-
WEALTH AND WELFAEE
BY
A. 0. PIGOU, M.A.PHOFESSOE OF POLITICAL ECONOMT IN THE
UNITEBSITT OF CAMBRIDGE
AUTHOR OF ' THE PRINCIPLES AND METHODS OF INDUSTRIAL PEACE '
'PROTECTIVE AND PREFERENTIAL IMPORT DUTIES,' ETC.
* Discontent, to be effective, must be shot with the colours of
hope.'Charles Booth.
MACMILLAN AND CO., LIMITEDST. MARTIN'S STKEET, LONDON
1912
-
rib77)
?\,^loD^^-COPYRIGHTV
-
TO
DR ALFEED MAESHALL
-
PREFACE
Several years ago I began to study the causes of unemploy-
ment. It soon became apparent, however, that these causes
are so closely interwoven with the general body of economic
activity that an isolated treatment of them is scarcely
practi-
cable. Hence the gradual growth and more extended scope of
the present volume. The drift of the argument is indicated
in a brief and incomplete manner in the Analytical Table of
Contents. This table, however, is not a summary, and, in-
deed, it deliberately ignores the more difficult parts of
the
discussion. It should, therefore, only be used as a guide to
the pages on which the different topics treated in the text
may be found. I have endeavoured so to arrange the book
that the main body of it may be intelligible to readers
other
than professional economists. Such readers, however, are re-
commended to omit Chapter III. of Part I., and Chapters I.,
VIII., X., XI., and XII. of Part II.
In a work covering so wide a iield, and attempting to
bring into an ordered unity so many different aspects of
economic life, I cannot hope to have avoided altogether
error
and ambiguity. For protection against these dangers I owe
-
viii WEALTH AND WELFAEEmuch to Mr. J. M. Keynes, of King's
College, Cambridge, and
Mr. Donald W. Corrie, of the Inner Temple, both of whomhave very
kindly read the whole book, either in manuscript
or in proof. I am also indebted to Mr. Corrie for the com-
pilation of the Index.
A. C. PIGOU.
King's College, Cambridge,
September 1912.
-
ANALYTICAL TABLE OF CONTENTS
PAET I
WELFAEE AND THE NATIONAL DIVIDEND
CHAPTER I
WELFARE AND ECOKOMIC WELFAREPAGE
1. Welfare can belong only to states of consciousness ... 3
2. Economic welfare is that part of total welfare which consists
insatisfactions connected with the earning and spending of the
nationaldividend ........ 3
3-6. Instances can be multiplied, in which economic causes
affectingeconomic welfare in one way afifeot total welfare in a
differentway. Nevertheless, there is a presumption that conclusions
abouteffects upon economic welfare will hold good also of effects
on totalwelfare ........ 4
7. And reasonably adequate conclusions about effect upon
economicwelfare can often be obtained by economic science, in spite
ofthe partial and limited character of that science . .11
CHAPTER II
ECONOMIC WELFARE AND THE NATIONAL DIVIDEND
1, For the most part economic causes act upon economic welfare,
notdirectly, but through the national dividend . . .14
2-3. The gross national dividend consists of the annual flow of
thosegoods and services that can be brought easily into relation
witha money measure ...... 14
-
WEALTH AND WELFAEEPAGE
4-7. And the true net national dividend consists of the gross
nationaldividend minus an appropriate allowance for depreciation
ofcapital .... ^^
8. If a cause is introduced, which makes for an increase in the
aggregatesize of the net national dividend, provided that the
absolute share
of no group of members, in terms of the commodities which
thatgroup is accustomed chiefly to consume, decreases, the
economicwelfare of the community as a whole is likely to be
augmented . 20
9-10. If a cause is introduced, which makes for an increase in
theabsolute share of relatively poor groups of persons (in terms of
the
commodities which those groups are accustomed chiefly to
con-sume), provided that the magnitude of the aggregate
nationaldividend does not decrease, economic welfare is likely to
beaugmented ........ 24
11. If a cause is introduced, which diminishes the" variability,
orinequality in time, of the dividend, and especially of that part
ofit which accrues to the poorer classes, economic welfare is
likelyto be augmented ....... 32
CHAPTER III
THE MEASUREMENT OP THE DIVIDEND AND IT8 PARTS
1. Since the dividend is made up, not of a single commodity, but
ofmany commodities varying in quantity, it is necessary to findsome
way of measuring changes in its magnitude . . .33
2. If we suppose the money income of the community to be
constant,such a measure is equivalent to the reciprocal of an index
numberof general prices ....... 34
3-16. The remainder of this chapter, therefore, is concerned
with theproblem of constructing, so far as that is possible, an
indexnumber, such that its reciprocal shall move in the same
directionas the quantity of economic welfare... .35
CHAPTER IV
THE NATIONAL DIVIDEND AND THE QUALITY OP THE PEOPLE
1. The conclusions reached in the second chapter must now be
reviewedin the light ofmodern biological knowledge
. . .52 2. That knowledge warrants the belief that general
welfare and
economic welfare alike could be increased by measures
restrictingpropagation among the obviously degenerate. This belief,
however,is additional to, and does not disturb, our results . .
.53
-
CONTENTS xiPAGE
3. It is sometimes held that modern biology, by demonstrating
thedominant part played by heredity, as compared with
environment,has proved economic inquiries, which are, in the main,
concernedwith environment, to be unimportant. Reasons are offered
forthe rejection of this view ...... 56
4. It is sometimes held, further, that the advantage to economic
welfare,claimed in the second chapter to result from (1) an
increase in themagnitude and (2) an improvement in the distribution
of thenational dividend, are cancelled by indirect biological
effect . 61
5-6. Reasons are offered for the rejection of these views . .
62
CHAPTER V
The method of discussion to be followed is outlined . . 66
PART II
THE MAGNITUDE OF THE NATIONAL DIVIDEND
CHAPTER I
PAEETO'S LAW
1. Certain statistical investigations by Professor Pareto seem,
at firstsight, to suggest that the magnitude and the distribution
of thedividend are rigidly bound together, in such wise that it is
im-possible for the dividend as a whole and the real income of
thepoor to move otherwise than in the same direction . . 71
2-3. This view, however, cannot be derived from Professor
Pareto'sstatistics ........ 72
4. In order to determine how far causes affecting the magnitude
andthe distribution of the dividend act harmoniously in respect
ofthese two things, we need, therefore, to investigate the
broadproblem of distribution ..... 77
CHAPTER II
PKODUCTION AND DI8TRIBDII0N
1. Distribution of income among people is analytically quite
distinctfrom the distribution among factors of production of
economictext -books. Nevertheless, no great error is introduced, if
we
identify the income of the poorer classes with the receipts of
thefactor labour ...... 78
-
xii WEALTH AND WELFAEEPAGE
2. The general run of causes affecting the magnitude of the
nationaldividend act through (1) changes in the quantity or
technical
efficiency of non-labour, (2) changes in the quantity or
technical
efficiency of labour ... '9 3-13. In an elaborate discussion, it
is argued that, broadly speaking,
causes acting through the supply of non-labour are
harmonious,
in the sense that they are likely to affect the magnitude of
thenational dividend and the aggregate real earnings of labour in
thesame direction . . . . 80
14-15. And the same conclusion is reached concerning causes that
actthrough the supply of labour . ... 92
16. Hence, we conclude generally that economic welfare will be
in-creased by causes which increase, and diminished by causes
whichdiminish, the magnitude of the national dividend . . 95
NOTE TO CHAPTER II
Uncertainty-Bearing as a Factor of Production . . 95
CHAPTER IIITHE MASNITUDE OF THE DIVIDEND AND EQUALITY OF
MARGINAL NET PRODUCTS
1-4. It is generally, but not universally, true that the
nationaldividend would be reduced below its natural size by
causesmaking the marginal net products of resources in different
usesunequal. In general, the more nearly equal marginal net
productsin aU uses are, the larger the dividend is likely to be .
