing firms. A new entrant would have to incur very large advertising expenditure to succeed in winning over buyers of the established brands to his brand.

For instance,

in the butter and chocolate

markets,

the

existing producers, Amul and Cadbury, enjoy a near monopoly due to strong differentiation. New entry in these markets is difficult although the initial investment is small and technology is simple. The new entrant may not wish to commit himself to very large advertising expenses, the effects of which are uncertain.

Classification of Determinants of Market Structure

For the purpose of theoretical analysis of different types of markets, we Classify each of the determinants of market structure into some broad categories which are summarized below.

(A) Number of independent firms The number of firms in an industry is classified into the following four categories: (i) Large, (ii) Few (ii) Two, and (iv) One

260

MICROECONOMICS FOR MANAGEMENT STUDENTS

(B) Degree of seller concentration The broad categories used for classifying the degree of concentration in any industry are: (i) Gi) (iii) (iv)

Non-existent Low Medium, and High

(C) Product differentiation The following four categories are generally used to classify the degree of product differentiation. (i)

Homogeneous or standardized products which are perfect substitutes. (ii) Slightly differentiated products sold under different brand names, i.e. close but not perfect substitutes. (iii) Different (and not differentiated) products satisfying the same want (e.g. tea and coffee, sugar and saccharin, radio and television, cooking gas and kerosene, etc.) — substitutes but not close substitutes. (iv) Products that have no substitutes (e.g. milk, salt, etc.)

(D) Condition of entry The three broad categories for classifying conditions of entry are:

(i) Free or easy entry. (ii) Entry difficult but not impossible. (iii) Entry impossible or prohibited by law.

Models of Market Structure

In microeconomic theory the behaviour (or conduct) and performance ofa firm is analysed in the context of five distinct theoretical models of market structure. The essential distinguishing characteristics of these models may be briefly summarized in terms of the following combinations of the four factors determining the market structure.

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

261

(1) Perfect or Pure Competition Number of

independent sellers Seller concentration Product differentiation

Large [A(i)]. Non-existent. All firms have insignificant and nearly equal market share [B(i)]. Non-existent. Homogeneous products

[C@)]. Condition of entry

Free or easy [D(i)].

(2) Monopolistic Competition Number of independent firms Seller concentration Product differentiation

Large [A(i)]. Non-existent or low [B(i) or B(ii)]. Slight. Products are close substitutes

[CGi)]. Condition of entry

Free or easy [D(i)].

(3) Oligopoly Number of

independent firms Seller concentration Product differentiation

Condition of entry

Few [A(i)}. Medium or high [B(iii) or B(iv)]. Non-existent or slight. Products may be homogeneous or close substitutes [C(i) or C(ii)]. Difficult [D(ii)]. Hence the number of firms remains small and concentration continues to be

medium or high even in the long run.

(4) Duopoly: A special Case of Oligopoly Number of

independent firms Product differentiation

Seller concentration

Condition of entry

Two [A(iii)]. Non-existent or slight. Products may be homogeneous or close substitutes. High [BCiv)]. Very difficult or impossible [D(ii)]. Hence two sellers continue to control the market even in the long run.

262

MICROECONOMICS

FOR MANAGEMENT STUDENTS

(5) Monopoly Number of sellers Seller concentration Product differentiation

Condition of entry

One [ACiv)]. Very high. One seller controlling almost 100 per cent of total supply [D(iv)]. The question of differentiation does not arise because competing firms and products do not exist. Products of related industries are only remote substitutes [C(iii)]. Barred or impossible. Legal bar in the case of public utilities. Hence the market remains in the control of one firm even in the long run [D(iii)].

In the conventional presentation of the theory of the firm the above five theoretical models of market structure are classified into three broad categories, namely, perfect competition, imperfect competition, and monopoly. Perfect or pure competition with a maximum degree of competition and monopoly with a total absence of competition are the two logically extreme cases. Using these two extreme cases — white and black — of market structure as standards for comparison, economists have developed models to analyse the behaviour and performance of the firm in the ‘grey’ area of ‘imperfect’ competition, including models of monopolistic competition, oligopoly and duopoly. This way of classifying market structure models is summarized in the chart below:

Market Structure Models |__,

Perfect Competition

Monopolistic Competition

Imperfect Competition

Oligopoly

Monopoly

»

“Duopoly

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

263

Alternative Classification of Market Structures

Another way of classifying the theoretical models of market structure is to use the criterion of Seller Interdependence. This criterion refers to the interdependence of the actions or behaviour of two or more competing firms. The behaviour of two or more firms is considered interdependent if one firm cannot act (change price or output) independently, without considering the possible reaction of its rival firms. Whether or not the rival firms will react to a price or output change by one firm depends upon the extent to which this action affects the rivals’ sales. If the number of firms is large and seller-concentration very low then one firm’s action will only negligibly affect the sales of its rivals. The latter will not therefore react or retaliate. But when there are only a few sellers (oligopoly), one firm’s action will significantly affect the sales of its rivals and hence the latter will in all probability react or retaliate. Thus, when the number of firms is large and concentration is low, any individual firm can behave independently without bothering about its rivals’ reactions. On the other hand, if the number of firms is small or seller concentration is high, the behaviour of competing firms is interdependent

in the sense that every firm must take into account the possible reaction of rivals to its actions. The degree of seller-interdependence is also indicated by what is described as ‘conjectural variation’, which measures the change in output or price of one firm (seller) arising or expected to arise in reaction to the change in the output or price of its rival (seller). Seller interdependence or conjectural variation is equal to zero when the number of firms in a market is large and seller concentration is non-existent or low, i.e. the market share of every firm is insignificant. Its value is numerically more than zero when the number of firms is small and/or seller concentration is medium | or high. On the basis of this criterion of seller interdependence we can classify the models of market structure into two broad categories of Atomistic and Non-atomistic competition. Atomistic Competition prevails in those types of market structure where each individual firm is like an atom; a minuscule part of the universe consisting of innumerable atoms. Like an atom the

264

MICROECONOMICS

FOR MANAGEMENT STUDENTS

individual firm is an insignificant supplier (or buyer) competing with a large number of equally insignificant suppliers (or buyers). Just as every atom moves in its own orbit independently, an individual firm acts independently without worrying about the reactions of its competitors. Any change in the output or price by one firm will have only an insignificant effect on the sales of competing firms, and hence competitors are not expected to and will not react to any action taken by it. Under atomistic competition, seller interdependence and conjectural variation between any two firms is zero or non-existent. Perfect competition and monopolistic competition are models of atomistic competition because in both these cases the number of sellers is large and concentration is low or non-existent. In contrast, non-atomistic competition prevails in those types of market structure where a firm is not like an atom, in the sense

that an individual firm is not insignificant. It controls an important proportion of total supply (or demand) and the market consists of only a small number of such suppliers (or buyers). Any individual firm’s action in the form of changes in output or price will have considerable effect on the sales of one or more of its rival firms who will react or retaliate by changing their own output or price. In non-atomistic markets no firm can act independently without worrying about the reactions of rivals. When altering output (or price) every firm has to conjecture or anticipate or assume the expected reaction of its rivals. Thus seller-interdependence or conjectural variation is not zero under non-atomistic competition. Oligopoly and Duopoly are models of non-atomistic competi-

tion because in both these cases the number of sellers (or buyers) is small and concentration is medium or high. Monopoly is an extreme case of non-atomistic competition.

The alternative classification of market structure based seller-interdependence is delineated in the chart below:

on

MARKET STRUCTURE: CONCEPTS AND DETERMINANTS

265

Market Structure

Atomistic Competition

Perfect or Pure Competition

Monopolistic Competition

Non-atomistic Competition

Oligopoly

Duopoly

Monopoly

Chapter 11 Equilibrium of the Firm

Business enterprises operate with the objective of making profits. We assume that the entrepreneurs who manage business firms are rational. Just as a rational consumer buys a commodity with a view to maximizing his satisfaction, a rational entrepreneur (i.e. his firm) will produce the output of a particular commodity in such a way that he earns maximum profits by selling that output. In short, the theory of the firm assumes that irrespective of the

market in which a firm may be operating, the objective of the firm is maximization of its profits and minimizing its losses when market demand is very depressed. The profits of a firm are equal to the difference between the total revenue that it earns by selling a given level of output and the total costs (money costs) it incurs in producing that level of output. Thus if 7R indicates total revenue and 7C shows total costs then Total Profits = (TR — TC). We have already seen the relationship between total costs and output. 3

Total Revenue

The total revenue of a firm is the total amount of money that the firm receives by selling a given quantity of output of its product. Total revenue is thus equal to the quantity of output sold, multiplied by the price that the firm charges for its product. (We assume that the sales volume of a firm is equal to the level of output produced).' Thus if Q is the output produced (and sold) by a firm and P is the price of the firm's product then 1 In the real world the sales of a firm are equal to current year’s production plus opening stock minus the closing stock of finished goods.

EQUILIBRIUM OF THE FIRM

267

TR=PXQ3 Clearly, the 7R of a firm is the same as the total amount that buyers are willing to expend on the firm’s product. The 7R of a firm at various levels of output is thus derived from the demand curve for the product of that individual firm. Note that the TR of a firm is not derived from the demand curve for the (whole) market.

Average Revenue (AR) The average revenue of a firm is equal to its total revenue at a given output level divided by that level of output. Thus, if AR shows the average revenue of a firm, 7R shows its total revenue, and Q and P show the output and price of that firm then,

AR = TR/Q =(PXQ)//Q

AR=P Thus average revenue is equal to or identical with price by definition. The AR curve showing different values of AR at various levels of output is same as the demand curve faced by an individual firm because the latter shows the various quantities of output of the firm’s product that buyers are willing to purchase at different prices.

Shape of an Individual Firm’s Demand Curve The relationship between AR and output Q is shown by the shape of an individual firm’s demand or AR curve. The AR curve can take three possible forms, namely (i) upward sloping or rising from left to right, (ii) downward sloping or falling from left to right, and (iii) horizontal or parallel to the X-axis. The first possibility is in most cases unlikely. Assuming that the income and tastes of buyers and prices of substitutes remain unchanged, an individual firm will not be able to sell larger and larger output by charging higher and higher price. Thus, it is

268

MICROECONOMICS

FOR MANAGEMENT STUDENTS

generally improbable that an individual firm’s demand or AR curve would be rising (except when the firm is selling a ‘snob’ good). Whether an individual firm’s demand or AR curve will be falling or horizontal will depend upon the type of market structure in which that firm is operating. If the market for the firm’s product is perfectly competitive, then an individual firm will have no control over its price. It being only a price-taker, the individual firm can sell as much as it wants

to at the given market (equilibrium) price. The firm does not have to lower its price in order to sell a larger output. Thus, under perfect competition, any individual firm’s demand or AR curve will be horizontal and parallel to the X-axis. Its AR or price will remain constant at OP even as output increases. See Fig. 1(a). Under imperfect competition or under monopoly the individual firm can influence or control the price by its own actions. It will have to charge a lower price if it wants to sell a larger output and AR or Price

AR = Price

output O

Figure I(a)

269

EQUILIBRIUM OF THE FIRM

AR

$$$

—————

> output

Figure 1(b) price will rise if the firm reduces output. Hence, an individual firm’s demand or AR curve will be downward sloping or falling under monopolistic competition, oligopoly, duopoly, and monopoly. See Fig. 1(6). In short, an individual

firm’s demand

or AR curve

is either

horizontal or is falling, but is not likely to be rising.

Marginal Revenue (MR) Marginal Revenue means the addition made to total revenue by selling one extra unit of output. Thus if 7R (NV) shows the total revenue for producing (and selling) NV units of output, 7R (NV + 1) shows total revenue for producing (V+ 1) units, and MR shows the marginal revenue, then

MR = TR (N+ 1) — TR (N). In general, if ATR shows change in 7k, and AQ shows change in output, then

270

MICROECONOMICS

FOR MANAGEMENT STUDENTS

MR=ATR/A

Q

or, in terms of simple calculus,

MR = dTR/dQ. From the meaning of MR it is clear that MR is equal to the sope of TR curve.

Relation between AR and MR We have seen that

MR =A TR/AQ MR= A(PxQ)/AQ =A (ARxQ)/AQ = AR(A Q/A Q) + Q(AAR/A Q) MR = AR + QA ARZA Q) “. (MR — AR) = Q x (Slope of AR curve) Here Q is positive and the AR curve can either be horizontal with its slope equal to zero or it can be downward sloping with negative slope. From the above relationship between AR and MR we find that (i) MR=AR when AR is horizontal and (ii) MR is < AR when AR is downward sloping. | Under perfect competition AR or Price is constant as output increases. Hence MR is also constant and equal to AR or Price at all output levels. In other words, the MR curve coincides with the

horizontal AR curve under perfect competition. See Fig. 2(a). On the other hand, under imperfect competition or monopoly, because an individual firm’s demand or AR curve is falling, the MR curve is also falling and MR is < AR (or Price) at all output levels. Geometrically, the falling MR curve is situated below the AR curve and its position is halfway between the AR curve and

Y-axis.” In Fig. 2(b) we find that the falling MR is below the falling Ak, and distance ab = bc and de = ef. Detailed proof of this statement is not given here.

EQUILIBRIUM OF THE FIRM

rg |

Price

PP

:

AR = MR

OULPUt

nw

O

Figure 2(a) Price

MR Note that ab = bc

Fre

Ser ce

&

and de = ef

b d

!

— 0 when e is < l. 1, MR is = 0 when

e = 1 and MR is

The above relationship between MR and Price or AR explains the shape of tae TR curve as derived from the demand curve for an individual firm. For a downward sloping demand curve, the price elasticity is more than one at lower output and higher prices, it becomes equal

EQUILIBRIUM

OF THE FIRM

Zh

Price

DD, = Demand curve

D

DB = BD, At B, e=1

Demand

Figure 3(a)

output

Figure 3(b) Note that OM of 3(a) = OQ of 3(b). Slope of TR curve is declining till it becomes zero at OQ output.

214

MICROECONOMICS

FOR MANAGEMENT STUDENTS

to one at the mid-point of the demand curve, and at higher output and lower prices elasticity is less than one. (This statement holds good for both linear and non-linear demand curves. ) MR, which is the slope of TR curve, is positive but decreasing at lower levels of output where e is > 1. MR reaches the value zero when elasticity is equal to one. At higher output levels MR becomes negative because e is < 1. Thus, the 7R curve is an inverse U-shape, i.e. a rising curve with decreasing slope which starts falling after reaching the maximum value at the output level where MR is zero. In Fig. 3(a), DD, is a falling demand curve with e= 1 at point B or output M. Here, e is > 1 for prices above OP and e

is < 1 for prices below OP. The line DMN is the marginal revenue curve which cuts output axis at where e= 1 and so MR = 0. For output larger than M, MR is negative since e is < 1. Fig. 3(b) shows the 7R curve derived from the demand and MR curves of Fig. 3(a). Here, output Q corresponds to M in Fig. 3(a), TR is rising, but with decreasing slope because MR is positive but falling upto Q. 7R is at its maximum at Q where e= 1 and MR, the slope of TR is zero. TR is falling beyond output Q because e is 0. It

will reach its maximum when MR,=0, and will decline when

276

MICROECONOMICS FOR MANAGEMENT STUDENTS

MR; is negative. When MR, is constant 7R will keep on rising by the same amount.

Maximization of Profit

As stated earlier, the objective of a rational firm is to maximize its total profits. The firm will therefore stabilize its output and reach equilibrium at that level of output where its profits are at their maximum. Given its demand or AR curve, the firm will charge that price which corresponds to the equilibrium output, i.e. output which maximizes profit. The question therefore is when, given the firm’s revenue and costs as functions of output, will a firm maximize its profits? The

firm will maximize its profits when its MR from an additional unit of output ts equal to the MC from that unit. We know that profits are = (7R—-7C). Hence profits depend upon how much is added to 7R and how much is added to TC by one extra unit of output. Total profits will tend to increase so long as one extra unit of output adds more to 7R than what that extra unit adds to TC. At the output level where MR is equal to MC, the addition to 7R by one extra unit is the same as addition to 7C, and so at that output level profits will reach the highest value. Production of one extra unit beyond the level where MR = MC will add less to 7R than it will add to 7C. That is, MR will be less than MC and so total profits as given by (7R — TC) will start declining. To understand this argument, note that the first unit will be produced only when MR from the first unit exceeds MC of that unit. Now as output increases 7R increases either by a constant amount, if the firm’s demand curve is horizontal (under perfect competition), or 7R will increase by decreasing amounts if the firm’s demand curve is falling (imperfect competition and monopoly).

That is, MR is either constant or it is falling as output increases. On the other hand, with increasing output, MC initially falls, reaches a minimum level, and ultimately increases. Thus, as successive additional units are produced, the difference between MR and MC will, after certain level of output, tend to diminish because while MR tends to remain constant or fall, MC tends to rise. Finally, an

EQUILIBRIUM OF THE FIRM

ot

output level is reached when this difference between MR and MC becomes zero. At this level the difference between 7R and 7C, i.e. profits, will stop rising and reach its maximum value. It can also be simply explained as follows: We have seen that,

MR.. Here, i= 1, 2, 3,..., n are successive units of output.

TR =

Similarly

Profits

TC=2 MC,

= TR TC

= 2 MR,- 2 MC..

Profits

= 2 (MR, — MC).

As successive

additional units are produced, the summation

on the right hand side showing profits will continue to increase so long as MR exceeds MC for an additional unit. Now, MR of additional units of output is either constant or is falling, while the MC of additional units finally starts rising. Consequently, the difference between MR and MC or (MR — MC) becomes smaller and smaller. That is, the summation, X~UMR — MC) increases by smaller and smaller quantities. When MC rises to a point where it is equal —- MC) becomes to MR, the difference between the two, i.e. (WR zero. Hence the summation, =(MR— MC) stops increasing and reaches its maximum value. If output is raised still further, the WRK of an additional unit will be < MC and so the difference between MR and MC will be negative. That is, the summation, Xx(MR — MC) will tend to decline. Thus profits, as given by summation, (MR — MC), will be maximized when MR = MC. This condition for profit maximizing equilibrium output can also be explained by using simple calculus.

Let

Q-= Output, S = Profits = Surplus TR = Total Revenue,

aS (TR

7C = Total Costs

TC)

We want to maximize the value ofS in relation to variations in

output. Two conditions have to be satisfied for S to be at its maximum

value. These are: (i) the first derivative of S with respect to Q is

278

MICROECONOMICS FOR MANAGEMENT STUDENTS

equal to zero and, (ii) the second derivative of S with respect to Q is negative or < 0. That is

(i) dS/dQ = 0 and (ii) d2S/dQ? is < O

(i) dS/dQ = 0 “ ATR — TC)/dQ=0 “. d(TR)/dQ — d(TC)/dQ = O ¥. GIN 7a0 =a fC\7de But d(TR)/dQ is = MR and d(TC)/dQ = MC

*. Profits or S is maximum when MR = MC. In other words, the firm maximizes its profit and reaches equilibrium at that level of output where MR = MC and the MC curve intersects the MR curve.

(ii) d2S/dQ? SATC and (iii) Profits are > Normal.

294

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Short Run Equilibrium in Depressed Demand Conditions When an individual firm under perfect competition is facing the uncommon market situation of very depressed demand, its primary concern will be to charge a price that will cover at least the average variable cost and forget about covering its fixed cost. In such a situation, the firm will not earn the normal profit because the price will be less than SATC. But it will not close down in the short run because the period is not long enough for the firm to change its fixed factors of production. This situation is shown in

Figs 5(a) and 3(6) and also in Figs 4(a) and 4(6). Fig. 3(a) shows SS, as the total market supply curve. DD, shows the total market demand curve which is situated very much to the left, showing that the total demand is very depressed. The market equilibrium price P,,, is therefore quite low. In Fig. 3(b),

corresponding to the market price P,,, the individual firm’s demand curve is situated at the level of P, which is = P. The firm’s

horizontal AR and MR curves correspond to the level P,. Fig. 3(b) also shows the U-shaped AVC as the firm’s average variable cost. SMC shows the firm’s short run marginal cost curve which

cuts AVC and SATC at their minimum

points, N, and N,.

The equilibrium output will be at the point G where SMC intersects the horizontal MR curve from below. This point is situated below the SATC curve showing that the firm is not earning even its normal profits. It covers some but not all its fixed costs in the short run because of demand deficiency. The firm covers its AVC but not its SATC. Figs 4(a) and 4(6) show the change in the firm’s short-run equilibrium output if the demand in the market were to decline even further so that the market equilibrium price falls even further. The horizontal demand or AR curve of the individual firm is so situated that it forms a tangent to the U-shaped AVC at its minimum point, N,. In this case the firm’s equilibrium price P, corresponds to the minimum point of AVC. The firm charges an

equilibrium price that just equals its AVC. This price may be considered to be the firm’s short run reservation price in the sense that if this minimum price is not covered the firm will stop production even in the short run.

O

Md

JaysDy 40

W

A4jsnpuy

ainsiy(DE

|

Ajddng

wuiniuqyinby

7 pupwag pup

Ad = AV YN SLYVS DAY. DIS

jonpiai wy puy §,

ans](Qe

«

QAV

UV=

ALS

wNLUgyInby

IWS

jndjno

YW

Ajddng pup pupwaq

W

:a]0N =4/YOQ WU] $

UOAeAIISOY dod

uiniugijinby Aa4jsnpu] 40 Jay4DW

O

wuniuquinby s way jonpiaipuy

AV YN

:

(D)p a4nsly

40)

bs

IN

td

(Mp aanshy

CN

IWS ‘DAV ‘DLVS Pd =

jndjno

YN = HV

IJAV

ILVS

INS

O

PERFECT COMPETITION

297

»

Long Run Equilibrium of the Firm We have seen that except in a situation of very depressed market demand, the short run equilibrium output of a typical individual firm under perfect competition is determined at that level where the price is greater than SATC and the firm earns more than normal profits, i.e. super-normal or positive economic profits. However, in the long run, entrepreneurs will have sufficient time to set up new firms or existing firms will have the time to diversify into new profitable lines of business. Because entry is free in a perfectly competitive industry, the super-normal or positive economic profits of established firms in the short run will induce new firms to enter the market in the long run. Moreover,

in the long run, the existing firms will expand the

size of their plants by changing their fixed factors. Thus the scale of their output will change on the long run average cost or LAC curve. The operation of these two processes, namely, increase in the

number of firms resulting from new entry, and expansion in the scale of output by existing firms, will have the effect of shifting the

total market supply curve towards the right. Given the total market demand curve, the rightward shifting of the total supply curve will bring down the market equilibrium price. Consequently, every individual firm’s demand or AR curve will

slide downwards, remaining parallel to the X-axis. This process of increase in total supply and downward movement of the market equilibrium price will continue as long as a typical individual firm earns super-normal profits, i.e. as long as economic profits are more than zero. It will come to a halt only when the number of firms in the industry and total supply in the market increase up to a point where the market equilibrium price is just equal to a typical individual firm’s LAC, and the firm would earn only normal profits. In the long run, therefore, the individual firm’s horizontal demand or AR curve will continue to slide downwards until it forms a fangent to the U-shaped LAC. Now we find that a horizontal AR can be a tangent to the U-shaped LAC only when the LAC is at its minimum. (The point

of tangency between AR and LAC can occur to the left of the

‘aJO

O

:?10N

YO =

[gO =

"wo=

WO =

U =

ee

Af

lq

ee

Ss

yoyre| 40

BU]

jaye

‘G17 (D)¢ st e

wnqiyinby a0ug

wniuqyinby ddUg

Aysnpuy]Ajddng

Aysnpu] Ajddng

SUL] JO JOquUINY aTeosUO ay} yndjno sIxeUl

suoyTuni

unIyOYSJaye

suo]un’

uUNAOYSyoyreY 40

unssuo7T wnuqyinby jo JaYAD LO AlJSNPuU]

(og

AjddngO

puDWOAG

a

ey

i'5

eet gn

JOxYIDW

tw pup

aansly

i

xu='WOlOO “WO = U x £00

un4js40ys pup

IW

ae

eazy 90 =

£40 =

Lo =

“OO=

Oo =

la

|

lows lovS

)

unsyoys

£&—

ajdyInuI Jo ay

Beds UL

(QS

winuqyinby 20g = {¢O

wnuqyinby 90g = WO

winuqyinby yndjno

‘Gl

jndjno

ie

IVT

wniuqyinby jo

LS

wnuqyinby yndjno

unssuoT JDNplaipuy ultl.y

lyDy'D = '00 '40) ('90 avsJo nding 00

suo]uni

unsyoys

"OVS

COWS

INT

aunsi](Mg

aes |

rN

Lu

un4soys pup

lip=

suo]uni

|

add = uD Au! ‘DVT INT OVS OWS

PERFECT COMPETITION

299

minimum point of LAC only if AR is falling, which is not possible under perfect competition). But when LAC is at its minimum, LMC intersects LAC from below. Hence, at the point of tangency of AR and LAC, the LMC cuts the MR (or AR) curve from below. Thus, this point shows profit maximizing output. In the long run, therefore,

the firm will reach

a profit maximizing

equilibrium

output at the minimum point of LAC. At this point profits are at their highest because MR=LMC. Profits are also only normal because the price or AR of the firm is just equal to its LAC. Hence, normal profits are the maximum profits that any firm can earn in the long run. Under perfect competition economic profits are zero in the long run. The process by which the long run equilibrium of an individual firm is determined under perfect competition is shown in Figs 5(a) and 5(0). In Fig. 5(a), DD, is the total market demand curve. The initial market supply curve is SS, which cuts DD, at the market equilibrium price P, and output ™,. In Fig. 5(b), ar, = mr, is the individual firm’s initial horizontal average and marginal revenue curve, and ar, will be positioned at price F, which is equal to P,. In Fig. 5(b), LAC and LMC show the long run average and marginal cost curves of the firm. SAC, and SMC, show the initial short run average and marginal cost curves for the short run plant size.

Here we find that the short run equilibrium output will be reached at point G where SMC, intersects horizontal mr, from below. Therefore, Q, is the equilibrium output of the firm in the

short run. For this output SAC, shows that average cost is equal to C,. At output Q, the firm makes super-normal or positive economic profits equal to the rectangular area C,KGF,. These super-normal profits in the short run will lead to the entry of new firms in the long run. The two simultaneous processes of increase in the number of firms and expansion in the scale of output by existing firms will continue until the total market supply curve, by shifting to the right, will become S55 1k Mersects the total demand curve DD, at market equilibrium output M, and price P,. Therefore, the individual firm’s ar and mr curves will continue to slide downwards until it becomes ar, = mr,, which is

300

MICROECONOMICS FOR MANAGEMENT STUDENTS

horizontal, at price = F, which is equal to P,. At this price the curve mr, = ar, is a tangent to LAC at its minimum point V. Here LMC intersects mr, from below. Therefore the individual firm’s equilibrium will be determined at output Q,. This equilibrium output is the optimum output because LAC is at its minimum at VN. In the analysis of long run costs we have seen earlier that at the optimum scale of output SAC is a tangent to LAC and SAC is also at its minimum. Consequently, SMC cuts SAC at this point and SMC is equal to SAC. Thus, at the optimum scale of output shown

by the point NV, minimum LAC = LMC = minimum SAC, = SMC,. To sum up, under perfect competition the long run equilibrium output of a typical individual firm is determined at that level where the following conditions are satisfied:

1) 2)

Price = AR = MR = LMC = minimum LAC = minimum SAC = SMC. Maximum profits = Normal profits and Economic profits = 0.

Note that the scale of output axis in Fig. 5(a) is a multiple of the scale of output axis in Fig. 5(6). Since, in the long run, the maximum profit a typical firm can earn is only the normal profit, there is no inducement for new firms to enter the market, nor is there any pressure on the existing firms to leave the market. Thus the number of firms in the industry is stabilized so that output of industry as a whole ts also at equtlibrium level.

Diminishing Returns and Perfect Competition Equilibrium An analysis of the determination of an individual firm’s equilibrium under perfect competition in the short and long run shows that marginal cost (MC) must be rising at the equilibrium output. This is because the MC curve cannot intersect a horizontal MR curve from below except when the MC curve is rising. While, in the short run, this equilibrium condition is satisfied when both MC ~and AC are rising, in the /ong run the equilibrium output is reached

when LMC is rising but LAC has reached its minimum, i.e. has stopped falling. In cther words, under perfect competition, because the individual firm’s demand (or AR) and the MR curve is

PERFECT COMPETITION

301

horizontal, equilibrium of the firm is logically incompatible with

increasing (marginal) returns or falling costs. At equilibrium output the marginal returns must be diminishing and marginal cost must be rising under perfect competition. This conclusion is clearly shown by Figs 2(a) and (6). The fact that the equilibrium ofa firm under perfect competition cannot occur when returns are increasing can also be shown by applying the condition of profit maximization. The second condition of profit maximizing equilibrium is: (Slope of MR) - (Slope of MC) is < 0. But since MR is horizontal, the slope of MR = 0. Hence, the

slope of MC is > 0 or MC rising at equilibrium.

Perfect Competition, Welfare, and Growth Adam Smith, in his classic work, The Nature and Causes of Wealth

of Nations published in 1776 described the forces of perfect competition as the ‘invisible hand’ which ensures that even without any planning or instructions/guidance from any authority, the functioning of innumerable independent producers and customers is so finely coordinated that the total quantity of any commodity produced is just equal to the total quantity demanded, buyers pay the price that just covers the cost of production, and its producers earn only the profits that they deserve to earn. In short, the resources are most efficiently allocated. Adam Smith believed that in a capitalist economy, if the operation of competitive forces is not obstructed either by the state or by private monopolies and cartels, the working of the competitive process in the long run will maximize social welfare and create conditions which are most conducive to economic growth. The reasons why Smith put so much faith in perfect competition can be easily explained as below. (i) Under perfect competition, in the long run, the price charged by a typical firm is equal to minimum LAC. Thus buyers pay the price that just covers the costs, i.e. they are not exploited. (ii) Producers in the long run earn only normal profits. But they have no reason to complain because they earn at least what

302

MICROECONOMICS

FOR MANAGEMENT STUDENTS

they deserve and, more importantly, do not earn more than their opportunity cost as entrepreneurs.

(iii) Each firm produces the optimum level of output because the long run equilibrium price is equal to the minimum LAC. The number of firms in the industry is also optimum because all the firms earn only normal profits. Thus both within each firm and for the industry as a whole resources are most efficiently allocated. If all industries are perfectly competitive, then for the economy as a whole the allocation of resources would be most efficient. (iv) We have seen in the chapter on costs of production that if MC = marginal cost, W = factor price, and MP = marginal productivity of any factor, then MC = W/MP. But MC = price of product under perfect competition.

, Price of product = W/ MP.

. Mx PCE of Produc =.

, Value Of Vir =-W, Hence each factor is paid that reward or price that is equal to the value of its marginal productivity, i.e. the price that corresponds to what a marginal unit of that factor contributes to production. Thus the owners of the factors are not exploited. (v) Under perfect competition, every producer is only a pricetaker and nota price-maker. Hence a firm can make profits higher than its competitors only by lowering its costs. This motivates a firm to reduce its costs through technological changes. At the same time, firms failing to keep abreast of the prevailing technology will be obliged to leave the industry. Under perfect competition only the fittest can survive and grow. Thus the pressures to improve technology and maintain efficiency which operate at the firm level lead to speedy economic growth when competition is keen (and tends to be perfect) in most markets.

Conclusion

A word of caution is, however, necessary to avoid getting carried away by this very appealing scenario presented by Adam Smith

PERFECT COMPETITION

303

and his followers in favour of the social and economic superiority of a laissez-faire capitalist system. The free-market economy will create socially and economically desirable results only if most industries are perfectly competitive. But this is a condition which in reality is not satisfied in any capitalist economy. In most free-market economies, perfect competition is an exception and not the rule. A majority of modern industries operate under conditions of oligopoly or monopolistic competition. Consequently, the results which economists associate with perfect competition are rarely encountered in the real world of business. In this world the guiding principle is the survival of the fattest and not survival of the fittest. What then is the relevance of the model of perfect competition? Why study it? The relevance of perfect competition is that it shows what the ‘ideal’ market structure is. It serves as a reference point or a benchmark. It helps us to understand how imperfect the markets and industries of the real world are and what are the effects of market imperfection.

Chapter 13 Monopoly and Price Discrimination

Monopoly means a market where a single firm controls the entire supply of a product which has no close substitutes. Just as perfect competition is one extreme case of market structure with maximum degree of competition, monopoly is the other extreme case where competition is altogether non-existent. Clearly, the distinction between an individual firm and the industry is not relevant in the case of monopoly because the industry consists only of one firm. Thus the market or industry demand curve and that of the individual are the same under monopoly. In the case of a monopolized product — assuming that its imports are not allowed — the total or market supply changes to the same extent as the change in the output of the monopoly firm. Hence,

unlike the individual

firm in a competitive

market, the

monopoly firm has power to control the price of its product. If the demand for the product remains unchanged, the monopoly firm can raise the price as much as it wishes by reducing its output accordingly. On the other hand, if the monopoly firm wishes to sell a larger output of its product, it must lower the price because total supply in the market will increase to the extent that its output increases. Thus, the principal difference between competition and monopoly is that in a keenly competitive market any individual firm is a price-taker whereas a monopoly firm is a price-maker. It needs to be noted that just because a firm is the only producer of a particular product this does not endow it with the power to set the price if close substitutes are available for its product. The buyers will switch over to these substitute products when the monopoly producer raises the price more than marginally. For instance, though Coca Cola Corp. is the sole producer of coke and

Colgate-Palmolive is the only producer of Colgate toothpaste, we do not describe these firms as monopolies because their products have close substitutes and their power to raise price is quite limited. For a firm to become a monopoly with price-setting power

MONOPOLY AND PRICE DISCRIMINATION

305

it must not only be the sole supplier of a product but also sell a product which does not have any close substitute. The demand curve for the product of a monopoly firm is downward sloping from left to right because it can raise its price by restricting its output and must reduce the price when it increases the output. Since the monopoly firm’s demand curve is falling, its marginal revenue curve is also falling, and the marginal revenue curve is situated below the firm’s demand curve at all levels of output. When this downward sloping demand curve is a straight line, the corresponding marginal revenue curve will be situated half-way between the demand curve and the price or Y-axis.

Equilibrium of the Monopoly Firm We assume that the monopoly firm decides its price and output with the objective of maximizing its total profits. In the case of a monopolist this assumption is justified because: (i) the monopolist’s demand and marginal revenue curves are known since its demand curve is the same as the market demand curve, (ii) the monopolist acts independently without any concern about the uncertainty of his rival’s reaction to his own actions since, by definition, it has no rival, and (iii) generally, an established monopoly firm does not have to iace any serious threat of new entry competition. The monopoly firm will maximize its profit and determine its equilibrium output at that level where its marginal cost curve intersects its downward sloping marginal revenue curve from below. Since the marginal revenue curve of the monopoly firm is situated below its average revenue or demand curve at all output levels, and since at the equilibrium output marginal revenue is equal to marginal cost, the profit maximizing monopoly price is greater than marginal cost. (In contrast, the profit maximizing price under perfect competition is equal to marginal cost.) Moreover, since in most cases the major portion of the demand curve of the monopoly firm is lying above the firm’s average cost curve, the price at equilibrium output is also greater than average cost (LAC). Hence super-normal profits or positive economic

306

MICROECONOMICS

FOR MANAGEMENT STUDENTS

profits is a distinguishing feature of equilibrium under monopoly. Such super-normal profit can be maintained even in the long run because, in the case of a monopolized market, new entry is very difficult, if not impossible. It needs to be noted here that the distinction between short run and long run equilibrium is not relevant in analysing the monopoly firm’s behaviour because, by definition, competition from entry

of new firms does not take place in a monopolized market. Fig. ] explains the determination of profit maximizing equilibrium output and price under monopoly. Here DD, and MR are the downward sloping demand (or average revenue) and marginal revenue curves of the monopoly firm. AC and MC are its U-shaped cost curves. MC cuts MR from below at point £. Output OQ, corresponding to £, is the profit maximizing equilibrium output. The price corresponding to OO is OP = OR. At OQ output average cost = OC = OK.

