Price Theory and Business Behaviour
Author(s): R. L. Hall and C. J. Hitch
Source: Oxford Economic Papers, No. 2 (May, 1939), pp. 12-45
Published by: Oxford University Press
Stable URL: http://www.jstor.org/stable/2663449
Accessed: 13/09/2010 16:44
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=oup.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to Oxford
Economic Papers.
http://www.jstor.org
PRICE THEORY AND BUSINESS BEHAVIOUR
By R. L. HALL and C. J. HITCH
FoR several years a group of economistsin Oxfordhave been studying problems connected with the trade cycle. Among the methods
adopted is that of discussion with business men, a number of whom
have been kind enough to submit to questioningon theirprocedure
in various circumstances:and among other mattersin the questionnaire were inquiriesabout the policy adopted in fixingthe prices and
the output of products. Mr. Harrod and Mr. Hall have given some
of the resultsof these questions in papers read to the BritishAssociation, Section F, in 1937 and 1938. Neither of these papers was published, and the present paper includes the evidence on which they
were based as well as what has been collected since: it also extends
and modifiesthe theoreticalstructurewhichhas been emergingfrom
the facts. The data which it containshave been collectedby various
members of the group, and the results have been discussed at its
meetings. The authors are responsible only for the form of their
presentationand forthe speculative part of the paper.
The purpose of the paper is to examine, in the light of the interviews, the way in which business men decide what price to charge
for their products and what output to produce. It casts doubt on
the general applicability of the conventional analysis of price and
output policy in terms of marginal cost and marginal revenue, and
suggestsa mode ofentrepreneurialbehaviourwhichcurrenteconomic
doctrinetends to ignore. This is the basing of price upon what we
shall call the 'full cost' principle,to be explained in detail below.
1. Significance and limitations of the evidence
The method followed has been to submit the questionnaire to
business men who were willingto answer it, and to discuss the questions and answersat lengthin an interview. The authors are acutely
conscious of the shortcomingsof an inquiryof this kind. We considered the evidence of only 38 of the entrepreneursinterviewed,
which is far too small a sample to warrantany finalconclusions. Of
these, 33 were manufacturersof a wide variety of products,3 were
retailers,and 2 builders. The sample is thus stronglybiased in favour
of manufacturers,and any conclusionsrelate particularlyto this type
R. L. HALL AND C. J. HITCH
13
of entrepreneur. It is also biased by the fact that most firmswere
approached throughpersonal introductions,and it is probable that
the entrepreneursinterviewedwere more successfuland more intelligent than the average business man. In the light of the smallness
and thebiased characterofthe sample,no significancecan be attached
to the precise percentagesof firmsbehaving in particularways. But
on some questions the replies are so nearly unanimous that it is
impossible to ignore their implications; and in general the answers
fall sufficiently
clearly into patternsto leave no doubt in the minds
ofthe authorsthat currenteconomictheorytends to regardbehaviour
that is of small practical importanceas typical, and what is a wellmarkedmode as unusual.
In the body of the paper only the summaryresultsof the evidence
will be given. This evidence has been taken in large part fromthe
specificanswers to the relevantquestions,but it has been necessary,
in some cases, to supplement the informationin the light of subsequent discussion and subsidiary questions. The answers of the
entrepreneurshave been paraphrased, under appropriate headings,
in the Appendix, pp. 33-45: these should be regarded as an integral
part of the present paper, some sections of which will be much
clearer when illustrated by the actual expressions on which the
argumenthas been based.
2. Current doctrine on price and output policy
The basis of currentdoctrineon the price and output policy of the
is that he expands productionto the point wheremargientrepreneur
nal revenue and marginal cost are equal. In the special case of
perfect(or 'pure') competitionin the marketforthe product,marginal revenue is equal to price, to which marginal cost is equated. In
the special case of pure competitionin the market for the factors,
marginalcost is equal to the cost of the additional factorsnecessary
to expand output by one unit, and this is equated to marginal
revenue. In all other cases (except where discriminatingprices may
be charged),marginal revenue is less than price, and marginal cost
is greaterthan the cost of additional factors,and the only rule of
equilibriumwithinthe firmis that marginal revenue and marginal
cost are equated.
The equation of average cost and average revenue, if it occurs at
all, is assumed to take place as the result of the entryof new firms
14
PRICE THEORY
AND BUSINESS
BEHAVIOUR
where average revenue exceeds average cost, and by the dropping
out of old ones where the reverseis the case (i.e. under the stimulus
of profitor loss). It is not an equation which any particular entrepreneurattempts to bringabout, or indeed one which he desires. It
is customary to distinguish, somewhat unsatisfactorily,between
industriesin which 'free entry' is possible, in which there is a longrun tendencyfor average revenue and average cost to be equated;
and others in which there are obstacles to free entry, where this
tendencydoes not exist.
The precise content of the terms 'marginal and average revenue'
and 'marginal and average cost' is usually left in obscurity.' Most
writers,includingProfessorChamberlin,have concentratedon longrun equilibrium, where the difficultiesof findingthe appropriate
content for the curves is least. Even here, however, there are the
extremelyimportantquestions: on the cost side, of the allocation of
selling costs; on the demand side, of the functionalrelationbetween
sellingcosts and the demand curve, and of the nature of the demand
curve. Is the relevant demand curve 'real', i.e. does it show what
actually happens when price is changed; is it hypothetical,in the
sense of being based, like Marshallian demand curves, on some
particular assumption regardingthe behaviour of other firms;or is
it imaginary,i.e. does it merelyshow what the entrepreneurbelieves
will happen when price is altered? ProfessorChamberlinis the only
writerwho has attempted a systematicsolution of these difficulties,
and it cannot be claimed that his treatmentis definitive.
In the short run, which has been relatively neglected, the same
ofinterpretationremainand othersappear. Here the only
difficulties
rule of equilibrium is the equation of marginal cost and marginal
revenue. But are the relevantmarginalcurves those drawn fromthe
short-or the long-runaverage curves? Probably most economists
1 Compare, for example, Joan Robinson, Economics of ImperfectCompetition,
p. 21: 'Complications are introducedinto the problemof the individual demand curve
by the existence of advertising,but these have been ignored'; and, on the same page:
'In an industrywhich is conducted in conditions of imperfectcompetitiona certain
difficultyarises fromthe fact that the individual demand curve for the product of
each of the firmscomprisingit will depend to some extent upon the price policy of
In drawing up the demand curve for any one firm,however, it is
the others....
possible to take this effectinto account. The demand curve forthe individual firm
may be conceived to show the full effectupon the sales of that firmwhich results
from any change in the price which it charges, whether it causes a change in the
prices charged by the othersor not. It is not to our purpose to considerthis question
in detail.'
R. L. HALL AND C. J. HITCH
15
would say that it was short-runmarginalcost whichthe entrepreneur
would considerin deciding how much to produce with given plant,
and that long-runmarginal cost would be relevantonly when he was
consideringthe desirabilityof expanding or contractingthe plant.
Probably they would ignore altogetherthe very importantdistinction betweenshort-and long-rundemand curves,because it has long
been customary to assume in analysis that demand conditions,in
some sense, remain constantover time. The fact that demand in the
futuredepends upon presentas well as futureprices,which makes it
impossibleto derivemarginalrevenuefromany singledemand curve,
is usually dismissed,ifit is consideredat all, with a briefreferenceto
'maintaininggoodwill' or 'spoiling the market'.