104
CHAPTER IV
HINDRANCES TO EQUALITY OF MARGINAL NET PRODUCTSDUE TO IMPERFECT
MOBILITY
1. In the present and two following chapters it is assumed
provisionallythat self-interest, if not interfered with, tends to
make themarginal net products of resources equal in all uses ; and
inquiryis made as to how far this tendency is checked by
"impedimentsto mobility" ....... 109
2. The popular conception of "impediments to mobility" and of
theireffect is confused and inadequate ..... 110
3-10. A somewhat elaborate analysis of the problem is attempted
110
11. We conclude that anything, which, either improves judgment
asto the comparative advantages of production in different parts
of
-
CONTENTS
the industrial field, or lessens the cost of movement from
onepart of that field to another, is likely to bring about
greaterequality among marginal net products, and, therewith, in
accord-ance with the reasoning of the preceding chapter, an
enlargementof the national dividend. Illustrations are given . . .
122
12. The above result, however, does not justify the inference
that cheapening, whether of knowledge or of movement, in the
senseof a transference of the cost of it from the persons who reap
thebenefit to the State or other public body, is likely to lead to
anenlargement of the dividend . ... 125
NOTE TO CHAPTER IVMobility and Unemployment ...... 127
CHAPTER VHINDBANCBS TO EQUALITY OF MARGINAL NET PRODUCTS DUB
TO
IMPERFECT DIVISIBILITY OF THE UNITS IN TERMS OF WHICH
TRANS-ACTIONS ARE CONDUCTED
1. When the units in which transactions are made are large, or
whenthey are compounded of two factors in a fixed proportion, the
ten-dency of self-interest to make the marginal net products of
eachfactor equal in all uses is obstructed .... 129
2. In modem times the size of the unit, in which transactions in
respectof capital take place, has been reduced, partly with the
help ofthe Stock Exchange....... 130
3. The compound character which formerly belonged to this unit
hasalso, in great part, been eliminated, by arrangements which
thegreat growth of securities adapted to serve as collateral
hasfacilitated ..... . . 131
CHAPTER VI
HINDRANCES TO EQUALITY OF MARGINAL NET PRODUCTS DUE TOTHE
RELATIVE VARIABILITY OF INDUSTRY
i1-2. A given system of impediments to mobility is likely to
involve
greater inequality between the marginal net products of
resourcesatdifierent points, the more variable, relatively to one
another, are
the demands for the services of resources at these points . .
135
j 3-8. As a first step towards an analysis of the influences by
which therelative variabilityof a country's industries is governed,
a discussion
is attenipted of the influences by which the absolute
variability of
-
xiv WEALTH AND WELFAREPAGE
the demand for labour and waiting in an individual industry
isdetermined.... l^^
9. It is shown that the influences laid bare in this discussion,
in con-
junction with certain further influences, control also the
extentto which the demands for labour and waiting at different
pointsvary relatively to one another ..... 145
CHAPTER VII
HINDRANCES TO EQUALITY OF MARGINAL NET PRODUCTS DDE TO
DIVER-GENCES BETWEEN SOCIAL NET PRODUCT AND PRIVATE NET PRODUCT
1-2. Private net product and social net product are
distinguished. Inso far as self-interest tends to bring about
equality of marginalprivate net products in different fields,
equality among marginalsocial net products is obstructed, whenever
marginal private andmarginal social net products diverge. Some
sorts of divergenceoccur under simple competition, further sorts
under monopolisticcompetition, and yet further sorts under
bilateral monopoly . 148
3. Under simple competition divergences may arise out of the
customaryform of certain contracts ...... 149
4-6. Illustrations are given from the various kinds of contracts
that aremade between landlords and tenants and between employers
andemployed ........ 150
7-9. Divergences may also arise independently of contractual
relations,in respect of forms of investment which render incidental
uncom-pensated services or disservices to the general public.
Illustrationsare given . .... 158
10. These divergences can be mitigated by a judicious employment
oftaxes and bounties ....... 164
11. Under monopolistic competition further divergences arise in
con-nection with resources invested in advertisement
. . . 165
12-14. Under bilateral monopoly divergences arise in respect
ofresourcesdevoted to " bargaining " and to fraudulent practices.
Thes^ canbe mitigated to some extent by penal laws.... 168
CHAPTER VIIITHE EQUALITY OF MARGINAL NET PRODUCTS IN DIFFERENT
USES
UNDER SIMPLE COMPETITION
1. Hitherto it has been provisionally assumed that self-interest
tends,so far as it is not obstructed, to make the marginal private
netproducts of resources in different uses equal. In passing to
an
-
CONTENTS
investigation of the validity of this assumption, we may
con-veniently neglect the distinction between marginal private
andmarginal social net product, and speak, without qualification,
ofmarginal net products ...... 172
2-5. A difficult argument shows that, under conditions of simple
com-petition, in industries of increasing returns, the marginal
netproduct tends to exceed, and in industries of diminishing
returnsto fall short of, the marginal net product yielded in
industries ingeneral ........ 174
6. And, when the various producers in an industry are not
situatedsimilarly, the marginal net products of resources invested
atdifferent centres within the same industry tend, except
underconstant returns, to be different ..... 177
7. It is possible, by resort to taxes and bounties,
diflFerential andotherwise, to mitigate these divergences ....
178
CHAPTER IX
THE CONDITIONS OF MONOPOLISATION
1. It will presently be necessary to inquire how far marginal
net productsin different uses tend to equality under monopoly. As a
pre-liminary to this inquiry, the conditions which determine
theappearance of monopolistic power need to be considered . .
180
2. Circumstances which, when the aggregate scale of an industry
isgiven, make it structurally economical for the typical
individualestablishment to be large, are favourable to the advent
ofmonopolistic power...... 180
3. So also are circumstances which make it structurally
economical forthe typical individual unit of business management
(embracing,perhaps, a number of establishments) to be large . . .
182
4. So also are conditions under which amalgamation is able
greatly toreduce expenses by cutting down competitive advertisement
. 184
5. So also is the existence of a highly inelastic demand for any
com-modity, since this implies the possibility of large gains if
mono-polisation takes place. The influences by which the degree
ofelasticity of demand for different commodities is determined
arediscussed ........ 185
6. On the other hand, anything that renders it difficult for
negotiationsaiming at amalgamation to be opened up is unfavourable
to theadvent of monopolistic power . . . . .189
7. And the same thing is true of those obstacles to agreement
whicharise out of conflicting claims about " participation " . .
19Q
-
xvi WEALTH AND WELFAEE
CHAPTER XMONOPOLISTIC COMPETITION
PAGE
1. Under monopolistic competition, self-interest does not tend
to makethe marginal net product of resources in the industry
affectedequal to the marginal net product of resources in general .
. 192
2-4. Kather, it leaves this marginal net product indeterminate
over arange, the extent of which depends upon certain influences
that arebriefly discussed ....... 192
CHAPTER XISIMPLE MONOPOLY
1-3. Under simple monopoly ^Zms restriction of entry to the
industryaffected, it is probable that the marginal net product of
resourcesin that industry will diverge from the marginal net
product ofresources iu general more widely than it would do under
simplecompetition . ...... 195
4. And, if the entry to the monopolised industry is not
restricted, thenational dividend will suffer in another way .