O

Q

Figure |

output

MONOPOLY

AND PRICE DISCRIMINATION

307

monopoly profits are = (OP-—OC) x OQ= The maximum PC x CK = Area of the rectangle PCKR. Since the AC curve includes normal profit, the area PCKR also represents the super-normal or positive economic profits of the monopoly firm. An important condition for equilibrium under monopoly is that price-elasticity of demand must be greater than one at the equllibrium output. That is, the monopolist cannot reach equilibrium at any price for which the elasticity is less than one. And at the price where elasticity is equal to one, the monopolist can reach equilibrium only if his production costs are zero — a situation which in the real world is very unlikely but not impossible. | The reason for this condition is to be found in the relationship between marginal revenue (MR), price (P), and elasticity (e). As already noted in Ch. 11, MR=P(1-1/e). Therefore, if at any price e is less than one, MR will be negative. But at profit-maximizing equllibrium output, MR is equal to marginal cost which is generally positive and in rare cases zero. Thus, the equilibrium condition cannot be satisfied when MR is negative, i.e. when e is less than one. When e is equal to one, MR is zero and so equilibrium output can be reached only if marginal cost is zero. This condition is explained in Fig. 2 where DB and MR are the monopolists’ demand and marginal revenue Curves. MR cuts the outputaxis (X-axis) at point Q. Point NV is the mid-point of the demand

curve DB so that

BN= ND. As we know, at price = i,

elasticity is = (BP, /P,D), which is > 1, and at price = F,, elasticity is ae Poe which is.< 4. At price = N, elasticity = (BN/ND), which is = 1. Hence MR is positive up to output Q corresponding to price = Non the DB curve. For any price below N, i.e. for output higher than OQ, MR is negative (below the X-axis).The monopolist can maximize profit only if price is greater than ON assuming marginal cost is positive. In Ch. 11 we have seen that increasing returns are incompatible with the equilibrium ofa firm under perfect competition. The short run as well as the long run marginal cost curve (SMC and LMC) must be rising at the equilibrium output under perfect competition | One commonly cited example of production with zero costs is that of ownership of a natural spring of mineral water with some healing quality.

308

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Demand

Figure 2 because the marginal revenue (MR) is constant at all output levels (i.e. the MR curve is parallel to the X-axis). In contrast, the monopoly firm can determine its profit-maximizing equilibrium output when marginal costs are falling or constant or rising. Since the monopoly firm’s demand and marginal revenue curves are downward sloping, the marginal cost curve can intersect the marginal revenue curve from below even when the former is falling or constant. That is, the monopoly firm’s equilibrium output can be less than or equal to or more than the optimum output. This feature of equilibrium under monopoly is illustrated in panels (a), (b), (c), and (d) of Fig. 3. In panel (a) of Fig. 3, DD,, MR,, MC,, and AC, are the monopoly firm's demand, marginal revenue, marginal cost, and average cost curves, respectively. The U-shaped MC, intersects the falling MR, from below at point

E,, even while MC, is declining because near this point MR, is falling relatively faster as output increases.

Here equilibrium

output OQ, is less than the optimum output ON, (with minimum AC,). At equilibrium

output the price is P,, average

and super-normal profits are given by area P,R,K,C,.

cost is C,,

MONOPOLY

AND PRICE DISCRIMINATION Price, MR, AC, MC

Price, MR, AC, MC

O

N, | output

Q,

Qs

Q,

output

Price, MR AC, MC

Price, MR, AC, MC

N,

O

Figure 3(b)

Figure 3(a)

O

309

output

Figure 3(d)

Figure 3(c)

In panel (b), MC, = AC, show the constant marginal and average costs. The downward

sloping MR, is intersected by horizontal

MC, from below at point E.. Equilibrium output is OQ,. At this output, price is P, and super-normal profits of the monopoly firm

are equal to area P,R,K,C). In panel (c) the falling MR, of the monopolist is intersected by

SLU

MICROECONOMICS

FOR MANAGEMENT STUDENTS

U-shaped MC, from below at point £, when the latter is rising (i.e. returns are diminishing). Here equilibrium output OQ, is greater than optimum output ON,. At this output, price is P,, average cost is C,, and the monopoly firm’s super-normal profits are equal to area BK AGC,. It is possible that a monopoly firm’s MC curve may intersect from below its falling MR curve at that output at which the AC is at its minimum.

That is, by some coincidence, the monopolist’s

equilibrium output can even be at the optimum level. But such accidental cases are quite rare. This is shown in Fig. 3(d). It will be seen from the above explanation that, regardless of whether the costs are falling, rising, or constant (i.e. whether returns to scale are increasing, diminishing, or constant), the

monopoly firm earns super-normal profits and its profit-maximizing price tends to be greater than its marginal and average cost. Also, generally the monopolist’s equilibrium output is either below or above the optimum level. As we know, the supply curve of a firm shows the various output levels which it is willing to produce at different prices. Under perfect competition a firm’s marginal revenue and price are equal at all output levels and so price is equal to marginal cost at profit-maximizing output both in the short run and in the long run. Thus, the firm’s marginal cost curve is the same as its supply curve because it shows the various quantities of output that the firm is willing to supply at different prices. In contrast, for a monopoly firm, since the price is above marginal revenue at all output levels, at the equilibrium output marginal cost is equal to marginal revenue but less than price. The marginal cost curve shows the output levels that the monopolist is willing to produce at different values of marginal revenues and not at different prices. The marginal cost curve of the monopoly firm ts therefore not the same as its supply curve.

Monopoly Power Our analysis of the profit-maximizing equilibrium under perfect competition and under monopoly shows that whereas under perfect competition, price is equal to marginal cost and in the long

MONOPOLY AND PRICE DISCRIMINATION

31]

run profits are only normal. But under monopoly price is greater than marginal cost and profits are more than normal even in the long run. Thus, the monopolist has power to charge a price that exceeds marginal cost and thereby earn super-normal profits. Keeping in view these contrasting features of perfect competition and monopoly, A.P. Lerner devised an index to measure the degree of monopoly power, known as the Lerner index.’ According to this index,

u=(P-—MC)/P. Here P is the The monopoly power of a firm = price of the firm’s product and MC is the firm’s marginal cost. Clearly, in a perfectly competitive market U is zero because P is = MC. We have seen that at equilibrium output MC = marginal revenue = MR. And MR=P(1-—1/e), where e is the price elasticity of demand.

u=(P—MC)/P “u=(P—MR)/P=1-—-(MR/P) But (MR/P) =(1 - 1/e) .u=1-(1-1/e)

jae We take only the numerical value of e, into account.

Thus the monopoly power of a firm varies inversely with the elasticity of demand for its product. The less elastic the demand for its product, the greater its monopoly power, and vice versa. The most important factor upon which elasticity depends is the number and closeness of the substitutes available for a product. The closer the substitutes for a product, the higher its elasticity, and lower the monopoly power of the producer of that product. Under perfect competition, the elasticity of demand for a firm’s product is equal to infinity and therefore its monopoly power is zero. The Lerner measure of degree of monopoly is in conformity

with what we see in the real world. We do find that in a country like India the suppliers of some essential goods and services, such

1 See Lerner (1934).

312

MICROECONOMICS FOR MANAGEMENT STUDENTS

as life-saving medicines, low-cost housing, banking, petroleum, cooking gas, electricity, railways, telephones, etc., enjoy unlimited monopoly power because the demand for these products is highly inelastic. In the case of petroleum, cooking gas, housing, and lifesaving medicines, monopoly power is revealed by the wide divergence between price and costs. In the case of railways, telephones, banking, and electricity, however, the prices are not very much higher than the marginai cost because these are either statutorily controlled or the producer firms are owned by government. The unlimited monopoly power of these public sector enterprises is revealed by the very poor quality of the goods and services they supply and the callousness with which they treat their customers.

Welfare Implications of Monopoly The fact that the monopoly firm can raise price above costs by restricting output and enjoy super-normal profits has led many economists to examine the welfare implications of monopoly in comparison with competition. If the monopoly firm and the perfectly competitive firm face the same marginal cost curve then price will be higher and output lower under monopoly than under perfect competition. This follows from the fact that, for a monopoly firm, price is greater than marginal cost whereas for a competitive firm price is equal to marginal cost. Figs. 4(a) and 4(6) explain this point in relation to constant and rising marginal costs. In Fig. 4(a), DD, is the demand curve for the product and MC=AC shows the constant marginal and average costs. Since under perfect competition price is = MC, the equilibrium output and price will be determined at point A

where MC cuts DD,. Hence Q, = competitive output and P, is competitive price. Let MR be the marginal revenue curve derived from demand curve DD,. Here the monopoly equilibrium output is determined at point B where MC cuts MR from below. The monopolist’s output is Q, and its price is P,. As P,>P, and Q, < Q,, under monopoly price is higher and output lower than under perfect competition. In Fig. 4(b), MC is the U-shaped marginal cost faced by the monopoly and the competitive firm. The demand curve for the

MONOPOLY AND PRICE DISCRIMINATION

313

output

Figure 4(a)

peo,

a

Q;

Figure 4(b)

output

314

MICROECONOMICS

FOR MANAGEMENT STUDENTS

product, DD, intersects MC at point F. Corresponding to F, perfect competition output and price are Q, and P,. MR is the marginal revenue from DD,. MC cuts MR from below at G. Hence Q, is the

monopoly firm’s equilibrium output and P, is the corresponding

monopoly price. We find that Q, is < Q, and P, > P,. Thus when a large number of small firms selling a nearly homogenous product agree to form a cartel, the perfectly competitive market is transformed into a monopoly and, as a result, price increases and output declines. The welfare implications of monopoly and competition in terms of consumer’s surplus is explained in Fig. 5(a). This concept is

defined in Ch. 4. In Fig. 5(a), DD, is the demand curve for the product, MR is the marginal revenue curve derived from DD, and MC = AC show the constant marginal and average costs. As already explained, P, and Q, are price and output under perfect competition. P, and Q, are price and output under monopoly. Here we find that at perfect competition price P,, output Q, is bought with

output O

Q,

Figure 5(a)

Q;

MONOPOLY AND PRICE DISCRIMINATION

315

the result that the consumer’s surplus is = Area DAP,. Under

to monopoly, the price rises to P, and output purchased declines Q,. Consequently, the consumer’s surplus diminishes to Area

DRP,. The loss in consumer’s surplus is = Area P,RAP,. This area is made up of Area P,RBP, plus Area RAB. But Area P,RBP, also

shows the super-normal profit that the monopoly firm would make by selling lower output, Q, at a higher price, P,. Thus there is a gain in the welfare of the monopolist as shown by Area P,RBP, and

loss in consumer’s satisfaction shown by Area (P,RBP, + RAB).

Now, in analysing the economics of welfare we do not make any value judgement about whether the gain by one section (or member) of society is preferable (or more desirable) than the welfare loss of another section (or member). Because such comparison involves interpersonal comparison of the utility of two persons. Utility being a subjective concept, we cannot compare the increase in utility of one person with the decrease of that of another. We thus find that the total loss of consumer’s surplus shown by the Area P,RAP, is partly offset by the gain in monopolist’s welfare shown

by Area P,RBP.. Hence,

net loss of welfare that

occurs when price rises from P, to P, is given by the Area RAB. Identification of welfare loss as a consequence of monopoly becomes quite complicated when the marginal cost curve is up-

ward sloping, because

here, in addition to consumer’s

surplus,

we have also to consider producer's surplus. The concept of producer’s surplus refers to the difference between the amount of revenue that the producer actually obtains by producing a given output and the minimum amount (of revenue) that he expects to earn by selling successive units of that output (i.e. the summation of marginal costs as given by the area under the marginal cost curve). The welfare effect of monopoly can be measured by finding the difference between the sum of consumer’s and producer's surplus under perfect competition and the sum of consumer’s and producer’s surplus under monopoly. This is explained by Fig. 5(b), where DD, is the product’s demand curve, MR is the corresponding marginal revenue curve, and OS is the simplified upward sloping marginal cost Curve. As explained before, P, and Q, are the price and output under perfect competition, and P, and Q, are the monopolist’s price and output.

316

MICROECONOMICS

leh

FOR MANAGEMENT STUDENTS

ae i

Q,

output

Figure 5(b) Now under perfect competition with price P, and output Q,, we find that:

(a) Consumer’s surplus = Area ODAQ, — Area OP, AQ, = Area DAP, = Area DRP, + Area P, RBP, + Area RAB.

(b) Producer’s surplus = Area OP, AQ, — Area OAQ, = Area P, AO

= Area P, BTO + Area BAT. And for the monopoly price P, and output Q, we find that:

MONOPOLY AND PRICE DISCRIMINATION

(c) Consumer’s surplus

ori

= Area ODRQ, — Area OP, RQ, = Area DRP,

(d) Producer’s surplus

= Area OP, RQ, — Area OTQ,

= Area OP, RT

= Area P, BTO + Area P, RBP, The welfare loss arising on account of rise in price to P, and decline in output to Q, under monopoly is given by the sum of (c) and (d) minus the sum of (a) and (b). After cancelling the areas common to these two sums we find that the net loss of consumer's and producer’s surplus is = Area RBA + Area BAT = Area RAT. Thus the welfare loss attributable to the monopolization of a market is shown by the Area of the triangle RAT in Fig. 5(b). The decline in output and loss of welfare in terms of consumer’s and producer’s surplus that occurs under monopoly is often referred to as ‘dead weight’ loss of monopoly.

Price Discrimination under Monopoly The monopoly firm can discriminate between different buyers differently by charging them different prices because it has the power to control price by changing its output and because the buyers of its product have no choice but to buy from it. Price discrimination refers to the situation where a monopoly firm charges different prices for exactly the same product.’ For a discriminatory monopolist, while the cost curves are the same in relation to all buyers, the demand and revenue curves are different for different groups of buyers. There are three types of price discrimination, namely, First Degree discrimination, Second Degree discrimination, and Third

Degree discrimination. First degree or perfect price discrimination refers to a situation 2 Clearly, price discrimination cannot arise when the number of sellers is more than one. This is because the competitor of the discriminating seller can attract away that group of buyers being charged a higher price by offering them the same lower price that the discriminator is charging others.

318

MICROECONOMICS

FOR MANAGEMENT STUDENTS

where the monopolist charges different price for each different unit of the output he sells. Thus for the same product (and perhaps even for the same buyer) price differs from unit to unit. For instance, in a city if there is only one cardiac surgeon, he may

charge different operation fees for bypass surgery to each different patient, depending upon the latter’s capacity to pay. If the perfectly discriminating monopoly firm charges a lower and lower price for each successive unit then the price charged for each unit is the same as the firm’s marginal revenue. Here the firm’s demand curve is the same as its marginal revenue curve. The firm’s equilibrium output will be determined at that output where the price charged for the last successive unit is equal to marginal cost of that unit. Fig. 6 shows first degree price discrimination. Here MR is the firm’s marginal revenue and its demand curve. MC, the marginal

cost curve cuts MR at E. AC is the average cost curve. The firm will produce OQ output, charge price QF for the last successive unit of output, and profits will be = Area MEGC.

outpul

O Figure 6

319

MONOPOLY AND PRICE DISCRIMINATION

Note that perfect discrimination can be adopted by a monopoly seller mostly when each unit of product is bought by a distinct buyer. This generally happens in the case of personal services like surgery, teaching or music tuition, lawyer’s services, etc. Second degree price discrimination refers to a situation where the monopolist charges different prices for different batches or blocks of units of the same product. For instance, a monopoly electricity company may charge one rate per unit for the first 1000 KWH of power consumed, another (lower) rate for the additional 1000 KWH, and so on. Here, within each batch the price per unit is the same, though price varies from batch to batch. Railway passenger fares often involve such discriminating prices accord-

ing to the distance travelled. For the same passenger and for the same type of comfort, the per kilometre fare is the highest for the firstX kilometres, and gradually declines as the distance increases. Fig. 7 illustrates the nature of second degree price discrimination. Here X-axis shows the output in batches of units, and Y-axis

O

900

1000

1500

Figure ¢

2000

2500

320

MICROECONOMICS FOR MANAGEMENT STUDENTS

shows price per unit in each batch. For the first 500 units the price is Rs 5 per unit, Rs 4 for the additional 500 units (i.e. for purchases

ranging from 500 to 1,000 units), and so on. The firm’s marginal revenue as well as demand curve is a step function given by the curve ABCDEFGH ... Here MC and AC are the marginal and average cost curves. The monopolist will determine the profit maximizing equilibrium output in such a way that the per unit price charged in the last unit produced is equal to the marginal cost. Here MC cuts the step line at point R so the equilibrium output Q is 2,000 units. The price or marginal revenue for this is QR or Rs 2 per unit. For output Q the average cost is QK. Hence the maximum profits are equal to the shaded area between the step line ABCDE .. RK and line KC. Note that the total cost here is OCKQ and total

revenue is the sum of areas of each shaded bar. The most commonly found cases of price discrimination relate to Third Degree price discrimination. When the monopoly firm divides the total market for its product into two or more markets (groups of buyers) and charges a different price in each different market, this is known as third degree price discrimination. This form of price discrimination is commonly found in public utility and social services. For example, in India electricity rates are lower for farmers in rural areas than those charged to factories in urban areas or a medical college charges, for the same education, lower fees to students from Scheduled castes and Tribes and higher fees to others or railway passenger fares are lower for senior citizens.

When is Price Discrimination Possible, Profitable and

Socially Desirable ? Just because a firm possesses a monopoly of a product does not mean that it can charge discriminatory prices. Even third degree price discrimination is possible only when the following conditions are fulfilled. (a) The total market must be clearly divisible into two or more sub-markets. That is, the sub- markets are separated because: (i) of political (national) or geographical boundaries, (ii) the buyers belong to different income or ethnic or social groups, (iii) the buyers make distinctly different uses of the product (e.g. the use

MONOPOLY AND PRICE DISCRIMINATION

321

of electricity by households, farmers, and factories), and (iv) the product sold is a personalized service. (b) Resale from one market to another is either not possible or involves costs that exceed the price differential between two markets. Price discrimination in personalized specialist services is quite common because one recipient cannot transform himself into another merely to take advantage of a lower price. If a library charges lower fees to female and child members, adult male members cannot become females or children in order to benefit from lower fees. Similarly, a farmer receiving power at a lower tariff cannot resell it to factories. When the markets are divided by national or geographical boundaries, then the cost (tariffs and/or transport) involved in buying goods in a low price market for

resale in a high price market may exceed the price differential. Even when price discrimination is possible (because resale between two markets is not possible), the monopolist will charge discriminatory prices only when it is profitable to do so, and price discrimination is profitable only when the price-elasticities of demand in the two markets are different. That is, if e, is elasticity in

one market and e, in another, then price discrimination is profitable only if e, and e, are unequal. For example, suppose the monopolist finds that at given price P the value of elasticity of demand is 2 in both the sub-markets. Thus, if the monopolist were to reduce price in one market by x per cent and raise it by x per cent in another, the demand in one market would increase by 2x per cent and would decrease by 2x per cent in the other. The monopoly firm will therefore not gain anything in terms of sale by practising price discrimination. However, when the value of elasticity at price P in one sub-market is 2 and in the other it is 4, then price discrimination will increase the firm’s total sales. This is because an increase of x per cent in the first sub-market will reduce demand by 2x per cent but a reduction in price by x per cent in the second sub-market will increase demand by 4x per cent. There will be a net increase of 2x per cent in the total demand for the monopolist’s product. Thus price discrimination becomes profitable only when, at a

given price, elasticities of demand in the two markets are unequal. Since the total demand increases by reducing price in the

322

MICROECONOMICS

FOR MANAGEMENT STUDENTS

relatively more elastic market and raising it in the less elastic market, the monopoly firm will increase supply in the more elastic market and decrease supply in the less elastic one. As it continues to do so, its revenue in the more elastic market will increase, but by smaller and smaller increments, while its revenue in the less

elastic market tends to decline but by larger and larger amounts. Since the firm is a monopolist in both the markets, its marginal revenue curves in each market is downward sloping. As a result, with increased supply, its marginal revenue in the more elastic market decreases and, with reduction in supply, the marginal revenue in the less elastic market increases. This process of reducing supply in the less elastic market.and increasing supply in the more elastic one will continue till a point is reached where the marginal revenues in the two markets become equal. The discriminating monopolist will maximize his profits when the marginal revenues in the two markets are equal. After this point the gain in marginal revenue in the more elastic market will become smaller than the loss of marginal revenue in the less elastic one. Geometrically speaking, with each extra unit of output pushed into the more elastic market the firm moves downwards from left to right on the marginal revenue curve for that market, and with each unit withdrawn

from the less elastic

market the firm moves upwards from right to left in the less elastic market. And both these opposite movements will continue till the marginal revenues in the two markets become equal. Like any other monopolist, the discriminating monopoly firm is also interested in maximizing its profits from its total sales from both the markets taken together, i.e. from the sum of outputs sold in the two markets. Even though the firm divides its total market into two sub-markets, and its price behaviour is different in each of these, so far as its profit motive is concerned it considers itself as a single producer (and not two different producers). It is therefore concerned with maximizing its total profits in relation to its combined demand and marginal revenue from the two markets taken together and its marginal cost of producing the total output (i.e. sum of outputs) sold in the two markets. In other words, the discriminating monopolist is not interested in maximizing profits in each individual market separately; he is interested in maximizing profits from the total of revenues from the

MONOPOLY AND PRICE DISCRIMINATION

323

two markets taken together and total cost of the sum of the outputs supplied to the two markets. The maximize profits at that level of bined marginal revenue for the equal to the marginal cost of the

discriminating monopolist will total output at which the comtwo markets taken together is sum of the outputs for the two

marginal cost curve. markets, as shown on the common Moreover, as has already been explained, to maximize profits by

selling in two markets at different prices, the monopoly firm will determine outputs for each market in such a way that marginal revenues in the two markets are equal. Thus, for determining profit maximizing equilibrium output the discriminating monopolist must satisfy the following two conditions: G)

MCT=CMR

Gi)

CMR=MR,=MR, Here,

MCT is the marginal cost curve for various levels of total output for the two markets as shown on common MC curve. CMR is the combined marginal revenue for two markets. MR, is marginal revenue of Market 1. MR, is marginal revenue of Market 2.

Clearly, output Q, for Market 1 is determined at that point where

MCT = CMR = MR,. And output Q, for Market 2 is determined at the point where MCT = CMR = MR. Price P, for Market 1 corresponds to output Q, on the demand

curve DD, for Market 1, and price P, for Market 2 corresponds to output Q, on the demand curve DD, for Market 2. The determination of profit maximizing monopoly equilibrium output under price discrimination is shown in Figs. 8(a), 8(b) and 8&(c). Here, in panel (c), the combined marginal revenue, CMR, is obtained by adding MR, and MR, horizontally. The common marginal cost curve, MCT, for total output cuts CMR at point E so that equilibrium fofal output is = Q,. At this point E, CMR = ON = MCT. In panels (a) and (b) we find that for MR, = ON, = ON, the equilib-

rium output is = OQ, and for MR, = ON, = ON in Market 2 equilibrium output is = OQ,. From demand curve, DD, for Market 1, we

©

ie

‘aM "YWOV

Jay4D |

Jaudd (D)

ani](D)§

jndjno “”)

aansiy(q)g

jndjno

Jay44DN Z JauDd (q)

‘aud “YW QV

O

‘SOJON

aan$1.4(2)g§

#40.

(2)jaubg

jndjno

=2¥’Q=YO 'y'O

‘00

=NO=7%Q ='NO‘NO

>%dO “IO aourisiq

ed

aalig Oe
(1/e.),

imererore,.()

tre is at

Poh e)).

Now at profit maximizing equilibrium output,

MR, = MR,

yd

bn dey

Pe (liz, lV eo).

P/F, = [11 — 17e,)7(1 — T/e,)].

326

MICROECONOMICS

FOR MANAGEMENT STUDENTS

But (1 — 1/e,) is > (1 - 1/e,). cd yi a} te PP jis

4: For & is SP:

Thus the price charged by discriminating monopolist in the more elastic market is lower than that charged in the less elastic market. A special case of price discrimination is that of a monopolist selling the same product in the domestic as well as the export market. What is special about such a monopoly firm is that in the export market he is one among a large number of very small suppliers (from several nations). This is particularly the case in world export markets for textiles and garments, leather products,

light engineering goods, processed food products, etc. Product differentiation in the export markets is not significant for these commodies because a large variety of brands are available. Consequently, the firm which is a monopolist in the domestic market has to face perfect competition in the export market. The demand for its products is therefore perfectly elastic (with elasticity equal to infinity) in the world market at internationally competitive price. The monopoly firm is thus a price-setter in the home market but only a price-taker in the export market. Examples of Indian companies facing this type of home and export market situations are Bata in footwear, Raymonds in woollen textiles, Liberty in garments, Lipton in tea, Maruti in small cars,

and Telco in trucks. Here the monopoly firm will charge the perfect competition price, P, in the export market which will be lower

than the price P, charged in the home market. This special case of price discrimination between home and export market by a domestic monopolist is illustrated in Fig. 9. Here line P,W shows the perfectly elastic, horizontal demand as well as

the marginal revenue curve for the export market. P, is the internationally competitive price. DH is the monopolist’s demand curve in the domestic market. DR is the marginal revenue curve derived from demand curve DH. The curve DH cuts P,Wat A and DR cuts

PWat B. Hence horizontal addition of DH and P,W gives us DAW as the combined demand curve and DBW as the combined marginal revenue curve for the two markets, home and export. The common marginal cost curve, MC cuts the combined marginal

MONOPOLY AND PRICE DISCRIMINATION

el

at point E. Output Q, corresponding to point revenue curve DBW E shows the profit maximizing equilibrium total output for the two markets taken together. Here, after pointB,the combined marginal

revenue, CMR is the same as MR, which is the marginal revenue for world market. Thus CMR=Q,E

is equal to Q,B on the DR

curve. Thatis, Q,B is that marginal revenue in home market which

is = CMR. Thus, the firm will sell OQ, output at home. Corresponding to this output OQ,, the price shown on the home market demand curve is = P,. Thus the firm will charge a higher price, P,, in the home market which has lower elasticity and lower price P. in the export market which is perfectly elastic. The difference

between equilibrium total output Q, and output Q, in the home

market shows the output supplied to the export market. If AC is

output

Figure 9 Price Discrimination in Export and Home Market

328

MICROECONOMICS

FOR MANAGEMENT STUDENTS

the common average cost curve, then for output Q, the average cost is = Q.K= OC. Thus the total maximum profits are = Area

CKEP.,. Profits in the home market are = Area CFNP.,,. In the export market profits are = FKEB. The practice of price discrimination by monopoly enterprises can be socially beneficial in two ways. First, the shiftofthe demand and marginal revenue curves towards the right under price discrimination leads to increased output and makes it possible for a monopolist to convert non-viable production into viable. Second, it enables the monopoly supplier of a socially meritorious good or service to offer price incentives in order to attract those needy or deserving buyers who cannot purchase the product at the single monopoly price. Two real world instances of particular relevance to developing countries illustrate these two points. (1)

A modern engineering testing and research laboratory equipped with many imported, expensive, and state-of-theart pieces of equipment would involve large initial capital costs,

but once

installed,

there would

be significant eco-

nomies of scale. However, the laboratory is economically non-viable because the existing less elastic demand curve of large engineering enterprises with funds to spare, and also the more elastic demand curve of the small enterprises, is situated below the long run average cost at all output levels. So at a single monopoly price the supply of this important, socially desirable service becomes uneconomic. But when the promoters of such a laboratory introduce price discrimination, the combined

demand

curve

is such that its estab-

lishment and operation becomes economically viable. This is explained in Fig. 10.

Here LAC and LMC are the long run average and marginal! cost curves of the engineering testing and research laboratory. DD is the demand curves of the large enterprises and BB is the demand curve of small factories. MR, and Mk, are the corresponding marginal revenue curves. We find that both DD and BB are situated below LAC at all output levels. Thus, any single monopoly price derived from either DD or BB for the supply of this laboratory’s services is economi-

cally non-viable. However, CDD and CMR are the combined

MONOPOLY

AND PRICE DISCRIMINATION

329

Price, MR, LAC, LMC

Figure 10

(2)

demand and marginal revenue curves obtained by horizontally adding DD and BB. LMC cuts CMR at point E or at output Q,. At this output, the combined demand curve CDD is above LAC. Hence, by charging prices P, and P, to the two groups of enterprises, the testing and research laboratory can become economically viable. The supply of medical education in a country like India is the responsibility of the government. However, a wellequipped medical college attached to a large hospital involves heavy investment.

As a result, the fees for such edu-

cation have to be quite high and can be afforded only by students from rich families, and is priced beyond the means of deserving students from economically weak families. If the government, as a monopoly supplier, charges a single monopoly price, then only a few rich students can join the medical college. On the other hand, if the government adopts

330

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

discriminatory prices, the college can enroll bright students from poorer families. The college would charge a fee to the rich students that is above the monopoly price and to poor students one that is lower than the monopoly price. Such price discrimination will make it possible to enrol a larger number of ‘deserving’ students (i.e. increase output) and provide a socially desirable product to those who deserve it but would not have been able to purchase it at the single monopoly price.

Natural Monopolies The term natural monopoly refers to an industry whose economic characteristics are such that the most efficient use of resources is possible only if the entire market is supplied by a single enterprise. In other words, it refers to an industry which by its very nature is such that competition would lead to wastage of scarce resources. Industries that provide public utility services are generally described as natural monopolies. Services such as electricity, telephones, post and telegraph, water supply, railways, broadcasting, etc. that are essential for most members of the society are called public utility services. The basic characteristics which make public utility industries ‘natural’ monopolies are as follows:

(1)

The proportion of fixed costs to total costs is very high in such an industry. Since the proportion of variable or operational costs is very low, the marginal cost of supplying one extra unit is also very low. This may be illustrated by following examples: (i) The incremental operational cost of supplying one additional unit of electric power to a factory or a household is negligible in comparison to the very high fixed costs of setting up a power plant, laying down transmission lines, installing a power connection, etc. (ii) The additional cost of supplying drinking water to an extra household is very low when compared to the fixed costs involved in constructing a new water supply plant and laying down pipelines. (iii) Carrying one extra letter or a telegram to a village is negligible in comparison to the fixed costs of setting

MONOPOLY AND PRICE DISCRIMINATION

ao |

up of a post office with telegraph cables. (iv) Taking one extra passenger from a town or village by train costs nothing, whereas the fixed costs of constructing railway track, a station, signal posts, etc., are very high. These examples show that in the case of public utilities it makes sound economic sense to allow only one enterprise to supply the entire market. (2)

Because of the predominance of fixed costs In these industries, the economies of scale and optimum scale of production are very high. Marginal and average costs tend to

decline even as output becomes very large. Consequently, given the demand, there is scope for only one enterprise to operate with an optimum scale of production. Allowing more than one enterprise in such an industry would mean that all the competing enterprises would have to operate at sub-optimum scale, with underutilized capacity. It is in this sense that competition becomes wasteful of scarce capital resource in industries supplying services such as electricity, posts and telegraph, telecommunication, railways, etc.

(3)

Establishment of these industries involves huge initial capital investment in fixed assets. As a result, when an enterprise operates below the optimum scale, society is obliged to bear a very heavy burden in terms of the high opportunity costs of unexhausted scale economies and underutilized fixed assets.

The above three characteristics show why the supply of public utility services is economically viable only when they operate as a monopoly and establishment of more than one competing enterprises would result in a wastage of resources. This explains why, in the case of public utilities, new entry Is generally prohibited by law. Enterprises supplying public utility services, such as electricity, railways, etc., are very often owned and managed by the government because their monopoly rights are legally protected and because these services are essential to enable the vast majority of households and enterprises to carry on their normal activities. State ownership of public utilities is justified on the ground that it safeguards public interest against monopolistic exploitation. In

332

MICROECONOMICS

FOR MANAGEMENT STUDENTS

countries like the USA and even India, however, there are many

examples of public utility enterprises that are privately owned and managed (e.g. Ahmedabad Electricity Company, Bombay Suburban Electric Supply, etc.). In all countries (including the USA), regardless of whether they are government or privately owned, the prices charged by public utility monopolies are regulated either by some autonomous statutory authority or by the government. For instance, in the USA the pricing of electricity companies is regulated by the Federal Energy Regulating Commission under the Public Utility Regulating Policy Act, while in India this task is performed directly by the state governments concerned. Even those economists and politicians who are against state intervention in economic activities tend to concur with the view that public utilities as natural monopolies should not be left free to charge a profit maximizing monopoly price. They also agree that the price of a public utility service should be equal to the cost of production and that it should not include any element of ‘excess’ profit, i.e. profit in excess of ‘fair’ or ‘normal’ profit paid as cost of entrepreneurship. This proposition is based on the view that monopoly profit is in the nature of ‘rent’ or unearned surplus that results from scarcity. Hence, inclusion of any profit in excess of ‘fair’ profit in the monopoly price would imply ‘taxation without representation’. However,

the question of which cost should be the basis for

determining the public utility price has been a subject of controversy amongst economists. There are two views on this subject. One view holds that the price of a natural monopoly or public utility should be so determined that it is equal to the marginal cost. The other view is that the price charged by a public utility should be equal to the average cost. The reason for using marginal cost as the basis for pricing a

public utility service is that such a price would have prevailed under the ‘ideal’ situation of perfect competition. Also, when price is equal to marginal cost, the supply curve of the enterprise becomes equivalent to its marginal cost curve. Consequently, as is the case under perfect competition, the price based on marginal cost brings about an equilibrium between demand and supply. (When price is greater than marginal cost, output is less than the output which would equate demand with supply. The opposite is

MONOPOLY

333

AND PRICE DISCRIMINATION

the case when price is lower than marginal cost.) The advocates of marginal cost pricing also argue that in a situation where the total demand is less than the optimum scale of production, and economies of scale are very high, output would increase and costs would decline when price equalled marginal cost. This is shown in Fig. 11. Here AC and MC are the (long run) Price, AC, MC

MC ZAC

O

Q,

=> ouput

'

—— Q»

(Os,

Figure 1]

average and marginal cost curves. MC cuts AC at the optimum output, Q,,. The market or total demand curve DD,, is located to

the left of Q,. The demand curve, DD, cuts MC at point M. If price is equal to MC then output is Q,, which is larger than output Q, which would be produced if price was made equal to AC. This additional output leads to increase in welfare, as shown by Area

ANM.