It has become customary in recent years to distinguishvarious
'conditions' in which firmsproduce on the basis of the nature of the
markets in which they sell. The followingclassification,which is
chieflybased on ProfessorChamberlin's,appears to the authorsto be
exhaustive.'
(1) Pure competition,in whichno singleproducercan significantly
affectthe market price by varyinghis output.
(2) Pure monopoly, in which the demand curve of the firmis
negativelyinclined, and in which, because there are no close
substitutes,the entrepreneurassumes that a change in his
price or output will cause no other producerto change his.
(3) Monopolisticcompetition,or 'polypoly', in which the demand
curveofthe firmis also negativelyinclined,because its product
from others, and in which the entrepreneur
is differentiated
assumes that his demand curve is independentof the reactions
of other producers,not, as in the case of monopoly, because
thereare no close substitutes,but because there are so many
competitorswithin his 'group' that no one is affectedto a
significantextent by a change in his price or output.
(4) Oligopoly (including,as a special case, duopoly), in which a
fewfirmsproduce an identical product,and each realizes that
a changein its price or output may induce a change in the price
or output of one or more competitors.
(5) 'Monopolistic competition with oligopoly', or 'monopolistic
1 The definitionof monopoly does not correspond to Professor Chamberlin's.
(See Table 9, p. 30, in which the firmsinterviewedhave been arrangedaccordingto
our classification.)
16
PRICE THEORY
AND BUSINESS
BEHAVIOUR
competitionin the small group', which is like polypoly in that
the product is differentiated,but like oligopoly in that the
producerdoes not assume that his competitors'price policy is
independentof his own.
In technical termsa monopolist (or a monopolisticcompetitor)is
distinguishedfroman oligopolistby the fact that the cross elasticity
of demand between his product and the product of any other one
firmis negligible,and his own demand curve therefore'determinate'.
If any cross elasticities between his and other firms'products were
not negligible,he ought to take into account the possible reactions
of these otherfirmsto any change in his own price,and the situation
would thus be oligopolistic.
There are two factorswhich, if present,tend to make these cross
elasticities small. One is the smallness of the proportion of consumers1 (or potential consumers) for whom the elasticity of substitution is high between the firm's product and any other, a
conditionwhichwill tend to make the firm'sdemand curve inelastic.
For the smallerthe numberof consumerswho transfertheirallegiance
afterany change in price,the less likely is any one otherfirmto find
affected. The second factoris the range and
its demand significantly
evenness of 'scatter' of the affectedconsumersamong the products
of other firms. There are two cases in which the range of scatter
would be great and the distributioneven:
(a) That of monopoly. Here there is only one firmin the 'group'
or 'industry'. If its price is raised it will lose some customers,but,
therebeing no close substitutes(in the ordinarysense), the customers
who desert are likely to choose such varying alternative ways of
spending their income that no single firm'sdemand will be affected
to a significantextent.
(b) That of monopolistic competition. Here there are many
competitorsin the 'group', and in generalthe elasticitiesof substitution between any firm'sproduct and those of some otherfirmsin the
group are high for a significantproportionof that firm'sconsumers.
If it raises its price, the customersit loses will, for the most part,
choose alternative products within the group. But because there
are many such products, and because the preferencesof consumers
are fairlyevenly divided among them, the number gained by any
particular firmis likely to be negligible.
1 Properly weighted,of course, by the number of purchases each makes.
R. L. HALL AND C. J. HITCH
17
With these definitionsit is clear that therecan be border-linecases
between monopoly and monopolistic competitionwith determinate
demand curves. The 'group' is a vague and unsatisfactorydivision,
and can onlybe definedin termsofthe highelasticitiesofsubstitution
among products of 'many' or 'typical' consumers. In general, the
smaller the proportionof consumers forwhom elasticitiesof substitution are high between products of firmsoperating in the 'group',
the smaller can be the number of firmswithin the group consistent
with determinate demand curves. The evidence of the interviews
suggests that in the case of certain luxury and fashion goods it is
possible for cross elasticities to be negligible,and competitors'reactions to be ignored,despite the fact that only a very few firms
are operating within the 'group' or industry as ordinarily conceived.'
The 'current doctrine' of the equilibriumof the firm,which runs
in terms of marginal cost and marginal revenue, is held to apply in
its simplerformonly to the firstthreeof the categoriesin our classification,i.e. to pure competition,pure monopoly, and monopolistic
competition. It breaks down in the remainingtwo, i.e. oligopolyand
monopolisticcompetitionwith oligopoly; these, as special cases, are
relegatedto footnotesor leftto mathematicians,because the demand
curve forthe product of the individual firm,and thereforemarginal
revenue,is indeterminatewhere the price and output policies of the
firmsare interdependent. Attempts have been made to solve the
problem of equilibriumin these last two cases by complicated variations of the simplerrule, but no one attempt has met with sufficient
approval to be considereda part of currentdoctrine.
Subconsciously,when dealing withotherproblemsand when teaching, most economistsprobably consider the case of oligopoly to be
exceptional,and assume the general relevance of the simple analysis
in termsof marginal cost and marginal revenue. They assume that
the elasticityof demand forthe product of the firmis a good measure
of the 'degree of monopoly', that productionis carried to the point
price
where this elasticityis equal to the ratio2
, that
price-marginal cost'
ifthe elasticityis less than this ratio,price is raised, if more than this
I The authors intend to develop the implications of the classificationin a subsequent article.
2 Since this will be the point where marginal cost will equal marginal revenue.
4520-2
c
18
PRICE THEORY
AND BUSINESS
BEHAVIOUR
ratio, price is reduced.' They assume that each factor is hired up
to the point where its marginal product is equal to its wage or,
more generally,where its marginal cost (dependent on its elasticity
of supply) is equal to its marginal revenue (dependent on the elasticity of demand for its product).2
For the above analysis to be applicable it is necessary that entrepreneursshould in fact: (a) make some estimate (even if implicitly)
of the elasticityand positionof theirdemand curves,and (b) attempt
to equate estimated marginal revenue and estimated marginal cost.
We tried,withverylittlesuccess,to get fromthe entrepreneurswhom
we saw, informationabout elasticityof demand and about the relation between price and marginal cost. Most of our informantswere
vague about anythingso precise as elasticity,and since most of them
produce a wide variety of products we did not know how much to
rely on illustrativefiguresof cost. In addition,many, perhaps most,
apparentlymake no effort,even implicitly,to estimate elasticitiesof
demand or marginal (as opposed to average prime) cost; and of those
who do, the majority consideredthe informationof little or no relevance to the pricing process save perhaps in very exceptional
conditions.