197
CHAPTER XIIDISCRIMINATING MONOPOLY
1. Under certain conditions monopolists are able to charge
discriminat-ing prices ........ 199
2-4. The nature of these conditions, which depend essentially
uponthe non-transferability of the commodities afifected, is
discussed . 199
5-7. Of three distinguishable forms of discriminating monopoly,
theonly one of practical importance is discriminating monopoly of
thethird degree, under which different prices are charged
betweenmarkets, the composition of which is determined otherwise
thanby the monopolist's own choice .... 202
8-16. Under discriminating monopoly of the third degree, plus
restric-tion of entry to the industry affected, the output may, not
improb-ably in certain circumstances, approach more nearly to the
quantitycalculated to make the marginal net product of resources in
thatindustry equal to the marginal net product of resources in
general,than it would do under simple monopoly ; but it is unlikely
toapproach more nearly to this quantity than it would do
undersimple competition . . ... 206
17. Circumstances are, however, conceivable, in which it may do
this 212
-
CONTENTS xvii
mi - PAGE) 18. The effect is examined of a peculiar class of
agreement concerning
price discriminations, entered into by sellers who are in
competitionwith one another . ..... 213
CHAPTER XIII
THE SPECIAL CASE OF RAILWAY RATES
1. The discussion of the preceding chapter throws light on the
con-troversy between advocates of " the cost of service principle "
and" the value of service principle " in respect of railway rates .
215
2-4. This controversy is obscured by the popular view that the
carri-age of copper and the carriage of coal, or the carriage of
coaldestined for A and the carriage of coal destined for B, over a
givenpiece of line, are "joint products." This view is
incorrect
. 215
5. The meaning in practice of "the costof service principle" is
explained 219
6. The meaning in practice of "the value of service principle"
isexplained ........ 222
7. " The cost of service principle " corresponds to simple
competition,and "the value of service principle" to discriminating
monopolyof the third degree....... 228
8. In general, the former is the more advantageous to the
nationaldividend ; but, as stated in the preceding chapter,
circumstancesmay arise in which the latter is the more advantageous
. . 229
9-10. These circumstances, however, are much less common
thanwriters on railway economics usually suppose . . . 231
NOTE TO CHAPTER XIII
Thb Zone System. ..... .235
CHAPTER XIV
purchasers' ASSOCIATIONS
i 1. The preceding chapters have shown that, in many
industries,neither simple competition, nor monopolistic
competition, norsimple monopoly, nor discriminating monopoly will
make themarginal net product of resources invested in them equal to
themarginal net product of resources in general. We have next
toinquire whether this result can be secured by resort to the
deviceof Purchasers' Associations...... 237
i2. The answer is clearly in the affirmative ; but no inference
follows
as to the effect on the national dividend, until the
comparative
advantages in respect of productive efficiency of Purchasers'
Associa-
tions and ordinary commercial businesses have been ascertained .
237b
-
xviii WEALTH AND WELFAEEPAOE
3. Not much light can be thrown on that matter by
historicalexamples ........ 238
4-5. Purchasers' Associations hare advantages in respect of
production,so far as they save costs in advertisement, bargaining
and safe-guards against fraud ...... 239
6. They are, however, at a serious disadvantage, in so far as
the sizeto which they individually attain must be determined, in
greatpart, by considerations irrelevant to economic efficiency . .
241
7-9. The importance of this disadvantage in regard to various
sorts ofindustry is discussed ...... 242
10. It is concluded that the field practically open to
Purchasers' Asso-ciations is strictly limited, and that the study
of further remediesfor the imperfections of ordinary business forms
is, therefore, stillrequired ........ 244
CHAPTER XV
STATE INTERVENTION
1. The present and two following chapters are concerned with
tlieremedy of State intervention in industry .... 246
2. There is no reason why this should be confined to those
industrieswhose operation requires resort to the right of eminent
domain . 246
3. The mere failure of private industry, when left free from
Stateinterference, to maximise the national dividend, does not of
itselfwarrant intervention, for this might make things worse . .
247
4. Certain modern developments have, however, rendered
govern-mental agencies better fitted for intervention than they
were informer times........ 249
CHAPTER XVI
PUBLIC CONTEOt OP MONOPOLY
1. This chapter is concerned with attempts by the State so to
controlprivate monopoly that supply shall be adjusted to demand,
and thenational dividend shall, therefore, be as large as it would
be undersimple competition....... 251
2-3. Among indirect methods of control, after reference to a
device ofminor importance, the policy of maintaining actual
competitionby refusing to allow rival concerns to combine is
discussed
. 251
4. A second indirect method of control is embodied in the policv
ofmaintaining jpofewMaZ competition. This is worked by
penalisingthe "clubbing" devices of cut-throat competition, or
destructivedumping, and of boycott
. .... 257
-
CONTENTSPAGE
5-6. The nature and effect of these" clubbing " devices are
explained . 258
7. In spite of many difficulties, it is probable that
legislation directedagainst them, if carefully prepared, may, at
all events, lessen theextent to which they are employed... .
261
8. But such legislation, even if successful in its immediate
object,would not serve completely to maintain potential competition
. 264
9-11. The inadequacy of indirect methods of control leads to the
sug-gestion of direct methods designed to fix price and output at
thelevel proper to simple competition..... 265
12. It is, however, exceedingly difiBcult for the State to
enforce theobservance of the conditions it imposes .... 267
13-16. And it is equally difficult for it to decide rightly what
conditionsto impose ; though certain devices are available by which
this diffi-culty can be partially overcome ..... 268
CHAPTER XVII
P0BLIC OPERATION OF INDnSTRIES
1. The difficulty of satisfactory public control over
monopolistic andcertain other industries suggests that the national
dividend mightbe increased by the public operation of these
industries, providedthat this would not involve serious loss of
economy in production 273
2. The economic efficiency of public and of joint-stock
operation ofindustry cannot be successfully compared by reference
to statistics 274
3. The efficiency of management in the two cases is likely to
bepretty much the same .... . 275
4. But, first, under public operation there is a danger that the
operat-ing authority may be tempted to maintain its enterprise by
theuse of unfair uncommercial methods at the expense of rival
enter-prises capable of satisfying the same wants more cheaply . .
276
5. Secondly, under public operation efficiency is likely to
sufferthrough an undue restriction of the supply of the factor of
produc-tion uncertainty-bearing ...... 281
6-7. Thirdly, efficiency is likely to suffer through the
establishment ofunits of management of an uneconomical size ;
though, in certaincases, where the normal state of things is
monopolistic competition,
public operation is, in this respect, superior to joint -
stockoperation ... .... 284
8. The inferiority in respect of productive efficiency, which
thusattaches, in general, to public operation, has to be
balanced
against the inferiority of public control in respect of the
most
economical adjustment of supply to demand. The result worksout
differently for different industries and at different times .
287
-
XX WEALTH AND WELFAEE
PART III
THE DISTRIBUTION OF THE NATIONAL DIVIDEND
CHAPTER I
INTRODUCTORYPAGE
1. The following part is concerned with attempts to improve
thedistribution of the dividend by the deliberate transference
ofresources from the relatively rich to the relatively poor,
firstthrough interference with the natural course of wages, as paid
toparticular poor persons, and secondly through taxation and
soforth. ....... 293
2. It can be shown that, in general, transferences brought about
ineither of these ways will only increase the real income of the
poorin the long run, provided that they do not diminish the
nationaldividend ........ 293
3. The order of topics to be discussed in the succeeding
chapters isindicated ...... 294
CHAPTER IIPOEMS OF INTERFERENCE WITH THE NATURAL COURSE OF
WAGES
1. Interference with the natural course of wages at any point
may takethe form of the establishment of an artificial wage-rate,
involvingeither a small or a large element of differentiation . .