For this additional output Q, Q,, the portion AM of the

demand curve is lying above the portion NM of the MC curve, which means that the additional amount buyers are willing to

334

MICROECONOMICS

FOR MANAGEMENT STUDENTS

pay for Q, Q, units is greater than the extra amount it costs to produce this extra output. Note that the additional amount buyers are willing to pay is indicative of the additional consumer satisfaction whereas the additional cost indicates expenditure of extra resources for incremental output. When the former exceeds the latter, there is a net gain in welfare. However, when price is equal

to MC, the total revenue at output Q, is given by Area OPMQ,

which is equal to OP x OQ,. And this revenue is less than the total cost of this output, as shown by Area OCKQ, which is equal to Q,K x OQ,. Thus, in a situation where DD, cuts AC at output less than the optimum output, the problem of marginal cost pricing is: who will bear the burden of paying for the excess of total cost over total revenue? On the other hand, we find that marginal cost pricing would result in total revenue being greater than total cost when the total demand is greater than optimum scale output, i.e. when the relevant portion of the market or total demand curve cuts AC to the

Price,

AC, MC

O

|

Oh

Figure 12

Q,

Q;

MUNOPOLY AND PRICE DISCRIMINATION

335

right of optimum output, Q,. This is shown in Fig. 12. Here the demand curve cuts MC at output Q,, which is greater than optimum output Q_. In this case the price, based on marginal cost, is OP which is above the average cost of producing output Q, as given by Q,K. Hence, the total revenue given by Area OPMQ, (which is equal to OP x OQ,) exceeds the total cost of producing Q, given by Area OCKQ, (which is equal to OQ, x Q, kK). In this case the price includes an element of ‘excess’ monopoly profit which, as noted earlier, is not desirable in the case of public utility

pricing because such profit is in the nature of rent or unearned surplus. Moreover, in this case output Q, is less than output Q, which could be produced if price were to be made equal to AC. The argument in favour of average cost pricing is that whatever may be the demand situation, total revenue would be equal to total cost. Hence there would neither be any subsidy nor any excess profit. However, the problem that is faced in average cost pricing is that it leads to either ‘underproduction’ or ‘overproduction’. When total demand is less than optimum output, average cost pricing results in ‘underproduction’, in the sense that some buyers will remain unsatisfied even though what they are willing to pay for additional output exceeds what it would cost to produce that output. As already explained by Fig. //, for additional output Q, Q,, what buyers are willing to pay is higher than the marginal cost of producing that extra output. Hence, when price is equal to AC, welfare is less than what it would be if price were to be equal to MC to satisfy extra buyers. On the other hand, when

the demand

curve cuts AC to the

right of optimum output Q,, we find that average cost pricing leads to ‘overproduction’. Fig. 12 shows that when price is equal to AC, for the extra output Q, Q, the portion MN of MC curve is lying above the corresponding portion MA of the demand curve.

That is, for the extra output Q, Q,, the amount that buyers are willing to pay is less than the additional cost (MC) of producing this additional output. That is, the additional! consumer satisfac-

tion obtainable from Q, Q, is less than the additional resources spent on this extra output. In this sense there is ‘overproduction’ and loss of welfare. We thus find that even though average cost pricing is simpler to apply and understand, it also has adverse welfare implications.

336

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Another problem faced in average cost pricing is the appropriate rate of profit to be included in the average cost to provide for the ‘normal’ or ‘fair’ profit margin to pay for the cost of entrepreneurship. Once this fair or normal profit rate is decided, there is the

additional problem of finding out whether or not the capital costs to which this fair rate of return is applied has been overestimated by the promoters. All these issues create complex practical problems for policy-makers, particularly in developing countries if the public utility enterprises are foreign-owned. In short, public utility pricing policy involves many controversial issues and there is‘no generally acceptable solution to this practical problem.

Monopoly in the Real World Enterprises supplying public utility services are perhaps the only cases of real world monopolies closely resembling the theoretical definition of monopoly. This, however, does not mean that monopolies are not to be found in industries other than public utilities. Even when a particular firm is not the only producer of a specified product, it may exercise effective monopoly power if it controls a predominant part, say three-fourths or more, of the total supply of that product and if the rest of the market (i.e., one fourth or less of total demand) is being supplied by a few relatively small firms. Therefore, to determine the existence of monopoly in the market for a particular product it is necessary to find out: (i) whether the largest producer of that product controls at least 75

per cent of the total output; (ii) the number of firms contributing to the rest of total output, and (iii) whether any of the smaller firms is a significant minor producer supplying about 10 per cent or so of the total output. Before establishing these facts care must be exercised in defin-

ing the particular product and finding out the conditions under which imports of that product are allowed. Ifthe productis defined too broadly as, say, food products or electrical equipment, then this would underestimate the extent of output concentrated in the hands of the largest producer and hence its monopoly power. On the other hand, if it is defined too narrowly as, say, glucose biscuits

MONOPOLY

AND PRICE DISCRIMINATION

aot

or autosheet (thermostat) electric kettle, then this would miss out on the fact of the availability of close substitutes. Consequently, there would be overestimatation of the degree of concentration and monopoly power of the producers of these narrowly defined products. Secondly, the figure of the proportionate share of a firm in the total domestic output of a specified product would overestimate its market power if the imports of that product were not subject to quota controls (through import licensing) or high tariff duties.

In short, although

in economic

theory the definition

of

monopoly is quite simple, identifying monopolies in the real world is a very complex task. Monopolistic price-output behaviour may even exist in a market consisting of a few or many suppliers, none of which controls a predominantly large share of the market and product differentiation is insignificant. In other words, even though the structure of a market may be competitive, its behaviour can be monopolistic if the competing firms voluntarily, by explicit or implicit agreement decide not to compete. For example, when competing independent suppliers of a particular product agree to quote a common price or not to quote a price below a certain minimum level they have voluntarily agreed to behave as a mono-

poly, acting as though they constitute a single supplier. Explicit agreements or tacit arrangements (collusion) by which competing firms decide not to compete with one another are commonly known as cartels. The Organization of Petroleum Exporting Countries (OPEC) is a well-known international cartel that rudely shocked the economies of the oil-importing countries by drastically hiking oil prices in 1973. In India, the rubber tyre producers had been alleged to have formed a cartel. As explained later in Ch. 16, in many industries rival oligopolistic firms suppress competition through the tacit arrangement of price leadership. Competing firms in an industry may avoid competition and follow monopolistic price behaviour by opting for maximization of joint (combined) industry profits rather than maximization of individual profits. Thus, in the real world, we find that business firms voluntarily adopta variety of restrictive trade practices to eschew competition and behave like a monopoly through concerted action. Besides the examples mentioned above, there are many other trade practices restricting competition among rival firms.

338

MICROECONOMICS FOR MANAGEMENT STUDENTS

Regulation of Monopoly We have seen that in comparison to free and intense competition, monopoly is considered to adversely affect economic welfare and technical progress in the following ways:

(a)

(b)

When a particular product is produced under conditions of monopoly its price is higher and output lower than its price and output would be if it was produced under conditions of free and keen competition. Since a competitive firm is a price-taker, it can make higher profits than its rivals only by shifting downwards its long run average cost (LAC) curve through increased efficiency. Also, a firm will be priced out of the market if its unit cost is higher than that of its competitors.

(c)

That is, under free and keen

competition super-normal profits are cost-determined. In contrast, the monopoly firm is a price-maker and so it can restrict supply, raise the price and make super-normal profits even though its costs may be high. That is, under monopoly, profits are price-determined and not cost-determined. In order to ensure long-run, stable profits a monopoly firm uses its power and resources to obstruct the entry of new, relatively low-cost producers and thereby thwart innovations.

It is on account of these unfavourable effects of monopolies and trade practices restricting competition that anti-monopoly laws have been enacted and enforced in free-market, capitalist economies. Even the staunchest believers in laissez-faire are not opposed to strict enforcement of anti-monopoly laws because they are aware that: (i) in a capitalist economy, free and unrestricted competition alone can ensure the highest degree of social welfare and optimum allocation of resources, and (ii) state intervention becomes necessary when monopolies and restrictive trade practices lead to loss of consumer welfare and inefficient use of resources. In the USA, laws prohibiting actions resulting in monopolization of a market or restriction of competition have been in force almost for a century. They are known as anti-trust laws because the objective of the first antimonopoly law called the Sherman

MONOPOLY AND PRICE DISCRIMINATION

339

Anti-Trust Act of 1890 was to deal with the trust and holding companies that had been created with the purpose of monopo-

lizing markets. In the UK, in Japan, anti-monopoly World War. In India, the tive trade practices came

in some West European countries, and laws were instituted after the Second law to regulate monopolies and restricinto force in 1970.

There are two basic approaches in dealing with the problem of legal control over monopolies and restrictive trade practices. The per se or American approach is based on the belief that free competition alone can serve the wider public interest and so any action to create monopoly (or enhance monopoly power) and any practice to restrain competition are per se or intrinsically prejudicial to public welfare. Therefore, the task before the court of law is only to establish whether the said action of a business enterprise would lessen competition and/or enhance its monopoly power. The court would not be concerned with the potential effects of such an action or practice on the public interest because any diminution of competition is considered inherently or per se against the public interest. Thus, under American anti-trust laws all actions such as mergers and takeovers that are likely to create or increase monopoly power and all practices leading to restriction of actual or potential competition are treated as anti-trust offences punishable with fines, damages, and even imprisonment. The American per se approach to the control of monopolies is concerned with the structure of markets and conduct or behaviour of firms; they are not concerned with the effects of this on the ‘performance’ of the concerned enterprises. Without citing any specific action or any type of practices, the Sherman Act of 1890 makes illegal any attempt to ‘monopolize trade’ and any ‘combination or conspiracy in restraint of trade’. The Clayton Act of 1914 is less general and prohibits the

following actions or practices if they restrict competition. (i) Tying contracts

(e.g. tying up purchase

of computer

stationary with

purchase of computers). (ii) Price discrimination. (iii) Exclusive dealings (i.e. a retailer being obliged to stock and exclusively sell the products of a particular company). (iv) Interlocking of directorships of different companies. (v) Merger of competing companies. The Federal Trade Commission Act outlaws ‘unfair methods

340

MICROECONOMICS FOR MANAGEMENT STUDENTS

of competition and advertising that is deceptive and/or makes false claims. The alternative approach to legal control of monopolies and restrictive practices is known as the ‘case-by-case’ or the ‘abuse’ approach adopted by the British anti-monopoly laws. Here, the law is non-committal with regard to whether monopolies and restrictions on competition are harmful or beneficial to the public interest. It requires that each case of monopoly or restrictive trade practice be examined on its own merits in relation to the criteria of public interest laid down in the law. If, in the context of these criteria, a particular case of monopolistic behaviour or restrictive practice is found to be prejudicial to public interest then only that particular form of behaviour or practice is required be discontinued. Under the British approach, monopolistic actions and practices restricting competition are neither good nor bad per se. In some situations they may harm the public interest but in others they may be beneficial. Hence, each case must be examined and

tested against the criteria of public interest as applied to a specific context, and only if it is adjudged to be detrimental to the public should it be prohibited. The Indian Monopolies and Restrictive Trade Practices Act of 1970 is based on the British ‘case-by-case’ approach. Following an amendment introduced in 1991, the MVRTP Act sets out the following three objectives:

(1) (2)

(3)

To control such monopolistic trade practices as are found to be harmful to the public interest. Tocontrol those restrictive trade practices that are adjudged to be prejudicial to the public interest. To disallow those unfair trade practices that are proved to be against the public interest.

Under this law, an independent semi-judicial body known as

MRTP Commission has been established to facilitate the implementation of the MRTP Act. In relation to the first of the objectives,

the MRTP Commission has powers of investigation but it has no power of enforcing its verdict. That is, it merely recommends the remedial action that the government may or may not enforce. But in relation to the second and third objectives the commission is akin to a court of law with suo moto power of investigation as well

MONOPOLY AND PRICE DISCRIMINATION

341

as power to enforce remedial action. The MVR7TP Commission takes up for investigation any case of monopolistic or restrictive or unfair trade practice on which it has received a complaint from the public or on its own initiative (suo moto). All restrictive trade agreements are subject to compulsory registration. | The Commission examines each case of monopolistic or restrictive trade practice in the context of the criteria of public interest laid down in the Act. If a particular practice is found by the Commission to be prejudicial to the public interest it issues a ‘cease and desist’ order. Those cases of monopolistic practices and restrictive trade agreements which have been adjudged by the MRTP Commission as not being detrimental (i.e., beneficial) to the public interest are allowed to continue. Similarly, those cases of monopolistic and restrictive practices that are known to exist, but which have yet not been taken up for investigation and inquiry by the MRTP Commission are also allowed to continue. Thus the Indian law is permissive and the businessmen are not certain whether or not a particular practice adopted by them will

_be struck down American

by the MRTP Commission.

In contrast, under

law there is no such uncertainty; businessmen

know

for certain that if any of their actions restricts competition or increases monopoly power, it will be treated as an anti-trust offence. The superior performance of the American per se approach compared to the British and Indian case-by-case approach is clearly suggested by the fact both in india and England an overwhelming majority (about 90 per cent) of the cases of monopolistic and restrictive practices, when investigated, were proved to be detrimental to the public interest. Thus experience shows that, barring a few rare, exceptional cases, most cases of monopolistic and restrictive practices are per se detrimental to public welfare.

Chapter 14 Monopolistic Competition

Introduction

Monopolistic competition is a form of market structure in which a large number of independent firms are supplying products that are slightly differentiated from the point of view of buyers. Thus, the products of the competing firms are close but not perfect substitutes because buyers do not regard them as identical. This situation arises when the same commodity is being sold under different brand names, each brand being slightly different from the others. For example, Lux, Lyril, Rexona, Hamam, Glory, etc. brands of toilet soap, or Colgate, Cibaca, Prudent, Promise, etc.

brands of toothpaste. Here, so far as the individual buyers are concerned, the ‘product’ does not mean ‘toilet soap’ or ‘toothpaste’ but a particular brand of soap, say Lux, or a particular brand of toothpaste, say Promise. This is so because, except for

the rare buyer who has never used soap or toothpaste, or a foreigner in India who has no knowledge of Indian brands of soap or toothpaste, any buyer wanting to buy soap or toothpaste will ask the shopkeeper not for a soap or toothpaste but for the particular brand he prefers. For instance, he will ask for ‘Lyril’ soap or ‘Colgate’ toothpaste. Each firm is therefore the sole producer of a particular brand or ‘product’. It is a monopolist as far as a particular brand is concerned.

However,

since the various

brands

are close sub-

stitutes, a large number of ‘monopoly’ producers of these brands are involved in keen competition with one another. This type of market structure, where there is competition among a large num-

ber of ‘monopolists’ is called monopolistic competition. The theory of Monopolistic Competition was developed by Edward Chamberlin of Harvard, USA, in 1933 (E.H. Chamberlin, The Theory of Monopolistic Competition, Harvard University Press, Cambridge,

Mass., 1933).

343

MONOPOLISTIC COMPETITION

The differentiation among competing products or brands may be based on real or imaginary differences in quality. Real dif ferences among brands refer to palpable differences in quality such as shape, flavour, colour, packing, after sale service, war-

renty period, etc. In contrast, imaginary differences mean quality differences which are not really palpable but buyers are made to imagine or are ‘conditioned’ to believe that such differences exist and are important. Advertising often has the effect of making buyers imagine or believe that the advertised brand has different qualities. This happens when the quality characteristic advertised is such that buyers cannot check its existence or are unaware of its usefulness. For example, a toothpaste brand advertises that its toothpaste contains a substance called fluoride. But a lay person as a consumer cannot check whether it really contains this substance. Nor is he/she aware of the usefulness of ‘fluoride’ for dental health. The advertisement, however, persuades him/ her to believe that this brand is qualitatively different from other brands of toothpaste. Similarly, advertisements for ‘solid state multi-channel’

television,

detergent with

lemon,

or toilet soap

with ‘cream’. When there is product differentiation, each firm has some degree of control over price. Since products are sold under brand names, the buyers become attached to particular brands. Consequently, a commodity is known and bought on the basis of its brand name rather than its commodity or generic name. The customer asks the shopkeeper for Lux or Hamam and not just for a toilet soap. If, therefore, there is a slight increase in the price of a particular brand of soap, say, Lux then the demand for it will not decline to zero because other brands are only close substitutes and not perfect substitutes. The buyers whose most preferred brand is Lux would rather pay a slightly higher price for Lux than buy another brand they prefer less. The same is true for the other

brands of soap. The producer of each competing brand has, therefore, some control over the price of his product and the extent of his power to control price depends upon how strongly the buyers are attached to his brand. As a result, under monopolistic competition, the demand

or

average revenue curve of an individual firm is a gradually falling

344

MICROECONOMICS

curve.

FOR MANAGEMENT STUDENTS

It is highly elastic but not perfectly so. Therefore,

the

marginal revenue curve of the firm is also falling and lies below the average revenue curve at all levels of output. It is in this respect that monopolistic competition differs from perfect com-

petition.! In addition to product differentiation, the other three basic characteristics of monopolistic competition are:

(i)

There are a large number of independent sellers (and buyers) in the market. (ii) The relative (proportionate) market shares of all sellers are insignificant and more or less equal. That is, seller-concentration in the market is almost non-existent. (iii) There are neither any legal nor any economic barriers against the entry of new firms into the market. New firms are free to enter the market and existing firms are free to leave the market. In other words, product differentiation is the only characteristic that distinguishes monopolistic competition from perfect competition. Firms selling slightly differentiated products under different brand names compete not only through variations in price but also through variations in product quality (product variation) and changes in advertising or selling costs. Thus, under monopolistic competition, an individual firm has to maximize profits in relation to variations in three policy variables, namely, price, product quality, and selling costs. (In contrast, under perfect competition there is competition only through price variation.)

Selling Costs, Advertising, and No-Price Competition Selling costs refer to costs of advertising and sales promotion. Conceptually, selling costs are different from production costs. The ! Under perfect competition, since the products of competing firms are homogeneous and perfect substitutes, a slight increase in the price by one firm will reduce its demand to zero. Since the individual firm has no control over the price, the demand or average revenue curve is perfectly elastic and parallel to the X-axis.

MONOPOLISTIC COMPETITION

345

latter refer to costs incurred to satisfy existing demand by creating ‘form’, ‘place’, or ‘time’ utility. That is, production costs relate to costs incurred in satisfying existing demand in the following ways: (a) by transforming one thing (e.g. clay) which has no (or less) utility into another thing (brick) which has utility (or has high utility), or (b) by transporting a good from a place (factory) where it has no utility to another place (city) where it has utility, or (c) by storing it (e.g., blankets) at a time (e.g., summer) when it has no utility up to a time (i.e., winter) when it acquires utility. Selling costs, on the other hand, refer to costs incurred in changing the nature of existing demand by changing the shape (i.e. the slope) of the demand

curve

and/or by creating new

demand and shifting the position of the demand curve. By increasing the attachment of buyers to particular advertised brands, advertising costs reduce the elasticity of demand and make the demand curve steeper than what it would have been without advertising. Moreover, when the distinguishing quantities and the price of any brand of a product (e.g. washing machine or cornflakes) are advertised, the awareness about the usefulness and affordability of a product increases. As a consequence, new buyers enter the market, shifting the demand curve to the right. Advertising has two facets: informative and competitive. The informative aspect is more important in the case of innovative products. That is, new types of products that satisfy an existing want qualitatively differently (e.g. battery operated quartz watches in place of mechanical watches) and new types of products that create a completely new want (e.g. radio or television when first introduced). With regard to such innovative products, the purpose and effect of advertising is to inform the audience about their | usefulness and affordability with the objective of a creating a new market or filling a ‘gap’ in the existing market. Competitive advertising, on the other hand, is more important

for products whose been established.

usefulness and acceptability have already

Here, the advertiser’s aim is to convince the

audience that the advertised brand is superior to other competing brands of the same commodity, e.g. advertisements for various brands of soap, toothpaste, etc. Continuous repetition of this message is supposed to impel buyers get strongly attached to particular brands. The purpose and effect of such advertising is

346

MICROECONOMICS

FOR MANAGEMENT STUDENTS

to increase the degree of product differentiation, reduce price elasticity of demand for advertised products, and thereby increase the monopoly power of the advertiser firm. Over time, this will also have the effect of raising the barrier to new entry competition. As already mentioned, under monopolistic competition an individual firm can compete with its rivals by slightly changing the quality ofits product or increasing its advertising expenditure while keeping its price constant. The kind of competition among the brands of a commodity where firms do not change their prices but only vary the quality of their products and/or change the selling costs is called non-price competition. Thus non-price competition is a distinguishing characteristic of monopolistic competition.

Assumptions in Analysing Firm Behaviour We analyse the conditions and process of long run equilibrium under monopolistic competition with the assumption that competing firms keep their selling costs and product quality constant and compete only through price variation. We then assume that:

(a)

The demand curve of each individual firm has the same shape (elasticity) and position (distance from the y-axis). That is, we assume

the demand

curves

of all firms to be

symmetrical. This assumption implies that market share of every firm is the same and equal to a constant proportion of total market demand. That is, if total market demand is Q and an individual firm’s demand is g then g = KQ, where K_

(b)

is a constant fraction for all firms. The cost curves, both average and marginal, are symmetrical for each firm.

Chamberlin admits that these two assumptions are ‘heroic’ or unrealistic but he needs to make them for logical convenience in order to analyse the long run equilibrium of a typical firm under monopolistic competition. When firms are competing only through price changes, there are three cases of long run equilibrium of a typical firm under monopolistic competition.

347

MONOPOLISTIC COMPETITION

Case 1. Case 2. Case 3.

When competition takes place only through the entry of new firms. When competition takes place only through price variation (price cutting). When competition arises through price variation and new entry.

Case 1: Long Run Equilibrium through New Entry Competition ent Under monopolistic competition, the number of independ firms selling differentiated products or brands of a given comis modity is large and the relative market share of every firm et insignificant. Therefore, the entry of a new firm into the mark will not have any noticeable adverse effect on the sales (or demand) of any of the established firms. Established firms will have disno reason to react to new entry by adopting practices to c) courage this. Moreover, there are no legal or non-legal (economi barriers against new entry. Hence, when high profits of the existing firms attract new entry, new firms will in fact enter the market. Price, MR, SAC, SMC, LAC, LMC

ce IMIG sha SS

\

D

SAC 139

a"

;

LMC

|

, LAC

E,

Re

i i

|

output

348

MICROECONOMICS

FOR MANAGEMENT STUDENTS

The process by which competition from the entry of new firms leads an individual firm’s long run equilibrium is explained with the aid of Fig. /. In Fig. J, the initial downward sloping demand curve of the firm is DD, and MR, is the corresponding marginal revenue curve. SMC and SAC are the short run marginal cost and short run average cost curves. We see that the SMC curve cuts MR, from below at point £,. The firm maximizes profits at output Q, and charges price OP, or Q,D. At Q, output SAC = OC,. It makes super-normal profits = area P,DKC,. The super-normal profits of existing firms induce new firms to enter this market. As the number of firms and brands increases, the market share of each firm declines and each firm

is able to sell less at the same price. Hence, the demand curve of every individual firm slides downwards, remaining parallel to itself. This process of competition from new entry continues so long as the profits earned by a typical firm are more than normal, i.e. so long as the demand curve lies above the AC curve. : The competition from new entry will stop and every firm will reach its /ong run equilibrium output when profits are only normal and price is just equal to long run average cost. This happens when the demand curve ofan individual firm becomes DD,, which

is at a tangent to the LAC curve at point £,. The marginal revenue curve MR, corresponds to demand curve DD,. Here LMC cuts MR, from below at point G at output Q,. Thus, the maximum profit that each firm can earn is only normal profit which is included in LAC. The point of tangency E,, is therefore the position of the long run equilibrium ofa firm where output is Q, and price is P,. When there is competition only from new entry, the long run equilibrium of the firm under monopolistic competition is reached under the following conditions:

(i) Price =AR=LACG= OP, (in Fig.dy} Gi). MR = LMC= GO, (inefie 1): (iii) Maximum Profit = Normal Profits. 2 Note that we assume the market or total demand curve (i.e. the demand curve for the commodity, e.g. toilet soap) to remain unchanged. Also, we have already assumed that at each given price demand for an individual firm is a constant fraction of total demand. When the number of firms increases, this fraction declines, but the slope of demand curve remains unchanged.

MONOPOLISTIC COMPETITION

349

However, because the firm’s demand or average revenue curve is falling, the price is higher than marginal revenue. Hence, under monopolistic competition, even though the long run equilibrium price is = LAC, it is greater than LMC. This is because, at equilibrium, MR = LMC but price is > MR. (Under pee competition Price =Minimum LAC= LMC.) Moreover, since the firm’s demand or average revenue DD, is falling on account of product differentiation, it can be a tangent to the U-shaped ZAC curve only when LAC is also falling. As shown in Fig. J, the long run equilibrium position £, will be at a point which is to the left of the minimum LAC. Thus, the long run equilibrium output Q, is less than optimum output, Q,, (where

LAC

is at its minimum).

The

difference

between

Q, and

Q, =(OQ,,- OQ,) shows the extent of excess or underutilized capacity. Equilibrium with excess capacity is therefore the necessary consequence of product differentiation and monopolistic competition.

Case 2: Long Run Equilibrium when Competition is through Price Variation

For the purpose of explaining the process of competition through price changes, Chamberlin used the device of two demand curves for every individual firm. Chamberlin argues that, since the group of sellers supplying differentiated products (or brands) of a given commodity consists of a large number of firms and since the market share of each firm is insignificant, every individual firm is justified in making the assumption that when it changes its price the sales of any of its rivals will not be noticeably affected. Every firm assumes that when it changes its price, its competitors will not react by changing their prices and so the prices of its competing or substitute products will remain unchanged. The changes in demand resulting from the changes in price undertaken on the basis of this assumption is shown by what is

called the ‘perceived or ‘assumed or ‘imagined’ demand curve of he firm. The assumption that its competitors will not follow suit when it reduces its price leads the firm to expect that the increase

390

MICROECONOMICS

FOR MANAGEMENT STUDENTS

in its demand will be proportionately greater than the reduction in its price. The perceived demand curve is therefore highly though not perfectly elastic. lt falls, but falls very gradually, showing why a firm is induced to cut its price. It is the decision-making demand curve because the firm decides to cut price on the basis of the change in demand it perceives or assumes to occur as a result of the change in price. However, because every firm’s market share is equally insignificant, each firm acts on the assumption that when it lowers its price, the prices of its competing firms will remain constant. Each firm therefore reduces its price on the basis of the same assumption, and consequently all firms in the market reduce their prices simultaneously but independently (i.e. not in retaliation). Each firm acts on the basis of its perceived demand curve. As a result, the actual increase in demand resulting from a reduction in price is much less than has been ‘imagined’ by each firm. The actual changes in demand arising from such simultaneous reduction in Price

O

|

M, N Figure 2

M,

Demand

MONOPOLISTIC COMPETITION

oat

price by all firms is shown by what is called the actual demand curve of an individual firm. Fig. 2 shows dd, as the assumed or perceived demand curve

and DD, as the actual demand curve. When price is lowered from P, to P, the firm assumes the demand to increase from M, to M,, but as is shown by DD, it actually increases only to M,N. The assumed demand curve is much more elastic than the ‘actual’ demand curve. This is because the former ‘assumed’ or ‘perceived’ changes in demand based on the assumption that only one firm changes its price, while its competitors keep their prices constant. The actual demand curve, however, shows the real chan-

ges in demand when all firms simultaneously but independently change their prices acting on the basisof same assumption. The actual demand curve DD, is also known as the firm’s ‘shareof-the-market’ demand curve because it shows the market share of a firm when all firms are changing (lowering) their prices simultaneously. Its position or distance afrom the Y-axis depends upon the number of firms in the market. As this number increases, the actual demand curve DD), gets closer to Y-axis because the absolute market share of each firm diminishes. And as the number of firms decreases, DD, moves to the right and further away from the Y-axis. Fig. 3 explains the process by which a typical individual firm’s long run equilibrium is determined when there is competition only through price variation. In Fig. 3, dd, is the perceived or assumed demand curve. DD, is the actual demand curve, AC the average cost curve, and MC

the corresponding marginal cost curve. Suppose the initial price

is P, at which initial output is Q, and C, is the average cost. Super-normal profits are therefore = P,R,K,C,. Since dd, is highly elastic, each

firm assumes

that it can

further increase

its total

revenue and profits by reducing its price from P, to P’ and increase demand by FM, as shown on dd,. However, every firm simultaneously reduces its price on the basis of the same assumption. The actual increase in demand for the firm is only FN, as shown by point NV on the actual demand curve, DD.. We therefore assume that guided by the perceived demand curve, dd, every firm will continue to reduce its price even though it finds that the actual increase in demand is smaller than the assumed or expected increase.

392

MICROECONOMICS FOR MANAGEMENT STUDENTS

Price, MR, AC, MC

S

MC AC

E,

ee

i

a Q,

Q,

d,

.

output

ae

Figure 3 The price at which the perceived demand curve dd intersects the actual demand curve DD shows the price at which assumed or expected demand is equal to the actual demand. Following Chamberlin, we assume that every firm, because it is insignificant and acts independently, is myopic (short sighted) but not irrational. Each firm is guided in its decision-making by the perceived demand curve dd and not its actual demand curve DD. The process of price cutting by every firm on the basis of its perceived or assumed demand curve dd and the simultaneous price reductions by all firms along the actual demand curve DD, continues as long as the firm finds that its assumed or perceived demand curve is above the AC curve and perceived profits are more than normal. The firms will stop competing through price reductions only when they find that perceived as well as actual profits are reduced to the normal level. In Fig. 3, this happens when the perceived demand curve dd

MONOPOLISTIC COMPETITION

353

becomes a tangent to the AC curve, i.e. when it becomes dd, which is at a tangent to AC at point E,. Note that at point £, sie actual demand curve DD, intersects AC. At point £, price = average cost and profits are wea on the basis of the epeeiven demand curve dd, and also on the basis of the actual demand curve DD. in Fig. 3, the marginal revenue curve mr, Paecspenas to dd, For profit maximizing conditions, we have mn consider the marginal revenue corresponding to perceived demand curve, dd. because, as has already been stated, the firm’s decision-making is based on the perceived demand curve. The MC curve cuts the mr, Curve from below at point G. The long run equilibrium output

is theretéte Q, and the long run equilibrium price is P,. At Q,, the firm’s profits are maximized though the maximum spelits are only normal profits. Under monopolistic competition the conditions of the long run equilibrium of a typical fi-™ when competition is through price variation are as follows:

1. 2.

The perceived demand curve, dd is a tangent to AC. MC curve intersects frorn below the marginal revenue curve derived from that perceived demand curve which is a tan-

gent to AC. (Here, MC cuts mr, from below at G in Fig. 3). Thus profits are maximized. Price = AR = AC= OP, (in Fig. 3). (Here AC stands for LAC)

3.

4.

..

| i

Profits = Normal. Profits are at their maximum but only normal. The actual demand curve DD, cuts the AC curve at the a point where perceived curve dd, is a tangent to AC. (DD, cuts AC at E, where dd, is a enna to AC in Fig. 3.) The price is greater than MC. This is because at equilibrium iC = mr, and mr, is < Price. (Since, dd, is a falling curve,

mr, is below dd, at all output levels.) We find that even in Case 2 the long run equilibrium of the firm is reached when AC is falling, i.e. at the output level which is less than optimum. This is because the falling perceived demand curve dd, can be a tangent to U-shaped AC only when AC is falling. In

354

MICROECONOMICS

Fig. 3, the optimum

FOR MANAGEMENT STUDENTS

output is Q,, where AC is at a minimum.

Therefore, the difference between equilibrium output Q, and optimum output Q,, or (OQ, — OQ,) is the excess or underutilized capacity.

Case 3: Competition through Price Variation and New Entry We have seen that the actual demand curve DD shows the absolute market share of an individual firm. Because we assume that the position and shape of demand curve are symmetrical for every firm, the market shares of all firms are assumed to be equal in terms of absolute quantity or size of output. It is given by a ratio of total market demand divided by the number of firms. The larger the number of firms in the market, the smaller the absolute market share of each firm. The position of DD, i.e. its distance from the Y-axis therefore depends upon the number of firms in the market. The actual demand curve DD will shift nearer to the Y-axis as the number of firms increases and will move further away from the Y-axis as the number

of firms decreases.

That is, DD

will shift

towards the left as new firms enter the industry and it will shift towards right when the existing firms leave the industry. As shown in the Fig. 4 the initial, actual demand curve is shown

by DD,. It cuts the AC curve at point J. Let dd, be the initial perceived demand curve cutting DD, at point B,. As explained in Case 2, competition among firms through price variation will con-

tinue until the perceived demand curve dd, becomes dd,, which is a tangent to AC at point £. Point E shows price to be = OP, However, point FE is not situated on the actual demand curve DD,. Hence, the firm finds that corresponding to point £, the actual

demand on DD, is =P,R. Now point k on DD, is above the AC curve. Therefore output P,R indicates super-normal profits shown by area P,RGC,. These super-normal profits induce new firms to enter the industry. As the number of firms increases the absolute market share of each decreases and the actual demand curve

DD, shifts towards the left. This

process

will continue

till DD, shifts to the position

of

DD, which intersects the AC curve at point E where the perceived

355

MONOPOLISTIC COMPETITION Price MR,

AC

MC

D

5,

d

By qd

MC AC

ad =~.

d;

output

Figure 4

demand curve dd, is a tangent to AC. At this point profits are normal on the basis of perceived demand curve dd, as well as on the basis of actual demand curve DD,. That is. actual demand and perceived demand are equal when profits are normal. The point of tangency between dd, and AC is at point E where DD, cuts AC. Here the long run equilibrium output is Q, and price is P,,. Here the competition through price variation is shown by the

downward shifts in the perceived demand curve along the actual demand curve. (From position dd, to position dd,, which is a tangent to AC at point £). And the competition through new entry is shown by the shift in the position of actual demand curve (DD, shifts to the position of DD, which intersects AC at the point of tangency of dd, and AC, i.e. at point E.). Under monopolistic competition, when there is competition through price variation as well as new entry (or exit) the long run equilibrium of the firm will be reached when following conditions are satisfied.

396

(1) (2) (3) (4) (5)

(6) (7)

MICROECONOMICS FOR MANAGEMENT STUDENTS

Perceived demand curve, dd, is a tangent to AC. Price is equal to AC. Maximum Profits = Normal profits (Economic profits = zero). MR=MC. Here the relevant marginal revenue is derived from the perceived demand curve. The actual demand curve (or ‘market share’ demand curve) DD cuts AC at the point where perceived demand curve (dd) is a tangent to AC. Price is > MC because Price is > MR The equilibrium output is less than the optimum output.

Here also we find that the long run equilibrium output is determined at the level where AC is falling and therefore the equilibrium output is less than the optimum output, Q,,. That is, excess capacity exists at long run equilibrium output.

Excess Capacity and Monopolistic Competition We have seen that under monopolistic competition the long run equilibrium output of the individual firm is /ess than optimum output in all cases, namely when there is competition through price variation (Case 2), or when competition is through new entry (Case 1), or when there is competition through price variation and new entry (Case 3). Equilibrium under monopolistic competition results in sub-optimum output with excess or underutilized capacity and unexhausted economies of scale. This excess capacity is due to two reasons. First, under monopolistic competition, product differentiation by competing firms causes the demand curve of the individual firm to slope downwards. Secondly, since the number of competing firms is large, every firm behaves independently, with the result that competition through new entry and price variation continues until every firm earns only normal profits. The typical firm will not reach long run equilibrium until its demand curve is a tangent to AC. In other words, the falling demand curve and long run tangency solution are the reasons which account for excess capacity equilibrium. (Note that at the point of tangency the slope of demand curve is equal to the slope of the average cost curve.

MONOPOLISTIC COMPETITION

B04

But the slope of falling demand curve is negative. Hence the slope of cost curve is also negative.) From the above it follows that price is higher and output lower under monopolistic competition rather than under perfect competition. This is because, under perfect competition, long run equilibrium output is determined at the optimum level and so price is equal to the minimum LAC. In contrast, under monopo-

listic competition, long run equilibrium output is less than optimum and price, though equal to LAC, is greater than the minimum LAC. Monopolistic competition and product differentiation are therefore considered

to result in loss of social welfare because,

in

comparison to perfect competition, consumers pay a higher price and are supplied a lower output. Secondly, under monopolistic competition, firms do not utilize their resources as efficiently as they would if the market was perfectly competitive. Chamberlin argues that this conclusion about the welfare implication of product differentiation and monopolistic competition is not justified because the consumers are willing to make a sacrifice for the wider choice of varieties offered by differentiated products. That is, the loss in welfare suffered by consumers from higher price and lower output is compensated for by the gain in satisfaction they obtain from the choice of varieties offered by competing brands with different qualities. Thus there is no loss of welfare in terms of consumer satisfaction. Critics point out that Chamberlin’s argument in defence of product differentiation is not valid because the consumers are not offered two options from which they have to choose one. They are not asked whether they want to pay a higher price for the lower output of differentiated products offering a wide choice of varieties or they want to pay a lower price for a larger output of identical, homogenous products. The former option is imposed upon the consumers by producers through advertising and sales promotion campaigns that create, in buyers’ minds differences regarding the qualities of competing brands, which are very often imaginary or insubstantial.