3. The 'full cost' policy
The most strikingfeatureof the answers was the numberof firms
which apparently do not aim, in their pricing policy, at what appeared to us to be the maximization of profitsby the equation of
marginal revenue and marginal cost. In a few cases this can be explained by the fact that the entrepreneursare thinkingof long-run
profits,and in terms of long-rundemand and cost curves, even in
the short run, rather than of immediate profits. This is expressed
to some extent by the phrase commonly used in describingtheir
policy-'taking goodwill into account'. But the larger part of the
explanation, we think, is that they are thinking in altogether
differentterms; that in pricing they try to apply a rule of thumb
which we shall call 'full cost', and that maximum profits,if they
1 See, for example, Joan Robinson, Economics of ImperfectCompetition,passim;
R. F. Harrod, The Trade Cycle, chaps. i and ii; and A. P. Lerner, 'Monopoly and
the Measurement of Monopoly Power', The Review of Economic Studies, vol. i,
no. 3.
2 See, forexample, J. M. Keynes, The GeneralTheoryof Employment,
Interestand
Money, p. 5 and passim; A. C. Pigou, The Economics of Welfareand The Theoryof
Unemployment;and J. E. Meade, Introductionto Economic Analysis and Policy.
R. L. HALL AND C. J. HITCH
19
result at all fromthe application of this rule, do so as an accidental
(or possiblyevolutionary)by-product.
An overwhelmingmajority of the entrepreneursthought that a
price based on full average cost (includinga conventionalallowance
forprofit)was the 'right' price,the one which 'ought' to be charged.'
In some cases this meant computingthe fullcost of a 'given' commodity,and charginga price equal to cost. In othersit meant working
fromsome traditional or convenientprice, which had been proved
acceptable to consumers, and adjusting the quality of the article
until its full cost equalled the 'given' price. A large majorityof the
entrepreneursexplained that they did actually charge the 'full cost'
price, a few admitting that they might charge more in periods of
exceptionally high demand, and a greaternumber that they might
charge less in periods of exceptionally depressed demand. What,
then, was the effectof 'competition'? In the main it seemed to be
to induce firmsto modify the margin for profitswhich could be
added to directcosts and overheads so that approximatelythe same
pricesforsimilarproductswould rule withinthe 'group' of competing
producers. One common procedure was the settingof a price by a
strongfirmat its own full cost level, and the acceptance of this price
by other firmsin the 'group'; another was the reaching of a price
by what was in effectan agreement,though an unconscious one, in
which all the firmsin the group, acting on the same principle of
'full cost', soughtindependentlyto reach a similarresult.2
The formulaused by the differentfirmsin computing 'full cost'
differsin detail, as will be seen by referringto the informationin
column B in the chart of evidence; but the procedure can be not
unfairlygeneralized as follows: prime (or 'direct') cost per unit is
taken as the base, a percentageaddition is made to cover overheads
(or 'oncost', or 'indirect' cost), and a furtherconventionaladdition
(frequently10 per cent.) is made forprofit. Selling costs commonly
and intereston capital rarely are included in overheads; when not
so includedthey are allowed forin the addition forprofits.
It would be useful foreconomic analysis if the magnitude of 'full
cost' in any case could be deduced fromthe technical conditionsof
I For a classificationof firmsaccording to the strictnesswith which they adhered
to the 'full cost' policy see Tables 6, 7, and 8, pp. 26 and 29.
2 In several cases trade associations published 'standard' figuresof costs in an
attemptto secure equal prices; firmsin the industrywere urged to use the 'standard'
costs in applying the full cost principle.
20
PRICE THEORY
AND BUSINESS
BEHAVIOUR
production and the supply prices of the factors. This is in fact impossible, forfourreasons. The firstis that the firmis not necessarily
of 'optimum' or ofany othersize, so that the extentto whichinternal
economies or diseconomiesare reflectedin the figuresdepends upon
historical accident.' The second is that the addition for overheads
varies according to the policy adopted for calculating the output
over which total overheads will be distributed. As Table 1 shows,
somewhat more than half the firmsused figuresof actual or estimated output,the others(including,in general,the more competitive
firms)full or conventionally 'full' output. The third is that the
conventional addition for profitvaries fromfirmto firmand even
withinfirmsfordifferent
products.2 The fourthis that sellingcosts,
which depend upon the demand, are included.3
TABLE 1
Outputassumedfor Distributionof Overheads
(Firms classified according to strictness of adherence to full cost principle.)
Conventional
orfull
.
Not adhering
.
Rigidly adhering
Normally adhering
Adhering in principle
Total
.
.
.
..
Actual or
forecast
..
Ambiguousor
no information
8
4
5
1
7
6
2
1
4
10
15
13
Why do entrepreneursbase price on 'full cost', as defined,rather
than attempt to equate marginal cost and marginal revenue? The
informationrelevant to this question given by the thirtyentrepreneurs adheringto the full-costpolicy is paraphrased in column C in
the Appendix and is tabulated in Tables 2-5.
I The informationon the slope of the cost curve at the point of equilibrium will
be found in column E in the Appendix. In many cases it is inadequate because the
question was frequentlynot asked. Thirteenfirmswere apparently operatingunder
conditionsof decreasing costs and fourunder conditionsof constant cost. Two firms
stated that costs were increasing because labour became more expensive as output
was expanded; but it was not clear whether the diseconomy in question was an
internal or an external one.
2 This allowance, when given,is stated in column B in the Appendix.
3 Informationabout the magnitude of selling costs, when given, will be found in
column E in the Appendix.
21
R. L. HALL AND C. J. HITCH
REASONS FOR ADHERING
TO FULL-COST PRINCIPLE'
TABLE 2
General
Belief that this is the 'right' price
Loyalty to Association .
Experience proved its advisability
.
.
.
.
.
.
52
2
2
TABLE 3
Reasonsfornotcharging
morethanFull Cost
Fear of competitors or potential competitors (including
.
belief that others would not follow an increase) .
.
.
They do not go in for a high profit
.
They prefer a large turnover
Buyers technically informed regarding costs
11 plus 6 of the 7
textile firms.
2
2
3
TABLE 4
Reasonsfornotcharging
lessthanFull Cost
.
Demand unresponsiveto price
.
Competitorswould followcuts
Difficultto raise prices once lowered
Trade Associationminimumprices
Conventionwith competitors
Quasi-moralobjectionsto sellingbelow cost
Price cuts not passed on by retailers
.
.
.
.
.
.
.
9
11
2
3
1
8
1
TABLE 5
oncesettled
Prices(howeverfixed)
Reasonsfornotchanging
.
.
Conventionalprice in minds of buyers .
.
.
.
Price changes dislikedby buyers
Disinclinationto disturbstabilityof marketprices .
.
.
5
4
3
A study of the replies confirmsthe existence of a strongtradition,
already referredto, that price 'ought' to equal full cost. This tradition is accountedforto some extentby an idea offairnessto competitors and is undoubtedlyone of the reasons forthe adherence to the
full cost policy. The otherfactorswhich seem to be most important
1 Little significancecan be attached to the actual numbers in each category,since
in most cases only those reasons volunteeredby the entrepreneursare included. Thus
the fact that only three mentioned the technical informationof buyers as a reason
fornot chargingmore than cost does not mean that in the other twenty-sevencases
this reason was not operative.
2 This is exclusive of the two in Table 3 who 'do not go in for a high profit', the
eight in Table 4 who had quasi-moral objections to selling below cost, and the three
in Table 5 who are disinclinedto disturb stability.
22
PRICE THEORY
AND BUSINESS
BEHAVIOUR
in inducing entrepreneursto follow this policy may be summarized
under six heads.