296
2, Even piece-wages in general imply some differentiation in
favour ofinferior workpeople....... 296
3-4. Under time-wages, when no excessive rigidity is maintained,
butattempts at adjustment of payment to efiiciency are made, a
some-what larger element of differentiation is present . . .
297
5. Under a rigid time-wage of so much per hour for all workers,
a muchlarger element of differentiation is present.... 301
CHAPTER IIIMETHODS AVAILABLE FOR RAISING THE WAGE-RATE AT ANT
POINT
ABOVE THE NATURAL LEVEL
1. The principal means by which attempts are made to raise the
rateof wages of any group of workpeople above the natural level
aretrade -union diplomacy, legislative action and pressure on
thepart of consumers ....... 303
-
CONTENTS xxiPAGE
2. These attempts may easily succeed in their immediate object
ofmaking the real wage-rate enjoyed by the group in questionhigher
than it would naturally be. They are, however, liableto direct
evasion ... . . 305
3. They are also liable to indirect evasion, through the
migration ofbetter workpeople to districts where higher wage-rates
are paid . 307
4. The real question is whether, and under what conditions, they
willsucceed in their ulterior object of transferring resources
fromrelatively rich to relatively poor persons. In Chapter V.
thisquestion will be discussed in reference to artificial
wage-rates of anon-differential character ...... 308
CHAPTER IVMETHODS OP ENGAGEMENT OP LABOUR
1. The answer to the question just proposed depends partly on
themethods in accordance with which workpeople are engaged in
theindustry where the artificial wage-rate is established . 309
2. These methods may be distinguished according as they involve
theconcentration of the work that is available upon a small
numberof men, or the spreading of it among many men . . . 309
3. The influences determining the degree in which concentration
orspreading shall prevail may be divided into three groups,
relatedrespectively to (1) the nature of the work performed, (2)
thenature of the demand for labour as regards iiuotuations, and(3)
the range covered by the typical individual centre of engage-ment.
........ 311
4. Skilled occupations are more favourable to the concentration
methodof engagement than unskilled occupations.... 311
5. In respect of skilled and unskilled occupations alike,
fluctuations inthe demand for labour tend to discourage this method
. . 315
S 6. But the existence in the occupation of a few large centres
ofengagement, as distinguished from many small centres, tends
toencourage it . . . . . . . . 316
7. This last circumstance suggests a consideration relevant to
thepolicy pursued by Labour Bureaus... . 318
CHAPTER VTHE POWER OP A NON - DIPPERENTIAL ARTIFICIAL WAGE-RATE
IN A
PARTICULAR OCCUPATION TO TRANSFER RESOURCES PROM THERELATIVELY
RICH TO THE RELATIVELY POOR
1. The question, whether the establishment of a non-
differentialartificial wage-rate at any point can succeed in its
ulterior objectof transferring resources from the relatively rich
to the relatively
-
xxii WEALTH AND WELFAREPAGE
poor may be considered first on the assumption that the
com-modity produced at this point is consumed exclusively by
othersthan members of the working classes . . . 320
2. The real earnings of that group of workpeople for whom the
arti-ficial rate has been secured will be increased, if the demand
fortheir services has an elasticity less than unity . . . 320
3. The magnitude of the elasticity of demand for the services
ofdifferent groups of workpeople is determined by various
circum-stances, the principal among which are discussed . . 321
4. If there is perfect mobility between different parts of the
industrialfield, and if the attractiveness to labour of each part
is measuredby the average earnings of the workpeople assembled
there, thereal earnings of the w^orkpeople as a whole are increased
by theestablishment of an artificial wage-rate at any point, when
theelasticity of the demand for labour at that point is less than
unity 325
5. In an occupation where the methods of engagement are such
thatwork is spread evenly among all the men attached to it,
theattractiveness of the occupation is measured by the
averageearnings of the workpeople assembled there . . . 326
6. In an occupation where the work is spread more or less at
randomamong the men attached to it, attractiveness may be larger,
ascompared with average earnings, than it is elsewhere. In thiscase
the establishment of an artificial wage-rate is likely to beless
favourable to the real earnings of the workpeople as a wholethan it
would be under the even spreading method .
. 327
7. In an occupation where the concentration method of
engagementprevails, the establishment of such a rate is likely to
be morefavourable than it would be under the even spreading method
. 329
8. The same proposition holds good of an occupation entrance
intowhich is obstructed by obstacles in the way of mobility . .
330
9. When account is taken of the fact that a great part of the
productof one group of workpeople is usually consumed by other
work-people, it becomes clear that the establishment of an
artificialwage-rate at any point is less likely to be successful in
bringingabout a transference of resources from the relatively rich
to therelatively poor than the preceding considerations, taken
alone,would suggest ...... 332
CHAPTER VIa?HE POWER OF A DIPFERBNTIAL ARTIFICIAL WAGE-RATE IN A
PARTICDLAB
OOCDPATION TO TRANSFER RESOURCES FROM THE RELATIVBLT RICHTO THE
RELATIVELY POOR
1. An artificial wage-rate containing a differential element
cannotmake the aggregate earnings of the workpeople as a whole
-
CONTENTS xxiiiPAGE
SO large as they might be made by some non-differential
artificialwage-rate ........ 334
2. Further, differentiation in the wage-rate in favour of
inferior work-people implies a concentration of work and wages upon
thosewho are not inferior....... 335
3. This proposition, which can be deduced from general
considerations,is supported by statistical evidence .... 336
4. It follows that a differential artificial wage-rate is much
less likelythan a non-differential artificial wage-rate to
advantage economicwelfare ........ 339
CHAPTER VIITHE ULTIMATE EFFECTS OF TRANSFERENCES BROUGHT ABOUT
BY
ARTIFICIAL WAGB-BAIES
1. It has now been established that a real transference of
resourcesfrom the relatively rich to the relatively poor can, in
somecircumstances, be effected through the establishment of
anartificial wage-rate. Will such transference react
beneficiallyupon the national dividend, and, therefore, ultimately,
on thereal income of the poor ? . . . . . . 341
2. Such a result is possible, but is unlikely. In general, the
reactionwill be injurious ....... 341
3. Even, however, when this is the case, it is still possible to
defendthe establishment of an artificial wage-rate for the lowest
class oflabour as a supplementary form of Poor Relief . . 343
CHAPTEK VIIIDIRECT TRANSPERBNOBS OF RESOURCES FROM THE
RELATIVELY RICH
TO THE RELATIVELY POOR
1. We now pass to attempts to transfer resources from the
relativelyrich to the relatively poor by direct action on the part
of philan-thropists or the State ..... 346
2. It is sometimes denied that such transference is possible, on
theground that any absorption of resources from the rich must
causethem to diminish to a corresponding extent their expenditure
inemploying labour. This is invalid .... 346
3. It is also sometimes objected that such transference is
impossible,unless the beneficiaries are at the same time prevented
from con-tinuing to work at industry. This also is invalid . .