Chapter 15 Oligopoly: Characteristics and Models

Introduction

Oligopoly is a situation in which only a few firms (sellers) are competing in the market for a particular commodity.! The distinguishing characteristics of oligopoly are such that neither the theory of monopolistic competition nor the theory of monopoly can explain the behaviour of an oligopolistic firm. These characteristics are briefly explained below.

(1)

(2)

(3)

Under oligopoly the number of competing firms being small, each firm controls an important proportion of the total (industry) supply. Consequently, the effect of a change in the price or output of one firm upon the sales of its rival firms is noticeable and not insignificant. When any firm takes an action its rivals will in all probability react to it (i.e. retaliate). The behaviour of oligopolistic firms is interdependent and not independent or atomistic as is the case under perfect or monopolistic competition. The demand curve of an individual firm under oligopoly is not known and is indeterminate because it depends upon the reaction of its rivals which is uncertain. Each theory of oligopoly therefore makes a specific assumption about how rivals will (or will not) react to an individual firm’s action. In view of the uncertainty about the reaction of rivals and interdependence of behaviour, oligopolistic firms find it advantageous to coordinate their behaviour through explicit agreement (cartel) or implicit, hidden, understanding (col- lusion). Also, because

the number

of firms is small, it is

feasible for oligopolists to establish a cartel or collusive arrangement. However, it is difficult as well as expensive to 1 When there are only a few competing buyers on the demand side of the market, it is called an oligopsomy.

-

OLIGOPOLY: CHARACTERISTICS AND MODELS

(4)

399

monitor and enforce an agreement or understanding. Very few cartels last long, particularly when oligopolistic firms significantly differ in their cost conditions. Under oligopoly, new entry is difficult. It is neither free nor barred. Hence the condition of entry becomes an important factor determining the price or output decisions of oligopolistic firms, and preventing or limiting entry an important objective.

(5)

Given the indeterminancy of the individual firm’s demand and, therefore,

the marginal

revenue

curve,

oligopolistic

firms may notaim at maximization of profits. Modern theories of oligopoly take into account the following alternative objectives of the firm:

Sales maximization with profit constraint. Target or ‘fair’ rate of profit and long run stability. Maximization of the managerial utility function. Limiting (preventing) new entry. Achieving ‘Satisfactory’ profits, sales, etc. Thatis, the firm is a ‘Satisficer’ and not ‘Maximizer’. (f) Maximization of joint (industry) profits rather than individual (firm) profits.

(a) (b) (c) (d) (e)

In view of the fact that the characteristics of oligopoly renders collusion (explicit or implicit cartel) advantageous and feasible, theories of oligopoly are divided into three broad groups, namely, models of non-collusive oligopoly, models of collusive oligopoly, and managerial theories. The important models of non-collusive oligopoly are: (a) Curnot model, (b) Kinked demand curve models, (c) Average Cost or ‘Mark-up’ Pricing model and, (e) Limit pricing model. The two major theories of collusive oligopoly are: (a) Joint profit maximization and, (b) Price leadership. Emphasizing the distinguishing characteristics of joint stock enterprises are the three models of managerial theory, namely,

(a) Sales maximization with profit constraint, (b) Maximization of managerial utility function, and (c) Firm as a satisficer (Behaviourist theory).

360

MICROECONOMICS FOR MANAGEMENT STUDENTS

Models of Non-Collusive Oligopoly

Curnot Model

Antoine Curnot, a French economist, presented his model in 1838 in a book entitled Researches into the Mathematical Principles of the Theory of Wealth. The Curnot model is in terms of duopoly (two sellers) but it can be easily extended to an oligopolistic situation. This model analyses the process of equilibrium in a duopoly situation when each duopolist assumes that his rival will not react when he changes his output to maximize profits.

Specific or Simple Curnot Model Assumptions (1) (2) (3) (4)

There are two sellers in the market. The products sold by these two sellers are homogeneous. The market, or total demand curve, is known and it is a straight line. Each duopolist assumes that his rival’s output will remain constant when he changes his output. Thus, each duopolist assumes his rival will not react to his action. That is, for each

(5) (6)

duopolist the conjectural variation or seller-interdependence, as given by dQ, /dQ, or dQ,/dQ,, is assumed to be zero. (Q, and Q, are the outputs of two sellers). Each duopolist produces output of which the profits are at the maximum. The cost of production is zero for both the sellers. For example, two natural springs of mineral water with healing qualities, each owned by one seller. The average and mar-

ginal costs for each seller are zero and these curves coincide with the X-axis. We explain the Curnot duopoly model with the help of Fig. /. In Fig. 1, CD = known straight line total (market) demand curve. Note that under pure competition, Price = marginal cost which is zero by assumption.

OLIGOPOLY: CHARACTERISTICS AND MODELS

361

Figure 1 Curnot Model

Hence, demand or output at zero price shows the competitive output.

..OD = Competitive output. Let the two duopolists be denoted by X and Y. Let Q,. and Q, be their respective outputs. Suppose seller X enters the market first, followed by seller Y. We analyse the behaviour of X and Yin stages. In Stage I, sellerX acts as a monopolist. He faces demand curve CD so that CA is his marginal revenue curve which must be situated halfway between the Y-axis and demand curve. CA cuts

OD at A, such that OA=AD=12 OD. At output OA, marginal revenue = marginal cost = zero and profits are at their maximum. Seller X charges price P, and makes profit = OARP.. Thus at Stage I, we find Q, = % OD.

Now seller Y enters with the assumption that X will keep his output constant at 12 OD. In other words, Y considers his demand

362

MICROECONOMICS

FOR MANAGEMENT STUDENTS

curve to be RD which shows the leftover demand after X has supplied OA output. — Hence at Stage I, seller Y finds his demand curve to be RD with RB as his marginal revenue curve. RB cuts the X-axis at B. For seller Y, marginal revenue = marginal cost = zero at output AB.

Thus profit maximizing output of Y at Stage I, is AB='12AD= Y (14 OD) = YW OD. Thus in Stage I,

and

= iOA =o 00

Q, = AB = 2 AD= V2 (¥2 OD) = Y4 OD

Seller Y charges price BT = OP, and makes a maximum profit = ABTK. However, since X and Yare selling homogenous products, the price will decrease from OP, to OP, for both of them. Profits of X

will thus decline to OAKP.. Assuming seller Y will keep his output Q, constant at Y4 OD seller X will have to reduce his output so as to raise the price and his profit. In Stage II Seller X will produce profit maximizing output on the basis of the demand leftover after assuming Q, to be = 14 OD. Therefore, in Stage Il, xX will produce output Q,.= 42 (OD — ¥4 OD) = (¥2— ¥8) OD. Note that the leftover demand is = (OD — AB) =(OD- BD) and BD is = 2 AD = V4 OD. At Stage II, Y decides his profit maximizing output assuming Q,. will remain constant at (42 — 8) OD. Hence, Q,, at Stage II will be = 4 [OD—- (2- %) OD].

Thatis,

Q,='%2(OD- 2 O0D+ % OD)

=(Y-

+ Vig) OD = (4 + Mie) OD.

We can carry on this reasoning further, to Stage III, Stage IV,

etc., to find Q, and Q, at each stage. In short, we find that at each stage, seller X will decrease his output in such a way that it will be equal to one-half of OD minus

the output of Y in the previous stage (which is initially zero). On the other hand, Ywill increase his output Q, at each stage so that it will be equal to one-half of the difference between OD and Q, | at the same stage. The stagewise changes in Q,. and ce are summarized below.

OLIGOPOLY: CHARACTERISTICS AND MODELS

Stage Values for Q,

Values of Qy

| in (Q, 2440p

Q, =%(1-%)OD

363

= Y4 OD

I

Q, =% (1-4) OD

Q, =% [1-(%-1/8)] OD = ()- 4+ 1/16) OD

= (4-1/8) OD OD= 3/8

= (744+ 1/16) OD =:(9/ 16)-O0.D

Ml

§=Q,=%[1-(44+1/16)]} OD = (%- 1/8 - 1/32) OD

Q,=% [1-(%- 1/8- 1/32)] OD = (4+ 1/16 + 1/64) OD

We find that at each stage, Q, declines by smaller and smaller quantities. In contrast, om increases by smaller and smaller amounts

at each stage. Both Q. and Q, will, therefore, arrive at

some finite values which will give the equilibrium values of Q, and Q,. — equilibrium values of Q, and Q, are found in the following way:

Q, =[4-1/8 - 1/32... JOD =[-Y (V4)-2 (M)?... JOD =[(Y-Wi(4+ (44)? +(%)....]OD

=—

[(Va)/(1 — Va)] OD

= (2-1/6) OD = (1/3) OD “. Equilibrium

Q,=(1/73) OD

O,= (4s i710 + 1/64)OD =[{[(4+ (“4 +(%)y....] OD 2 Note that the summation of a declining geometric series given by [atartar....]is=a/(l—r), where a =the first term and r is the constant ratio

by which a is changing. Here the series [14 + (%4)* + (4)*....] can be written as [Ya + (Y4)(') + Y4 (Y%)*)... J, so that the first term = % and r=.

364

MICROECONOMICS

FOR MANAGEMENT STUDENTS

= [(Y%)/(1 — Y%)] OD = (48) OD -. Equilibrium

Q, = (3) OD

Hence the total equilibrium output of the two duopolists X and Yis = (Q.+ 4 as ¥3 OD + 3 OD = % OD.

Since OD = competitive output, the duopoly equilibrium output is 44 of competitive output, and the equilibrium output of each duopolist is = 43 of competitive output.

We can write 4% as = (2)/(2 + 1) and Yas = 1/(2 + 1), where 2 is the number of sellers in duopoly.

|

Extending this duopoly case to oligopoly with the number of firms (sellers) to be = NV, we can say that according to the Curnot model, the equilibrium output of each of the WN oligopolists is = 1/(N + 1) x Competitive output. And total equilibrium output of N oligopolistic firms is = V/(N + 1) x Competitive Output.

Generalized Curnot Model In the simple Curnot duopoly model explained above we have found an individual firm’s equilibrium by using the very restrictive assumption that the cost of production is zero. However, the Curnot model can also be generalized to cover

cases where costs for the duopolist firms are not zero. For this purpose we use the device of Reaction Function for each duopolist. The Reaction Function of a duopolist firm shows how it reacts by determining its profit maximizing output for the different given quantities of outputs of its rival firm.

According to Curnot each duopolist firm assumes that its rival firm’s output will remain constant as it changes its output. Hence the Reaction Function (curve) of duopolist Firm X shows the profit maximizing output of X for different given (or constant) levels of output of rival Firm Y. Similarly, the Reaction Function of duopolist Firm Y shows the

OLIGOPOLY: CHARACTERISTICS AND MODELS

365

profit maximizing outputs of Y for the different given (constant) levels of output of X. In Fig. 2, the output of X is shown on the X-axis and output of Y on the Y-axis.

po

a

X,

X,

b

output of X

Figure 2 Curnot Model

Here the lines aX and bY are the Reaction Functions of Firm X and Firm Y, respectively. We find that aX and bYintersect at point C. This point C shows the Curnot duopoly equilibrium.

Let Q, = output of X and Q, = output of Y. Corresponding to point C, the output of X is = X, and output of Yis= Y,. Hence X, = Equilibrium output of Firm X and Y, = Equilibrium - Output of Firm Y.

366

MICROECONOMICS

FOR MANAGEMENT STUDENTS

The process by which this equilibrium is reached is as follows: Reaction Function aX shows that when the given Q, is zero, profit maximizing Q, (i.e. reaction of X) is = OX. And when given S, is = Oa, the profit maximizing Q, (reaction of X) is = zero.

Similarly, Reaction Function bY shows that: when given Q, is = zero, profit maximizing oe (reaction of Y) is = OY. And when given Q, is = OD, profit maximizing Q, (reaction of Y) is = zero. Suppose X starts with Q, = X,, then reaction function bY shows that Firm Ywill react by producing Q, = ¥,. But when given Q, is

= Y,, then reaction function aX of Firm X shows that X will react by producing X,,. Again, when Q, = X,, reaction function bY shows that ot chan-

ges to Y,, but when Q) is given as Y, then aX shows that Q, will change to X,.

We find that output X, is < X, and X, is < X,. Thus Q, goes on decreasing from X, till it converges at point C where Q, = X,. On the other hand, Q, goes on increasing from the initial quantity Y, to Y, to Y,, and so on, till it converges

at point C when

2 ee At point C, where aX and bYintersect one another, we find that, (i) for given Q,=Y,, Firm X reacts by producing output X, as shown by aX, and (ii) for given Q,. = X,, Firm Y reacts by producing ); = Y, as shown by OY. Thus outputs X, and Y, are consistent with one another in the sense that when Q, is X, and Q, LS, Firm X and Firm Ywill have no tendency to change their outputs any further. Point C, where the reaction functions of two firms, X

and

Y, intersect one another, shows the Curnot duopoly equi-

librium when Q, =X, and Q, = Y,.

Kinked Demand Curve Model of Oligopoly There are two versions of the Kinked Demand Curve model. One is called the Sweezy Version and the other is called the Hall and Hitch Version. Both models were conceived independently in 1939. The essential difference between these two versions is that Sweezy’s model is based on the marginalist approach, with the hypothesis that even an oligopolistic firm aims at profit maximization. In contrast, the Hall and Hitch version rejects the marginalist

OLIGOPOLY: CHARACTERISTICS AND MODELS

367

approach of profit maximization. It argues that, under oligopoly, firms aim at ‘fair’ profit and follow the Full Cost principle in

determining the price.” Sweezy’s Model of Kinked Demand Curve According to Sweezy, the most distinguishing feature of oligopoly is that an individual firm does not know (and cannot determine) the exact nature (functional form) of its actual demand curve because of the uncertainty and indeterminacy of rivals’ reactions to its own actions. An oligopolistic firm is therefore guided in its decisions by the ‘imagined’ demand curve which is based on what it expects to be the most likely (probable) reaction of its rivals. Under oligopoly, a firm expects that when it raises its price, it is most likely that rival firms will not follow suit by raising their prices. The rivals will keep their prices constant in order to increase their sales at the expense of the firm that raises the price. Hence, when a firm increases its price, its demand is expected to

fall much more than it would if its rivals were not to keep their prices constant. That is, for upward changes in price, a firm’s demand is expected to be highly elastic. In contrast, when the firm lowers its product price, it is most likely that its rivals will follow suit because if they did not do so they would lose sales to the firm that lowered the price. Hence, when a firm reduces its price, its demand is expected to increase much less than would otherwise have been the case (because its rivals will also reduce their prices). That is, for downward changes in the price, a firm’s demand

curve is expected to be Jess elastic

than it would have been had the firm’s rivals were not to follow suit by reducing their prices. Consequently, for an oligopolistic firm, the demand curve is

3 The Sweezy version was first presented in an article entitled ‘Demand under Conditions of Oligopoly’ by Paul Sweezy of the USA (Journal of Political Economy, 1939). The Hall and Hitch version was presented by two Oxford economists, R.L. Hall and C.I. Hitch in their article entitled ‘Price Theory and Business Behaviour’ (Oxford Economic Papers, 1939).

368

MICROECONOMICS

FOR MANAGEMENT STUDENTS

highly elastic and gradually falling for prices above the current or existing price, and for prices below the current price the demand curve is less elastic and steeply falling. Because of the differences in elasticity (and slope) at prices above and below the current price, the demand curve of the firm

has a corner or a Rink at the current or existing price.

In Fig. 3 the firm’s demand curve is APB, which has a kink or corner at current price P and output ON. The upward segment AP is relatively more elastic than the downward segment PB. That is,

if e, shows the elasticity of AP and e, shows the elasticity of PB, then e, is >e,. In Fig. 3 dotted line PB, shows the decrease in the firm’s demand that would have occurred if the rivals were not expected to keep their prices constant when the firm raised price above P. Dotied line PA, shows the rise in demand if rivals were expected not to follow any fall in price below P. Since the elasticity for a change in price above P is more than, and different from, elasticity for a change in price below P, there are two values of marginal revenue for current price = P. Thus the marginal revenue curve has a discontinuity or gap at price = P. For

Price MR,MC

op ]

A

\

> output

Figure 3 Sweezy’s Kinked Demand Curve Model

OLIGOPOLY: CHARACTERISTICS AND MODELS

369

the upper AP portion of the demand curve the marginal revenue (MR,) curve is QC and for the lower portion PB, the marginal revenue (MR,) curve is DE.

The marginal revenue curve corresponding to APB is shown by QCDE with discontinuity or gap = CD. Note that both e, and

e, have to be > 0 for MR, and MR, to be positive at P. The magnitude (or length) of this P(1/e, — 1/e,). This follows from the fact that find that MR, = Pre, — 1)/e, and MR, = P(e, — Hence MR, — MR, = Ple,e, — e, — Cre ree ey= Pre =e, mde.) Since enise> e,, the gap MR,

gap is given by MR = P(1 — 1/e). We 1)/e,. =e, 7e!, — MR, is positive.

The marginal cost curve, MC of the firm passes through the discontinuous gap CD in the marginal revenue curve QCDE. Though the current, existing price = P is not precisely equal to the profit maximizing equilibrium price (as there is no unique MR at price P), this price P is consistent with profit maximizing, mar-

ginalist equilibrium. For output less than ON we find MC is below marginal revenue and for output more than ON we find MC is above marginal revenue. That is, MC cuts the discontinuous MR curve from below. Since, under oligopoly, demand curve is kinked at the existing price (P) and marginal revenue curve has discontinuity CD at the existing price, any upward or downward shift in the MC curve will not bring about any change in the current or existing price so long as the new MC curve passes through the gap (CD) in the marginal revenue curve (QCDE).

In Fig. 3 the new higher marginal cost curves MC, and MC, are passing through the gap CD with the result that the current price = P continues to be consistent with profit maximization even while remaining constant at the existing level. Thus the most important conclusion of Sweezy’s kinked demand curve model of oligopoly is that price remains unchanged and rigid or ‘sticky’ at the existing level P when, in the short run, the marginal cost increases due to a rise in raw material prices or

hike in wages through trade union pressure. Thus Sweezy’s Kinked demand curve model

explains

the

370

MICROECONOMICS

FOR MANAGEMENT STUDENTS

rigidity or stickness oligopolistic prices in the face of short-term

increases or decreases in variable input costs. When costs of raw materials or labour rise, profits will get squeezed and when these costs fall, the benefit of lower input costs will not be passed on to

the consumers. Thus the Sweezy model of Kinked demand curve under oligopoly explains why prices of oligopolistic firms are inflexible and fail to reflect short run changes in variable costs of raw materials and wages. The principal shortcoming of the Sweezy model is that it does not explain how the existing or current price is determined, and this is a criticism that Sweezy accepts.

Hall and Hitch Version of Kinked Demand Curve

The Hall and Hitch model of the Kinked demand curve is based on an empirical survey of a sample of 38 well managed firms in England. The survey was conducted by these two Oxford economists to find out how firms in the real world determine price and output. The principal findings of the study were as follows:

(a) (b)

Inthe real world, most manufacturing firms operate in oligopolistic markets. Contrary to what is assumed by economic theory, in reality oligopolistic firms do not know their demand curve because of uncertainty regarding their rivals’ reaction. They do no therefore know their marginal revenue curve. Since most large firms tend to be multi-product firms, they also do not

know the marginal cost curve. Thus in the real world, firms cannot determine equilibrium price and output by marginalist calculations, i.e. by equating marginal revenue and marginal costs.

(c)

Oligopolistic firms in reality determine their price on the basis of the Full Cost principle. They charge that price which not only covers variable and fixed costs but also yields a

fair profit margin. The Full Cost is the sum of average variable cost (AVC) and average fixed cost (AFC) at Normal output

OLIGOPOLY: CHARACTERISTICS AND MODELS

371

level and a predetermined percentage of this sum added for ‘Fair’ (reasonable) profit. In short, according to this principle AFC) at Normal Output + ‘Fair’ +C Price = Full Cost = (AV Profits as a percentage of (AFC + AVC). For instance, if normal output is 1000 units, Total Fixed

(d)

(e)

and Total Variable Costs at this output are Rs 8000 and 2000, respectively, and fair profit is considered to be 10 per cent, then Full Cost Price = 8+2+1= 11 Rs/Unit. The demand curve has a Kink at the price which is equal to Full Cost price. If a firm charges a price higher than Full Cost, its rivals will not follow suit but will keep their prices constant. Hence, for prices higher than the Full Cost price, the demand curve of an oligopolist has high elasticity. If the firm charges a price lower than Full Cost price, its rivals will follow suit by lowering their prices. Hence, for prices less than the Full Cost price, the oligopolist’s demand curve has relatively low elasticity. Oligopolistic firms adopt Full Cost pricing rule because it not only covers AFC at normal output but also earns a reasonable rate of profit. The objective of oligopolistic firms is to have long run stable profits and a ‘quiet life’, free from uncertainties. If profits exceed what is regarded as a ‘reasonable’ or ‘fair’ rate, it may attract new entrants and accusation of ‘excessive’

profits from customers as well as distributors. Both these consequences will cause instability of long run profits and make life difficult (unquiet) for firm’s decision-makers. Similarly, charging a price below full cost will be considered ‘unethical’ by competitors and create a threat of price war. Also, it is difficult to raise price later to the full cost level. Thus, for oligopolistic firms, price tends to remain rigid or sticky at the full cost level, and short run changes in costs

and demand will not cause changes in the oligopolistic price.

The Full Cost version of the kinked demand curve is shown by Fig. 4 where ON, = Normal output, P, = Full Cost price and A,P,B, is the eed demand curve. Piaaticitg e, for AP, is greater

than elasticity e, for P,B,. The kink occurs at the full cost price. Thus unlike the Sidaby version, this version explains how the existing price is determined.

372

MICROECONOMICS FOR MANAGEMENT STUDENTS

Price

A,

O

N, Normal output

Notes:

output /

OP= OP, = Full Cost Price at Normal Output. ON, = Normal output, e,fs > e,

Figure 4 Hall and Hitch Version of Kinked Demand Curve

Average Cost Pricing or ‘Mark-up’ Pricing Theory of Oligopoly According to this theory, the objective of an oligopolistic firm is not the maximization of short run profits but sustaining a ‘target’ rate of return on investment in the long run. The firm charges a price on the basis of its average variable

cost (AVC). The price is ‘marked up’ to cover a certain ‘gross’ profit margin above the AVC. This gross profit margin consists of _ average fixed costs (AFC) at ‘normal output level plus the predetermined ‘target rate of profit. The problem faced by an oligopolistic firm in deciding the price on the basis of long run maximum profit is that the Jong run demand curve for an individual firm is indeterminate and unknowable for the following reasons:

(i) Reactions of rival firms to the firm’s action of increasing or lowering price or output cannot be known.

OLIGOPOLY: CHARACTERISTICS AND MODELS

(ii) (iii)

373

Market share may change as a result of new entry in the long run. Relative prices of substitutes, and tastes and fashions and income, may change in the long run.

Secondly, even the long run cost curve is indeterminate as technology and factor prices do not remain constant in the long run. The firm has definite knowledge only about its short run costs, as indicated by AVC (average variable cost), AFC (average fixed cost), SMC (short run marginal cost) and SATC (short run average total costs). Also, the firm knows that its long run average costs (LAC) will be less than or equal to (but not higher than) its SATC. The AVC curve is not U-shaped but saucer-shaped, and is constant for a wide range of output. This is because most manufacturing enterprises keep ‘reserve’ capacity to meet fluctuations in demand and breakdowns in equipment. Consequently, as output varies, the reserve or standby equipment are brought into (or kept out of) use without any change in AVC. The significance of reserve capacity for the shape of the AVC curve can be understood from the fact that the efficiency of a variable factor like labour decreases (and its average cost rises) after the given fixed factor is used optimally, because this fixed factor gets used inefficiently through increased breakdowns, wastage, etc. For instance, a basic item of

equipment is ideally used for one shift. If it is put to use for two shifts then in the second shift the same number of workers can produce less from it, and AVC rises through diminishing returns.

If however there is an additional! basic machine in reserve to be used in the second shift then AVC will not rise. Note that excess capacity is different from reserve capacity. Excess capacity means underutilized, sub-optimum capacity. Reserve capacity means maintaining flexibility in fixed factors in order to raise or lower output without raising AVC. Reserve capacity allows AVC to remain constant and hence allow MC also to be constant and equal to AVC for a wide range of output.

Since the firm does not know and cannot determine its long run demand curve, it does not take into consideration the demand curve for determining the price. It determines the price solely on

374

MICROECONOMICS FOR MANAGEMENT STUDENTS

the basis of the ‘cost-plus’ principle and sells whatever output it can, depending upon the prevailing demand conditions. Thus demand determines the output and not the price. In order to determine the price, the oligopolist firm takes three factors into consideration: (i)

The AVC, which is constant and equal to MC for a wide range of output. (ii) The AFC at the standard or normal output level. (Note that the AFC is based on total fixed costs inclusive of fixed costs of reserve (equipment) capacity.) (iii) The ‘target rate of profit per unit, i.e. the rate of return on investment it intends or plans to earn. The ‘target’ rate is decided on basis of three factors, namely, (a) compensation for risk involved, (b) threat of new entrants, and (c) inducement to a continuing flow of long term funds (from the capital market). The target rate of profit may be higher than, equal to, or lower than the ‘fair’ rate of profit. It is generally less than the maximum profit. Since the firm has no idea of ‘maximum profits’ (because it does not know its MR curve) it does not and cannot compare targetted profits with maximum profits. The firm determines the price by using the following formula: Price = AVC at normal output + AFC at normal output + Targetted Profit per unit of normal output. Here the sum of the last two terms, i.e. AFC plus Targetted Profit

per unit at ‘normal’ output level is called the Gross Profit Margin (GPM), and the Per Unit Targetted Profit (on investment) at normal output level is called Net Profit Margin (VPM). This method of pricing is also known as ‘Mark-up’ pricing because here the price is arrived at by ‘marking up’ (hiking up) the AVC (or MC). The above-mentioned formula can be also given as the ‘markup’ formula in the following way: Price = (1+ K) AVC Normal.

Here AVC (Normal) means AVC or MC at Normal output. More important, K shows the ratio (or percentage) by which the AVC

gio

OLIGOPOLY: CHARACTERISTICS AND MODELS

is ‘marked up’ to cover AFC at Normal output plus Per Unit Target rate of return at normal output level. This ratio K is equal to AFC plus targetted profit per unit divided by AVC at normal output. Suppose a firm invests Rs 1,00,000 to produce normal output of 10,000 units. Suppose its target rate of return on investment is 10 per cent, i.e. Rs 10,000 on Rs 1,00,000. Thus, at Normal output

the Per Unit Targetted Profit is 10,000/10,000 = 1 Re per Unit. Suppose the Fixed Costs at normal output is Rs 40,000 and variable costs is Rs 60,000. Hence AFC and AVC (= MC) at normal output are Rs 4 and Rs 6, respectively. Hence, the price per unit is = Rs 6 +Rs 4+Re 1=Rs 11. The gross profit margin (GPM) is = Rs 5 per unit and Net Profit Margin (WPM) is Rs 1 per unit. The ‘mark-up’ proportion K = 5/6. Hence price = (1 +5/6)6=6+95= 11 rupees per unit.

Note that K = GPM/AVC and NPM = (t) x (/). Where t= targetted rate of return expressed as a fraction and /=amount of Investment. (Here tf= 0.1 and investment is = 1,00,000). Price and costs

SATC

AVC

AFC (including payment for reserve capacity) output Q,

Qh

Q)

Figure 5 Mark-up Pricing Model

G

376

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Fig. 5 explains diagramatically the mark-up pricing principle. The AVC curve is saucer-shaped showing constant AVC for the output range OF to OG. SMC is the marginal cost derived from AVC. SMC is below AVC upto OF output and rises above AVC from output OG. It coincides with AVC in the range FG. SATC is the short run average total cost curve derived by adding continuously falling AFC to AVC. SMC cuts SATC at minimum

SATC. Here, OQ, is the normal output. The AVC (or MC) at output Q, is = OH=BQ,. The AFC at OQ, output = CB which is = KQ, on the AFC curve. FC = Targetted Profit per unit at OQ, output. Price = OP= Q_F is the ‘mark-up’ price. The ‘markup’ proportion, K, is = HP/OH=BF/BQ, and ‘Mark-up’ Price =FQ,=(1+ BF/BQ,) BQ,. Here BF= gross profit margin and FC = net profit margin. The range of output FG shows the effect of reserve capacity to keep AVC constant. It will be seen from Fig. 5, that when demand increases, at the

same price P an output larger than OQ, will be sold. When output rises to Q,, AFC falls to Q,K, = C,B, and NPM rises to F,C,. Although the gross profit margin remains the same (at FB, is = FB) the firm earns more than ‘targetted’ profits as NPM rises. Similarly, when demand decreases, at the same price P an output smaller than OQ, is produced. AFC increases and Net Profit Margin falls and the firm earns less than ‘target’ profit. Suppose output falls to Q,,

then AFCrises to Q,K, = C,B,, Net Profit Margin falls to F,C,, though Gross Profit Margin is the same at F,B, = FB. However, oligopolistic firms revise the GPM to AVC ratio, i.e. they change upwards or downwards the ‘mark-up’ proportion in response to long term or cyclical changes in demand. During depression, to take into account the overall or general decline in

demand, oligopolistic firms may reduce the mark-up. And during boom period, mark-ups are revised upwards. Similarly, when due to some macroeconomic policies, the threat of competition from imports or from foreign firms increases, mark-ups may be reduced and so also the price of oligopolistic firms. Moreover, when the

products of oligopolistic firms are differentiated, the mark-up price refers to a cluster of narrowly differing prices of competing firms. The model we have explained relates to the case of oligopoly with homogenous products.

OLIGOPOLY: CHARACTERISTICS AND MODELS

gid

Limit Pricing Theory of Oligopoly Under oligopoly the condition of entry becomes an important structural variable determining the price and output decisions of firms because new entry is neither free, as it is under perfect or monopolistic competition, nor is it impossible, as under monopoly. We have seen that under perfect competition, when entry is free, profits in the long run are levelled down to normal and price is lowered to the level of minimum long run average cost (LAC). Hence when entry is difficult and not free, it means that an existing firm can, to a certain degree, raise the price above minimum LAC

without attracting new entrants. On the other hand, the fact that entry is difficult but not impossible means that if an existing firm raises its price above the minimum LAC by more than a certain amount, it will induce new firms to enter the industry. Consequently, in the long run, profits of the existing firm will decline. When determining its price or output an oligopolistic firm takes into account not only the actual competition from existing firms but also the potential competition from new entrants. In order to ensure sfable profits in the long run it adopts a price—output policy which will deal with the threat of competition from new entrants. An oligopolistic firm therefore determines the price which is called the limit price because its objective is to discourage or limit the entry of new firms. Limit price is the maximum price that a firm can charge above minimum LAC, without inducing new entry. If the actual price is higher than the limit price, then new entry will be attracted. That is, if the actual price exceeds the limit price, then after entry the new entrant’s profit will be more than normal inducing entry. On the other hand, if the actual price is lower than the limit

price, then after entry new entrant’s profit will be less than normal, discouraging new entry. If the actual price is equal to the limit price then the new entrant’s profit after entry will be just equal to normal. At this price, while new entry is not attractive, existing firms would not want to leave the industry. In order to define Limit price we need to define the concept of condition of entry. For this purpose we use perfect competition price as the standard of comparison because at this price profits of a firm are normal and so entry is not induced. Moreover, the

378

MICROECONOMICS

FOR MANAGEMENT STUDENTS

perfect competition price is determined in the long run by the free entry of new firms. We can therefore say that when entry is difficult and not free the existing firm’s price can and will exceed the perfect competition price. The concept of condition of entry measures the extent to which new entry is difficult. It’s value is zero when entry is free. The factors that determine condition of entry have already been explained earlier in Ch. 10.

Let Condition of Entry be denoted by £. Then, Baie p/l- oars. Here P= Perfect Competition Price which is = Minimum LAC. P, = Limit Price = Maximum price that a firm can charge without inducing new entry. Here P, is > P. when entry is difficult and not free, i.e. when E is >0

The Limit Price will be less than the profit maximizing price of a monopolist, but it wili be higher than the normal profit price of perfect competition. The determination of Limit Price under oligopoly will depend upon whether new entry is difficult because of the absolute cost disadvantage faced by the new entrant or because of the relative cost disadvantage of the new entrant. Accordingly, there are two models of Limit Pricing. In both these we assume that the long run average cost (LAC) curve of the oligopolistic firm is L-shaped and not U-shaped.

Absolute Cost Disadvantage and Limit Price Absolute cost disadvantage of the new entrant firm means that its LAC curve is located above the LAC of the existing firm at all levels of output. To explain the determination of Limit Price when the new entrant suffers from absolute cost disadvantage we make the following assumptions:

(a)

Existing firms keep their output constant when new entry occurs. The new entrant can supply only that portion of the given market demand that is not already being supplied by

19

OLIGOPOLY: CHARACTERISTICS AND MODELS

the existing firms. In other words, the demand curve that the new entrant has to consider is only that portion of market demand curve which is to the right of the (constant) output of existing firms. (b)

LAC of existing and new entrant firms is L-shaped so that beyond optimum scale, LAC is constant at all scales of output, i.e. no diseconomies of scale exist.

(c)

LAC of new entrant is L-shaped and situated above the LAC of any existing firm and the vertical difference between the two is same at all levels of output. Products of the new and existing firms are homogeneous.

(d)

The determination of Limit Price is explained by Fig. 6.

Price,

LAC D

LAC, LAC,

LAC,

dye Pe Lint p, POM tis! castaavbagiaha E.

ae

k;

Rk.

i

price

or. |

Competitive price

P-_——

—----=-==

:

R, LAC,

: | l

|

O

Optimum

|

|

i

|

bake

|

Q

Q-

|

Qn

|

D

Q;

output

(min, LAC)

Figure 6 Limit Pricing Models Here, we find that the L-shaped LAC of any existing firm is given

by LAC,. The LAC of new entrant is LAC,. DD, is the market demand curve which cuts LAC, at point. k, where output is =

380

MICROECONOMICS FOR MANAGEMENT STUDENTS

Q.. The minimum LAC, is the competitive price, P.. DD, shows

that at price P., the competitive output is Q.. Now DD, cuts LAC, at R, when output is = Q,. The existing firms will keep their output constant at Q,. Hence the left over portion of the demand curve that the new entrant can supply is = k,D,. The existing firms will therefore determine

Limit price = P,=R,Q,. The demand curve R,D, faced by new entrant shows that it is lying below the new entrant’s LAC, for any price below P,. And so, for any price < PL new entry is not attracted as profits would be less than normal. On the other hand, if the existing firm keeps output constant at Q, which is < Q, then

the price is = P,=R,, which is above P,. Here DD, shows that the new entrants’ demand curve is Rk,D,. Since R, is located above LAC,, the new entrant’s profit would attracts entry is induced. For any price < P, new entry is not for any price > P,. P, is the maximum price P..(or minimum LAC) which can without attracting new entry. Hence,

be above normal and hence attractive, but itis profitable price above the competitive be charged by existing firms P, ts the limit price.