(i) Producers cannot know their demand or marginal revenue
curves, and this for two reasons: (a) they do not know consumers'
preferences; (b) most producers are oligopolists,and do not know
what the reactions of their competitorswould be to a change in
price.
(ii) Although producers do not know what their competitors
would do if they cut prices,they fear that they would also cut.
(iii) Althoughthey do not know what competitorswould do ifthey
raised prices, they fear that they would not raise them at all or as
much.
(iv) Prices are not lowered by actual or tacit agreement among
producers because of the conviction that the elasticity of demand
to make this course pay.'
forthe group of products is insufficient
(v) If prices are in the neighbourhoodof full cost, they are not
raised by actual or tacit agreementbecause it is thoughtthat, while
this would pay in the short run, it would lead to an underminingof
the firmsby new entrantsin the long run.2
(vi) Changes in price are frequentlyvery costly, a nuisance to
salesmen, and are disliked by merchants and consumers. Several
entrepreneursreferredexplicitlyto the fact that there are conventional prices to which customersare attached, and that these have
to be charged,which means that in these cases only large changes in
price which are clearlyunprofitableare possible.
All these reasons militateagainst changingprice fromthe conventional level. In addition, (i) is a reason fornot adopting the alternative price policy of equating marginal cost to marginal revenue;
(vi) makes it undesirable and almost impossibleto equate short-run
marginal cost and marginalrevenue; and (v) is a reason forthe conventional price level being no higher than 'full cost' including a
'reasonable' addition forprofit-a tendencyreinforcedby tradition.
If it is desiredto illustratethe positionofequilibriumgeometrically,
this may be done for the typical case where oligopoly elements
are presentby the use of a kinked demand curve,the kink occurring
1 In this they are, in most cases, certainly right,since the elasticity of demand
for the products of the group as a whole is less than that for the product of any
one firm.
2 If prices are below what entrepreneursconsider the full cost level they will be
raised by agreementprovided that it is possible to secure one.
R. L. HALL
AND
C. J. HITCH
23
at thepoint wherethe price,fixedon the 'full-cost'principle,actually
stands. Above this point the curve is elastic, because an increase in
pricewill not be followed(or so it is feared) by competitors,who will
be glad to take any extra sales. Below the point the demand is much
less elastic because a reductionin the price charged will be followed
eventually by competitorswho would otherwise lose business. If
A
FIG. 1.
this is the characterof the demand curve it followsthat over a wide
rangeofmarginalcoststhe existingpriceis the mostprofitable.It also
followsthat,withgiven costs,this priceis most profitableover a wide
range of possible fluctuationsof the demand curve, since wherever
the demand curve may be the kink will occur at the same price.
The two diagrams (Figs. 1 and 2) are intended to help the reader
to grasp one point in the argument; like all diagrams,they are much
more precise than the circumstancesthey purportto explain.
In Fig. 1 AA representsthe demand curve forthe product of one
firmof the 'group' if all other firmsmaintain their prices at P:
24
PRICE THEORY
AND BUSINESS
BEHAVIOUR
BB representsthe demand for the product of the firmif all other
firmsvary theirprices as it does, this being a proportionateshare of
the market demand.
If competitorsare forced to cut prices below P when any firm
begins such a movement, but do not raise their prices above this
point if only one firmdoes so, then it is very likely to pay any firm
a~~~~~
Fix. 2.
A
to maintain price at P. For its own demand curve will have the
shape ofthe heavy line, kinkedat P. If MRa is the marginalrevenue
curve to the curve AP, and MRb the marginalrevenue curve to the
curve PB, then the marginal revenue to any firmis discontinuous
below the point P. And as long as the marginalcost ofthe firmintersects the line PQR at any point between Q and R, P must be the
most profitablepoint and price thereforestable.
If the demand curvesshift,but the kink remainsat the same price,
there will still be a range between the two marginal revenue curves
on the perpendicularbelow the actual position of P: and price will
be stable fora wide range of marginalcosts.
R. L. HALL AND C. J. HITCH
25
In Fig. 2 let AC representa section of the short-runaverage cost
curve for a firm,excluding profits. If the firmassumes that it will
sell an output OA, and adds 10 per cent. of its average cost at that
point forprofit,it will set the price at OB, and be willingto sell, in
the firstinstance,whateveris demanded at that price. If otherfirms
act in the same way, the price will be stable forthe reason explained
in connexion with Fig. 1. The curves d'd', d d, and d"d" represent
various positionswhichmay be actually taken by the demand curve:
only at P are the profitswhich are made those whichwere expected,
but the pricewill not be changedforthe otherpositions. Any circumstance which lowers or raises the average cost curves of all firmsby
similaramounts, on the otherhand, e.g. a change in factorprices,is
likely to lead to a re-evaluationof the 'full cost' price OB.
If the demand curve shiftsmuch to the left of d'd' and remains
thereforsome time,the price is likelyto be cut in the hope of maintaining output. The reason for this cannot be explained geometrically except in the special circumstanceswhere the lower part of the
demand curve becomes much more elastic when it moves to the left
or where marginal costs fall considerably as output is reduced.
Usually one entrepreneuris overcomeby panic: 'there is always one
fool who cuts'; and the rest must follow. If the demand curve shifts
much to the rightof d"d" the price is likelyto be reduced in the long
run, because the long-runaverage cost curve is likely to be falling
and entrepreneursfear that the high profitswill induce competition.
(In Fig. 2 the long-runaverage cost curve would lie below the shortrun curveAC on eitherside of the point below P.)
4. Extent and strictness of adherence to 'full cost' policy
From so small a sample it is difficultto generalize concerningthe
strictnesswith which firmsadhere to the full cost policy, but an
examinationof the answers summarizedin the Appendix, pp. 33-45,
and of Tables 6, 7, and 8, indicates that it is the rule ratherthan the
exceptionto attemptto do so.
Of the 38 firmswhich we investigated, 12 maintained that they
adhered to the 'full cost' policy, with negligibleexceptions,' at all
times and in all circumstances. Of the remainder,15 adhered to it in
normal times, most times being 'normal' in this sense. In addition
1 Selling below cost on 'loss leaders', shading prices on one line and making it up
on another,and cuttingslightlyon large ordershave not been consideredinconsistent
with 'rigid' adherence.
PRICE
26
THEORY
AND
BUSINESS
BEHAVIOUR
TABLE 6
Degree of Adherenceto Full Cost Principle (Classified accordingto
Types of Market)
Not
adhering
.
.
Monopoly .
.
.
Oligopoly .
Monopolistic competition
Monopolistic competition
.
with oligopoly
Total
.
.
.
.
1
1
Adhering Adhering
Adhering
in
rigidly
nornzally
principle
2
1
3
2
1
..
5
.
3
.
3
5
9
2
.
8
12
15
3
TABLE 7
Adherenceto Full Cost Principle (Classified according to Types of
Product)
Not
adhering
Consumers' goods
.
.
Textiles
Intermediate goods
.
Capital goods
.
Retailers .
.
Builders .
Total.
.
4
1
1
1
Adhering Adhering
in
rigidly
normally
principle
4
2
3
3
7
1
3
.