347
4. Transference of this kind being thus obviously possible,
inquirymust be made into the effect on the magnitude of the
nationaldividend of (1) the fact of it, (2) the expectation of the
fact of it . 349
-
xxiv WEALTH AND WELFAEE
CHAPTER IX
THE EFFECTS OF THE FACT OF DIRECT TRANSFERENCES FROM THE
RELATIVELY ETCH TO THE RELATIVELY POORPAGE
1. From the point of view of this chapter a transference of
resources
from the relatively rich to the relatively poor means a
substitution
of commodities consumable by the poor for commodities
consum-
able by the rich and for machines . ... 361
2. The reduction in the consumption of the rich will leave
their
efficiency practically unchanged. AVhether the national
dividend
will be increased, therefore, depends partly on how large a
fractionof the resources transferred to the poor is taken from
machines andpartly upon the comparative return obtainable from
investment in
machines and in the persons of the poor .... 351
3. The fraction of resources taken from machines is larger
whenresources are absorbed by death duties than when they
areabsorbed by income tax ...... 352
4. Concerning the absolute magnitude of this fraction we are
not, how-ever, able to make any statement, except that the fraction
willalways be somewhat less than unity .... 354
5. Whether the rate of return from resources turned over for
consump-tion by the poor is greater or smaller than the rate which
wouldhave been yielded by them in the form of machines depends
uponthe conditions by which a transference of resources is
accompanied,and the categories of poor persons to whom it is made .
. 354
6. The return to the marginal pound invested in the efficiency
of thepoor is probably much greater than the return to the
marginalpound invested in the efficiency of machines. This suggests
thatresources, if transferred, would obtain an increased return .
355
7. But the suggestion depends on the assumption that the poor
wouldinvest new resources in themselves in a competent manner ;
and,in fact, they are not likely to do this .... 356
8. Hence, if transferred resources are to yield an increased
return, the
act of transference must be accompanied by control and
direction,through the enforcement of conditions on assisted persons
. 358
9. Certain propositions concerning the nature of these
conditions arelaid down ........ 360
10. Even when control is exercised, resources transferred to
thedefective and the old are not likely to yield a return as
largeas they would have yielded in the form of machines.
Butresources transferred in such wise as to preserve normal men
inmiddle life from sickness and unemployment, and, still
more,resources transferred to build up the bodies and minds of
normalchildren, are practically certain to yield a return much
larger thanthis ......
.362
-
CONTENTS
CHAPTER X
THE EFFECTS OF THE EXPECTATION OP DIRECT TRANSFBEBNCBSFROM THE
RELATIVELY RICH
PAGE 1. The expectation of voluntary transferences from the
relatively rich
implies the discovery of a further desired use for resources,
and sostimulates the supply of effort and waiting . . . 365
2-3. Voluntary transferences are involved in many schemes of
mutualinsurance ..... . . 366
i. They can also be induced by a judicious employment of honours
anddecorations ... . . . 369
5. The expectation of coercive transferences from the relatively
richwill not check the supply of effort and waiting, if the
trans-ferences are effected by the taxation of true rents or of
windfalls . 869
6. The expectation of transferences, effected through taxes that
strikeresources which are consumed and resources which are saved
atthe same rate, will check the supply of effort, but not,
directly,the supply of waiting ..... 370
7. Nor will they, if reasonably arranged, check the supply of
effort toany large extent . . . . . 372
8. The expectation of transferences effected through taxes
thatdifferentiate against resources which are saved may,
however,seriously check the supply of waiting .... 373
9. But this statement does not apply to transferences effected
throughpostponed taxes, such as death duties . . . 375
CHAPTER XITHE EFFECTS OP THE EXPECTATION OP DIRECT TRANSFERENCES
TO
THE EBLATrVELT POOR
1. The effects of the expectation of transferences to the
relativelypoor differ according as the transferences (1) do not
differentiateeither against or in favour of slackness on the part
of the poor,(2) differentiate against slackness, or (3)
differentiate in favour ofslackness ........ 379
2. Examples of non-differential transferences are cited . . .
379
3. The expectation of some kinds of non-differential
transference does,but the expectation of other kinds does not,
check the supply ofindustrial effort on the part of the poor ....
382
4. Examples of transferences which differentiate against
slackness, andthe expectation of which will, therefore, tend to
increase thenational dividend, are cited ..... 383
-
WEALTH AND WELFAEEPAGE
5. The main part of the transferences that are made under the
PoorLaws of modern States differentiate iu favour of slackness on
thepart of the poor ; for their amount varies inversely with the
pro-
vision which certain poor persons make for their own sustenance
. 384
6. Differentiation in favour of small provision through
insurance is not
equivalent to differentiation in favour of slackness in
contributing
to build up the national dividend ..... 385
7. The injurious effect on the dividend of the expectation of
trans-ferences that differentiate in favour of slackness can be
mitigatedby deterrent conditions ...... 388
8. The character proper to these conditions is discussed . .
389
CHAPTER XII
A NATIONAL MINIMUM
1. The establishment of a national minimum, may at the same
timediminish the national dividend and increase the aggregate
realincome of the relatively poor ..... 393
2. By a national minimum is understood an objective minimum
ofconditions in all departments of life, below which the fortunes
ofno citizen are allowed to fall ..... 394
3. Economic welfare is best promoted by the establishment of
anational minimum at such a level that the direct good
resultingfrom the marginal transference to the poor just balances
theindirect evil brought about by the consequent reduction of
thedividend ........ 395
4. It follows that the national minimum should be set higher
when acountry is wealthy than when it is poor . . 396
PART IV
THE VARIABILITY OF THE NATIONAL DIVIDEND
CHAPTER IECONOMIC WELFARE AND THE VARIABILITY OP THE INCOME
OP
THE REPRESENTATIVE "WORKING MAN
1. The economic welfare of a group of individuals is larger, the
moreevenly the consumption of the representative member of
thatgroup is distributed through time..... 401
2. And the joint economic welfare of two groups, each of which
is moreor less homogeneous, is increased if a system of
reciprocal
-
CONTENTS xxviiPAGE
transferences is established between them, in such wise as
todiminish the variability of consumption of the poorer of the two
. 402
3. It follows that any arrangement which, other things being
equal,diminishes the variability of consumption of the
representativeworking man, even though it be by way of reciprocal
trans-ferences which increase the variability of the consumption of
theother classes, tends to make economic welfare larger . . 402
4-6. This conclusion is fortified by a consideration of the
indirectevil incidents, including the evil of intermittent
employment,which variable consumption on the part of a poor man
usuallyimplies ........ 403
7. A proposition analogous to that laid down as regards the
cmiswmption,holds good also as regards the real income of the
representativeworking man ....... 406
CHAPTER II
INSURANCE
1. Diverging from the main argument, we observe that, when
thevariability of the income of the representative working man
isgiven, arrangements which diminish the variability of his
con-sumption increase economic welfare .... 408
2-3. One such arrangement is mutuality and another is saving .
408
4. These two methods are often combined in practice under the
formof insurance........ 410
5. The field over which insurance can be made operative is
limited byimportant technical considerations..... 410
6. And it is further limited by lack of forethought and so forth
onthe part of the poor....... 412
7. This circumstance suggests that the State could increase
economicwelfare by either encouraging or compelling poor persons to
insure 415
8. The respective advantages of encouragement and compulsion
arecompared . ...... 417
CHAPTER III
VARIABILITY IN THE REAL INCOME OP THE REPRESENTATIVE WORKINGMAN
IN RELATION TO VARIABILITY IN THE AGGREGATE REALINCOME OF THE
WORKING CLASSES
1. We now revert to the main argument .... 420
2. The variability of the real income of the representative
working manis not necessarily diminished by a cause which
diminishes thevariability of the demand for, and earnings of,
labour as a whole . 420
-
xxviii WEALTH AND WELFAEEFAOE
3. It is, however, diminished by the generality of ordinary
economiccauses which have this effect, and throughout the following
Part,
with the exception of the last chapter, attention is confined
to
these causes . . ..... 421
CHAPTER IV
THE VARIABILITY OP GENERAL PRICES
1. Since the variability of the real earnings of labour as a
whole is
modiiied by the fact that the purchasing power of money
isvariable, the causes of this variability need to be studied . .