Relative Cost Disadvantage and Limit Price New entry under oligopoly may be difficult also because of the new entrant’ relative cost disadvantage. Relative cost disadvantage arises when the optimum scale of output is very large in relation to the existing total demand,

and consequently

the

new entrant can enter only with sub-optimum output. And for any output below the optimum, the unit cost of new entrant will be higher than that of an existing firm. In other words, it would mean that either the price after entry would fall (below LAC) if the new entrant decides to enter with optimum scale of output or the new entrant’s output would be sub-optimum if the price is to remain unchanged. Both these alternatives would place the

new entrant in a disadvantageous position vis-d-vis the existing firm. In explaining the determination of limit price, when new entry

is difficult because of the relative cost disadvantage of the entrant, we make the following assumptions:

OLIGOPOLY: CHARACTERISTICS AND MODELS

a) b) c) d) e)

381

The existing firms keep their output constant. New entrant enters with the optimum scale. LAC is L-shaped for all firms. Both the new entrant and existing firms have the same LAC. Products of existing and new firms are homoegenous. The determination of Limit price, P, is explained with the aid

of Fig. 7.

O

Qn

Q»

Q

2;

Q

AW

Qs;

output

Optimum (min. LAC)

Notes: Q, Q¢= OQ

Qy Qs = OQm

Q, Q3 = OO 7.

Figure 7

Here the LAC of both the existing and the new entrant firm is LAC. DD, is the given market demand curve. It cuts LAC at point

kK, where output is = Q.. The minimum LAC=P.. Hence P, = competitive price and Q, = competitive output. Q,, = optimum output with minimum LAC. The existing firms will keep output constant at OQ, which is arrived at by deducting OQ, from OQ..

J8Z

MICROECONOMICS

FOR MANAGEMENT STUDENTS

That is, on X-axis, distance Q,Q,.= OQ,,, At output OQ, the demand curve shows that price is = QR, =OP,. Thus the limit price

= OP,. We find that if the new entrant enters with optimum output = OQ, when existing firms are producing OQ, then total output after entry will rise to OQ, +OQ,,=OQ. because OQ, = Q,Q.. Hence price after entry will fall to Q.R = OP. and profit lowered to normal because OP..= minimum LAC. Thus entry is not induced. Suppose existing firms charge a price OP, which is higher than Limit price, OP,. At this price the demand curve shows that output

is OQ,. If the existing firms’ output is constant at OQ,, and the new entrant enters with output OQ, then total output rises after entry

to OQ, = OQ, + OQ, = OQ, + Q,Q, since OQ,, = Q,Q,. After entry the price would therefore be P, shown by the point R, on demand curve DD. Since Price P, is above LAC, the after-entry price would earn more than normal profits. Hence, for any price >P,, new entry is profitable and attractive.

If the price charged by existing firms is P, which is < OP, then DD, shows the existing firms’ constant output to be OQ,. Suppose the new entrant enters with optimum output = OQ,. Then after

entry output will increase to OQ, = OQ, + Q,Q;, where OQ,, is = Q,Q.. For output OQ., price after entry on DD, is OP., as shown by point R,. But price OP, = RQ, is below LAC and so profits would be less than normal. Hence, for any price below P,, new entry is not attracted as it is not profitable. Thus we find that for any price greater than P, new entry is induced because after-entry profits are above normal, whereas for any price below P, new entry is discouraged because after-

entry profits would be less than normal. Hence P, is the limit price.

Chapter 16 some Non-Traditional Models

of Oligopoly

A. Managerial Theories of the Firm

Alternative Objectives of Oligopolistic Firms We have seen that the characteristics of cligopoly are such that the objective of the firm is not necessarily maximization of profits. According to the full cost pricing model, the oligopolist firm’s objective is fair and stable profit; according to the mark-up pricing model, it is target rate of return; and for the limit pricing model it is discouraging new entry. However, there are objectives which are peculiar to oligopolistic firms which are organized as large, widely-held joint-stock companies with their shares (stock) listed on the stock exchange. The most distinguishing feature of the joint-stock company asa form of business organization is the separation of ownership from management. The owners of a company comprise largely a vast number of small and widely scattered individual shareholders. They receive a share in the profits of the company in the form of dividends and also elect the board of directors that appoints the top management which, in turn, takes all decisions affecting prices, costs, advertising, investment,

profits, etc. In the election of the

board of directors, each equity (ordinary) share (and not each shareholder) carries one vote and voting by proxy is permissible. Those who control the maximum number of proxies for the shares held by small, scattered, indifferent shareholders control the com-

position of the board and the top management. On the other hand, the top management, which takes all the pricing, output, cost, and profitrelated decisions have only a nominal or minority stake in the ownership of the company. Moreover, the top management earns a contractual salary and not a share in profits. Thus the

384

MICROECONOMICS

FOR MANAGEMENT STUDENTS

shareholders, who are the owners and earn profits in the form of dividend, do not participate in decision-making; while the top management which is entrusted with the decision-making authority are hired executives earning a contractual salary and not profits. It is in this sense that ownership and management are separated in a large joint-stock company. Thus, in a large joint-stock corporation, decisions regarding price, output, costs, profits, etc. are taken on the basis of the objectives set by the hired, salaried top management for whom the interests of the shareholding-owners are secondary, though not to be neglected. The

interests

of the top management

are

related

to salary,

perquisites, discretionary expenditure, power, prestige in the business world, size of staff controlled, and, equally important, jobsecurity. Of these, salary, power, staff under control, prestige etc. are objectives that are closely connected with the quantum or srowth of sales. While, security of tenure is related to profits, in the sense that a company not earning sufficient profits will be taken over and the new owners are likely to dismiss the existing top management from their jobs. For the security of their jobs in the company, the management must keep the shareholders satisfied by earning a certain minimum profit in order that the shareholders receive the expected, satisfactory dividend. Thus the management has to serve its interests (salary, prestige, staff expenditure, etc.) with the constraint imposed by shareholders’ expectation of profits (and, therefore, the market value of shares).

Bearing in mind these objectives of the top management and the constraint it imposes, W. Baumol, R. Marris, and Cyert and March have developed what are called the managerial theories of the

(oligopolistic) firm. These theories of the firm are termed managerial theories because the behaviour and performance of large, widely held public limited companies (corporations) are deter-

mined by the objectives of the top management.

Objective of Sales Maximization with Profit Constraint Accordingto William Baumol, the objective of the top manage-

ment of a large joint-stock company operating in an oligopolistic

SOME NON-TRADITIONAL MODELS OF OLIGOPOLY

389

industry is to maximize sales revenue with a (minimum) profit constraint.' This is because the interests of the top-management in the form of salary, prestige, staff expenditure, etc. are closely and directly related to the monetary value of the sales (turnover) of the company. However, the job security of the management is related to the threat of takeover which, in turn, depends

upon

whether the company earns the minimum profit necessary to keep

Eine Profit

shag

iniumum

rofrit

F

ynstraint

| 59).

O

Notes.

F

6 ee OF

Q3;

G

output

Qo9B = Minimum Profit Constraint = OF Q ky = Constrained Maximum Sales

Q3h3 = Unconstrained Maximum Sales Q|A =R,C= Maximum Profits FAG = Profit Curve

Figure 1 Baumol’s Constrained Sales Maximisation Model W.J. Baumol, Business Behaviour, Value and Growth (Macmillan, New York,

386

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

Shareholders satisfied. Thus, the oligopolistic joint-stock company decides its output or price to attain the objective of maximizing sales with a profit constraint. This is illustrated by Fig. /. In Fig. 1, curve TR shows variations in the money value of sales (price multiplied by output), and curve TC shows variations in total costs as output changes. As noted earlier, the slope of 7R shows the marginal revenue and slope of TC shows the marginal cost. 7k and 7C intersect at points Mand Nand so profits are zero at the corresponding outputs F and G. The tangent to 7R at point FR, is parallel to the tangent to 7C at point C. Hence, at these points the slopes of 7R and TC, showing marginal revenue and marginal costs, are equal. At these points the vertical distance between 7R and 7C measuring profits is at its maximum. Thus profits are maximized at output Q, corresponding to points A, and C. Here maximum profit are Q,A on the curve FAG showing variations in | profit with output. TR is highest at point RX, where marginal revenue is zero. Here sales are maximized regardless of the profit level. Corresponding to R, is output Q, which shows sales maximization without profit constraint. If the firm is interested only in maximizing sales without any concern

about earning a certain minimum

profit, then the

curve FAG shows that at output Q, it would earn profits = Q, D. Suppose the shareholders expect a minimum profit, shown by an arbitrarily chosen point B on profit curve FAG, than profit = Q,B is the minimum profit constraint imposed on the management's behaviour to achieve sales maximization. At output Q,, corresponding to profit level B, we find that sales revenue on 7R curve is =

R,. Hence R, is the maximum sales that is attainable with the constraint of earning minimum profit = Q,B. Thus Q, is the output to be produced to achieve the managerial objective of sales maximization with profit constraint. At this output, the price is = R,Q,/O0Q,. At profit maximizing output Q,, price is = R,Q,/0Q,.

But since the slope of 7F falls from k, to R,, the price shown at OQ, is less than that at OQ,. We find that the price is lower and output (Q,) is higher under the constrained sales maximization hypothesis, than under traditional hypothesis of profit maximization. Note that the minimum profit constraint, Q,B, is decided arbitrarily here, but in practice the prevailing interest rate would determine the opportunity cost of investment in equity shares.

SOME NON-TRADITIONAL MODELS OF OLIGOPOLY

387

What is also important to note is that sales maximization with minimum profit constraint results in lower output than output that would result if sales were maximized without the constraint of the shareholders’ expectation of a certain minimum profit. Hence the management will decide to produce cutput Q, and not Q,,.

Objective of Maximization of the Managerial Utility Function Even according to Robin Marris,° in a large modern joint-stock corporation, the decisions regarding price, output, quality, advertising, technique, inputs, etc. are taken by the top management with the objective of maximizing their own satisfaction or utility from salary, prestige, perquisites, and job security. But Marris argues that these sources of the management’s utility depend upon the rate of growth of the firm’s sales and not on the absolute size of its sales (as assumed by Baumol). It is not the stagnant company but the fast-growing firm that is willing to pay high salary and carries prestige even though the former may have a larger absolute volume of sales. Secondly, the management also wants security of tenure in their job which depends upon profit distributed to shareholders as dividends. A very high growth rate would entail large-scale borrowing and a higher level of interest charges. High growth also necessitates an increased proportion of retained profits and hence lower dividend rates. Beyond a point both these consequences of high growth lead to the threat of takeover and loss of job for the management. On the other hand, an excessive concern to ensure job security, as reflected in a high rate of profit, would result in minimizing reliance on borrowed funds. These would lead to a slow growth in sales and hence, relatively lower salary, prestige, etc. Thus, the management has to choose the ideal or optimum combination of growth and job security. The managerial utility function consists of a map of indifference curves, each curve showing those combinations of growth and job security that yield equal levels of utility. 2 R. Marris, ‘A Model of Managerial Enterprise’, Quarterly Journal of Economics, 1963 and Theory of Managerial Capitalism (Macmillan, New York, 1964).

388

MICROECONOMICS

FOR MANAGEMENT STUDENTS

On the other hand, the firm as an organization and the share-

holders as its owners also have the twin objectives of growth and profitability that are interrelated. Up to a certain point, higher growth would lead to higher rate of profit through larger market share, economies of scale, etc. But after that point, higher growth

would entail an increasing burden of interest costs and hence lower rate of profit. Thus we have, on the one hand, a relationship between growth

and profits for the firm and its owners, and, on the other, there is the utility function of top management consisting of indifference curves showing equal utility combinations of growth and job security which is dependent upon profit rate. The management will try to reach the highest of these indifference curves, bearing in mind the growth and profit relationship obtaining in the firm. Fig. 2 explains how the management of a large oligopolistic joint-stock corporation determines the equilibrium combination

Profit Rate (%)

EA

:

NOE ee

BOG,

eat

ICs

(an | O

G

P Growth rate

of sales (%)

OBP is Growth-profit relation for Firm /Cp, IC), /Co, IC3 . . . etc. are the Indifference Curves

Figure 2 Marris’ Models of Maximisation of Managerial Utility

Function

SOME NON-TRADITIONAL MODELS OF OLIGOPOLY

389

of growth and profit that would maximize the managerial utility function. Here, Y-axis shows the rate of profit (percentage) and growth rate of sales (percentage) is shown on X-axis. The curve OBP shows the growth-profit relationship for the firm. Up to point B, the profit rate increases as growth increases. Thereafter, a higher growth rate involves higher costs and hence a lower profit rate. /C,, IC,, IC,, and IC, are the indifference curves, each curve

showing equal utility for different combinations of growth and profit rate for the management. The utility of management is maximized at point E where JC, is a tangent to the growth-profit curve. IC, is the highest utility that management can achieve. Here G = equilibrium growth rate that maximizes the managerial utility function and equilibrium profit rate for the shareholders is = P.

The Firm as ‘Satisficer’

The hypothesis that the objective of the firm is to maximize profits or sales or the managerial utility function implies that the ‘rational firm is a ‘maximizer’ of one or the other variable. This means that the firm has complete knowledge of all the options and constraints, and also the probability of various outcomes. In other words, any variety of maximization hypothesis implies ‘global’ rationality on the part of the firm. Global rationality requires unlimited information regarding all the variables and constraints involved in the maximizing exercise. This ‘global’ rationality, therefore, assumes that information is costless and obtainable without expending time. In reality, however, obtaining information is costly, requires expertise, and it is time-consuming. Hence the firm has to make do with limited information. In other words, it acts with ‘bounded’

rationality, in the sense that its behaviour is rational bearing in mind the realities of limited information and the uncertainties of the real world. Moreover, when the firm is a large joint-stock company, its organization involves several interests groups, such as shareholders, management, workers, customers, creditors, etc. The in-

terests and aims of these groups in the firm are often conflicting. The firm has to operate as a coalition of interest groups, each group having its own aspirations. The firm has, therefore to

390

MICROECONOMICS

FOR MANAGEMENT STUDENTS

behave as a coalition organization bound by constraints imposed by uncertainty and limited information. The management of the firm sets before itself certain goals based on the ‘aspiration levels’ of different interest groups and conducts its behaviour to satisfy these ‘goals’ or aspiration levels in terms of profits, sales, market share, managerial reward, etc.

Given the constraint of limited information and the divergent interests of the groups with stakes in the firm, the objective of the firm is to achieve satisfactory values of profits, sales, managerial benefits, etc. rather than the maximum values of these variables. In this sense, behaving within bounded rationality, the firm is a satisficer and not a maximizer. This model of a large oligopolistic joint-stock corporation is

known as the behavioural theory of the firm.’

B. Models of Collusive Oligopoly As already mentioned, under oligopoly, every individual firm is faced with the uncertainty about how its rivals will react to any change in its output/price and how this reaction of rivals will, in turn, affect its demand (sales/price). Moreover, the action-reaction process among its oligopolistic rivals may result in cut-throat

price competition (or an advertising war) that would ultimately lead to losses for all the firms in the industry. Hence, oligopolistic firms find it advantageous to give up their independent behaviour and coordinate their actions either through an explicit agreement (cartel), if permissible under the law, or through an implicit, hidden understanding (collusion) if such restrictive trade practices are prohibited under the anti-monopoly law. Since the number of firms in an oligopolistic industry is small, it is feasible to work out and enforce an arrangement of coordinated behaviour through a restrictive trade agreement. The two commonly used models of collusive oligopoly are: (i) Price Leadership and (ii) Joint Profit Maximization.

3. R.M. Cyert and J.G. March, A Behavioural Theory of the Firm (Prentice-Hall, 1963).

SOME NON-TRADITIONAL MODELS OF OLIGOPOLY

391

Models of Price Leadership under Oligopoly Implicit collusion in the form of Price Leadership is quite common in oligopolistic industries. Instead of independently deciding their own individual profit maximizing output or price, oligopolistic firms follow the leadership of one firm in setting the price and charge the same price as is charged by the leader. If products are differentiated, then they follow the leader firm by changing their prices in such a way that the relative price differential between different products remains unchanged. This form of consciously adopted ‘parallel’ behaviour by oligopolists to avoid competition is often termed conscious parallelism. The firm that is accepted as the price leader may either be the most efficient one with the lowest cost or the dominant firm with the largest market share. The adoption of price leadership practice is preferred by oligopolistic firms because they do not have to fear the uncertain consequences of retaliation by the most efficient or the dominant firm. Moreover, price leadership practice allows firms to avoid making an explicit agreement which in many countries is illegal under the anti-monopoly law.

Dominant Firm as the Price Leader

In many oligopolistic industries it is found that one firm has the dominant share of the market and each of the rival firms are relatively small, both in terms of their market share and productive capacity. In such oligopolistic industries, the relatively small firms accept the dominant firm as the price leader and, assuming homogeneous products, set their prices at the same level as the profit maximizing price charged by the dominant firm. They will together supply whatever output they can sell at this price given the total demand curve and the leader’s profit maximizing output. This model assumes that the market (total) demand curve and the demand curve of the dominant firm are known. The marginal revenue curve of the dominant firm is derived from its demand curve. Given its cost curves, we get the profit maximizing output and price of the dominant firm which is the lead price followed

O

aansly(D)g

JUDUIWIOG dapDaTUdy §

O

LaPDAT S,WAL] DIU ‘YW IV OW

wuiniuquinby

WAL] S jndjno

AOPDIT O

04

je}UOZLIOYaoueysip 'qly =

*4O =

dO

dO

'g's =

“q'yw

(ME

jaysppy

winiiqyinby

Ajddng

4amojjo. 4 swii1.f

(D)E = dO Ul

Ajddng puv

pudwia pubg

jays

ojo.

a1nsiy(ME

Lg=

peons9, a2uR\sip '4O=

Od

[eONIaA eBuRSIP YOUl

49M0]]sutsy ,

ln

49N0]]SUL ,

y =a" ‘ed

—:sajoAy

O04

d

SOME NON-TRADITIONAL MODELS OF OLIGOPOLY

393

by the smaller rival firms. On this basis, we get the combined supply curve of the smaller rival firms when following the price leadership practice. Fig. 3, with panels 3(a) and 3(b), explains the determination of dominant leader firm’s profit maximizing output and price and the derivation of follower firm’s combined supply curve from the market demand curve. In Fig. 3(a), Fb, is the dominant firm’s known individual demand curve and MR is the corresponding

marginal revenue curve. In Fig. 3(b) DD, is the total market demand curve. At price F in panel (a), the dominant firm’s output is zero and so at the corresponding price P in panel (b), the entire market demand PE on DD, is supplied by the follower smaller firms. Thus point £ is one point on the combined supply curve of the smaller firms. Similarly, at price F, in panel (a), demand and output of the leader firm is F,b,. At the corresponding price P, in

panel (b) the horizontal distance P,B, on DD, is = F,b,. Thus the entire demand is supplied by dominant leader firm and so the supply by the smaller follower firms is zero. Hence P, on Y-axis of panel (b) is another (starting) point of the combined supply curve of the follower firms. Joining the two points P, (on Y-axis) and E (on DD,) in panel (b) we get the combined supply curve P,S of the follower firms. In panel (a) the dominant leader firm’s marginal revenue curve is MR, and AC and MC are its average and marginal cost curves. MC cuts MR at point e. Hence output OQ and price OF, are the profit maximizing output and price of the dominant firm in panel (a). The smaller, rival firms adopt this price as the leader price. As shown in panel (b), price OP, corresponds to OF, in panel (a). The follower smaller firms supply output OM, at the leader’s price OP, = OF. This price leadership arrangement can be maintained only if . the smaller rival firms stick to the arrangement that they will _ supply only OM, output. If they supply more or less than this agreed output, the profit maximizing calculation of the price leader dominant firm will be upset and hence the price leadership arran_ gement will break down.

394

MICROECONOMICS FOR MANAGEMENT STUDENTS

Price Leadership by the Low Cost Firm In this case, firms with relatively higher costs fear that effective competition by the most efficient firm will erode their market shares. Hence they accept the leadership of the lowest cost firm rather than compete with it. Let X, Y and Z be the three oligopolistic firms. In Fig. 4, the average and marginal cost curves of firms X, Y, and Z are given by AC,, AC,, and AC,, and MC,, MC,, and MC,,. Price, MR, AC, MC

O Notes.

Q)

Qo

Q3

output

Output: P3R3 = OQ3 : P3R2 = OQ2 and P3R; = OQ; Price: OP3 = QR) = QoRo = Q3k3

Average Cost: Q)K, = OC); OC» = QoKo and OC3 = Q3K3 MC, cuts MR, at £y; MC> cuts MR» at E> and MC3 cuts MkR3 at E3

Figure 4 Low Cost Firm as Price Leader

SOME NON-TRADITIONAL MODELS OF OLIGOPOLY

395

Here Firm Z is the lowest cost firm. Let dd,, dd,, and dd, be the demand

curves

and MR,, MR,, and Mk, be the corresponding

. MC, marginal revenue curves of Firms , Y, and Z, respectively

cuts MR, at point £,. Hence OQ, is the profit maximizing output of

for Firm Z and OP, is its profit maximizing price as shown on dd, ted output OQ,. This is the leader’s price which is followed or adop by by the Firms X and Y. Thus OP, is the common price changed ucts. all the three firms, assuming that they sell homogeneous prod supply Given demand curves dd, and dd,, firms X and Y will Firm Z outputs OQ, and OQ, at the common leader price = OP,. n will make the maximum profit shown by area P, R, K, C; as show by price OP, and average cost OC, at output OQ,. At price = OF. firms X and Y will make profits shown by areas P, R, K,C, and at Pic, respectively, because on AC, average cost is C,K, output OQ, and on AC,, average cost is C.K, at OQ, output. er It will be noticed that Firm X and Y would have charged high penprices and supplied lower outputs had they behaved inde case, dently by maximizing their own individual profits. In that price since MC, cuts MR, at E,, Firm X would have produced OX at MC, OF, and firm Y would have produced OY at price OG since cuts MR, at E,. This price leadership is maintainable only if the ly follower firms X and Ystick to the understanding that they supp would not less or more than outputs OQ, and OQ,, otherwise they Z. This, upset the profit maximizing calculation of the leader Firm in turn, would result in a breakdown of the explicitly or implicitly agreed price-leadership arrangement. In both cases of price leadership it must be noted that if the not follower firms accept the price set by the leader firm but do , follow the understanding on the outputs to be supplied by them the price leadership arrangement will not work.

Maximization of Joint Profits

Under oligopoly, each firm may find that in the long run, instead of pursuing the goal of maximization of its own profits, it is advantageous to work towards the goal of maximization of the joint (combined) profits of the industry. This latter goal has the further advantage that each individual firm need not know its demand

396

MICROECONOMICS FOR MANAGEMENT STUDENTS

curve. What they require to know is the demand curve for the industry, i.e. the market or total demand curve. This task can be undertaken by the industry association. From the individual firms’ marginal cost curves (MCs), the combined MC for the industry is derived through a horizontal — summation of the MCs of individual firms.* Since the industry demand curve is known, we derive from it the industry marginal revenue curve. The point at which the Combined MC cuts The Industry MR curve (from below) gives the industry equilibrium output where the joint (industry) profits are maximized. The price on the industry demand curve for this output is the common price that is charged by each individual oligopolist firm. Each firm then equates its individual MC with the equilibrium value of industry MR to determine its own output. Its profit is. derived by multiplying that output with the difference between the: common price charged by each firm and its own AC for the — output at which its own MC is = common equilibrium MR. Fig. 5 explains the determination of an industry's common equilibrium price at which joint profits are maximized as well as individual oligopolist firm’s output and profits. X, Y, and Z are three oligopolist firms. In Fig. 5, panel (a) shows MC, and AC, are the cost curves of Firm X. Similarly, MC, and ACy show cost curves of Firm Y in panel (b). MC, and AC; are cost curves of Firm Z in panel (c). Their horizontal summation

gives the combined marginal and average cost curves CMC and CAC in panel (d). In panel (d), DD is the industry or total demand curve and MR is the marginal revenue curve for industry derived from DD. Here, CMC cuts MR at E and so Q, is the joint profit

maximization equilibrium output for the industry. P,, is the common (monopoly) price corresponding to output Q,, as shown on

DD’. All the three firms X, Y, and Z charge the same price = OP. Distance Q,E is the industry equilibrium Mk. Firms X, Y,

and Z find that this common MR is equal to MC, at point Ey, it is equal to MC, at Ey and equal to MC; at E,. Firm X will thus 4 A horizontal summation of the cost curves is obtained in the following way. If for one firm at given cost OC on Y-axis the output shown on the cost curve Céaiae OO, and for second firm at the same cost OC (on Y-axis) the cost curve CCrsshows output OQ, then the combined cost curve for the two firms is obtained by ‘shoorine for cost curve OC output OQ, + Q,Q,, such that Q,Q,= OQ,.

aansly(D)G

O

‘sajONy —

© O

aansiy(Q)g

aour\siq 4gO wi (Dg = aourrsiq 4OOul (Ps=

a0UeISIq “gO =

jndjno jo A

aansiy(Q)g

(UOWIWIOD) ‘90g

FyogndNne Om

=7770 ="7'0“TO +*00 +00 700

7h

(Aajsnpuy) IWo0slg Jepoyp

AJOdouOp

UOIDSIUNXDP JO Julor

IVI ‘INI YW ‘Md

ainshy(P)G

=O

LO{J

joioy

AAJSNP

jndjno

398

MICROECONOMICS

FOR MANAGEMENT STUDENTS

produce Qy output, Firm Ywill produce Qy output, and Firm Z will produce Q, output. From panel (d) we find that at the industry price = P, the industry average cost on CAC is = OC = Q,K. Hence the maximum joint profits are = CKRP.. From panel (a) we find that at price P,, and output Q. profits of X are = P_ FKC, because for output Q,, average cost of X is = OC,,. Similarly, from panels (b) and (c) we find that profits of firms Y and Z at price P,, are given by

P,, GJC, and P,, MNC,, respectively. By adopting the objective of maximization of joint profits, the colluding oligopoly firms are forgoing independence of behaviour in return for certainty of price and stability of profits.

Chapter 17 Taxation and Performance of the Firm

The theory of firm provides the necessary insights into the pricing and output decisions of firms working under different market structures. It is important to understand how the government can influence these decisions through the traditional fiscal measures of taxation and subsidy. In this chapter, we propose to examine the effects of taxation and subsidy on the output and price decisions of a firm and the industry working under perfect competition and monopoly in the short run and the long run. Subsidies are treated as negative taxes. We consider four different types of taxes: (a) lump-sum tax, which is a tax of a fixed amount levied on the firm per unit of a time period irrespective of the price or output or profits of the firm; (b) profit tax, which is levied on the firm as a fixed percentage of its profits per unit time period; (c) specific commodity tax, which is levied per unit of output produced/sold by the firm during the unit time period; and (d) ad

valorem tax, which is levied as a fixed percentage of the price of

the product sold by the firm. In reality, the tax subsidy systems are more complicated, e.g., a composite tax mixing the fixed and the variable tax, progression in the rate structure, or in investment subsidies of different types exist in reality. It is hoped that this analysis will provide the basic framework on the basis of which analyses of these complications in real life situations can be made.

Lump-sum Tax Since the lump-sum tax is levied as a fixed amount independent of the firm’s output, price, or profits, directly or indirectly, it is considered by the firm as a fixed cost both in the short and long run. According to the theory of firm, with profit maximizing entrepreneurs, the price and output decisions are not affected in the short run. This is because a profit maximizer always operates at

400

MICROECONOMICS

FOR MANAGEMENT STUDENTS

the output rate (x) where his marginal cost (MC) equals marginal revenue (MR), with the former exceeding the latter at higher output rates. Since the lump-sum tax is considered as a fixed cost, it affects neither the MC nor MR of the firm, and hence equilibrium output and the price of the firm in the short run remain unaffected by such a tax or subsidy. This is true irrespective of the market structure within which the firm operates. Such a tax (subsidy) only reduces profits (losses) of the firm in the short run because

it shifts the average cost (AC) curve upward (downward).! In the long run, however, there is a possibility of the firm going out of business on account of the lump-sum tax. Let us consider a monopoly firm making only normal (or relatively small supernormal) profits because of the level and shapes of its AC and AR (average revenue) curves. Now if a substantial lump-sum tax is imposed, the firm would have continued in the short run to produce the same output and charge the same price as before although incurring losses. In the long run, however,

AC

curve

with

the

tax

remains

well

above

if the new

the AR

throughout the range, the firm has no alternative

curve

but to close

down (See Fig. 1). Exactly the opposite effect can be expected of a subsidy to a firm incurring losses. Unlike monopoly, in the case of a perfectly competitive industry, only a few firms would go out of business in the long run due to the imposition of a lump-sum tax. As a result, the supply curve would shift upward and leftward although the MC curve of each of the surviving firms remains the same as before. The industry would therefore experience an increase in price and a decline in output in spite of an expansion in output of the surviving firms in the industry. If some of the existing firms were making super-normal profits before the imposition of the tax, a few of them would continue doing so even after its imposition. This is because

| It may be noted here that a subsidy of a fixed amount can affect the firm’s

output decision in the short run if it is linked to a firm producing some output. This can happen when the firm would otherwise have decided to shutdown because it was unable to cover even its variable costs, but the availability of the

subsidy induces it to remain in production. There are several cases of sick units, which are institutionally assisted for a turnaround, and who find themselves in such a situation (e.g. BJFR companies).

TAXATION AND PERFORMANCE

OF THE FIRM

40]

P & Costs

LMC

LAC! LAC

Q/t

Figure | Lump-sum Tax on a Monopoly Firm The monopoly firm is in equilibrium producing OQ output, charging OP price, and making AFGP super-normal profits. A substantial lump-sum tax shifts its LAC to LAC’ with LMC remaining unchanged. It would make PGHB losses even in the long run, compelling it to close down.

the industry experiences a rise in the price to an extent that allows the marginal firm to cover its average cost after imposition of the tax (See Fig. 2). If all the firms in the industry were earning super-normal profits before the imposition of the tax, a lump-sum tax reduces their profits and thereby discourages new entrants to the industry. This would restrict the expansion of the industry’s output. A lump-sum subsidy would have exactly the opposite

“bo pure ,'bQ 0} syndjno

WINs-xe]

sylysey} DVT JO [[e au}

suy

‘:premdn OU],

SUL

SulyeU AjTUO

yewuou syyoid s10jaq sty

uOnIsodull

A194} Ul BSeaIDU! Ue BDUSLIadxe ‘Ia@ABMOY ‘SUL SUIAIAINS dy) UOYg ‘peonpad oie syyoid ewuou-redns s/y way pue syyoid Jew4OU ATUO SUTUIVS WAY [eUISIeU! aU] SI fY Wd MON OO 0} Ayquenb ety sulonpad pue jqO 0} edd ou} BUISCIIOUI ‘}¢ O} piemdn syiys Asnpul auq jo aaino Ajddns ayy ‘Ansnpul oy} Ava] VDUDY PUe xv} Jaye Sasso] SUTYeW Iq P[NOM Xe} OU} JO

5

ee

IVT

if

1 WL]

co

|

| |

1ars

y_.

es

O

olddOVv ap5

6

UONNAdWO, JIajsad Ul XDI. wins-dwinT

Z ainsl]

!

RED

fui]

1M

| |

|

a OD

Ae}

EA Wee *

RA

SISOD

O

baneta

d ¥

Bd

Bae

OVI

OVT

OY

SISO)

/

VI

ta

/

oF VO)

Aajsnpuy

/ Ps

QIW'1

xe}-o1d e ye yndyno “6G pue 'b_ Bulonpoad pue syyoid jewsou-radns Suryew wuNqTInba uni suo] Ul ere “Y pure |‘yy SWI

d

1/0

S

TAXATION AND PERFORMANCE

OF THE FIRM

403

effect of encouraging new entrants, expanding the output, and reducing price in the perfectly competitive industry. It is important to note here that the lump-sum tax or subsidy need not be only in monetary terms. A law or regulation or incentives of various types may also have the same effect as the lump-sum tax or subsidy. Imposing a law requiring firms to instal a certain type of pollution control mechanism or government providing specific infrastructural facilities to firms in an industry

are examples of a lump-sum tax or subsidy in kind rather than cash.

Profit Tax

The short run effects of a profit tax is very similar to those of the lump-sum tax. It reduces the profits of the firm without disturbing the profit-maximizing output or price decision of the firm — whether operating under perfect competition or monopoly. This can be shown with the help of the simple algebra of profit maximization: tm = TR-—TC where 7 is profit.

When profits are taxed @ 100 p per cent then profits become:

(al

2) Aaah @ et 2S

d(1-—p)r

Theref erefore, res ~~~ io

ae2 dn

Doel = Cl.=ks Dp)

(l=

pit pins C) MC).

(since p is constant w.r.t. x.) Therefore, Profits are maximized at an output rate where, MR = MC and a dx

< a dx

Since the profit tax applies to accounting profits, it is possible that firms earning only normal profits may also be required to pay the tax. Under such circumstances, the firm may have to leave the industry in the long run on account of the imposition of a profit tax. In the case of firms incurring losses, since the profit tax does

404

MICROECONOMICS

FOR MANAGEMENT STUDENTS

not apply, it does not matter. Firms making super-normal profits would experience a decline in their profitability. They may start earning a normal or sub-normal profits. This not only discourages new entrants but encourages the exit of some of the marginal firms. Under competitive conditions, sub-normal profits lead to the exit of firms, shifting the industry supply curve upwards and leftwards. Price rises and output contracts in the industry, but the surviving firms experience an increase in their output levels.

Specific Commodity Tax Since a specific commodity tax is levied as a fixed amount per unit of acommodity sold/produced during a given period of time, we may refer to it, for brevity, as a unit tax. Unlike the previous two types of taxes, a unit tax affects the MC directly, with implications for the price and output decisions of the firm even in the short run. This can be shown as follows:

m= TR — TC before tax

t' = TR—(TC+k-x) after unit tax of Rs R.

Therefore,

dn'/dx =MR—-MC—k=0

for maximization

i.e. MR= MC+R is the necessary condition of equilibrium.

Since after the imposition of a unit tax, the equilibrium occurs at higher value of MR, which is always negatively related to output under monopoly, the imposition of a unit tax on monopoly firms leads to higher price and lower output than earlier. This is irrespective of the behaviour of the MC with respect to x, i.e. rising,

falling, or constant (See Fig. 3). As can be seen from Fig. 3, the price after tax cannot rise by more than the tax R unless the MC is falling with respect to x at the equilibrium point. Thus, a profit maximizing monopolist would not shift the entire burden of the unit tax to the consumers if his marginal cost is constant, or

increasing with output over the relevant range.” In the case of a firm working under perfect competition, a unit tax results in lower output being supplied by each firm since its 2 This can be shown algebraically by assuming linearity of AR and AVC.