.
.
.
.
.
.
.
.
.
.
.
..
..
..
.
.
8
12
1
Adhering
3
2
2
15
3
3 firms(all in textiles) professedto adhere 'in principle', but not in
fact because of the severe and chronic depression in the trade. Of
these 18 firmsadhering normally or 'in principle', 12 said that if
business were very depressed they would cut prices below full cost,
and 6 of these expressed a reluctantwillingnessto cut all the way to
prime cost if that proved necessary to 'keep going'.' Only 2 of the
30 firmsadheringsaid that they would charge more than full cost in
1 These cases of cuttingto prime cost present a difficulty
forthe analysis in terms
of marginal cost and marginal revenue as well as forthat in terms of full cost, since
none of the firmshad a perfectlyelastic demand curve. The explanation in all six
cases is that the producersare workingto contract-most are contractorsand capital
goods manufacturers-which means that each unit produced is unique. This allows
price discriminationand makes price and marginal revenue identical.
R. L. HALL AND C. J. HITCH
27
exceptionallyprosperous times when they were having difficultyin
fillingorders,and it is doubtfulwhethereven these two would charge
in such circumstancesas much as the marketwould bear in the short
run: rationingby refusingto take orders or to fill them promptly
was preferredto 'excessive' prices.
The behaviour of a certainnumberof firmsis clearlynot explicable
in terms of full cost, and in the Appendix these have been omitted
fromthe table and discussed separatelyon pp. 43-5. One seems to
be a monopolistwho behaves more or less in the text-bookmanner.
Four had deliberatelycut prices in fairlynormal times because they
estimated (explicitly in three cases) that the demand was elastic
enough to make this course pay.'
5. Stability and instability
We may distinguishtwo main cases, which we shall call those of
price stability and instability,since the terms equilibrium and disequilibrium have a connotation too precise to be warranted here.
The distinction correspondsfairly closely to that made by some
economists recently between non-aggressive and aggressive price
policies.2
(i) In cases of relative stabilityeach firmadheres as closely as it is
able to its own formula. Where costs do not differwidely withinan
industryall firmswill charge similarprices, and the consumerswill
be distributedamong them accordingto the factorswhich make the
marketimperfect,such as the proximityor attachmentof customers
to particularfirms.The price may be set by the strongestfirm,or by
a process of trial and errorwith all firmsmaking some adjustments;
in any case it is unlikely to pay small or new firmsto make such
departuresfromit as to call attentionto themselves. We cannot say
preciselywhat this pricewill be, forreasons already explained; ifit is
set anywhereover a fairlywide range it will have a tendencyto stay
there. The nearestthat we can get to an exact statementis that the
pricerulingwherethese conditionsobtain is likelyto approximateto
the fullcost of the representativefirm;and that this price is reached
directlythroughthe communityof outlook of business men, rather
than indirectlythrougheach firmworkingat what its most profitable
I Two firmsin the 'full cost' group said that they would cut prices in the (rare)
elastic.
cases in which they thought demand sufficiently
2 See, forexample, J. M. Cassels, 'Excess Capacity andMonopolistic Competition',
QuarterlyJournal of Economics,May 1937.
28
PRICE THEORY
AND BUSINESS
BEHAVIOUR
output would be if competitors'reactions are neglected,and if the
play of competitionthen varied the numberof firms.1
The occasions on which prices stabilized at the full cost level will
be changed have been summarized in column D of the table in the
Appendix and in Table 8. A change in price may or may not upset
stability' in our sense. The price is likely to be changed without a
departurefromprice stabilityif thereis a change in costs which will
affectall firmstogether,such as a change in wages or the price of
materials,or if a new process is generallyadopted; in these cases the
idea of the rightprice will change and with it the price itself. From
this point of view theremay be somethingto be said forbusinessmen
who assert that income-taxis added to price, since if all competitors
in any trade regardedit as a cost it would tend to become one.
If demand shifts,prices may be allowed to deviate fromfull cost
without disturbing stability. As trade conditions deteriorate,for
example in a slump, full cost, wherethis is computedby distributing
overheads over actual or estimated output, will oftenbe allowed to
rise above price. This tendency is strengthenedby the anxiety to
keep plant runningas full as possible, givingrise to a generalfeeling
in favour of price concessions. This may pass into a condition of
price cutting,and the industryis then in a position of 'instability'.
(ii) Prices in an industrybecome 'unstable' as soon as any of the
competitorsform an -idea of a profitableprice which is markedly
differentfrom the existing prices. From our inquiries this seems
most likelyto happen when a trade becomes really depressed,and is
a potent factor making for an agreement which will substitute a
formalarrangementforwhat was previouslyonly a sense of fitness.
Conversely,it may happen when ordersincrease to the point where
existing firmshave difficultyin fillingthem.2 Otherwise,it seems
most likelyto happen when thereis a new entrantwho is determined
to establish himselfon a large scale, perhaps because of the cupidity
or inefficiencyof the existing producers; or when one competitor
thinks he has a method in advance of those of his competitors; or
(more rarely) when one of the participants begins to act on the
I The variation in the number of firmsmay still serve the purpose of tending to
equate the rate of profitson capital to the normal level. Thus, if a 'normal rate' of
profitsof 10 per cent. on turnoverrepresentsan abnormallyhigh rate on capital, the
entry of new firmsmay reduce it, without affectingprice, by increasing 'excess
capacity'.
2 This circumstance will remove the kink in the demand curve and make its
elasticity above the old price similar to its elasticity below it.
R. L. HALL AND C. J. HITCH
29
assumptions of competitionin text-books. The price then becomes
'what the marketwill bear', and the size and numberof units in the
industryand its methods of productionare likely to be changed.
TABLE 8
Occasionson whicha DeparturefromFull CostPrinciplemightbe
made
(a) Price reductions:
.
.
.
Depressed Trade
.
Cyclically on competitive lines
.
When necessary to 'keep going'
Necessity to follow a competitive price
If competitor broke agreement .
.
.
Loss leaders
.
.
.
.
.
.
.
.
6
2
4
.
.
.
.
.
.
.
.
8
1
2
2
3
2
1
2
.
.
.
.
.
On expensivelines,to cover concessionson cheap ones .
1
1
1
1
1
.
.
.
.
.
Attemptto capturenew markets
To obtain a large contract
Seasonally to stimulate sales
.
.
To clear old stock .
Cases where demand was elastic
(b) Price increases:
.
In specialities .
To maintain unemployment
More work not wanted
Need for funds for expansion
.
.
.
.
.
.
.
.
.
.
.
6. Comparison of full cost analysis with current doctrine
The modificationof conventionaltheorywhichthe answersrequire
may be discussed under two heads: (i) modificationsin long-run
analysis; (ii) modificationsin short-runanalysis.
(i) In general: the answers as summarizedin Table 9 suggest that
pure competition,pure oligopoly, and pure monopoly (in the sense
definedabove) are rarelyfound in the real business world. Monopolistic competitionis more common, but the typical case is that of
monopolisticcompetitionwith an admixture,which is usually large,
of oligopoly. The answers indicate, moreover,that while Professor
Chamberlin'sanalysis of price determinationin this typical case is
correctin the sense that he has probably definedcorrectlythe limits
withinwhich price must lie, the process of its determinationwithin
these limits is more straightforwardand the resultingprice more
stable than he implies.' These limits may be described as (1) the
1 E. H. Chamberlin,The TheoryofMonopolisticCompetition,
pp. 100-4. Professor
Chamberlin'sDD' and dd' demand curves (see, e.g., pp. 90-1) are drawn on the same
assumptions as our illustrativecurves BB and AA on Fig. 1, p. 23.