423
2. The determinants of it are (1) the variability of the demand
formoney, (2) the variability of the supply of money, and (3)
theelasticity of the supply of money ..... 423
3. The causes to which variability in the demand for money is
dueare discussed .... . 424
4-9. The circumstances, including the currency and banking
arrange-ments, by which the elasticity of the supply of money is
deter-mined, are discussed in some detail . . . 425
10-11. The causes to which variability in the supply of money is
dueare discussed ....... 432
12-13. When a variation takes place in the demand schedule
formoney, which, it is known, will only last a short time, there
isalways a tendency for the supply schedule to vary in such a wayas
to reduce the consequent movement in general prices . . 435
14. This circumstance suggests that general prices, or, in other
words,the purchasing power of money, might be rendered much
lessvariable than it is by the introduction of a system under
whichthe supply of money should be authoritatively regulated in
theinterests of stability . . ... 437
CHAPTER VCAUSES THAT BRING ABOUT VARIATIONS IN THE REAL
INCOME
OF THE WORKING CLASSES
1. Reverting again to the main argument, we note that the
variabilityof the real income of the working classes is made
somewhat largerthan it would otherwise be by the fact that
working-class con-sumption largely consists of commodities of
especially variablesupply ........ 439
2. Apart from this, variations in the real earnings of the
working classesas a whole may be initiated either by spontaneous
variations inthe volume of the national dividend or by spontaneous
variations
-
CONTENTS xxixPAGE
in the comparative attractions exercised by the rival uses
ofimmediate consumption, storage and investment in the purchaseof
labour ...... . 440
3. The process by which these variations react to bring about
variationsin the real income of the working classes is explained .
. 441
4-5. And the manner in which the process is modified by
themediation of the monetary system is indicated . . . 442
CHAPTER VI
THE VAHIABILITT OF THE BODNTY OF NATURE AND OFFOREIGN
DEMANDS
1-2. The causes of spontaneous variations initiated in the
volume ofthe national dividend arise, from the standpoint of a
particularcountry, out of variations in the bounty of nature and
variationsin the desire of foreigners for its exports ; and the
resultingvariability is likely to be smaller, the more numerous are
thesources of production and of demand . . . 447
3. The variability of the bounty of nature is likely to be
smaller, thesmaller is the extent to which the things we sell and
buy are of akind largely affected by natural forces outside human
control . 449
4. The variability in the desire of foreigners to purchase our
exportsdepends upon certain circumstances, which are discussed . .
451
5. And attention is drawn to the influence of the elasticity of
thisdemand . . . . . 451
CHAPTER VII
THE VARIABILITY OF ERROR IN BUSINESS FORECASTS
1. The causes of spontaneous variations initiated in the
comparativeattractiveness of the investment use are, in the main,
equivalentto those which bring about variations in the
expectationsof business men concerning the absolute profitableness
ofinvestment ... . . 453
2. Among them are the causes of variations in the error of
businessmen's expectations ... . . 454
3. The magnitude ofthese variations is affected by (1) the
characteristicform of modern industry, (2) the quality of the
persons whoseforecasts are practically operative, (3) the reactions
upon oneanother of difTerent persons' forecasts, and (4) the
reproductivepower of the errors that are born . . . 455
-
XXX WEALTH AND WELFAEEPAGE
i. The characteristic form of modern industry includes the
practice ofexchange and the fact of " prospectiveness. " Both these
things
enhance the variability of error . . 455
5. The quality of persons whose forecasts are practically
operative isaffected by the admission, on the one hand, of
professionalfinanciers and, on the other, of speculating members of
thegeneral public ....- 457
6-7. Mutual reactions upon one another of the forecasts of
differentpersons increase the variability of error. In modern
industrythey are strengthened by that interdependence among
businessfirms which is set up by debtor-creditor relation . . .
459
8-9. The reproductive power of the errors that are born also
increasesthe variability of error. It is strengthened by the
influence of themonetary system and of rashness in the working of
the creditsystem . ... . 462
CHAPTEE VIII
THE EELATION BETWEEN THE CAUSES OF VARIATIONS AND THE
VARIA-BILITY OF THE REAL EARNINGS OF THE WORKING CLASSES
1. The variability of the resources invested in the purchase of
labourdepends, not only upon the variability of the dividend and of
theattractions exercised by different uses, but also upon (1)
theorder in which individual variations in each of these things
occurand (2) the correlation, if any, that exists between the two
seriesof variations , ..... . 467
2. In so far as individual variations tend to be arranged in
successionsof upward variations followed by successions of downward
varia-tions, the resultant variability of the earnings of labour
isenhanced ...
. .. 468
3. In fact there is some tendency towards this arrangement in
thevariations that occur' both in the bounty of nature and in
thecomparative attractiveness of different uses . . . 469
4. In so far as upward (and downward) movements in the bounty
ofnature are correlated positively with upward (and
downward)movements in the comparative attraction exercised by
theinvestment use, the resultant variability of the earnings of
labouris enhanced........ 470
5. In fact there is some tendency towards positive correlation
betweenthese two sets of movements
. .
. . .471 6. Some discussion is offered as to the relative
importance of the
different influences that are responsible for the variability of
theaggregate real earnings of labour
. . 472
-
CONTENTS xxxi
CHAPTER IX
PHILAKTHROPIC AND STATE ACTION DESIGNED TO LESSEN THEVARIABILITY
OF THE DEMAND FOB LABOUR
PAGE 1. Philanthropic or State action diminishing the
variability of the real
earnings of labour as a whole need not always diminish the
varia-bility of the real earnings of the representative working
man. Theconditions in which it will do this have now to be studied
476
2. The preliminary objection that philanthropists and the State
areimpotent to diminish the variability of the real earnings of
labouras a whole may be set aside as invalid .... 477
3. As was indicated in Chapter III. , devices, which diminish
thevariability of the real earnings of labour in all parts
(collectively)of the industrial field, otherwise than by increasing
it in somepart, necessarily diminish the variability of the real
earnings of therepresentative working man ..... 478
4. Such devices may be adopted through the dovetailing, by
groupsof associated consumers, of an essentially variable partial
demandinto the interstices of the general demand.... 479
5. They may also be adopted by employers of labour . . . 481
6. Devices, which diminish the variability of the real earnings
oflabour in all parts (collectively) of the industrial iield, by
intro-ducing variations in one part to compensate existing
variations inanother part, are not certain to diminish the
variability of thereal earnings of the representative working man,
but are morelikely to do this, the more perfect is the mobility of
labour . 483
CONCLUSION . . . . 487B
INDEX . . . .... 489
-
PAET I
WELFARE AND THE NATIONAL DIVIDEND
-
CHAPTEE I
WELFARE AND ECONOMIC WELFAEE
1. "If I am asked 'What is good?' my answer is thatgood is good,
and that is the end of the matter. Or, if Iam asked ' How is good
to be defined ? ' my answer is thatit cannot be defined, and that
is all I have to say about it." ^
"Welfare means the same thing as good. It, too, cannot
bedefined, in the sense of being analysed. At the same timewe can
say, and, indeed, it is the chief task of ethics to say,
whether, and in what way, particular things belong to
welfare.For the purpose of this volume it is sufficient to lay down
onthis subject two propositions : first, that welfare
includesstates of consciousness only, and not material things
orconditions : secondly, that welfare can be brought under
thecategory of greater and less. This is all that need be saidhere
concerning welfare in general.