SISOD

(D)

‘ELT “9'T

vold osu

aseai UT Dap yndjnpue o Ue

‘ynsol ouy

Mou

suisnasdul JSOD

Ajodoway uow st ut

00

"QWsye

‘@) dd


UaA YIM

wnuqi st yinbo

O (C)

Wun xe} JO

LI

Cg

Seseo

SYTYS24} OW

SuispJSOD assaq

240YM jdd>

ELL

[oued

SAINT

VO

Afodouow spes, 0} e

Sse] UL oly “xe]uy

Wun xe} uo

add asl4SI a[qiss 0} odpuy

‘LET “81

‘aoud Y

Sulsr JO eyd ‘ao1d Y

AjOdouop yndjnpure o iJO

Judjsuoy JSOZ

jiuy xv] uo

(q)

O ©

wintqymMbea Sulon YO poid yndjnpure o

O

406

MICROECONOMICS

FOR MANAGEMENT STUDENTS

MC curve shifts leftward and upward with the MR=AR curves remaining at the same level as before the imposition of the tax. For the industry, this results in decrease in the supply, shifting the industry supply curve upward by the same amountas the unit tax. The shift in the industry supply curve is on account of both the exit of firms as well as a shift in the individual MC curve of the firms in the industry. The unit tax thus results in sharing of the burden of the tax between the consumers and the producers through increased price and decreased quantity. To what extent the burden of the tax is borne by the consumers and the producers depends on the elasticities of demand and supply in the industry

(See Fig. 4). From Fig. 4, it becomes clear that if the firm A, earned only normal profits before tax, it would go out of business in the

long run with the imposition of the unit tax because it would shift the AC curve upward by the same amount asthe tax while the price would rise by less than the tax amount in the event of increasing cost conditions. Under such circumstances, only the firms earning high super-normal profits before tax would be able to survive after the imposition of the tax, and their profits would substantially decline. If, however, there are constant or decreasing cost conditions giving rise either to a horizontal or downward sloping supply curve for the industry, the magnitude of price response differs. This can happen if there are strong external economies in operation. If the industry supply curve is horizontal, the entire burden of the unit tax is shifted to consumers because the price rises by exactly the same amount as the tax. Similarly, if the industry supply curve is downward sloping, the price rises by more than the quantum of tax or, in other words, the consumers have to bear

a greater burden than the quantum of tax. The basic formula of sharing of the tax burden between consumers and producers is given by (see Fig. 4 for symbols):

P’P _ elasticityofsupply _

Ge TP

Ng

reies

libbrium) ee eg (at original equili e eee a s aic — and dem yof cit sti ela

Here e, would always be negative whereas e, may be positive or negative. If e, is negative, we have to make directional change in the formula, i.e.

SISO)

uy XDI,

aang] p 4aPU JIajlad

d

UONNaduo,)

Aajsnpuy

1/0

way stp ut winiuqyinba suionpoid bo yndyno pue SulSreyd gO ‘aod ayy suipuodsei0 Aysnpul d wnluqyinbs si OO jndjno pure YO ‘80Ud USYUM 34} run xeyst ‘pesodurt DyJo Wy‘py syrys Ol QW J] We /bO et WAY SIBAOD S$} B[GLHA‘SJSOD ayy yndyno prnom Ajayerpaurut je} Aq bb ym e sulpuodsaii0 [yeyut 9 ANsnputyndyno Aq *,OOD ‘aABMOH Ul BU} SUC] ‘UNI ay) Ansnpul wntqiyinbs st peuteye ye ,OOC pue idO ‘aoud way sy yndyno osye spuedxa 0} ,b¢E indyno ye jdO “e014dey [e}O} UN Xe}SI UPAIS AQ gj,Jo YDSIYM SIBUINSUOD Je9qidd pue ayy siadnpoid teaq‘dif ay], a10Ul onseya oy} Ajddns ‘aAdnd GU} 1aMOT VU} UapiNg UO s1vdNpoid pue dy] o1OU! INSe][s VY] PUeUWAP‘@AIND VY} JIMO] VU} UspiNG UO ‘SASWINSUOD

Yd

0 0

408

MICROECONOMICS

FOR MANAGEMENT STUDENTS

Le aah LE tng expen fe Page ss o> =

Peaer eet eee Pye Cee ett= ep

where 7 would lie between P! and P if e, is negative. Thus, the formula is applicable for all possible cases.

Ad Valorem Tax

In this case, a fixed percentage of the price of the product is taken as tax. Thus, the higher the price charged, the higher the absolute amount of tax per unit of output. Such a tax can be incorporated into the calculations of profit maximizing output and price in the following way: m= TR-—TC=P-X-—TC, When

where P= fx) is a demand function.

an ad valorem tax of 100 t per cent is imposed on price,

mi =P. X—(TC+1PX) l

2 = = MR —- MC - t MR = 0 for maximization 3 ari MC = MR at new equilibrium X. Since 1/(1 — f) is always greater than one, MR is higher at the new equilibrium output than at the earlier one, implying lower quantity and higher price for a firm working under monopoly. It can be easily seen that the case of an ad valorem tax is very similar to that of a unit tax in its effect on the price and output decisions of the firm, whether working under perfect competition or mono-

poly. The only difference between the two types of tax lies in the way they shift the supply curve of the perfectly competitive industry. In the case of a unit tax, the supply curve shifts upward vertically parallel, whereas the ad valorem tax shifts it in an anticlockwise direction, making it steeper at all output levels. An ad

TAXATION AND PERFORMANCE

OF THE FIRM

409

valorem tax, as in the case of the specific commodity tax, also reduces the super-normal profits of entrepreneurs. In real life situations, taxes such as sales tax, excise or customs

duty, are frequently imposed on an ad valorem basis. However, most subsidies are generally linked to output rather than the value and are, therefore, more closely akin to the case of unit tax con-

sidered here. Similarly, corporate tax resembles the profit tax considered in this chapter. Whatever real-life complications occur with regard to some of these basic taxes/subsidies can be effectively considered within the framework of the theory of firm with appropriate modifications.

Chapter 18 Pricing of Factors of Production: General Principles

Our study of the market mechanism presented in Chapters 10 to 16 is limited to the analysis of pricing of goods (and services) in the product markets and does not cover the functioning of factor markets. Product or output markets are markets where consumer goods, raw materials and intermediate goods are bought and sold. However, aS we have seen in Chapter 9, the production and distribution of these goods require employment of factors of production, broadly classified as land, labour, capital and entrepreneurship. Firms, which are sellers of goods and services in the product markets, are buyers of factors of production in the factor markets. In exchange for their contribution towards the production (and distribution) of goods and services, the factors of production are paid ‘factor prices’. Economists use special terms for these prices. Rent is the price paid for the use of land, wages are paid to labour, interest is the price of capital and entrepreneurs receive profits. The factor incomes earned depend on factor prices and the quantities in which the factors are employed. The owners of these factors of production spend their incomes in the product markets to buy goods and services. Firms operating in the product market spend their revenues in the factor markets to buy factors of production. | Thus, we can view the economy as composed of two sets of interrelated markets: product or output markets, and factor markets. The owners of factors are sellers in the factor markets and

buyers in the product markets; whereas the producers of goods and services are sellers in the product markets and buyers in the factor | markets. In this chapter we examine the general principles governing the pricing of factors of production.

PRICING

OF FACTORS

OF PRODUCTION

41]

Why Treat Factor Pricing Separately? Economic theory treats factor pricing distinctly from product pricing mainly because of two reasons. First, as we will discuss in detail in the following chapter, the supply of the factors of production is characterised by such peculiar features that elements determining factor prices are clearly different from those determining product prices. Second, factor demand is clearly distinguishable from product demand: it is a derived demand. Factors are demanded because they are required to produce goods and services. Hence the demand for a factor is derived from the demand for the product that empolys the factor in its production. A firm will buy more or less of a factor depending upon whether the demand for the product produced by employing that factor increases or decreases. The demand for factors of production is also influenced by the nature of competition in the product market. As explained in Chapter 13, a monopolist’s output is less than that of a perfectly compeletive firm facing the same demand and cost conditions for a particular product. Thus, demand derived for a factor by the monopoly producer is lower than demand from a perfectly competitive firm. This explains why employment of workers and managerial staff is most likely to decline when a horizontal merger brings two or more competing firms together to form single monopoly.

Marginal Productivity In modern economic theory, the general principles governing the pricing of factors of production revolve around the concept of marginal productivity. Just as a consumer decides how much of a product he consumes by comparing its marginal utility with price, a producer decides the quantity of a factor to be employed by comparing the factor’s marginal productivity with its price. The marginal productivity of a factor also decreases with its increased employment. A producer’s profits from the employment of two or more factors 1s maximum when they are employed such that the ratio of marginal productivity to factor price is.equal for each factor.

A firm demands a factor of production on the basis of profits expected by employing the factor. Hence factor demand depends on

4]|2

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

the contribution of the factor to the firm’s physical output and sales revenue. The contribution of a factor to a firm’s physical output is measured by its average physical productivity (APP) and its marginal physical productivity (MPP). The APP of a factor is the output per unit of factor employed. The MPP of a factor shows the extent to which the total output increases (or decreases) when the employment of that factor is increased (or decreased) by one unit. If O denotes the output and F indicates the quantity of factor F employed then APP of factor F is given by the ratio, Q/F. Further, if AQ and

AF indicate the changes in Q and F, then MPP of F is equal to the ratio, AO/AF.

It has been explained in Chapter 8 that due to the operation of the laws of returns, when the employment of a variable factor used in combination with one or more fixed factors is increased, the APP

and MPP reaching

of that variable factor first increase but ultimately, after their maximum,

both APP

and MPP

tend

to diminish.

Hence the curves representing APP and MPP have an inverted U shape. However, as we shall see later, for the equilibrium analysis, what matters 1s that portion of these curves which shows diminishing APP and diminishing MPP. So in the diagrams used in this chapter, we generally show only the downward sloping parts of the APP and ~ MPP curves. The contribution that a factor of production makes to the sales revenue of a firm is measured by average revenue productivity (ARP) and

marginal

revenue

productivity

(MRP).

The ARP

of a factor

shows the amount of revenue the firm earns per unit of that factor. The MRP of a factor shows the amount by which total revenue increases (decreases) as the employment of that factor is increased (decreased) by one unit. If TR dentes the total revenue from the sale of output, and F' indicates the quantity of factor / employed

by the firm, then ARP of F is equal to 7R/F. Further, if A7R and AF indicate changes in 7R and F' then MRP of factor F is equal to Faie9aa From the definitions of MPP and MRP, if follows that MRP is equal to marginal revenue (MR) multiplied by MPP, 1.c., MRP = MR x MPP. If TR, O and

F indicate

total revenue,

output of product

and

employment of factor Ff, and ATR, AQ and AF indicate small changes in TR, O and F. Then

by definition,

PRICING OF FACTORS

OF PRODUCTION

413

MRP. = ATRIAF. = (ATR/AQ) x (AQ/AF) = MR x MPP. Marginal Revenue Product and Value of Marginal Product The value of marginal productivity (VMP) is the money value of the additional output produced by employing one extra unit of a factor. This is obtained by multiplying marginal physical productivity with the price of the product, i.e., VMP = Px MPP, where P denotes the

product price. The relationship between MRP and VMP depends upon the nature of competition in the product market. When the product is sold in a perfectly competitive market, MRP = VMP: but when it is sold under conditions of monopoly or imperfect competition, MRP and VMP are unequal and MRP is less than VMP As explained in Chapter 12, in a perfectly competitive product market, price is equal to marginal revenue of the firm at all levels of output. Therefore, when a factor is employed by a firm which is selling its product in a perfectly competitive market,' MRP

S'2] "MR = MPP PrP x, MPP

=

VIVEP.

We have seen in Chapter 13, 14 and 15 that if the product market is characterised by the existence of monopoly or other forms of imperfect competition (i.e. monopolistic competition or oligopoly), the demand curve for the product is downward sloping. Since the price of the product of a monopoly or imperfectly competitive firm decreases as its output increases, MR is less than price at all levels ' Using the notations used earlier, MRP = (ATRIAF) = [A(PQVAF) = P (AQ/AF) + Q (AP/AF) But for a perfectly competitive firm, P remains constant as QO changes. Hence, ar =U, “. MRP = P (AQ/AF). But (AQ/AF) = MPP “Mar = ) (MPP) = VMP.

414

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

of output. Therefore, when a factor is employed by a firm which sells its product under conditions of imperfect competition, the MRP (= MR x MPP) and VMP (= Price x MPP) of the factor are unequal because MR and price of the product are unequal and, since MR is less than price, MRP is less than VMP* Thus when the product market is perfectly competitive, MRP and VMP of a factor are equal to each other. But when the product market is monopolistic or imperfectly competitive, MRP and VMP of a factor are unequal and MRP is less than VMP. As explained in Chapter 8, since the behaviour of MPP of a factor in relation to changes in employment of that factor is governed by the laws of returns, the curve showing MPP of any factor takes the shape of inverted U. Thus when a firm increases the employment of a factor beyond a certain quantity and, as a result, the firm’s output is raised beyond a certain level, the MPP of that factor ultimately decreases. Now MRP is a product of two variables, MR and MPP because MRP = MR x MPP. Of these two the second variable, MPP

decreases as a firm increases employment of the factor beyond a certain level. The first variable, WR, either remains constant (when the product market 1s perfectly competitive) or decreases (when the product market is imperfect). In both cases, MRP declines as employment, and hence output, increases beyond a certain level. Hence, the MRP curve in response to increasing employment slopes downwards. However, since MRP is less than VMP when the product market is imperfectly competitive, the VMP curve is also downward sloping and MRP curve is situated below the VMP curve. For perfectly competitive product markets, the MRP and VMP curves

coincide as MRP = VMP. »)

.

:

:

“ Using the notations already used earlier,

MRP

= ATRIAL = eo le, = P(AQ/AF) + Q(AP/AF)

But AOQ/AF = MPP and VMP = P x MPP

(1) .. MRP = VMP + Q (AP/AF) Now on the right hand side of equaltion (1) we find that the (algebraic) sign of (AP/AF) is negative because as the firm increases employment of factor F (by AF) and thereby increases its output Q, the product price decreases (i.e. demand curve is downward sloping with a negative slope). Thus, MRP = VMP plus a negative quantity. “. MRP < VMP.

PRICING OF FACTORS

OF PRODUCTION

The MRP curve slopes downwards as beyond a certain level for perfect and shown below. Fig. /(a) shows the MRP a perfectly competitive product market. VMP curves for a firm in a monopolistic product market. Here, MRP is less than employment.

a factor’s employment rises imperfect markets. This is = VMP curve for a firm in Fig. 1(b) shows MRP and or imperfectly competitive VMP at all levels of factor

VMP MRP

VMP = MRP

0

Employment of factor F

Figure I(a) VMP MRP

0

Employment of factor F

Figure 1(b)

415

416

MICROECONOMICS

Demand

FOR

MANAGEMENT

STUDENTS

Curve for a Factor of Production

A firm’s demand curve for a factor of production shows the quantities of the factor which the firm is willing to employ at different factor prices. In order to analyse the nature of a firm’s demand curve for a factor, we make the following two assumptions: (1) while increas-

ing (or decreasing) its output the firm changes the employment of only one variable factor and keeps the employment of the others fixed; (2) perfect competition exists in the factor market, such that the supply of any factor for every buyer-firm is perfectly elastic. This assumption implies that the price of the factor would remain unchanged whether the firm employs larger or smaller quantities of the factor. Moreover, since the buyer-firm is a price-taker, it cannot influence the market price. The firm pays the factor its market equilibrium price determined at the level total demand for and total supply of the factor are equal. At any given price, a firm’s demand for a factor is based on whether it can increase its profits from the use of additional units of the factor. Thus, the firm compares the additional revenue arising from employing one extra unit of the factor to the price it pays for that unit. However, additional revenue from each additional unit is the MRP of

a factor. Hence, a firm compares a factor’s MRP with its price. As long as the MRP of the factor is greater than its market price, the firm continues to buy additional units since profit from the use of that factor increases. As the employment of a factor increases, the MRP of additional units decreases whereas price remains constant since the firm is a price taker by assumption. At some level of employment, the MRP of the factor is equal to its price. At this point, the firm’s profit 1s maximum and ceases to increase. Thus under perfectly competitive factor markets, a firm’s demand for a factor is such that the price of the factor is equal to its MRP. If the firm increases

the employment

of the factor beyond

this, MRP

of the

additional unit is less than its price, and profit falls. If the firm employs less than this quantity, then MRP of an extra unit exceeds its price and the firm’s profit increases by using more of the factor. Hence, profits are maximum when MAP = P. From the above discussion, it follows that assuming perfect competition in the factor market, the downward sloping MRP curve

is the same as the firm’s demand curve for the factor.

PRICING OF FACTORS

OF PRODUCTION

417

In Fig. 2, the negatively sloped curve F'D, shows an individual firm’s demand curve for factor F. It also shows the MRP of factor F at different levels of employment. The horizontal straight line say, PF, shows the perfectly elastic supply of factor facing an individual firm. At the given market supply P,F,, the firm would demand Q, of F’ since MRP of F is equal to its price P,. As price of F falls to P, the firm increases employment of F to Q,. Conversely, an increase in price of F to P causes the firm to decrease its demand to Q3. Factor Price

and

MRP

0

Demand for factor F

Figure 2

Marginal Factor Cost and Factor Supply The price paid to a factor of production constitutes the firm’s unit cost of employing it. Assuming that firms change their output by varying only one factor at a time, a firm considers how much is added to its total variable costs by increasing the employment of that factor. The addition to total variable costs made by one extra unit of the variable factor is called that factor’s marginal factor cost (MFC). MFC of a factor shows the change (increase or decrease) in the total variable cost resulting from the change (increase or decrease) of one unit in the employment of that factor. Hence MFC is equal

41&

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

to the ratio of the change in the total cost of a (variable) factor (A7TVC) to the change in the employment of that factor (AF), i.e.,

MFC = ATVCIAF °

The behaviour of MFC in response to increase in the employment of a variable factor depends upon the price that the firm has to pay for every additional unit of that factor. Whether the firm pays the same price or it pays higher and higher price for each additional unit of the factor depends upon the nature of competition in the factor market.

MFC

under Perfect Competition

Competition in the factor market is perfect when (a) there exist a large number of independent, relatively small buyers and sellers of that factor;* (b) the service provided by all suppliers of that factor is homogenous 1.e. all suppliers of that factor are equally efficient; and (c) it is easy for any potential supplier of that factor service to enter the market and any exiting supplier can easily exit from the market. These conditions ensure that in a perfectly competitive factor market, every employer or buyer-firm and every supplier of that factor is a price-taker with no power to influence the market price of the factor. Hence, the supply of that factor is perfectly elastic for an individual buyer-firm. Neither is the firm required to pay a higher price for employing additional units of the factor nor can it bring down the price by reducing its employment. In a perfectly competitive factor market the factor-supply curve facing an individual buyerfirm is horizontal and parallel to the X-axis Gust as in a perfectly competitive product market the demand curve facing an individual seller-firm is horizontal and parallel to the X-axis). Fig. 3(a) and 3(b) illustrate the relationship between the market-equilibrium price and an individual firm’s factor-supply curve in the perfectly competitive market for factor F. * Note that change in total cost (ATC) is equal to change in Total Fixed Cost

(ATFC) plus change in Total Variable Cost (ATVC). But by definition, ATFC = 0. Hence ATC

MFC

= ATFC + ATVC = ATVC. Also, from above:

= (ATVGAG) = MCX Mr

(CIar}

+ Consequently, each individual buyer-firm’s demand constitutes a very small proportion

of the total (market) demand

for that factor and each individual

supplier provides only a small proportion of total supply of that factor.

PRICING OF FACTORS

OF PRODUCTION

419

In Fig. 3(a) total (market) supply of factor F is shown by the upward sloping curve FS, and the downward sloping curve FD, shows the total (market) demand for F. The perfectly competitive market for factor F is in equilibrium at price P,; which corresponds to the intersection of FS, and FD, at point E,. In Fig. 3(b) the supply Factor

price

0

Total demand for and supply of factor F

Figure 3(a) Factor

price

0

Supply of factor for the firm

Figure 3(b)

420

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

curve of factor F facing any individual firm is shown by the straight line P,F, which is parallel to the X-axis at the level of market equilibrium price P,. This shows that for every individual firm the factor-supply is peffectly elastic. The supply of factor F to an individual firm would decline to zero if the firm offers it a price even slightly lower than the market equilibrium price P,. On the other hand, the supply of factor F available to an individual firm would increase infinitely if the price offered is even slightly higher than P,. Fig. 3(a) shows that the market (equilibrium) price falls to P, or

rises to P; when an increase or decrease in the total supply of factor Fis such that the market supply curve shifts right to FS, or left to F’S,. Consequently, as shown in Fig. 3(b), the individual firm’s factorsupply curve, while remaining parallel to the X-axis, slides downwards to PF, or upwards to P3F3. Thus, a firm purchasing a factor of production from a perfectly competitive market pays the same price for employing smaller or larger quantity of that factor. As a firm increases employment of a (variable) factor every additional unit of the factor increases the firm’s total variable cost by a constant and this 1s equal to the price of the factor. In other words, when there is perfect competition in the factor apart the firm’s MFC is constant and equal to the price of the factor.” Fig. 4 shows that in the perfectly competitive factor market the curve showing MFC 1s a straight line parallel to X-axis and it coincides with the individual firm’s horizontal factor-supply curve since MFC is constant and equal to price of the factor at all levels of employment.

> This can be explained as follows? Let Ff = quantity of the (variable) factor employed by a firm P, price of factor F TVC = total variable cost of factor F = P,F Let A indicate a small change in each of the nae variables vize /, P, and 7VC. Now MFC = ATVC/AFP = (AP, FIAF) = (P, AFIAF) + (FAP, /AF) = P, + F (AP; IAF). But in a perfectly Lsinperities market, a firm’s factor- supply curve is perfectly elastic and so P, remains constant as F varies 1.e. AP, = 0. Hence MFC= P,.

PRICING OF FACTORS MFC

OF PRODUCTION

4?]

and

Price of Factor

MECC = Price of factor

Se ae

kL

0

aes Ss a

Quantity of factor employed

Figure 4

MFC under Monopsony and Imperfectly Competitive Factor Market We have seen that when competition in the factor market is perfect, no single buyer of the factor can influence its market price and every buyer-firm is only a price-taker. The opposite situation prevails when the factor market is characterised by monopsony. Being the sole buyer of a factor, the monopsonist firm can set and control the price of the factor by changing its own demand. An increase (or decrease) in the employment of a factor by the monopsonist firm leads to a corresponding increase (or decrease) in the total (market) demand for the factor and this in turn leads to a rise (or fall) in the price of the factor, provided there is no change in its supply. Monopsony in the factor market generally occurs in the case of highly specialised factors of porduction. For instance, highly specialised labour, such as locomotive drivers or air-traffic controllers are employed only by the railways or in airports. This is also the case with capital goods made for only one specific use, such as signalling equipment used by the railways. Since the price of a factor increases as the monopsonist firm buys more of it, the supply curve facing the monopsonist firm is upward sloping. The average factor cost

422

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

(AFC),° or the price of a factor, rises as the monopsonist firm increases its employment. Hence, MFC also rises with increasing employment and MFC is greater than AFC or price at all levels of employment.’ In Fig. 5 the upward sloping supply curve FS shows that for the monopsonistic firm the price of a factor F rises as it employs increasing quantities of factor. The MFC curve is also upward sloping and MFC is greater than price at all levels of employment. Note that the positive intercept on the Y-axis indicates the minimum supply price of the factor. The supply curve of a factor facing an individual firm slopes upwards not only in the case of monopsony but even when the factor market is characterised by imperfect competition. This happens when there is oligopsony on the demand side or there exists

Price of Factor

MFC

(AFC),

_

MFC

eee,

0

Quantity of factor employed

Figure 5

© Let TVC = total variable cost; F = quantity of factor employed and P = price of the factor. Then, AFC = TVC/F = PF/F = P = price of the factor. 7 Let A indicate a change in TVC, P and F. Then, MFC = ATVC/AF = ACP P\/AF = P + F (AP/AF). Now since supply curve of factor is upward sloping AP/AF is > 0. Hence, MFC = P + (a positive quantity). .. MFC > P.

PRICING OF FACTORS OF PRODUCTION

423

monopoly or oligopolistic competition on the supply side of the factor market. In an oligopsonistic factor market there are only a few buyers of a factor. For example, there are in India few domestic airline companies, such as Indian Airlines, Jet and Sahara, who hire domestic airline pilots.® Also, there are only a few oil refinery companies in any country who buy refining machinery. When there are only a few buyers of a factor, a single firm’s demand constitutes a significant proportion of the total demand. Hence when an oligopsonistic firm changes its demand for a factor the market demand and, therefore, price of the factor is significantly affected. The supply curve of a factor faced by an oligopsonistic firm slopes upwards because it has to pay higher and higher price for a factor as it increases employment of that factor. Monopoly on the supply side of a factor market arises when the supply of a factor is controlled by a single seller. For example, employees of a factory come under the banner of a single trade union which acts as the monopoly supplier of labour negotiate to wages, hours of work and other work conditions. The monopolist seller can and does charge higher prices for supplying larger quantity of the factor because there is no other source from which the buyer-firms can obtain its increased supply. Consequently, the supply curve facing an individual buyer firm is upward sloping when the supply is monopolised. In an oligopolistic factor market the supply of a factor is controlled by a few major sellers. Markets for many capital goods such as heavy machine tools, textile machinery, transformers, trucks, etc. are oligopolistic in a developing country like India. Here, a single supplier can charge a higher price as its employment by a buyer-firm increases because this will affect total demand significantly. The curve showing MFC is positively sloped and MFC is greater than the price of a factor when : (i) a firm is a monopsonistic or an oligopsonistic buyer and/or (ii) a firm is buying a factor from a monopolistic or an oligopolistic seller of the factor.

* Since the training required for international airline pilots is much more rigorous and expensive, many pilots are trained only to fly domestic airlines.

424

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

Equilibrium Employment of a Factor The profit maximising employment level of a factor by a firm is called the equilibrium employment of that factor. As a firm employs additional units of a variable factor, the sale of additional output produced by it adds to the firm’s total revenue, TR. At the same time the payments made to these additional units of the variable factor add to the firm’s total variable cost, TVC. Hence while deciding whether or not to employ one additional unit of a factor the firm compares the amount which that extra unit (of the factor) adds to firm’s TR with the amount which it adds to its TVC. The profits derived by the firm from the employment of a variable factor increases as long as the addition to TR made one extra unit (of that factor) is greater than the addition it makes to TVC. Now the addition to TR resulting from one extra unit of a factor is defined as MRP and the addition to TVC arising from one extra unit of a factor is defined as MFC. Hence the profit derived from use of a factor increases so long as MRP exceeds MFC. We have explained earlier that as the firm increases employment of a factor,

its MRP

decreases

whereas

its MFC

either

remains

constant (if factor market is perfect) or MFC increases (if factor market is monopolistic or imperfect). Consequently, with increasing employment of a variable factor, the gap between MRP and MFC becomes narrower and narrower till it finally becomes zero. Hence the profit derived from employment of a factor is maximum when MRP = MFC. If the firm increases employment of the factor beyond the level where MRP

= MFC,

then MRP

will fall below MFC

and

profit decreases. Thus the profit maximising equilibrium employment of a variable factor is determined by the firm at that level where MRP of the factor is equal to its MFC. This can be explained as follows. We have seen in Chapter 11 that a firm’s profits are maximized when MR = MC. As explained earlier, MRP = MR x MPP; or MR = MRPIMPP. Also,

or

MFC MC

SG x Aer = -MPFOIMPP

Hence, a firm maximizes profits when or

MRP/MPP =

MFC/MPP

MRP

Vere

=

PRICING OF FACTORS

OF PRODUCTION

425

Thus, in order to maximize its profit from the use of a variable factor, the firm must employ the factor such that its MRP = MFC.

Equilibrium Employment Under Different Combinations of Product and Factor Markets | In the previous section we have seen that irrespective of the nature of competition in the product and factor markets, the general condition a rational firm ought to satisfy for maximising its profits from the use of a variable factor is to employ it in such quantity that MRP of the factor is equal to its MFC. We now apply this general condition for analysing the determination of profit maximising equilibrium employment of a variable factor under following combinations (models) of product and factor markets. (1) Perfect competition in the product market as well as in the factor market. (2) Monopoly

(or oligoply)

in the product

market

but perfect

competition in the factor market. (3) Perfect competition in the product market but monopsony oligopsony) in the factor market.

(or

(4) Monopoly (or oligopoly) in the product market and monopsony (or oligopsony) in the factor market.

Case I: Perfectly Competitive Product and Factor Markets In the real world there exist many small and medium scale industries, particularly those producing non-durable consumer goods such as textiles, plastic goods, food products, etc. where an average firm sells its product in an intensely (almost perfectly) competitive product market and buys its factors like labour and capital goods from keenly (perfectly) competitive factor market. We have seen that when a firm sells its product in a perfectly competitive product market, the MRP of a variable factor is equal to its VMP. Ina perfectly competitive factor market, a firm maximizes

its profits from the use of a variable factor when MRP is equal to MFC. From this, it follows that when there is perfect competition

426

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

in the product market as well as in the factor market, a firm determines equilibrium employment of a variable factor at that level where VMP is equal to price of the factor. This is shown in Fig. 6. Here P,S, is the perfectly elastic supply curve of the variable factor.

The MFC curve coincides with P,S, so that MFC = Price of the factor = OP,. VMP curve which coincides with MRP curve intersects PS, at point £,. Equilibrium employment F, corresponds to point fF, so that at F’, level of employment VMP = price of the factor = OP...

VMP. MFC & Price of Factor

S, = MFC

Rigel

i ,

= Prige

VMP (= MRP)

0

ee

|

Employment of factor

Figure 6

Table 18.1 explains this case of equilibrium employment in numerical terms by using illustrative data for employment, F, output, QO, product price P,, and factor price P,, for an umaginary firm. The firm’s profit maximising equilbrium employment of factor F is 60 units because at this level of employment VMP = P, = MRP = MFC = 50 When 60 units are employed the firm makes maximum 1500 from the use of factor F.

profits of

0 Ol OC Of OV OS 09 OL 08

[

(4)

|

40j90{ gf

augfO

ee

OS OS OS OS OS OS OS OS OS

rs

£

nee

>a

1s09

a

0 OOT O61 OLC Ove OOP OSL O6P OcS

b

wnporgVY

=inding

(OLY

ayqouyA — JO

jwjoy —

re OOS I OOO I OOS OOOT OOSC OOO’ OOSE OOOr

ace ei

pakojd — ua (d)

aojan{f 4

supfo

= Ol Ol Ol Ol Ol Ol Ol Ol

i

(Cd)

I

=

9

jonposg

ag§=fo

OOO! 006 OOLT OOPE OOOT OOST OO06P OOCS

(O'd)

anuaaa — y

jDIOL

ager TSE

= OS OS OS OS OS OS OS OS

id

dV

/OALY

=34W

ca Ol 6 8 L 9 ¢ v ‘

8

AVWOV

=ddic

r OOT 06 O8 OL 09 OS OV Oe

6 OO 06 O8 OL 09 OS OV Of

Ol

ddW

=JWA

AVALY X'd

aN--

I 1

OOS O06 OOCI OOrI OOS OOS OOrI OOCI

[I

DAL-~

=Ssijfaly

428

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

Case II: Monopoly (or Oligopoly) in the Product Market and Perfect Competition in the Factor Market Firms supplying electricity, rail transport, telephone service etc. are, as explained in Chapter 13, natural monopolies in the product markets. But in the markets for factors such as clerical or unskilled labour, capital goods like office equipment or hand tools or land (for office buildings), each of these monopoly firms faces nearly perfect competition. As explained earlier, in the case of a monopoly (or an oligopolistic) seller in the product market, the MRP of a variable factor is less than its VMP and the downward sloping MRP curve is situated below the falling VMP curve. Since this firm operates under perfect competition in the factor merket, it faces perfectly elastic supply of the variable factor. The MFC of its variable factor is therefore equal to the price of this factor. Any firm has to equate MRP with MFC to determine the equilibrium employment of a variable factor. Hence, the firm which is a monopolist in the product market but buys its variable factor from a perfectly competitive factor market reaches profit maximising equlibrium employment of a variable factor at that level where MAP of that factor is equal to its price. At the equilibrium level of employment the price paid to the (variable) factor is less than VMP because MRP

is less than VMP.

This difference between VMP of a factor and its price which arises on account of monopoly in the product market 1s called monopolistic exploitation. This concept is based on the argument that under ideal conditions of perfect competition a factor deserves to be paid a price which is equal to the value of what it produces at the margin of employment i.e. its VM@P. But on account imperfections in the product market its price is lower than what deserves and in this sense it is exploited. | Figure 7 shows that as in Case I P,S, shows the firm’s perfectly elastic supply curve of the variable factor and MFC curve coincides with P,S,. The negatively sloped MRP curve cuts P,S, at £,. The firm’s equilibrium employment is fF’; which corresponds to point £). At F, level of employment, MRP = OP, = price of the factor = MFC

and VMP=OV,=F,M,. We find that at equilibrium employment the price OP, of the variable factor 1s less than its VM@P as shown

by

PRICING

OF FACTORS

OF PRODUCTION

429

distance M,E,. Hence M,E,=V,P, shows the extent of ‘monopolistic exploitation’.

MRP.

VMP

Price and

MFC Vv;

P,

0

Employment of factor

Figure 7

lable 16.2 illustrates in numerical terms the Case II of equilibrium employment for an imaginary firm. Here the data for units of factor employed F, price of factor P,, TVC, output of product, MPP and MFC (1.e. columns | to 4 and columns 7 and 8) are the same as those in Table 15./ Note here that being a monopolist, the firm reduces Its price as it raises its output. In this case, the firm’s profit maximising equilibrium employment is 40 units of the variable factor F. At this level of employment, MKP = 49.99. And MFC =P,=50. Thus MRP is approximatley equal to MFC and the price of factor F. At this equilibrium employment, the firm earns maximum profits of 795.We find here that when

40 units of factor F are used by Jihe stitmarVM

P= 57.5

whereas the price of factor F is 50. Thus,VMP exceeds price by 7.5 and this measures the extent of ‘monopolistic exploitation’.

vOchS = CLI 88P OOL C6L C6L OIL 00S is

[I

DAL-~YL =SIfOld

AVWALVY ddw

6

‘OZ 06 SL8I OO'OE fiat = hL BC tS Or 00'8r 666 PS LS ocss oso O01L CO48 00001 0000! . e

Ol

ddW X"d =dWA

€ v ¢ 9 L 8 6 Ol 5

OS OS OS OS OS OS OS OS °

€

OE'L 96LE SL9E = OSL espn - Ge 00'°8 OOZTE Cole = GOB C6ct. -. OSB 00°6 OIL O0‘01 O00! " -

p

OCS 06r OSp OOF OVE OLT 061 001 0

£

000r OOSE O00E 00ST 0007 00ST 0001 OOS ia

c

OS OS OS OS OS OS OS OS OS

/

08 OL 09 OS OV O€ OC Ol 0

9

\ nme

gq sojov{ fo symp

L

pakojdwa (Ode & CHL

fO — ajqniupA yf 4019Rf jpioy — fo aaug = inding

(4)

Saleid

(‘q) AV

aMnuaday jonporg Joloy fo aug

|

(OF Vo

S105 JOALY =JAW

8

AV/OV =ddW

C8l 9IGeL

PRICING OF FACTORS

OF PRODUCTION

43]

Case III: Perfect Competition in the Product Market and Monopsony (or Oligopsony) in the Factor Market In the real world one rarely finds a firm selling its product in perfectly competitive market but acting as a monopsonist in the factor market. A typical case of this nature is that of the international airline company in a developing country like India, Pakistan or Kenya. There exists only one international airliner in each country, Air India, Pakistan International Airlines, Kenya Airways, respectively. In the global market for international passenger traffic each one of these airline companies faces intense (nearly perfect) competition. However, in relation to the domestic supply of a highly specialised factor, air pilots trained tor international flights, each of these national carriers 1s a monopsonist due to the existence of restrictions like visas and work permits. Inter-country migration is not easy and this compels

air pilots in India, Pakistan, or Kenya to sell their factor

services to their own national carriers. Thus, in the factor market for air pilots, each of these international airline companies acts as a monopsonist. Since the product market in which the firm sells its product is perfectly competitive, MAP of the variable factor 1s equal to its VMP. But since the firm 1s a monopsonist in the factor market, the MFC of the variable factor 1s greater than its price. Now for any firm the condition for equilibrium employment of a factor is that MRP=MFC. Hence the firm which faces perfect competition in the product market but is a monopsonist in the factor market decides equilibrium employment at that level where VM@P = MRP = MFC. But since the price of the factor is less than its MFC, it is also less than VMP or MRP. This difference between MRP and the price of the factor is called ‘monopsonistic exploitation’ because here the factor is paid a price lower than what it deserves, i.e. MRP and in this sense it 1s ‘exploited’ because of monopsony in the factor market. 7 Fig. & shows VMP curve intersecting MFC at point £,. Corresponding to point &, the equilibrium employment of factor Fis F). At F, level of employment, VMP = OV, = FE, and price of factor is OP,=F,R,. Thus, the price paid to factor F by the firm is less than VMP = MRP as shown by the distance V,P,=E,R, Hence monoposonistic exploitation is V;P,;=E,h,.