30
PRICE THEORY
AND BUSINESS
BEHAVIOUR
polypoly price (i.e. the one which would be established if entrepreneurs assumed that no otherfirmwould change prices in responseto
an original change), and (2) the price which would be established if
the industryas a whole acted as a monopolist. Average cost (including normal profits)will be equal to price in the long run wherethere
is 'free entry', a condition unlikely to be fulfilledif the oligopoly
element is at all important.
TABLE 9
Consumers' goods .
.
.
Textiles
Intermediate pro-
ducts
Capital goods
Retailers
..
Builders
Totals
Monopoly
Oligopoly
1
..
..
..
..
..
3
2
..
.
.4
..
2
4
Monopolistic
competition
with
Monopolistic
competition'
oligopoly
6
3
8
4
1
..
1
11
3
2
2
19
But in the actual cases hereexamined the precisemethodby which
this resultis attained, and by whichprice is fixedbetweenthe limits,
is not what a reader of ProfessorChamberlin'sbook would infer. In
most cases no attempt is made to estimate marginal revenue from
either short period or long period demand curves, nor to estimate
marginalcosts. The heightof price (betweenthe two limits)is determined on the 'full cost' principle, conditioned by such historical
accidents as (a) the size and efficiencyof the firmsin the industryat
the time price stability was achieved, and (b) the extent of their
optimismand of theirfear of potential competitorsas measured by
the percentage addition for profits. Once this price has been fixed
price competition,except in highlyabnormal circumstances,ceases.
Profitsare reduced to normal, if at all, by an influxof firmswhich
raises costs by reducingoutput per firm(increasing'excess capacity')
or by competitionin quality and marketing.
The answers also suggest that the distinctionbetween monopoly
and monopolistic competition on the one hand and monopolistic
1 Some of the firmslisted as monopolistic competitive were on the border-line
between monopolisticcompetitionand monopoly. See p. 17.
R. L. HALL AND C. J. HITCH
31
competitionwith an admixtureof oligopolyelementson the other is
not of very great importance. Only where oligopoly elements are
presentis the demand curve 'indeterminate'in the economist'ssense,
but in the other cases it is unknown to the entrepreneur,and this
seems to be the essentialpoint. It is truethat in the case ofmonopoly
or monopolistic competition the possibility of findinghis demand
curve by experimentingis open to the entrepreneur;but there are
objectionsto experimentation,and the prospect of a quiet life seems
in many cases to have the greater appeal. Entrepreneursseem to
be somewhat less likely to fixprices on the full cost principle where
the demand curve is determinate,but there are some who do so.
It proved to be extremelydifficultin practice to distinguishbetween oligopolistic firmsand others. The distinctionseems to be
almost entirelyone of degree, for all firmswere conscious to some
extent of the presence of competitorsand the possibilityof reactions
to changes in their price and output policy. In some cases the
distinctionseems to rest upon the size of the price or output change
under consideration. While a small change, which stole few customers fromothers,would be overlooked, a large change would lead
to retaliation of some sort. In other cases the distinctionseems to
depend upon the size of the firmconsidered. In the same market
some firms-normallythe smallerones-would apparentlynot attach
much weight to possible retaliation by competitorswhereas others
-the largerones-would. In the classificationoffirmsin the Appendix those which seemed to be little influencedby the possible reactions of competitorsto small changes in prices have been included as
monopolistic or monopolisticallycompetitive. The test applied is
whetherthe firmis sufficiently
independentfora Marshalliandemand
curve to be drawn which,in the neighbourhoodof the actual price,
would,in conjunctionwiththe cost curves,forma reasonablyaccurate
guide to the most profitableprice policy.
(ii) The answersdo not confirmthe commonanalysis of short-run
equilibrium in terms of marginal cost and marginal revenue. It
seems to be much more nearlytrue (in the case of manufactured,and
particularly of finishedproducts) that, save in very exceptional
conditions when the attachment of producers to the conventional
price breaks down, the long-runanalysis of price, as given above,
applies in the short run. This does not mean that there will be no
tendencyforthe prices of these goods to fall in depressionsand rise
32
PRICE THEORY
AND BUSINESS
BEHAVIOUR
in booms, but simplythat therewill be no tendencyforthem to fall
or rise more than the wage and raw material costs.
These considerationsseem to vitiate any attemptsto analyse normal entrepreneurialbehaviour in the shortperiod in termsof marginal curves. They also make it impossible to assume that wages in
the shortrun will bear any close relationto the marginalproduct (or
marginalrevenue) of the labour employed.1 Perhaps the most important consequencesare forthe analysisofthetrade cycle2and especially
of the effectsof changes in money wage-rates,in which the assumption is ordinarilymade that employment is carried either to the
point where the marginal product is equal to the wage-rate, or
'if conditionsof imperfectcompetitionprevail', to the point where
the marginal revenue (computed fromthe elasticity of the demand
curve) is equal to the wage-rate. Certainlygreat doubt is cast on the
generalapplicabilityof a theorywhichplaces any weighton changes
in the elasticityof demand in the shortrun as a factorinfluencingthe
price policy of entrepreneurs.3
7. Recapitulation
If our sample is at all representativeof business conditions,we
suggest that the followingconclusionsmay be drawn:
(i) A large proportionof businesses make no attempt to equate
marginal revenue and marginal cost in the sense in which
economistshave asserted that this is typical behaviour.
(ii) An element of oligopoly is extremelycommon in markets for
manufacturedproducts; most businesses take into account in
theirpricingthe-probable reactionof competitorsand potential
competitorsto theirprices.
(iii) Where this element of oligopolyis present,and in many cases
where it is absent, there is a strongtendency among business
1 The 'Principle of Substitution' is, of course, not invalidated. The ratio
marginal cost of factor
factor will tend to be the same for all factors.
marginal
prodct of
of factor
marginal product
2 The authors intendto produce an article on this subject in the near future. They
suggest that the price policy here outlined partly explains J. T. Dunlop's statistics
in the Economic Journal, Sept. 1938, which indicate that real wages tend to vary
directlywith output during the course of the trade cycle.
3 The 'law of diminishingelasticity of demand' may be a partial explanation of
the price-cuttingin some depressions which leads to conditions of 'instability'.
Several entrepreneurstestifiedthat depressed markets tended to be 'price markets';
i.e. markets in which buyers are particularly sensitive to price changes.
R. L. HALL AND C. J. HITCH
33
men to fixprices directlyat a level which they regard as their
'full cost'.
(iv) Prices so fixed have a tendency to be stable. They will be
changed if thereis a significantchange in wage or raw material
costs, but not in response to moderate or temporaryshiftsin
demand.
(v) There is usually some element in the prices rulingat any time
which can only be explained in the light of the historyof the
industry.