2. Of welfare in general economic welfare is one part.It is
welfare arising in connection with the earning andspending of the
national dividend, or, in other words, of thoseparts of the
community's net income that enter easily intorelation with the
measuring rod of money. Economic wel-fare, however, does not
contain all welfare arising in thisconnection. Various good and bad
qualities indirectly as-sociated with income-getting and
income-spending are excludedfrom it. It does not include the whole
psychic return, whichemerges when the objective services
constituting the nationaldividend have passed through the factory
of the body ; ^
1 G. E. Moore, Principia Ethica, p. 6.2 Of. Fisher, Capital and
IncoTne, p. 168.
-
4 WEALTH AND WELFAEE parti
it includes only the psychic return of satisfaction.
Thus, economic welfare is, as it were, a part of a part of
welfare.
3. This heing so, it is plain that welfare may changewhile
economic welfare remains the same ; and that a givenchange in
economic welfare will seldom synchronise with an
equal change in welfare as a whole. This circumstance might
seem, at first sight, to render the study of economic
welfareunimportant. To conclude thus, however, would be to
mis-conceive the whole purpose of economic investigation.
Thatpurpose is not primarily scientific, if by science we intendthe
single-eyed search after knowledge for its own sake. Itis rather
practical and utilitarian, concerned chiefly to laybare such parts
of knowledge as may serve, directly orindirectly, to help forward
the betterment of social life.
Hence, the failure of economic welfare to act as a baro-meter of
total welfare is of but slight importance. For, what
we wish to learn is, not how large welfare is, or has been,but
how its magnitude would be affected by the introductionof causes,
which it is in the power of statesmen or privatepersons to call
into being. The failure of economic welfareto serve as an index of
total welfare is no evidence that thestudy of it will fail to
afford this latter information : for,though a whole may consist of
many varying parts, so that achange in one part never measures the
change in the whole,yet the change in the part may always affect
the change inthe whole by its full amount. If this is the case, the
practicalimportance of economic study is fully established. It
willnot, indeed, tell us how total welfare, after the
introductionof an economic cause, will differ from what it was
before;but it will tell us how total welfare will differ from what
itwould have been if the cause had not been introduced.
Thateconomic science may provide this important information isa
possibility. Is the possibility realised in fact ? To theview that
it is so realised two serious objections may beurged, and must be
studied.
4. The first objection is an obvious one. It is thatcauses
influencing the satisfactions and dissatisfactions, whicharise in
connection with income, are not confined in their
-
OHAP. I WELFAEE AND ECONOMIC WELEAEE 5
effect to these things, but modify also other parts of
welfare.Since the modifications thus brought about are not taken
intoaccount by our study, we are not entitled to infer that,
becauseone specified cause would add more to economic weKare
thananother specified cause, therefore it would add more to
totalwelfare. Nor may we even infer that, because a specified
causewould increase economic welfare, therefore it would increase
totalwelfare. For, the non-measured effect on non-economic
welfaremight modify, or even outweigh, the measured effect on
economicwelfare. Furthermore, full justice is not done to
thisobjection, when it is merely stated in terms of
possibility.For, we know definitely that, in many cases, economic
causes doinfluence non-economic welfare, both directly and
indirectly, ina very important degree.
Let us look first at direct effects. Every conscious state isa
complex of many elements, and includes not only satisfactionsbut
also cognitions, emotions and desires. This being so, it isnatural
to expect that causes operating to change satisfac-tions may,
either in the same act or as a consequence of it,alter some of
these other elements. This is, in fact, whathappens. For example,
if the commodity which satisfies somedesire becomes more common, it
will often result that thisdesire, through experience of
realisation, becomes more intense.But, if the desire is in itself
good or bad, a growth in itsintensity will alter welfare, and this
alteration will beadditional to any that takes place through
satisfaction. Norare desires the only things that may be
collaterally affectedby changes in the economic environment. The
core of feelingand purpose, which we call people's character, is
susceptible,especially in youth, of modification either towards
good ortowards evil. The surroundings of work react upon thequality
of life.^ Ethical quality is affected by the occupa-tionsmenial
service, agricultural labour, artistic creation,
^'
' The Plymouth Cordage Company of Massachusetts found that, in
conse-quence of surrounding their factory with lawns and
shrubberies, and covering itswalls with creepers, ' the employes
took home with them the lessons we wereendeayouring to teach. They
started to fix up their own grounds ; walks thathad never seen the
edging knife were edged, and lawns were carefully cut,which at once
began to lend an entirely different character to the homes of
theemployes.' "Meakin, Model Factories ami Villages, pp. 79-80.
-
6 WEALTH AND WELFAEE paet i
independent as against subordinate economic positions/
monotonous repetition of the same operation, and so on
into which the desires of consumers impel the people whowork to
satisfy them. It is affected, too, by the influencewhich these
people exert on others, with whom they may bebrought into personal
relations.^ The social aspect ofChinese labour in the Transvaal and
of the attempt byAustralian pastoralists to maintain the convict
system, as asource of labour supply,' have relevance to welfare.
Thehuman relations that arise out of industrial relations arealso
relevant. The esprit de corps and interest in the fortunesof the
firm, which animate the workpeople in establishmentswhere the
personal intercourse of employers and employed iscordial, besides
leading to increased production of wealth, isin itself an addition
to welfare. Nor is this all. Closelyinterwoven with the
quasi-patriotic sentiment, which arises inconnection with some
kinds of industrial arrangement, theremay be found another
important non-economic good
sympathetic and friendly feeling between man and man.
Aslarge-scale industry extended during the eighteenth andnineteenth
centuries, employers and employed became moredistant in station,
and their opportunities of meeting oneanother diminished. In the
wake of this inevitable physicalseparation, there followed in some
cases a moral separation
"the personal alienation of the employer from his fellow-menwhom
he engages to work for him in large numbers." * Thisspirit of
hostility was an obvious negative element broughtabout in
non-economic welfare by an economic cause ; and thepartial
suppression of it through Boards of Conciliation andCopartnership
arrangements is an equally obvious positiveelement. Yet again,
economic causes affect that non-economicpart of welfare, which is
associated with family life or inter-
1 Thus, it is important to notice that machinery, as it comes to
be moreelaborate and expensive, makes it, pro tanto, more difficult
for small men,alike in industry and agriculture, to start
independent businesses of their own.Cf. Quaintance, Farm Machinery,
p. 58.
2 Mr. Eowntree, for example, rightly emphasises the important
influencewhich the overlookers may exert on the moral tone of
persons employed in afactory. Cf. Industrial Betterment, pp.
10-11.
2 Cf. V. S. Clark, The Labimr Movement in Australia, p. 32.^
Oilman, A Dividend to Labour, p. 15.
-
CHAP. I WELFARE AND ECONOMIC WELFAEE 7
national comity. The unity of interest and occupation
whichcharacterises the farm family, as distinguished from the
town-dwelling family/ the sociability and the opportunities
formutual education, both formal and informal, attached
tomembership of co-operative stores, the coarsening influence
ofover-crowded houses, the possible destruction of sympathybetween
kindred peoples through pro-tariff or anti-tariffdisputation, the
possible contamination of judicial or ad-ministrative activity by
the poison of covert bribery, all thesethings are relevant. So
also, finally, is the reflex influenceupon people's characters
produced by such things as publicmuseums and even municipal baths.^
This very real elementin welfare will only be counted in economic
welfare in thespecial case, in which one group of people devote
income topurchasing things for other people. When they do this,
theyare likely to take account of the total effect, and not
merelyof the effect on the satisfactions of those people,especially
ifthe said people are their own children. As Sidgwick
acutelyobserves : " A genuine regard for our neighbour, when
nothampered by the tyranny of custom, prompts us to give himwhat we
think really good for him, whereas natural self-regard prompts us
to give ourselves what we like." ^ In thiscase, therefore, the gap
between the effect on economic welfareand the effect on total
welfare is partially bridged. In manycases, however, it is not so
bridged.