432

MICROECONOMICS

FOR

MANAGEMENT

MRP VMP Price and MFC

V,

STUDENTS

MFC

|

re po

3

E;

FS (= Price)

Lover R,

VMP = MRP

0

Fy

Employment of factor

Figure &

Table 18.3 given below illustrates Case III in numerical terms for an imaginery firm. This firm being a monopsonist buyer of factor F’, the price of F, i.e. Py increases as it increases employment of F. From Table /8.3 we find that our imaginary firm’s equilibrium employment is 50 units of factor F’ because at this employment level VMP = MRP = MFC = 50. When this imaginery firm employs 50 units of factor F, it pays the price of 57.2 per unit of F and makes the maximum profit of 1140 from the use of facotr ’. We find here that at equilibrium employment, while the price of F is 57.5, the VMP = MRP of Fis 60. Thus, price of factor F'1s lower than its MRP and so the extent of monopsonistic exploitation 1s (60 — 57.5) = 2.5.

Case IV: Monopoly in the Product Market and Monopsony in the Factor Market =

The case of a firm which is both a monopolist in the product market and a monopsonist in the factor market is not quite uncommon in the real world. In fact, most natural monopolies are not only the sole

[

(4)

oS

Ol OC Oe OY OS O9 OL O8

|

sojovf-— yf

aqfO

OCS Os gos S°9S CLS O'8S 06S 0-09

‘3

a

eee a

pasojdusa (dd)

vf 404 sf

spunfo

OCS O8OI S99 O9CC O98C O8Ve OCIL O08P

Ara

SOY =

OOT O61 OLC OVE OOP OSL O6P OCS

Vv

Camby =

9

|

¢

Ol Ol Ol Ol Ol Ol Ol Ol OOO 0061 OQOLC OOVE OOOP OOSP OO6P OOCS

(Od)

JMIO,

jonposg anuaaa — y

i

JINPOAd

(OVVE

-ajqvianaJO

Indin = g——nj adttg oy fo

eel8leC

Z OCS 09S g°8S S'6S 0-09 ee O'S9 Lo

AV

JOALY

-=J4W

6 OOT 06 08 OL 09 OS OV Of

8 Ol 6 8 L 9 S Vv e

Se ll

AV/OV

=Jd0.

OOT 06 08 OL 09 OS OV Of

OL

RSS Si a

ddW

aA =

AV/MLV Xd

duw

|

|

|

|

|

=

|

Ww

= o

I

Il

SS)

O8P OC8 SeOl OVIT OVI OcOl OLL OOP

2.

DAL~YL

=O

434

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

suppliers of their products but they are also the only buyers of certain highly specialised, industry-specific factors of production. For instance, Indian Railways has the monopoly of rail transport service and is the only buyer of a specialised factor, locomotive drivers. Each State Electricity Board is the only supplier (distributor) of electricity in the concerned state and it is the only buyer of industry-specific capital goods like transmission towers and turbines. Same is the case with organizations providing services like post and telegraph and drinking water supply. We have seen earlier that in the case of a firm which is a monopsonist buyer of a variable factor, MFC is greater than price. We have also explained that when a firm is a monopolist in the product market, MRP of the variable factor is less than its VM@P. Now

for any firm profits from the use of a variable factor are maximized when MRP of the factor is equal to its MFC. Hence in the case of a firm which is a monopolist as well as monopsonist, profit maximising equilibrium employment of a variable factor is determined when MRP is equal to MFC. But since VMP 1s greater than MRP and since MFC is greater than price of the factor, both VMP and MRP are greater than price of the factor. In Fig. 9 MRP curve intersects MFC curve at point £,. Corresponding to point &,, equilibrium employment of variable factor F’ is F, at which MRP = MFC. At F, level of employment, price of the factor is OP,;=F,N, as is shown on the supply curve factor FS and VMP is equal to V, as shown by ‘the VMP curve. Thus, at equilibrium employment F,, VMP exceeds price of the factor by distance V,P,= K,N,. Here ‘exploitation’ measuring the excess of VMP over price of factor arises on account of monopoly as well as monopsony.

Total exploitation

= VMP — P,; = OV, — OP, = V,P = K,N,

Here exploitation due to monopoly is V;H, =K,£, and exploitation arising on account of monopsony is H,P,;=E,N,. Table 18.4 illustrates in numerical terms equilibrium employment of a factor in the case of an imaginary monopolist-cum-monopsonist firm. Being a monopolist the firm has to lower price of product as output increases and being monopsonist the firm has to pay higher price as it increases employment of F The table shows. that equilibrium employment of variable factor F in our imaginary firm

PRICING OF FACTORS

OF PRODUCTION

VMP, MRP MFC, Price of Factor

435

MFC

FS-=. Prce

VMP

0

F,

Employment of factor

Figure 9

is 30 units. When 30 units of F are employed MRP = MFC = 58.5. Here, since VMP = 68 and price of factor = 55.5.Thus, VMP is greater than price of the factor. The firm earns maximum profits of 630 from the use of factor F when it employs 30 units. Here the total exploitation of factor Fis = VMP — Price of F = 68.0 — 55.5 = 12.5. This is made up of two parts: monopolistic exploitation which is = VMP — MRP = 68.0 — 58.5 = 9.5 and monopsonistic exploitation which is = MRP —P?= 58.5 —55.5 = 3.0. Note that this case of a firm being monopoly-cum-monopsonist should not be confused with ‘bilateral monopoly’. Here the same firm is a monopolist as well as a monopsonist. Whereas in the case of a bilateral monoply, one firm which is a monopolist faces another firm which is a monopsonist.

Marginal Productivity and Resource Allocation We have seen that the profits from the employment of a factor are maximized when the MRP of that factor is equal to its price. This suggests that assuming perfect competition in the factor market, the

0061 (-) O86(-) C8Z(-) OSTZ CIS O£9 O£9 O8P *

C6l 0'8Z CLE O'8b eis 0°89 O18 O00! a

oe) cS Ciel CcOr 0'0S CRE OIL O00! .

€ p ¢ 9 L 8 6 Ol =

0'L8 OSL O td, 0'L9 C19 C8 O'9¢ O'7S
), the /RR criterion fails. It is also important to note that the NPV and /RR criteria give the same select-reject answer if all R,’s of a project are positive (i.e. when the NPV curve does not intersect the horizontal axis in multiple points). However, the answer can differ if some K,’s are negative (i.e. when the NPV curve has multiple points of intersection

with horizontal axis). Moreover,

if the time profile

of the receipts, initial cost and economic life of different projects are similar, even the ranking of projects by the two criteria 7RR & NPV) would be the same; but if one or more of these differ among projects, the rankings given by NPV could be different from the ones given by /RR. This is because,

in the former case, the NPV

curves

of

different projects would not intersect one another in the positive

496

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

quandrant, while in the latter case they would (as is the case with projects X,Y, anid Z in Pig, 7.

Price Changes So far we have assumed that the general price level is constant over years or that all our calculations about the receipts and costs are carried out at constant prices. Now, if we drop this assumption, we can see what modifications are necessary. We shall illustrate it only with the NPV criterion. If the net receipts R,’s are evaluated at current prices (i.e. prices prevailing in corresponding years), we have to take note of the fact that such R,’s are not comparable to one another not only because they are received at different points, but also because the purchasing power of the units of their measurement differs. To make

them comparable,

therefore, we deflate the (R,’s) which

are

measured at the current prices, not only by the interest rate factor but also by the inflation rate factor. If the rate of inflation is given by 100p per cent

NPV = {{R\/(. +p) +0] + [R/C +p + iP] +... + Ree ee

cre eee oe

(9)

As a good approximation to this formula, we can merely take + p = 1' as a new discount rate and apply the old formula to calculate NPV which is not based on the assumption of constant price level. There is likely to be a definite difference in the result if the change in the

price of the final and intermediate products of the project can be accurately predicted. Moreover, if the price change for the products of the project is significantly different than the average rate of inflation or deflation in the economy, the NPV of the project would be definitely affected by considering the receipts at current prices and then deflating them by general inflation or deflation rate to take care of the purchasing power variations. However, in practice, in most of the cases it is difficult to envisage and predict the future prices of the products associated with the project vis-a-vis the general price trends. If one is not sure about the relative price movements over time, it is safer to calculate the costs and benefits of a project at constant base period prices. Considering price changes, particularly, the relative

INVESTMENT ANALYSIS

497

price changes, in future introduces an additional dimension of risk and uncertainty.

Risk and Uncertainty It is important to note that so far we have assumed certainty about future where important parameters in the project appraisal are known and unique. When we introduce elements of risk and uncertainty about the outcome of the project, we immediately realize that decision-making becomes a very complicated process because supernormal profits or losses as the reward or punishment for the right or wrong decision now arise. The methods of investment decision considering risk and uncertainty can be classified into two broad categories: informal methods and formal methods. We shall very briefly consider here these two methods. The informal method proceeds by classifying investments into various categories based on the nature and magnitude of the risk and uncertainty they entail. If the net returns from two projects or investments are the same, the one involving less risk and uncertainty would be preferred over the one involving more risk and uncertainty. However, even this criterion depends on the objective of the entrepreneur. If he has special preference for risky and uncertain situations, his preference may be quite opposite to what we have stated. More importantly, such projects are hardly found in actual practice. Very often, we have alternative projects where serious trade-offs between the profitability and risk and uncertainty are involved. Investments with high returns are generally more risky, whereas less risky investments tend to be yielding less but more assured returns. Well-defined trade-offs between these two aspects of the investments as perceived by the decision-taker holds the key. One way to incorporate the risk and uncertainty element into the project appraisal exercise is to consider a higher rate of discount by adding risk premium to normal discount rate otherwise considered. Such a procedure, however, would affect all the alternatives equally though they may not be equally uncertain or risky. The solution to this type of problem 1s to consider projectspecific discount rates. It only makes the analysis more complicated and less objective if the NPV criterion is used. However, if /RR is used as a criterion, some of these problems do not arise. Similarly,

498

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

the simple criterion of the pay-back period takes into account considerations of safety, security and certainty of the investment and its returns. The formal methods of analysis involves many computations and consideration of various alternatives. One such method is sensitivity analysis or what is known as the decisions tree method. The analysis is carried out for each project separately so that all situations which are possible can be fully considered. When the total number of possible outcomes is known and the associated probability of occurrence of each outcome can be estimated, the situation is said to be involving risk. Uncertain situation, on the other hand, would

arise when either the total number of possible outcomes is itself unknown and/or the associated probability of the occurrence of each " outcome cannot be estimated. For instance, the capacity utilization in an oil mill on account of production or procurement of groundnut involves risk, but the same on account of power failure or workers’ strike involves uncertainty. The sensitivity analysis involves as the first step identification of the most important and crucial variables which are subject to risk or uncertainty like capacity utilization, raw material prices, the exchange rate and the final product price. The second step is then to consider their possible different values. For example, if we identify capacity utilization and prices as the most important variables involving risk, we consider with definite probabilities, the different values which can be encountered in future.

Suppose there are two possible values of capacity utilization—one corresponding to an optimistic view and the other corresponding to a pessimistic situation. If there are four possible prices, then, we

have to examine in all 4 x 2 = 8 different situations and calculate the NPV for each one. Thus, we come to know the maximum and minimum possible profits out of the project. Moreover, since we can

also calculate the probability of each of the 8 outcomes, we can generate the expected NPV of the project by considering mean of the 8 NPV’s weighted by corresponding probabilities. For uncertain situations, we cannot calculate the expected NPV, but we could consider reasonable number of alternative scenarios. The most pessimist and the most optimist combinations of the factors would give the worst and the best expected NPV which can facilitate the decision-making.

|

INVESTMENT ANALYSIS

499

Social Benefit-Cost Analysis As we

have

seen

earlier, the basic criterion

used for investment

analysis in the private sector is the financial profitability of the projects. In the public sector, however, the criteria for investment decisions have to be broader than mere financial profitability. Since the public sector investments are usually very large in magnitude, they have significant effect on the whole economy. They are often undertaken to achieve more than one objective. The commercial profitability criterion cannot, therefore, be relied upon exclusively as a fair indicator for the desirability of the investment. For such investment decisions, say in the case of public utilities, we have to

consider wider implications of the proposed project on the society as a whole. In evaluating a public sector project, thus, we have to calculate the internal economic rate of return (IERR) and the social benefit-cost ratio (SBC. ratio) rather than internal financial rate of

return and the simple benefit-cost ratio. The basic difference between these two lies in identifying and evaluating the costs and benefits of the project. In SBC analysis, we not merely consider the profits accruing to the entrepreneur, but also the costs and benefits accruing to others in the society affected by the project. The SBC analysis becomes necessary because the project may have some externalities and the market price of the output and inputs do not necessarily reflect the true benefits or opportunity costs of the project for the society. By externalities of a unit, we mean the direct or indirect short-term or long-term effects of the operations of the unit on efficiency and/or functioning of other units, not accounted for in its price. Thus, if a big factory is established in a region, it will pollute air and will also generate demand for ancillary goods and services thereby contributing to the development of the surrounding area. The former is a real cost and the latter is a real benefit to the society. However, these things are not considered in calculating the commercial profitability of the factory. They represent the effects of the factory on the society, but are really external to the factory under the existing property laws. The SBC analysis, unlike the commercial profitability analysis, would consider these effects while deciding about the desirability of setting up the factory. If the amount of the investment involved is very large, 1t is most likely

500

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

that the magnitude, nature and importance of its externalities could be such that they can play a decisive role in determining the desirability or otherwise of the project. The second important justification for the SBC analysis is that the market price paid by the consumers of the final product of the project does not truly reflect the willingness of the consumers to pay for the product. The market price actually reveals the marginal valuation of the product by the consumers for the corresponding units of the product. Since the market price remains the same for all consumers,

it shows the minimum value that can be extracted by the producers from the existing consumers of the product. This is because the market demand curves are assumed to slope downwards implying an inverse relationship between the price of the product and the quantity demanded of the product. If 3 units are produced at a cost of Rs 10 and if they are sold at Rs 4 per unit, the total revenue would

be Rs 12 and the net profit would be Rs 2. However, the society has not gained only by Rs 2 because if only one unit of the product was sold, it might have fetched Rs 6. Then, if the second unit was sold

it might have brought Rs 5 and then, the third unit brings Rs 4. Thus, the 3 units are actually worth Rs 15 to the society and not Rs 12 as the market price of Rs 4 would suggest. The difference between what consumers are willing to pay and what they actually pay for the given amount of products is always positive and is known as consumer surplus. The ordinary commercial profitability criterion simply ignores this additional benefit which accrues to the society or the consumers. However, the SBC analysis duly recognizes this aspect because when we are considering certain major decisions for large investments such benefits may make all the difference in the decision. Similarly, the market price for the inputs does not always reflect the true opportunity cost of the input to the society. Government policies of indirect taxes and subsidies, administered prices and interest rates, import quota and tariffs protection, exchange controls and restrictions on movements of goods, services and people, etc. distort the price signals. Even when all these factors are absent, the market prices could still fail to reflect the true opportunity cost of

the input on account of imperfection of the market. Proper adjustments for all these factors are necessary in order to evaluate costs and benefits of a project from society’s point of view. The SBC analysis is especially useful in public utilities (or services) which

INVESTMENT ANALYSIS

may include projects related to power, transportation and telecommunication.

water

501

supply, sewerage,

Shadow Prices

It is clear from the above discussion that there are two basic problems underlying the SBC analysis: to identity the costs and benefits of the project to the society, and to evaluate them with the help of ‘appropriate’ prices. In practice, the first problem requires a complete acquaintance with all the technical aspects of the project and an imaginative foresight about the possible impact the project is likely to make on the economy. Here, we have to clearly demarcate various issues such as the domain of the project’s influence which can be determined reasonably by considering the nature of the project and the objectives sought to be achieved. The second problem assumes special importance in the SBC analysis because once the items of the social cost and social benefits are identified, it becomes necessary to compare the total social costs with total social benefits in order to judge the desirability of the investment from society’s viewpoint. The market prices, as we have seen, do not reveal the actual scarcity of resources and commodities. Some adjustments in the market prices of the resources and the products are necessary to estimate and evaluate the total social benefits and costs of the project. Instead of the market prices, we have to use a different set of logical accounting prices, known as ‘shadow prices’, to evaluate the benefits and costs of the project. Shadow prices of the inputs reflect the social marginal productivities of the inputs which are not always revealed through their market price because of distortions in the functioning of the price mechanism arising from factors such as market failures, controls, quotas. Thus, shadow prices are used in evaluating costs to remove such distortions and should not be interpreted as expected future prices. They are merely accounting prices indicating the realities of economic scarcity in relation to the demand in the society. The shadow prices of the inputs represent the true opportunity cost of the resource to the society. The opportunity cost of a resource to be used in the project can be measured as the product due to the resource foregone

in its best alternative occupation which is generally represented by

502

MICROECONOMICS

FOR

MANAGEMENT

STUDENTS

the value of its marginal product. On the other hand, the shadow price of the output represents the social average utility instead of social marginal utility of the product. This is because the true benefits of the project to the society are measured by the total utility generated by the product sold. There are several conceptual problems in estimating or determining the shadow prices of inputs and outputs. Shadow price of labour, shadow price of investment, social discount rate and shadow price of foreign exchange rate are particularly important parameters to be estimated in any detailed SBC analysis. We illustrate the issues involved in their estimation with the help of social discount rate and shadow wage rate for unskilled labourers. The social discount rate replaces the market rate of interest in the calculations of the NPV of the project in the SBC analysis. The social discount rate is supposed to reflect a true rate of time preference for society.’ It may be calculated as a weighted average of the rates of time-preference for different classes of individuals. Future generations for which they make everybody save. Thus future benefits may not be discounted heavily in SBC framework, the implication of which is to prefer relatively a lower social discount rate than the market rate of interest. Since NPV and the discount rate are inversely related, this would tend to raise the NPV of the project. To illustrate further, we may discuss briefly some conceptual problems in determining the shadow wage rate for a particular type of labour. If the conditions of perfect competition and relatively small variations in the supply of labour prevail, the market wage rate can be taken to reflect the true social marginal productivity of the given type of labour. However, in a labour surplus economy like India, it is argued, the opportunity cost of unskilled labour appears to be zero and the wages actually paid to them are consumed. The investment in the economy, therefore, declines by this total wage amount which could have been saved and should therefore be considered a cost to the society since it represents potential output loss in future for the society. But the wages to such labourers also raise their present consumption leading to an increase in their welfare which is a gain

to the society. Thus, a transfer of income from the employer to the ' See Appendix B at the end of Chapter 6 for one of the possible ways to estimate Social Time Preference Rate in India.

INVESTMENT ANALYSIS

503

unskilled workers in the form of wages in an essentially labour surplus economy is an increase in present consumption at the cost of future consumption. The shadow price of labour in the SBC calculations, therefore, has to be estimated by assigning proper weights to these aspects. Similarly, they are also conceptual problems encountered in determining the shadow price of investment and shadow price of the foreign exchange rate. After resolving these issues and appropriately estimating all these shadow prices, the social costs and benefits of a project can be evaluated reasonably satisfactorily. The actual method for computing the net social present value (NSPYV) or the internal economic rate of return (IERR), are the

same as described earlier for calculating NPV and IRR.

Appendix C Help Mr Brown to Invest

Mr David Brown is a qualified textile engineer having wide experience of working with several north-American firms for over 15 years. He has also developed excellent contacts with major chain stores and supermarkets operating in the continent. He has identified a clear and assured market for certain type of textile products in the USA and Canada. Some owners of the chain stores and supermarkets have offered to provide capital to him if he decided to start his own production anywhere in the world and enter into a long-term agreement of supplying the desired products. Mr Brown considered the offer very seriously and has already completed a plan and design for a $100 million production unit which has the potential of catering to their exact demand. When he discussed the project with his financiers, he found that the cost of capital for him would be 8 per cent per annum. Based on his vast exprience, he identified about 40 possible destinations all over the world where he can launch his project. After carefully considering the macroeconomic performance of the various economies, and other factors such as the availability of raw materials, skilled manpower, basic infrastructural facilities and political stability, he could short-list five countries which deserved a closer look as the possible destination for the project. He collected the required data and prepared projections necessary for appraising the investment proposal of starting the unit in each of the 5 countries. These figures are given in Table I. It can be seen that the 5 countries differ significantly from oneanother not only in terms of the cost of various items, but also in terms of the government policies regarding permissible depreciation, corporate income tax rate, and exit policy. Differences in the exit policy are reflected in terms of the salvage value of the project, which includes the scrap-value and the terminal cost, at the end of its 10 year economic life. Mr Brown carefully considered certain technical changes in the lay-out and design to suit the local conditions in the

APPENDIX C

505

case of each of the 5 countries. He also estimated the capacity utilization in the initial period before the unit settles down to its full utilization in each country. According to his estimates, the rate of capacity utilization in the initial years in different countries is given in Table 2.

Table | Estimated Annual Costs and Revenues at 100 per cent Capacity Utilization of the Project in each of the 5 Countries (in $ Million) Items

Countries A

B

&

D

E

/

2

&)

4

5

6

Sales Raw Materials & Labour Cost Fuels & Infrastructural Facilities Salaries & Rent Repairs and Maintenance Interest @ 8% Permissible Depreciation Rate of Corporate Tax Net Salvage value after 10 years

80

80

SO

80

80

20

20

15

2S

15

10 3

16 2

10 4

(5 2

10 5

2

eg

4

v3)

5

8

3)

8

8

8

12 40%

9 30%

S 50%

10 25%

8 20%

3

10

(—)40

15

(-)35

Table 2

The Rate of Capacity Utilization in each of the 5 Countries (as percentage) Years

A

B

S.

D

E

f

Z

)

4

a

6

|

50

50

60

70

10

2

70

80

80

100

50

3

100

100

100

100

80

4

100

100

100

100

100

S06

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

Questions for Discussion |You. may use lotus, excel or other similar packages for efficient calculations. In calculating cash-flows, treat $ 100 million as outflow during year zero.] 1. NO

3.

Considering all the information, what advice you will give to Mr Brown about the location of his project?

A well-known international agency assessing the ‘country risk’ or ‘systemic risk’ factors for international investors suggests risk premiums of 2 percent for countries A and C; 3 per cent for country B; 3.5 per cent for country D; and 4 per cent for country E. What would be your advice to Mr Brown after incorporating this information? If it is possible for Mr Brown negotiate with governments a tax holiday for the first one or two yars of operation, which countries should he consider. :

Appendix D

Social Cost Benefit Analysis of the Sardar Sarovar Project.

Narmada is the longest westward flowing perennial river in India. Historically, it was considered the dividing line between

and the south.

It covers

a total length of about

the north

1312 km

from

Amarkantak in Madhya Pradesh (MP) where it originates, to the Gulf

- of Cambay in the Arabian Sea, near Bharuch in Gujarat. It flows over a distance of 1077 km in MP, 35 km as the border between MP and Maharashtra, 39 km as the border between Maharashtra and Gujarat,

and finally 161 km in Gujarat. All along the route, a total of 41 small and large tributaries join the river. Over 20 million people, including tribals, inhabit the river basin. This amounted to almost 2.5 per cent of India’s total population in 1981. The annual flow of water in this perennial river is estimated to be approximately 27 Million Acre Feet (MAF) with 75 per cent dependability. Barely 2 MAF out of this flow 1s currently used. In terms of per capita income, Gujarat ranks fifth in the country after Goa, Punjab, Haryana and Maharashtra. Its per capita income (State Domestic Product) was about 20 percent higher than the national average in early eighties. Of late, its relative position has declined considerably. Agriculture and other primary activities play a very significant role in the state economy contributing 28-38 per cent of the total income generated and 60-65 per cent of the total working force. Progress of agriculture in Gujarat, however, is seriously hampered by the availability of water. Only 18 per cent of the net area sown Is irrigated, the rest being dependent on rains. Frequent occurrence of droughts and floods characterizes the economy. Sometimes droughts and/or floods occur for two to three years

in a row,

destabilizing

the

economy,

detracting

it from

the normal growth-path and demoralizing the population and enterprise. Out of the 18 percent net irrigated area, almost threefourths is accounted for by the well-irrigation. Further prospects of

508

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

ground-water development are not bright in Gujarat. The cropping intensity has been fluctuating at a very low level between 1.08 and 1.09. The crop yields are also not as high as they are in other states with assured water for irrigation. In terms of industrial development, Gujarat ranks second in India—only after Maharashtra. In terms of the necessary infrastructure for the growth of industry, Gujarat faces acute shortage of power. Currently, it relies on thermal power which makes it highly dependent on coal. Heavy transport cost of coal and the nature of technology makes power relatively very costly as compared to most other states. In fact, availability and cost of power are major impediments in the growth of industry. In this situation, major multipurpose projects on rivers are not merely very attractive but perhaps the only alternative for the state. The Sardar Sarovar Project on the Narmada river, is one such.

Historical Background The identification exercise for major irrigation projects in the Narmada river valley began in 1947. The Broach (Bharuch) project in Gujarat was one of the seven identified for detailed investigation. In 1948, priority was accorded to four out of the seven projects. These projects were: Tawa, Bargi and Punasa projects in MP and Broach project in Gujarat. After detailed investigations were completed for the Broach project in 1957, Navagam site was recommended for the dam. Other parameters of the project remained undecided because it was the terminal project for the whole Narmada system. The developments on the upstream were to dictate its scope. Between 1957 and 1960, the technical details of the dam could not be finalized owing to several alterations in the scope of the project. In 1960, Gujarat acquired the status of a separate state. It, then,

undertook a detailed study which recommended that the height of the dam should be raised. The MP state government considered this recommendation in 1963. However, the two state governments did not agree on sharing of water from the dam. In 1964, therefore, the dispute was referred to a high-level committee of ie ae headed by Shri A. N. Khosla.

APPENDIX D

509

The Khosla Committee Report submitted in 1965, and the various

rounds of discussions among Gujarat, MP, Rajasthan and Maharashtra could not resolve the issue. In 1968, therefore, the Gujarat govern-

ment finally launched a complaint under the Interstate Water Disputes Act. The central government then constituted Narmada Water Disputes Tribunal in 1969 for the purpose. The Tribunal took 9 years to give its award in which it gave clear directives on various matters, such as sharing of the Narmada water, hydel power, the height of the dam and release of water. After this, Gujarat government started

acting with firm determination in matters related to the Narmada project at Navagam. It was named as Sardar Sarovar Project (SSP). The process of planning and implementation of the project was streamlined by the creation of the Narmada Planning Group (NPG). It consisted of engineers, economists, agronomists and other experts, its basic task being formulating, planning and implementing the project. SSP involved massive resources. The project cost was estimated at Rs 5,000 crores at the time of its first economic appraisal in May 1983. Subsequently, it was revised at Rs 6,500 crores in 1986. With the July 1991 devaluation of the rupee, it was estimated at Rs 9,000 crores. The project is the biggest and the most expensive so far in India. This single project is likely to cost as much as all major and medium irrigation projects of the centre and all states taken together during the entire Seventh Plan period (1985-90). Since implementation of such a large-scale project only through the state budgetary resources was impossible, the state government decided to approach international agencies like the World Bank for financial assistance. The Bank, in turn, required various studies to be done before it could

consider the request. The NPG shouldered the responsibility for carrying out these studies quickly. Several of the required studies were done through the state government departments and the rest by research institutions, universities, academic bodies and consultancy

organizations. The NPG selected the Tata Economic Consultancy Services (TECS) in 1981 to undertake the first economic appraisal to get a broad idea of costs and benefits of the project by April-May 1982. In December 1981, TECS prepared an approach paper on ‘Methodological Framework for Economic Appraisal of Narmada Project’, which discussed the assumptions to overcome the data

510

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

gaps due to non-availability of primary data and the methodology followed. According to Mr Vijay Paranjpye, one of the foremost critics of the SSP, the TECS report made available in printed form in May 1983 is a document on which the justification of the project and its

financial approval from the Planning Commission of the Government of India rest. The World Bank’s approval of the project in 1985, however, was based on the economic analysis of the project carried out by its own experts in February 1985. However, the Bank’s analysis suffered from the major limitation of ignoring or underestimating certain environmental costs of the SSP. A lot of controversy, people’s movements and agitations took place in India against the SSP. A correct and full economic accounting of all costs and benefits including environmental aspects was considered essential and the Bank, therefore, sponsored a major report on environmental impact of the SSP and the Narmada Sagar Project (NSP) in Madhya Pradesh. The Report was submitted in 1987. It carefully reviewed the physical, social and environmental issues associated with these dams. The consultant, D. W. Lavenhagen, considered scenarios for both the dams with and without the projects. His report pointed to the immense environmental benefits from the project provided proper practices of environmental management were followed to minimize the cost. The Department of Forest and Environment in the Government of India took a tough stand on the two projects and delayed their final clearance. However, with the availability of Lavenhagen’s findings on the two projects, it agreed for a conditional approval

of both

the SSP

and NSP

in 1987;

and hence

the two

projects got their final clearance from the central Government in April 1987. The critics of the SSP made representations to the World Bank and various other international forums for a comprehensive review of the project particularly for integrating impact on environment. Consequently, the World Bank experts made field visits in April-May 1989 to modify and update their original economic analysis of the SSP. In December 1990, the Bank finalized its review of the SSP Economic Analysis which found the SSP economically

viable after incorporating environmental costs and benefits. The purpose of the present case is to generate discussion on important issues of the social cost-benefit exercise of multipurpose project of large dimensions. The case of the SSP in this context is

APPENDIX D

511

highly relevant and interesting because its economic appraisal has generated a lot of debate and controversy. It 1s also used as the main argument by the supporters and opponents of the project. Politicians, activists and even laymen talk forcefully about it in public forums within and outside the nation. The first economic appraisal of the SSP carried out by TECS (1983) assumes considerable importance in this context though, by now, it is generally considered a totally outdated and invalid exercise. The whole controversy leading to serious agitations against the SSP started with the critics finding several faults with this Report. In the next section, we discuss Paranjpye’s (1990) analysis of the SSP modifying and revising TECS estimates to obtain his estimates of the social costs and benefits of the SSP which are acceptable to the opponents of the project. In the final section, then, we raise a few issues on the economic appraisal of such projects.

Paranjpye’s Revision of the TECS

Estimates

The main limitations of the TECS report as perceived by its critics are: it is based on inadequate data, it makes inappropriate assumptions and it provides an improper treatment of costs and benefits of the project. Their perceptions were, however, based on the initial methodological note prepared by TECS. At that time, the results of the bench-mark surveys initiated by the NPG were not available, hence TECS proposed to substitute hard data with some ‘heroic’ assumptions. However, by the time the TECS report was finalized, all the survey results were available and were, therefore, used by TECS for estimating various benefit and cost items. Critics feel that the NPG sponsored TECS study overestimates benefits and understates costs. The costs considered by TECS were worked out by the Gujarat government officials who used shadow prices rather than market prices. The method used for estimating the shadow prices was popularized by Little and Mirrlees, by taking the international border prices of tradables coupled with the standard conversion factors and assuming some proportion of actual wages reflects the shadow wage-rates. The critics of the SSP have severely criticized the TECS report for understating the project costs. Their main argument is not so much against the methodology for

512

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

estimating shadow prices in general, but against ‘major omissions’ in calculating the total project costs. Paranjpye (1990) made an attempt to modify the estimates of the TECS report in the following way: (a) The cost of rehabilitation and resettlement (R&R) was grossly underestimated in the Report. Here the main contention against the Report is that it ‘used notional figures for estimating the probable costs that would have to be incurred as cost of land and other resources submerged under the reservoir. It is, therefore, obvious that the cost of R&R as recognized today had not

been anticipated in its full magnitude by the government of Gujarat’. (Paranjpye, 1990, p. 168). According to Paranjpye, this cost must now be revised and we must add Rs 244.74 crores (at 1981-2 prices) to the total cost of dam and dykes of Rs 621 crores reported in the TECS report. This is because the Chairman of Narmada Nigam announced in 1988 that the Gujarat Government would have to spend Rs 600 crores for R&R largely on account of increased prices of land to be transferred to the oustees and cost of providing infrastructural facilities on the new locations over and above the ones existing on the old locations which would be submerged. Applying an average inflation rate of 8 percent each year, this amounts to Rs 378.10 crores at 1981-2 prices. Paranjpye argues that Rs 133.36 crores were already considered for R&R in the Report. Hence, only Rs 244.74 crores (= 378.10 — 133.36) need to be added to the total cost. It may be mentioned here that TECS had estimated Rs 133.36 crores as the cost of the land and non-movable infrastructural facilities which would be submerged in the reservoir. Land would include all types of land, including cultivable land and forest land. The figure also included the cost of rehabilitating people displaced by the project. Economic cost of land under submergence is estimated by TECS as the economic value of output from land that would be foregone forever. Economic prices are used for this purpose. Similarly, income expected during the period of clearing the forest before submergence is subtracted from the project cost. (b) Since the World Bank experts subsequently changed the base

from 1984 to 1989 and revised the cost of construction upward

APPENDIX D

513

by Rs 590 crores, this baseline increase in the cost has to be added to ‘reflect the actual baseline cost’. This increase in the cost is due to varying rates of inflation for different items rather than any change in design or material content. At 1981-2 prices, the increase in the cost amounts to Rs 375.58 crores. (c) The SSP is dependent on the regulated water release by the upstream reservoir NSP. A part of the total benefits of the SSP would

not accrue in the absence of the NSP. As a result, the

tribunal recommended 18 per cent of the cost of the NSP main dam and works to be borne by the SSP. This element did not figure in the TECS report because the cost of the NSP was not known at that time. The total cost of the NSP is estimated as Ks 2167 crores at 1987-8 prices. The share of the SSP, however,

(d)

works out to Rs 146.47 which amounts to Rs 92.47 crores at 1981-2 prices. The catchment area treatment costs were not considered in the TECS report. This expenditure is necessary for maintenance of the dam and must be included in the economic analysis. (Paranjpye, 1990, p. 169). The inter-departmental expert committee estimated the cost to be Rs 1300 crores at 1984—5 prices. The share of water from the Narmada for Gujarat is 32.1 percent, and attributing a corresponding cost share to the beneficiaries,

the share of Gujarat would be Rs 495.3 crores.