APPENDIX
Analysis of Replies to Questionnaireon Costs and Prices
In the following pages is summarized the information on which the tables
in the text are based. This is sometimes rather vague, either because the
replies were vague or because, in the case of firms seen in the early stages of
the inquiry, the technique of questioning had not been mastered. We began
by expecting that the answers would lie along lines differentfrom those which
they actually followed, and we did not always press for information which later
we should have found of great importance.
The firms have been classed into monopolies, oligopolies, those working in
conditions of monopolistic competition, and those working in these conditions
with an admixture of oligopoly. The classification has been made by the
authors on the basis of all the information available. In the first part of the
Appendix the firmswhich followed what we have called the 'full cost' principle
are listed, including those firms which considered that this was the right
policy but had difficultyin adhering to it. In the second part the information
obtained from the firms which did not adhere to this policy is summarized.
The letter before the number' of a firmindicates the type of product:
a
Consumers' goods.
Textiles.
c = Intermediate products.
d = Capital goods.
e = Retailers.
f = Builders.
b
=
=
The information under each firm is given in the following order:
A. Price policy.
B. Method of calculating cost.
C. Reasons for adhering to the full cost policy, or to the modification
actually employed.
D. Circumstances in which this policy would be departed from.
E. Selling costs: and any information about whether costs to the firm
were increasing, constant, or decreasing.
1 The numbers do not correspond to those used by J. E. Meade and P. W. S.
Andrews in 'Summary of Replies to Questions on Effectsof Interest Rates', Oxford
Economic Paper8, No. 1.
4520.2
D
AND BUSINESS
PRICE THEORY
34
S~~~~;t-
BEHAVIOUR
0~~~~~~~~~~~~~~~'
Ca
~~~~~~eO
4flP
bmcCa0
0
0
45 cd
4-D
*0'0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
U
00
Q
~~~~~
~ ~ ~ ~ ~~
~ ~ ~ ~ ~ ~~~~~~~"
0 ~~~~~~~~~~~~~~~~~~~~~~~~~
0
44
0
0
~~~~~~~~~~
0~~~~~~~~~~~~~~~~~~00
c ~
0
0c~~~~~~~~~~~~~~~c
00
Z0
m
0 timHc
0~~~~~~~~~~~~~~cd
~~ ~ ~ ~ ~~~0+ac
Q~~~~~~~~~~
00
"
C
p0
35
R. L. HALL AND C. J. HITCH
0*
0
z 0
0
~~~~~~~~~~~~~~~i2~~~~~~~~~~~~~~
4--J0
4-D~~~~~~~
-
4-D
0~~~~~
to
C) k
0
0 0
P,
0
ca
o0
H--
c
0a
0 Ca
0
0
_
+D 5
0
0
C
C
lz~~~~~~~~~~~0
0 0.0
0~~~~~~~~
HDb04
S aC
0
410c0-
-
4-
0
ri20
Ca P.,
+Ca
Ca
0~~~~~~~~~~~~~~
o
D
b
-
36
PRICE THEORY
~~~~
AND BUSINESS
BEHAVIOUR
~O020:
MCa
~~~
~k ~ ~ ~ ~
P.
~~~
0O
C)
a
.w~~~~C
ri~~~~~~~~~
O~~~~~~~
S~~C
0~~O2
bD
~
0
0
*
0~~~~~
~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~M
a
~~~~~
a)-4.4~~~~~~~~~~~~~~~~~~~~~~c
+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
0
4
.
a
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
H~~~~~~~~~~~~~~C
H
M 4-D
Ca
Ca0~O* 0
ct~~~
d
o
0
0
0
0
0
0
- CH4D
H
~
.
cdcda) Ca
C)~~
C)
Ca~~~ri2~
ac 0
41
M 'o 41H
.,-4
C)
bLC
0,0
a
c
0~~~~~~~~~~~~~~~~~~~~~~~~01
~~~~~~~~~~~
0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0
c
t
C)
0
t0PA10.
)
2
a
C)
0
0
+D
d4
0
0
0 a
C 1-b
'o
0
C
-:
0
4-
C
w
-
E-
R. L. HALL
AND
37
C. J. HITCH
ca.
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0
~~~~~~~~~~~~~~~~~C
Ca
tH
CaP
.
~~~r
,S~S~W0
4)0
0
0~a
H
-4-)o
-4-)
~'
W0
0
)
-
--
0~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~
~~~~~~~~r
4-Z,
0M
02
~~4
.
0
4)C
0
0 ;-
4-4~~~~~~~~~~1
38
PRICE THEORY
0
4~~~0~
2
BEHAVIOUR
AND BUSINESS
-iC
+Ocl
_Ca
0
~ ~
*~~~~~~~~~~~
~
~
~
~
~
~
1
to
H
;:o
0
P-,
-
+
0
,
z
e
0)
~~+~0
~~0 c0~~~~
HO
4-1
0
to
CO~~~~~~~~~~~~~
4z
0
)
I~~~~~~
c
R. L. HALL AND C. J. HITCH
~~~0~~~~~c~~~~00~~~~~4
~ ~ ~ 4
.0c
00
4~~~~~~~~~~~~c
Id04
0)d ;Z-
+-
0
O0
O
Mi
-4-1
Ca
d
0
OO
0~~~~~~~~~~~~0)0
00
0.~
O
0)Q
J
~
;-q ce~c
0
0~
0
pq
00
39
.,-q
O0
~ ~ ~0 ~ ~ ~ ~
)
0
~
~
o C~0
0
~~~~~
P.
0
-4
0
000)
00~~~~~~6
-~~~~~~~~~~~~~~~~~
ri
0
'p
H~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
e4
to
~~ ~ ~
40
PRICE THEORY
BEHAVIOUR
AND BUSINESS
0
1-1
0
00
~~~~~~ ~~~~~~~~
0
ce
00c~
0
43
0
~~~~~
~~~~~~~
ro~~~~~~~~c
0.,
0.,
0~~~~~~~~~~~~
0 cq
~
C) o
; -,
- e -0
0
~
0O
~
-415 P-
4Z
~
~0
Q~~~~~~q0
~
0
~
24
S
;0-q
0
.
0r4e
0~~b0~~~~
0
0 00
0 o
ce4-
ct
~~
~~~~ H
~!
bo
0 0
0
P-
E5
~
b
~
~
~
4
0~~~~~~~
0
-
C
c
4
0
0
A
41
R. L. HALL AND C. J. HITCH
~~~~~~~D0~
~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~
0
M0
~~~~~~~~~~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~4
0 4-44-
00
-0
0
0
)
to
c
c-4PO0-'dc
0t:~
e
-4~
0 0C
4.
c
to ce
d
0 d
-O
P-
M0 '
O~~~O
c
0
0
D0
0e
0
M
0
P
M
0
ce
c
M~~~~~~~
P
la
_
ce000
,~~
45 0:
'd .,.A
o ce
fl
0 0
C)c~00
'-4~~~~~~~~~~~~~~~~~~~~~~~~~04~~.
~~~0bC~~~~~~~~~~~. 0~0
~~~~~
0
0C
P4l
0~~~~~~~~~~0;0
.0
0
0
q.