When adjusted for 1981-2 prices, this would be Rs 357.76 crores. (ibid). As against this, the World Bank argues that there are two choices with catchment treatment: both the costs and benefits should be-ignored or both should be included. This is important because catchment area treatment has substantial in situ agricultural benefits and is generally found to be paying for itself particularly in India. (€) Regarding the downstream losses particularly to the Bharuch district economy after the dam is built, Paranjpye's estimates are highly tentative and uncertain. He argues that loss of fish alone in the downstrem areas would amount to Rs 10 crores per annum and estimates a loss of Rs 500 crores at 1983-4 prices for 50 years on this account. At 1981-2 prices, this is equivalent to Rs 428.66 crores which should be added to the total cost of the SSP. However, the World Bank argues that the impact on estuary cannot be predicted with any degree of certainty. Since the SSP

514

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

controls less than 20 per cent of the river flow, it would depend on Aas HER: occurring upstream over next 30 years. Expe rience elsewhere does not suggest that total value of fish catch should inevitably fall substantially. However, in the absence of certain measures, it may fall significantly. ({) The cost for archaeological losses due to submergence which was also not included in the TECS report is estimated by Paranjpye to be Rs 1.5 crores at 1981-2 prices. (g) The compensatory afforestation cost is again not included by the TECS _ report. The total area for the compensatory afforestation is estimated 13,387 hectares, out of which 4600 hectares are in

Gujarat. The Gujarat government has set aside only Rs 1600 per hectares as its share. On the other hand, the per hectares cost of afforestation in MP was estimated to be Rs 18839.95 at 1986-7 prices. Deflating this amount to 1981-2 prices and applying it for the entire compensatory afforestation (i.e. 13,387 hectares), Paranjpye (1990) estimates the cost to be Rs 17 crores which according to him should be added to the total cost of the project. (h) Assuming the utilization rate of only 80 per cent, Paranjpye reduces the estimates of annual benefits by 20 per cent since it was recognized that 20 per cent of the proposed command area will not be irrigated by the project.

With these modifications and revised estimates, Paranjpye (1990) comes to an estimated ‘Internal (economic) Rate of Return’ for the SSP on the Narmada river. He considers two scenarios: (a) assuming

no time overrun and 100 per cent accrual of benefits from irrigation and power as assumed by the SSP authorities; and (b) assuming a

time overrun of 5 years, irrigation utilization of 80 per cent and 100 per cent generation of hydel power. In the case of (a) he finds an IRR (economic) of 10.8 per cent. In the case of (b) he gets an IRR (economic) of 9.2 percent. In the light of recommendations of the national committee of experts headed by Nitin Desai to consider 9 percent as cut off opportunity cost of capital (7 percent for drought-prone areas), Paranjpye (1990) draws the conclusion that ‘a

project involving such a huge investment which cannot cover the interest on capital is not worth undertaking’ (p. 173). He argues that

the project involves a gestation period of about 20 years and that the

514

APPENDIX D

real rates found to financing Paranjpye

of roturn from large-scale irrigation projects are usually be around 5-6 percent only. Moreover, the pattern of the SSP and the likely interest liability according to (1990) are given in Table 1.

Table 1 Interest Liability for the SSP (as percentage)

Source I a)

Planning Commission

b)

World Bank, IDA Loan, etc.

c)

Public Borrowing

Approximate Nominal Proportion of Interest Cost Rate

| Remarks

2

3

35-40

Nil

Plan funds to be used for critical development projects

10

7.6

To be returned in hard currency

50-55

14-15

Would impose tremendous public debt burden

-

d

As against this, the market rate of interest on long term investment may be taken to be 12 per cent. Since the opportunity cost of capital is very high in the Indian Capital market, Paranjpye does not find the SSP worth undertaking.

Issues for Discussion

Considering all the facts presented above, some specific issues for the discussion could be:

(a) Identification of costs and benefits In a social cost-benefit analysis of such mega projects, proper identification of all relevant costs and benefits plays a crucial role. Although very good guidelines are available from theory, it is

516

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

important for all concerned with the project, politicians, activists, technocrats, and laymen, appreciate and be aware about the justifications for including or excluding certain items of cost and benefits arising out of the projects. Ignoring this issue may not only be misleading but also be counterproductive. As we have already noted, the opponents of the SSP led by Paranjpye have case on ‘major Omissions’ in estimating total cost of the SSP. How many of them are genuine omissions?

(b) Valuation of benefits The costs of such mega projects are generally evaluated at the social Opportunity costs of the inputs for which appropriate methods of estimating the shadow prices are followed. The benefits of such projects should, therefore, be valued in terms of their contribution to the social welfare. Thus, valuation of the benefits must consider

incremental social welfare due to the increased output. This is applied particularly for valuing the significant benefits of the project like additional power and water supply for domestic industrial uses. The prevailing market prices are not likely to measure the additional social welfare generated by the project output since the government usually subsidizes such critical inputs for developmental purposes. Proper evaluation of these benefits incorporates other concerns and objectives of the society into the appraisal of the project. The TECS report, without assigning any reasons, uses opportunity cost approach rather than willingness-to-pay approach for evaluating the benefits. The World Bank (1990), on the other hand, uses the latter

approach. The opponents of the SSP are silent on this issue. The TECS report has followed the opportunity cost approach and estimated the power benefits as the cost averted following the supply approach. Thermal plants (210 megawatt capacity) for power supply on firm basis and gas plants (with 150 megawatt capacity) for peaking purpose (at margin) are considered as the practical and relevant alternatives for power generation in the Indian context. TECS estimated the value of power supplied from the project 65 paise per kilowatt-hour of non-peaking power and 115 paise per kilowatt-hour of peaking power at 1981-2 prices. The World Bank (1990) estimates of imputed surplus for different

categories of consumers of power to be generated from the SSP are

APPENDIX D

St?

presented in Table 2. The consumer surplus in each consumer category is estimated as half the value of the difference between the economic cost of private supply and the economic value of the tariff rate. This method is based on the assumption that price elasticity of demand for power is unity. In practice, price elasticities at low levels of electricity tariffs in LDCs are well below unity which implies that the estimated surplus would underestimate the true surplus and hence the value of the benefits. Table 2 Estimated Rates of Consumer Surplus (For FY 93)° Consumer

Category

Consump-

— tion Share

Auto-

Esitmated Average

Surplus

generation

Tariff (ps./kwh)

Imputed

(In %)

Cost

Z

3

17.1

2S

52

42

aw

5.3

2S

87

70

103

(LV + MV) Industry (HV)

8.3 46.3

207 207

99 121

79 97

64 55

Agriculture

16.4

34 1°

19

15

163

/ Domestic Commercial

Financial’ Economic’ 4

>

(ps./kwh)4 6

Industry

Other'

6.6

207

120

98

a5

100.0

244

89

di

87

All Uses

(Average)

‘The first year of incremental supplies from projects included in the programme time-slice. Estimated on the basis of FY88 actuals increased by 4.7 per cent per annum—the average annual real increase in tariffs in the Western Region between FY85 and FY88. © Financial rates multiplied by SCF of0.8. Surplus imputed at half the difference between the economic value of the average tariff level and the estimated cost of autogeneration. The electricity rate equating the economic costs of diesel and electric pumping. Comprises public lighting, public water pumping and bulk sales.

518

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

The main problem is with the estimation of the willingness to pay value of Municipal & Industrial (M & I) water supply. Without any scientific approach to estimate this critical output of the SSP, the World Bank (1990) takes the value to be less than 3 paise per gallon even after 30 years when the population would have almost doubled. However, it recognizes that during the drought years, the problems of water shortage in the project area have been acute. Very high price per unit of water is known to have been paid during the droughts which hit the area very often. Water scarcity in the project area, moreover,

leads

to severe

shortage

of fodder

and

starvation

of

livestock forcing the seasonal migration of people and livestock on a massive scale. Only the better off can afford such migration. The poorest of the poor are forced to stay back in starvation waiting for water and fodder to be transported through rail or road by the state. In the absence of the SSP, a large proportion of population in the project area will be forced to resettle with livestock in an unplanned and uncompensated manner. Expenditures required to preserve the welfare of the population should necessarily be considered while arriving at the estimate of the willingness-to-pay for M & I water use. Value of 3 paise per gallon 1s certainly a serious underestimation of the true value of benefit on account of M & I water output of the SOP,

(c) Treatment of environmental costs and benefits In projects of such magnitude as the SSP, the environmental impact is very significant. The basic issue 1s that in the light of high degree of uncertainty attached to the environmental costs and benefits of the project, how much emphasis should be placed on their estimation? How much weightage should be given to these aspects in taking final decision about the project? The World Bank’s approach after 1989 as contained in its publication [see Dixon et al. 1989; pp 45—9] on this issue is to integrate environmental aspects totally into project design and operation. The Bank’s critical look at its own past experience suggests that “measures to assure environmental protection can often be shown to have economic benefits that exceed their economic costs and that even in case where it 1s not practical to

quantify them, these benefits, especially the avoidance of irreversible effects may justify the cost of protection; and that preventive measures provide more effective and substantially less costly protection

519

APPENDIX D

than later remedial measures’ [ibid., p.47]. The opponents of the SSP have used the environmental aspects as an important argument only in terms of their impact on costs and not on benefits. The World Bank (1990)

has reviewed

their original economic

appraisal

(1985)

by

considering both the environmental costs and benefits of the SSP under each of the flowing heads:

L. 2. 3. 4.

“wildlife Parks and natural reserves Compensatory afforestation Minor forest produce, e.g. gums, smoke

>. 6. 7. 8. 9. 10. 11.

dyes,

liquors,

medicines,

leaves, étc.

Fublic health Navigation Monuments and village shrines Rare plants Carbon-oxygen balance through vegetation and biomass Micro-climate, i.e. moisture in the atmosphere Induced pressure on land

12. Flood control on the downstream

5,000 hectares

13. Salinization 14. Ground water 15. Tourism

The approach of the World Bank (1990) for estimation of costs and benefits

of all these items

has been

the one

of conservatism.

It,

therefore, estimates a net effect of all these heads considering the environmental aspects of the SSP to be an addition of Rs 6 to 8 crores (at 1989 prices) per annum over 50 years in the cost of the SSP.

(d) Resettlement and other socio-cultural aspects Large dams like the SSP, usually require involuntary resettlement of several thousand people. Apart from the cost of the land being inundated along with the replacement costs of facilities and other assets, the psychological costs inflicted upon the population being shifted is also considered relevant for social welfare. The SSP will create a 214 km long reservoir submerging 248 villages of which 245 villages will get only partially submerged. It will displace about 1 lakh people out of which about 70 per cent are tribals who have

been considered as socio-economically backward and vulnerable by

520

MICROECONOMICS

FOR MANAGEMENT

STUDENTS

the constitution. The 39 thousand hectares which would be inundated by the reservoir consists of sizeable degraded forest areas (approx.

14 thousand hectares.) and. agricultural land of Moreover, the gains of the draw down farming considered. Both the scenarios, with the project project need to be considered for the socio-cultural

inferior quality. should also be and without the costs as well as

the benefits of the resettlement and rehabilitation of the project affected population (PAP). The increased pressure on land due to 2

per cent population growth p.a. on the SSP command area of 1.8 million hectares has serious implications. So far the debate between the proponents and opponents of the SSP has centered around the costs of R&R only. The benefits of R&R as well as scenario without the SSP also need to be considered for an objective analysis of the project.

(e) Distributional aspects Usually, it is argued that large dam projects such as the SSP might Increase the gap between the rich and the poor in the society; and that the poor and the vulnerable sections bear all the costs while the rich get the benefits. Although distribution of income is a concern for fiscal action, large dams do not necessarily have undesirable distributional implications. If they are carefully planned, designed and implemented, the benefits flowing from such projects to the poor in terms of drinking water, water for irrigation, electric power, flood

control and fisheries could be substantial to improve their conditions. The whole issue of distribution of costs and benefits has to be viewed in the larger context of the complementarity or competitiveness of erowth and equity.

(f) Delay in decision making Justice

delayed

is, often, justice

denied.

An

opportunity

lost is

similarly a permanent cost. The time spent on taking decisions on such mega projects as the SSP is not without its costs since the project is ultimately economically viable. An external lending

agency

may

not be concerned

with this type of cost, but the

benefitting society should be.

(g) Sensitivity analysis To consider the various types of uncertainties and risks involved in the project benefits and costs, sensitivity analysis of the project

APPENDIX D

521

identifying crucial parameters has to be performed. Paranjpye (1990) has tried one such exercise. The World Bank (1985 & 1990) has also

attempted such exercises.

References

6.

Alvares, C. and R. Billorey (1988): Damming the Narmada, Natraj Publishers, Dehra Dun. Dixon J. A., L. M. Talbot and G. J. M. L. Moigne (1989): Dams and the Environment: Considerations in World Bank Projects, World Bank Technical Paper Number 110, Washington D.C. Levenhagen, D. W. (1987): Overview Environmental Impact Report: Sardar Sarovar and Narmada Sagar Complex Project Area, Narmada River Basin, India. Consultant’s Report, The World Bank, Washington n. Narmada Bachao Andolan (1990): Sardar Sarovar Project: An Economic, Environmental and Human Disaster, Bombay. Paranjpye, Vijay (1990): High Dams on the Narmada: A_ Holistic Analysis of the River Valley Projects, Indian National Trust for Art and Cultural Heritage, New Delhi. Sheth, Pravin (1991): Sardar Sarovar Project: Dynamics of Development, Vikas Bharati Institute of Policy Studies, Research and Futurology, Ahmedabad. Tata

Economic.

Consultancy

Services

(TECS)

(1983):

Leonomic

Appraisal of Sardar Sarovar Project, Bombay. World Bank (1985): Narmada River Development—Gujarat—Sardar Sarovar Dam and Power Project, Staff Apprisal Report 5017—IN, February 12. World Bank (1985): Narmada River Development—Gujarat—Water Delivery and Drainage Project, Statf Appraisal Report 5018-IN, February 12. 10. World Bank (1990): Review of Sardar Sarovar (Narmada) Projects Economic Analysis, India Agriculture Operations Division, December 16.

Questions for Discussion Consider critically the modifications suggested by Paranjpye in the estimates of costs and benefits of the SSP. What would be

ae

bo

MICROECONOMICS

STUDENTS

the additional social cost of the project at 1981-2 prices according to you? Why? In the light of Paranjpye’s calculations, do you accept his

conclusions? Why?

6.

FOR MANAGEMENT

|

What is the difference between the power benefits from the SSP as estimated by TECS and the World Bank? What are the benefits from resettlement and rehabilitation in the case of the SSP? How could they be considered in the appraisal? What would be the distributional implications of the SSP? Does the delay in decision-making about the SSP have the same implication as the delay in completing the project? Why? What are the important parameters to conduct a sensitivity analysis for the SSP?

Bibliography

American Economic Association (1951), Readings in the Theory of Income Distribution. Bain, J.S. (1956), Barriers to New Competition, Harvard University Press. Baumol, W.S. (1977), Economic Theory and Operations Analysis, 4th edn, Prentice-Hall of India Pvt. Ltd. Boulding Kenneth, E. (1966), Economic Analysts, 4th edn, Harper and Row. Cohen, K.J. and Cyert, R.M. (1975), Theory of the Firm: Resource Allocation in a Market Economy, Prentice-Hall of India Pvt. Ltd. Friedman, Milton (1962), Price Theory—A Provisional Text, revised edn, Aldire Publishing Company. Handerson, J.M. and R.E. Quandt (1971), Microeconomic Theory—Mathematical Approach, McGraw-Hill Kogakusha Ltd. Hicks, J.R. (1963), The Theory of Wages, 2nd edn, Macmillan (First published in 1932). (1964), Value and Capital: An Inquiry into Some Fundamental Principles of Economic Theory, ELBS. Koutsoyiannis, A. (1985), Modern Microeconomic, ELBS, Macmillan Publishers Ltd. Lancaster Kelvin (1971), Consumer Demand—A New Approach, Columbia University Press. Layard R. (ed.) (1972), Cost-Benefit Analysis, Penguin Modern Economics

Readings. Lerner, A.P. (1934), ‘The Concept of Monopoly and Measurement of Monopoly Power, Review of Economic Studies, June. Little ILM.D. and Mirrlees J.A. (1974), Project Appraisal and Planning for

Developing Countries, London, Heinemann Educational Books. Maddala,

G.S.

and

Ellen

Miller

(1989),

Microeconomics:

Theory

and

Applications, McGraw-Hill. Makridakis, S., Wheelwright, S.C. and V.E. McGee (1983), Forecasting: Methods and Applications, John Wiley and Sons, Inc. Marris, Robin (1967), The Economic Theory of ‘Managerial’ Capitalism,

Macmillan. Marshall, Alfred (1961), Principles of Economics: An Introductory Volume,

ELBS and Macmillan.

524

BIBLIOGRAPHY

Miller, R.L. and R.E. Meiner (1986), Intermediate Microeconomics: Theory,

Issues, Applications, 3rd edn, McGraw-Hill Book Company. Mills, T.-C. (1990), Times Series Techniques for Economists, Cambridge

University Press.

|

Reekie, W.D. and J.N. Crook Heritage Publishers. Rees

R.

(1974),

The

(1988), Managerial

Economics

of Investment

Economics,

Analysis,

Civil

3rd edn, Service

College, Occasional Paper No. 17, London, Her Majesty’s Stationery Office. : Robinson, Joan (1933), The Economics of Imperfect Competition, Macmillan. Salvatore, Dominick (1989), Managerial Economics, McGraw-Hill Book Company. Samuelson, P.A. and W.D. Nordhans (1995), Economics,

15th edn, McGraw

Hill Inc. Stigler, G.E. (1966), The Theory of Price, 3rd edn, The Macmillan Company. United Nations Industrial Development Organization (1978), Guide to Practical Project Appraisal: Social Benefit-Cost Analysis in Developing Countries, Project Formulation and Evaluation, Serial No. 3, New Delhi, Oxford & IBH Publishing Company.

Index

absolute cost disadvantage, of new entrant firms 378-80 ‘abuse’ approach, in controlling monopoly in Britain 340 accounting, and opportunity costs 226 ad valorem

tax 29, 399, 408-9

advertising, and brand names 343 competitive 345-6 costs 344—5 informative 345 aggregate purchasing power 181, 182 anti-monopoly laws 338 anti-trust laws 338-9 in-WSAe338, 339 Arc Elasticity, of demand for Cigaretios (152453 point elasticity and 130-1 auto-regressive integrated moving average (ARIMA) 180, 187 auto-regressive models 180 average (costs), and marginal costs 239-42 pricing, monopoly and 335-6 pricing model/‘mark-up’

pricing model, of non-collusive oligopoly 359, 372-6 variable cost, and SATC 238-9 average factor cost 421] physical productivity 412 revenue productivity 412

average annual rate of return (AARR) 487, 489-90 bandwagon effect 102 bandwagon goods 102, 103 barter 10, 14 Baumol, William 384 Becker) Gray -159 behavioural theory, of a firm 390 bidis, in informal sector 150, 151 bilateral monopoly defined 435, 437, 456 and trade unions 456—60 book rate of return (BRR) 487, 490 Box-Jenkin’s method 180 brand loyalty 77 monopolistic 342 Britain, controlling monopoly in 340 budget constraints, of a firm 68-71, 108, 109, 114, 216

budget line 72, 80, 112 buyers 17-18 and sellers 16, 44

capacity utilization 505, 514 capital, cost of a new firm 258 as a factor of production 12 fixed 13 intensive techniques 204 labour, and production 195, 197 working 13 cardinal utility approach/theory 48 demand curve and 56-9

526

INDEX

limitations of 60-1 cartel 314, 337, 358, 359 ‘case-by-case’ approach, in controlling monopoly in Britain 340, 341 cash-flows 488, S505 catchment

area treatment costs

st3 cement consuming sectors, in India 185 Cement Manufacturers’ Association (CMA) 178, 185 Chamberlin, Edward H. 288, 342,

346, 349, 352, 357 choice, as an economic problem 2-4 cigarettes, tax on I[50ff Clayton Act of 1914, USA 339 collusive oligopoly, models of 390 commodity, subsidy 106 tax 140} 12 versus income taxation 110-13 compensated demand curve 95-8 compensating variation, in income 89, 90 compensatory afforestation cost 514, 519 competition, demand curve of a firm under 289-9] with entry of a new firm 347-9 equilibrium price and output under perfect 287ff in the market, and potential few evitrant 256;°25-7)' 261, 263, 264 monopolistic 264, 304 perfect 264, 287ff and price variation, with new entrant firm 347, 349-56 and welfare and growth 301-2

condition of entry, concept of aT}, B18 conspicuous consumption 104, 105 consumer demand 45—6 theory 44ff, 106, 169 consumer(s’), equilibrium 53, 55, 56, 71-3, 110 equilibrium and demand curves 78tt expectation and market demand 21 surplus 98-100, 500, 516-18 surplus loss under monopoly 315, 316-17 surveys 172/175 tastes and preferences, and market demand 21-2 cost-difference method, of compensating variation 89, 90-1 costs, of production in the long run 245-8 of production in the short run 230-6 and productivity factor 242—4 theory of 223ff country risk 506 cross-elasticity of demand 144-6 Curnot, Antoine 360 Curnot model generalized 363-6

of non-collusive oligopoly 359, 360-6 customs duty 409 Cyert 384 dearness allowance 121 decisions tree method 498 demand curve

for a factor of production 416

anticipated 481

INDEX demand

curve, of a firm 44, 94,

95, 96, 267-9 actual, of a firm 351, 354, 355 assumed, of a firm 351 and cardinal utility 56-9 and consumers’ equilibrium TSE see also, market demand curve demand(s), of a consumer 45-6

elasticities 128ff factors affecting level of 19-22 for goods determinants of 101 forecasting level of 170-6 forecasting, methods of 169ff for cements in India 178-85, 187 functions 128 functions for cigarettes 157 law of 17-19 and supply, basic framework loft supply curves 40-3 and supply, interaction 34-8 depreciation 504—5 Dholakia, Bakul H. 178 diminishing returns, and perfect

competition equilibrium 300-1 direct income subsidy 106 discount rate 492-8 social 502 distributional aspects of SSP 520, See downstream losses 513-14 drawdown farming 520 duopolist firms, profit maximizing output of 364—5 duty, withdrawal of, on leaf tobacoo 151 econometric models, on demand forecasting 180-3 economic activity 7

5]

economic agent 2, 3 behaviour 4—5 choice-making by 6 economic life of project 488, 504 economics, as a science 4—5, 6 concept of Iff problems in 2 subject matter of 3 economic goods 8-9 economies of scale 248-50 economic profit 227-8 Edgeworth Box 119, 120 elasticity of demand, definition of 129 for food in India 165-7 slope versus 128-30 under competition and monopoly 311, 321 End-use method, on demand forecasting 177, 183-8 Engel curve 83 enterprise, as factor of production 12, 13-14 entrepreneur 14, 189, 224, 227 entrepreneur and business risk 482 and innovations 484 environmental costs and benefits 518-19 equilibrium employment of a factor 424—5 under different combinations of factor and product markets 425-38 equilibrium, of firm 266ff of monopoly firm 305-10 and output under perfect competition 287ff price, for a firm 36, 40, 42, 283 quantity 36, 38, 40, 42 equity issue 151

528

INDEX

definition of 189

ethics 5 Euler’s theorem, and law of variable proportion 212-15 exit policy 504 exploitation monopolistic 428, 429, 434,

435 monopsonistic 431, 434, 435 total 434—5 excess capacity 373 and monopolistic competition 354, 356-7 exchange, terms of 34, 44 excise duty 29, 409 on tobacco 152 excise policy, of government 150 excise revenue, from tobacco

equilibrium of 266ff marginal revenue of 269-76 objectives of 191-4 ownership and control of 192 performance of a 253 production and 189-9] profit maximization of 276-86 as “satisficer’ 389-90 structure-conduct-performance hypothesis of 253-7 taxation and performance of 399ff total revenue of a 266—7, 276 fixed factors, in production 229, 230 Full cost principle, oligopolistic firms and price determination under 370-1

152,

162 expansion path 221-2 export markets, monopoly firm selling in 326 externalities 438, 499 factor market competition, in 418 monopsony 42] factory supply perfectly elastic 418-20 factor prices, and choice of techniques 220-2 factor substitution, Iso-Quants and 203-9 Federal Energy Regulating ~ Commission, USA 332 Federal Trade Commission Act, USA 339 firm(s), average revenue of 267 behaviour, analyzing 346-56 conditions of entry by new 256-9, 260-2

choice of technique by 215-20

geometric formula, for elasticity of demand 133-7 Giffin good 91-3, 105 ‘global’ rationality 389 gross complements 146 Gross Fixed Capital Formation (GPCRS gross substitutes 146 habit forming goods 103 head tax 115 Hicks? FR. 86 Hicksian Demand curve 96, 98

income, as flow concept 9 and consumption 21, 83 and demand 84 demand curve see Engel curve effect 79-84, 87, 93, 115 elasticity, of demand 144, 146-9 index. 126, 427 of consumers, and market and demand 20

INDEX

wealth and 9-10 income tax 110, 112 commodity versus 110-13 and labour supply 113-16 index of real income 124, 125, 126 India, monopoly regulation and restrictive trade practices in 339 tax on cigarettes in LSOff indifferenee cunve..63,..65,67,. | corner solution and curvature of 74-7 indifference map 63, 68, 7] indirect taxes 29-30 Industrial Licensing Regulation and Reservation 258 industry, level forecasting 17] and market 252-3 see also, firms information, for demand forecasting 172 innovations 483 input price, and cost of production 29, 33 interest 13, 224 interest saving-investment theory of 470-5 liquidity preference theory of 475-7 | loanable funds theory of 477-80 market rate of 490-6 internal economic rate of return (ERR) 499-503, 514 internal rate of return (IPR) 488, 493-7, 514 Interstate Water Dispute Act 509 investment analysis 486I/f and price changes 496-7 independent 486—7 mutually exclusive 486-7

529

risk and uncertainty in 497-8, 506 Iso-cost curve/lines 216, 217-18, Zt Iso-Quants, and factor substitution 203-9

joint profit maximization 390, 395-8 joint stock enterprises 359, 383-9 Kinked demand curve models, of

non-collusive oligopoly 359, 366-72 Hall and Hitch version of 366, 370-1 Sweezy version of 366, 367-70

laboratory experiments, for demand forecasting 175 labour backward bending supply curve of 444-6 labour, and capital and production 1952 Ty 3 200 as factor of production 12, 13 intensive technique 204 skilled 13 supply, income tax and 113-16 unskilled 13 land, as factor of production 12, 229 land, charcteristics of 460-2 Laspeyre's-mdex, of price, 123, 20, che | of real income 125 law of demand 17-19, 128 and supply 38, 44 law of diminishing marginal returns 73, 74, 200, 202

530 law of diminishing marginal utility 50-2 law of equi-marginal utility 53-6 law of supply 28 law of variable proportion 214 Euler’s theory a 212-15 legal barriers, for new firms 257-8 leisure defined 44] and income, choice between AA]—6 Leraer ter. Ott limit pricing model, of non-collusive oligopoly 359, 377-82 theory of oligopoly 377-82 oligopoly, and absolute cost advantage 378-80 long run average cost and short run average cost 245-8, 250 long run equilibrium, of a firm 297-300, 347-54 long run marginal cost 250-1 low cost firms, as price leader 394-5 luxury goods, demand for 148 macro-level demand forecasts 170 management, and shareholders in joint stock companies 383-4 managerial theories, of firms 383-4 managerial utility function, maximization of 387-9 managers, in a firm 1, )92=4 March 384

marginal costs 236-8 and average costs 239-42 as basic for pricing public utility services 332-4

INDEX

and marginal revenue, of monopoly firms 305-10 of output to firms 219 marginal productivity 41] revenue productivity 412-15 value of 413 and resource allocation 435-7 social and private 437-8 marginal factor cost 417 and factor supply 417 under perfectly competitive factor market 418-2] under monopsony and imperfect competitive factor market 421-3 marginal product, of labour 198, 199-200, 202 marginal rate of substitution 66-8 and marginal utility 73—4 marginal rate of technical (MIRTS)' 205 *207¢ 215-20, 220 marginal revenue, of a firm 139, 269-76 and price elasticity of demand 272-7 and sub-markets 322 marginal utility, diminishing, and diminishing marginal rate of substitution 73-4 of goods 51, 52, 60 of money spent 54, 55 ‘Mark-up’ pricing model 373, 376 market(s), behaviour 337

clearing price 34 concept of, in economy 14-15, 16 demand curve 22-4, 100-5 demand curve, movement along and off 25-8 equilibrium 34, 35 function of 44

53]

INDEX

industry and 252-3 for managers 193-4 structure 252ff structure, Classification of determinants of 259-60,

263-4

|

structure, models of 260-2 M&l water use and output 518 Market Research Information Bureau (MRIB) 171, 173 marketable surplus 116-18 of foodgrains 117-18

Marris, Robin 384, 387 Marshall, Alfred 96 Marshallian demand curve 96, 98 meso-level demand forecasts 170 micro-level demand forecasts 170 minimum wage laws 440 and wage determination 446-54 money costs, of production 223-5 monopoly (monopolistic), competition 342ff competition, excess capacity and 354, 356-7 competition and price control 343 ; control over price and output oS ee bs: equilibrium of, firm 305-10 legal control of 340 lump-sum tax on 400-1 natural 330-6 power 310-12 and price determination 304ff price out-put behaviour 337 in product market 428—30, 432-5 in the real world 336-41

regulation of 338-41 selling in domestic and export markets 326 types of price determination 317-20 welfare implications of 312-17 Monopolies Inquiry Commission 254 Monopolies and Restrictive Trade Practices Act 340 Monopolies and Restrictive Trade Practices (MRTP) Commission

340-1 monopsony in factor market 431-5 Narmada Bachao Andolan 521 Narmada Planning Group (NPG) 209. 5, | Narmada Sagar Project (NSP) ny CUES Narmada Water Dispute Tribunal 509 National Sample Survey Organization (NSSO)

173

natural monopolies 330-6 net complements 146 net present value (NPV) 488, 490-8 curve 492-5 expected 498 net social present value (NSPV)

503

|

net substitutes 146 non-competing groups 439-40 non-price competition 344, 346

oligopoly, characterisucs and models of 358ff collusive 359 joint profit maximization under 395-8

Ie

INDEX non-collusive non-traditional

359, 360-82 models

of 383ff

oligopsony 422, 43] opportunity cost 500-3, 516 opportunity costs, of factor production 226-8 optimum allocation of resources 437 optimum plant size 249 optimum scale of production 258-9 ordinal approach, to utility 61 Organization of Petroleum Exporting Countries (OPEC) ae output, equilibrium and output, under perfect competition 287ff ownership of firms, and control 192 management and, of joint stock companies 383—4 Paasche’s index, of price 125, 120, bas of real income 124 Pande, I.M. 165 Paranjpye Vijay 510-16, 520-1 patent law 257 pay-back period (PBP) 487, 489 ‘perceived’ demand curve, of firms 349-50, 351, 353, 354-5 perfect competition in product market 431-2 perfect competition, long-run

equilibrium output under 357 price 377 perfect discrimination, under monopoly 319 perfectly competitive product and factor markets 425-30

point elasticity, and arc elasticity 130-1 population, and market demand 19-20 price, consumption curve for goods 93-5 cutting process by firms 352 determination of 38-9 determination in oligopoly firms 374 determination in monopoly firms 304ff determination of, and monopoly 317-20 determination, and profitability and socially desirable 320-30 STICCL Ol—71, F2,c 9 ty ha elasticity of demand 131-3, 135, 137-8 elasticity of demand for cigarettes 154, 158, 160 elasticity of demand and total revenue 138-42 elasticity of demand, from regression model for cigarettes 154—9 index and welfare 121-7 leader, dominant firm as 391-3 leadership under oligopoly 590, oo1 leadership by low cost firm 394—5 of commodities and demands and consumption 22, 24, 29, 30; 99; 122-4, 127 or related commodities, and market demand 20-1 and value 10-11 variation under monopoly competition 353, 355 private and social costs 228—9

$33

INDEX

producers’ surplus, concept of 315-17 product, differentiation 255-6, 259, 200 and monopoly competition 343-4, 357 quality 344, 346 production, costs 345 factors of 11-14, 224 and firm 189-91 functions 194—6 theory of 189ff productivity, and costs 242-4 profit maximization, of firm ro I. (O9" POA" 215, 2165224, 226; 276-86, 301, 374, 375 under monopoly competition 305, 306-7, 308, 310, 322-3 under oligopoly 391, 393, 395-8 profit meaning and characteristics 480-1 as monopoly rent 481 risk, uncertainty and 481-3

innovations and 483-5 project affected population (PAP) 520 public sector companies 19] Public Utility Regulating Policy Act, USA*332 public utility services (industries),

as natural monopolies 330-2 quasi-rent 467-9 Ramani, K.V. rent

178

concept of 460-2 as differential surplus 462-5 Ricardian theory of 462-5

scarcity (rent) 465-7

rate of exchange, measurement of 128, 129 ready-made garments, production of 190 rehabilitation and resettlement (R&R) benefits and costs of 519-20, 522. Gost of:312 relative cost disadvantage, of new entrant firms 380-2 rent 12, 223 reserve capacity 373 resource, allocation 4 alternative use of 3 scarcity 2-3 restrictive trade practices 337,

3567-339 returns of scale, of production 209-12, 248-50 revenue, of a firm 266—76 of government, from taxation [12 price elasticity and total 138-42 sales, forces 173, 174 maximization, and profit constraints 384—7 promotion 344 salvage value 488, 504 scale of production, definition of 221 scarcity, concept of 2 Schumpeter, Joseph 483, 484 scrap-value 488, 504 seller(s), and buyers 254 concentration 254-5, 260, 263 and producers behaviour 28 selling costs, monopoly competition and 344-6

534

INDEX

sensitivity analysis 491, 498, 520, S22 shadow prices 500-3 shareholders, management and, in joint stock companies 383-4, 386 Sherman Anti-Trust Act of 1890, USA 338-9 short run average total cost 238-9 short run equilibrium, of a firm 291-4 in depressed demand conditions 294—7 slope, versus elasticity 128-30 social benefit 438 social benefits, and monopoly firms 328-30 social benefit-cost analysis, 486, 499-503 of Sardar Sarovar Project SO7ff social cost 438 social costs, private and 228-9 social time preference rate

163-5,

502 socio-cultural costs and benefits 519-20 state domestic product 507 subsidies, in cash or kind 106-10 government 29-30, 399-401, AQ9 substitution effect 84-6, 87, 92

of income tax 115 sunk costs 229-30 supply, factors affecting level of 28-30 law of 28 supply curve 30-2 systemic risk 506 Talbot L.M. 521 Tata economic consultancy services (TECS) 509-16, 521

tax, ad valorem

29, 399, 408-9

burden, sharing between consumers and producers A406 on cigarettes ISOff commodity 399, 404-8 corporate 409, 504—5 holiday 506 lump-sum 399-403 policy of government 110 profit 399, 403-4 and government subsidies, and commodity supply 29-30 sales 409 taxation, and performance of firm 3001 Technical Efficiency 207, 208 technology, and commodity production 28-9 factor prices and 220-2 firms’s choice of 215-20 and production 189-204, 206-8, 210-12 test marketing, for demand forecasting 174-5 terminal costs 488, 504 Tewari, D.D. 165 theory of costs 223ff theory of demand, application of |O6ff theory of exchange 119 theory of monopolistic competition 342 theory of production 189ff time series models, on demand variables 177, 178-80 time value of money 488 tobacco consumption 150 see also, cigarettes total average, and marginal product 196-203 — trade union 440

INDEX

535

and wage determination 453-60 true elasticity, concept of 130

voluntary exchange, of goods 118-21

United States of America (USA), anti-trust laws in 338 tax on cigarettes in 158-9 utility, as the basis of demand 46-7 function 61-8 marginal 48-50, 51 measurement of 47-8 total 47-8, 50-1

wage piece and time 439 differences 439-40 wage-leisure consumption curve 445 wage bargaining and non-economic factors 459 and economic factors 460 wages 223 wealth, and income 9-10 tangible and non-tangible 10 willingness to pay 499, 516 welfare, implications, of monopoly 312-17, 357 price index and 121-7 World Bank 510-21

value, price and [0-1] variable factor(s), laws of veturn tO a. 232 in production 229, 230 | Veblen effect 104, 105 vertical integration 258