0
0
P~~~~~q
c44
0
ce
.~~~~~~~~~~~~~~~~0
.M 0 M
ce~~~~~~
;
0
PRICE THEORY
42
AND BUSINESS
0~~~~~~~~~~~~~w00.C
0~~~~~~~~~o
0
~~~~~~~CI72
0
tcj
12
0
BEHAVIOUR
~~~~~~~ 0ce
~
~~ ~ ~
4~~~~
t
;.4
0
0~
~~~~~~~~~~~~~
00~0ce' 0'
~
~
0o
~~~~~~~
~ ~~ ~ ~ ~ ~ ~ ~~~~1
0 ;-,
0
o
b
0
P-4
o
bbLJ
o
.
0
4-0 4-1
00
00
4-
~
ce~~~~~~
0
0~~~~~~~~~~~~P4p
0r
QO+~~~bi
~
R. L. HALL AND C. J. HITCH
43
The followinggives an account of the policies followedby the firmswhich
did not price theirproductsaccordingto any formof the cost principle.
Firm a 1. This firmproduced fashion goods, and though broadly affected
by competition,had almost a monopoly' in its own range, The demand for
its goods was a peculiar one, the price itselfbeing regardedas one of the propertiesof the commodity,and sales being oftenhigherat a highprice than at
a low one. Thus it was necessaryto discoverthe 'right' price,which might
not be related to cost at all: cost was the bottomlimitof price. Price might
be cut in depression,but this must be concealed fromthe customers. On
certainstaple, more competitivelines, price mightbe less than would cover
full overheads: partly to maintain full employment,partly because these
goods were made to train learners. Overheads about equal to works costs.
No agreementin this trade, but firmswere all anxious to know what their
rivals were doing.
Firm a 3. This firmsuppliesmainlyto multiplestores and 'has to take the
price as fixedwithinlimits' and work to it. But there is some play of price
and this is veryimportantas affecting
sales. Overheadstaken as 150 per cent.
of directlabour, this being added to materialand directlabour chargesto get
full costs. Their selling firmtakes their costs, adds 121 per cent. for its
(selling)costs,and what profitit can get. They will go down to directcosts to
to raise priceswhichhave been once
keep the businessgoing,but it is difficult
reduced. They discriminatein prices, accordingto the season, and forsuch
reasons as bulk orders.
Price reductionsstimulate sales, and they produce as cheaply as possible
forthis reason. There are no agreements,and bad trade leads to a tendency
to reduce prices.
Costs fall as turnoverincreases: it is surprisingeven to themselveshow
much.
Firm a 5. Described as almost a monopoly,but has to think about the
possibilityof new entrantsto the industry. What the marketwill bear is the
only considerationany business takes into account: bottom limit is prime
costs. Costing is carefullydone, and overheads are distributedon basis of
80 per cent. operation: but in any case, so long as overheads are covered
somewhere,it does not matter which line carries them. Prices are liable to
vary at shortnotice with cost of materialand with the competitiveposition.
Demand formost of theirproducts is inelastic: with a competitiveline, response to cuts is immediate. But a cut may reduce sales, if it leads consumer
to suspect quality.
Agreementsare the most importantfactoragainst competitivecuttingand
include 'expectation that competitorwill counterwith cuts'. Even when an
agreementhas been broken (as tends to happen with any sharp change in
tradingconditions)you try to keep in touch: you telephone to your competitorand inquire,and ifhe is cuttingyou say, 'You get the order,old man,
but do let 's get togetheron the price' (Apparent discrepancybetween these
lines in mind.)
paragraphs: informantmay have had different
I Though this firmhas been classifiedas a monopoly,the fact that the evaluation
was not independent
ofits pricemakestheanalysison
ofits productby consumers
was based inapplicable.
whichthe classification
44
PRICE THEORY
AND BUSINESS
BEHAVIOUR
Selling costs, includingwholesalers'margins,whichare considereda selling
cost, are about 20 per cent. of workscost.
Firm a 12. This firmis a member of a highlycompetitiveindustryto
which entry is very easy. What the market will bear is much the most
importantfactor in determiningprice. Any price which will make some
contributiontowards overheads is better than nothing. Demand is very
inelastic for the industryas a whole, and the sales of any particular line
depend mainly on whetherthe style becomes fashionable or not. To raise
priceshas been disastrous.
Agreementswith competitorsare no good, and though you expect your
cuts to be countered,you do not pay much attentionto this. The onlyfactors
restrainingthe tendencyto cut prices are beliefin the inelasticityof demand,
and a strongdisinclinationto cut below directcosts.
Sellingexpensesabout 16 per cent.,profitabout 4 per cent. ofpriceto wholesalers. Industryworkingshorttime,so that costswouldfallowingto spreading
of overheads.
Firm b 4. Until 1929 slump this firmhad rarelysold anythingbelow full
cost. Since then a more aggressivepolicy has been followed,both to bringin
new linesand to expand old ones. Prices are normallygot by adding overheads
on a 'full' basis to direct costs, and then adding marginsdepending on the
of
state of the market,which depends mainlyon the degreeof differentiation
product which they have achieved. Lower limit of price is direct cost plus
50 per cent. overheads,upper limitis fixedby fearof competition,so that it is
higherwherethey have a speciallygood article. Competitorsalways counter
with cuts. 'The disinclinationto sell below cost is all a matterof how much
you must do to keep runningon fulltime.'
Selling and warehousingcosts about 10 per cent. of price. Cost falls with
increase of output, even when they go into overtime.
Firm c 3. 'Price depends mainly on what the market will stand.' Each
line is expectedto earn a certainmarginover primecost: if it will not do this,
it is not produced,because resourcescould be used moreprofitablyelsewhere.
Once a price has been fixed,it is changed as rarelyas possible. They regard
themselvesas marketleaders: 'If we feel ourselves compelled to increase or
reduce prices, we are fairly certain that our competitorswill do likewise.
Unless they are prepared to accept a sacrificein their profitrates, they will
have to follow our procedure.' Reasons against raising prices are that it
disheartenssalesmen, and probably damages long-runcompetitiveposition.
Competitorswould probably follow any cuts, but doubtful if they would
followincreases(made apart fromchanges of cost).
About half their costs are distributional. They think that costs could be
reduced if therewere a considerablylargerdemand.
Firm d l. All ordersby contract. Agreementin one line. In this line the
firmsets the prices at total cost plus 10 per cent. This is its upper limit of
price: outputis estimatedand total overheadsallocated in proportionto direct
labour. The lower limitis direct costs, which include that part of overheads
which varies with output. Price not raised above upper limitbecause this is
found to be unwise with permanentbusiness: buyers technicallyinformed.
The price reductionsin depressiondo not have much effecton sales, but have
R. L. HALL AND C. J. HITCH
45
to be made to retain share of what business there is. Selling costs about
10 per cent ofprice.
Firm e 3. Market highlycompetitive. Close attention has to be paid to
what competitorsdo, and they must be eitherled or followedin price movements. Thus very little attentioncan be paid to costs. No rigid formulais
used for computingthese: if competitiveprice is too low, the line is discontinued. Two departmentsrun at a small loss because of advertisementvalue.
There are really no restrainingtendenciesagainst price reductions(but as the
firmis a retailer,the conditionsare those ofmonopolisticcompetition).