Let us now turn to indirect effects. Causes that modifyeconomic
welfare may influence other parts of welfare, notdirectly, but
indirectly through objective conditions of welfareother than the
national dividend. The most important ofthese conditions are the
services rendered by certain objectsof natural beauty, and those
rendered by certain people,for which money payment is not made. The
effect ofeconomic caxises on objects of natural beauty is
illustrated by
' Cf. Proceedings of the Americaih Economic Association, vol. x.
pp. 234-5.2 Cf. Darwin, Municipal Trade, p. 75.' Practical Ethics,
p. 20. Of. Effertz : "Ce que les int^ressds savent
gen^ralement mieux que les non-int&ess^s, ce sont les
moyeiis propres k r&iliserce qu'ils croient toe leur int^ret.
Mais, dans la determination de I'int^ret lenon-interesse voit
gdneralement plus olair."
Antagonismes ^onomiques, p.237-8.
-
8 WEALTH AND WELFAEE pakt i
the frequent desolation of beautiful scenery through the
hunt
for coal or gold, the desecration widely wrought by
uncon-trolled smoke from factories, and the degraded form
frequentlyassumed by public advertisements.^ Effects on the
servicesrendered by certain people occur whenever economic
causes,such as Poor Law or factory regulations, or fluctuations in
thedemand for men's labour, divert women workers from factorywork
or paid home-work to unpaid home-work, in attendanceon their
children, economical preparation of the family meals,repair of the
family clothes, thoughtful expenditure of house-keeping money, and
so on.^ It is clear that, in cases of thiskind, the changes that
appear in economic welfare result almostentirely from the
definition we have selected for that term,and do not represent any
considerable change in total welfare.It may be well, however, to
add that, though there occurs regu-larly a large absolute amount of
unpaid work, which does notenter into incomethe philanthropic work,
for example, ofunpaid organisers, church workers, Sunday school
teachers, thescientific work of disinterested experimenters, the
political workof many among the leisured classes, and so forthit is
probablynot the case that economic causes bring about any large
amountof transference between unpaid and paid occupations.
Conse-quently, the qualification proper to inferences concerning
totalwelfare from facts about economic welfare is probably not
solarge under this head as it might be thought to be at
firstsight.
5. The second objection is of a more recondite character.The
methodological principle at the basis of economic science,and that
which separates it from the other social sciences, isthe reference
which it makes to a measure, namely, money. This
' The Advertisement Regulation Act, 1907, allows local
authorities to frameby-laws designed to prevent open-air
advertising from affecting prejudiciallythe natural beauty of a
landscape or the amenities of a public park or pleasurepromenade.
It is not, we may note in this connection, a decisive
argumentagainst underground, and in favour of overhead, systems of
tramway power wiresthat they are more expensive. The London County
Council have deliberatelychosen the more expensive underground
variety.
2 It must be noted, however, that, according to the
investigators employedby the Royal Commission on the Poor Laws, a
plan of liberal relief to widowswith children adopted in Glasgow
sometimes led to mere gossip and bad habits."So many of the women
are devoid of domestic or other interest that work forwages IS a
positive safeguard."Report, p. 154.
-
CHAP. I WELFAEE AND ECONOMIC WELFAEE 9
measure, however, can only be brought into relation
withsatisfactions and dissatisfactions through the medium ofdesires
and aversions. Consequently, in order to make use ofour measure, we
are driven to define a given quantity ofsatisfaction as the
satisfaction of a desire of given intensity.When a desire of
greater intensity is satisfied, we say there ismore satisfaction
than when a desire of less intensity is satis-fied, and vice versa.
This is, of course, in itself an entirelylegitimate proceeding.
When, however, satisfaction is definedin this way, it is essential
to observe that, though satisfactionremains a part of welfare, an
addition to satisfaction, or to theeconomic welfare which is
comprised of satisfaction, does notnecessarily imply, even when
everything else remains thesame, any addition, much less an
equivalent addition, tototal welfare. In order that this
implication should holdgood, it would be necessary that the
satisfaction of desires ofequal intensity should always involve, in
themselves and apartfrom their effects, equal measures of "good."
It is easilyshown, however, that the required equivalence may fail
ineither of two ways.
First, let us suppose, to take the strongest case, that
the object of desire is always desired as a means to good
andwith an intensity proportionate to the good which it is
ex-pected to yield. In this case a given addition to
satisfaction,in the sense defined above, will not yield a
correspondingaddition to welfare, when the expectations that have
beenformed are erroneous ; and expectations are
frequentlyerroneous.
Secondly, as a matter of fact, objects of desire are notdesired
with an intensity proportionate to the "good" ex-pected to result
from them. The point is illustrated in aspecial case by Sidgwick's
remark : " I do not judge pleasures[and the case is clearly the
same with satisfactions other thanpleasures] to be greater and less
exactly in proportion as theyexercise more or less influence in
stimulating the will toactions tending to sustain or produce them."
^ In like manner,Franz Brentano writes : " The actual presence of
love by nomeans testifies unconditionally to the worthiness of the
object
' Methods of Ethics, p. 126.
-
10 WEALTH AND WELFAEE parti
to be loved. ... It frequently happens that a person, even
while loving something, confesses himself that it is
unworthy
of his love
:
Video meliora proboque,Deteriora sequor." ^
This point is obviously of large practical importance.
Together
with opinions as to the relative ignorance of the poor, it
lies
at the basis of arguments, such as those advanced by Mr. Leverin
favour of prosperity-sharingwhere employers decide in
what form the workmen shall take their extra gainsasagainst
profit-sharing,^ and of the kindred arguments oftenemployed in
defence of the sanitary clauses of the FactoryActs, governmental
prohibition of truck, and compulsoryinsurance. In wider fields its
significance is equally clear.Imagine, for example, that a
statesman is considering how farinequality in the distribution of
wealth influences welfare.
He will reflect that the satisfaction of some of the desiresof
the rich, such as gambling excitement or luxurioussensual
enjoyment, or perhaps, in respect of Easterncountries,
opium-eating, is ethically inferior to the satisfactionof primary
physical needs, to the securing of which thecapital and labour
controlled by the demand of the rich would,if transferred to the
poor, probably be devoted. On the other
1 Origin of the Knowledge of Right and Wrong, p. 17.2 Under
profit-sharing Mr. Lever holds "that most of the extra money
received by the men is either absolutely wasted personally, or
in luxuries whichdo not materially improve the living conditions,
so that the wife and familyoften fail entirely to benefit
therefrom. In place, therefore, of merely sharingits profits with
its employes, this firm endeavours to share with them the resultsof
prosperity, by laying aside yearly a proportion of its profits to
be invested forthe benefit of all, not to be divided among them and
thus lost sight of. " (Meakin,Model Factories and Villages, p.
428.) Again: "The benefits of prosperity-sharing may be applied to
labour in a great variety of ways. . . . One of thebest methods ...
is to be found in building cottages to be let at low rentals.This
plan is most effective in elevating and bettering the conditions of
labour,and has the additional advantage of ensuring that the wives
and children shallshare in it. . . . Contributions may be made
towards the building of clubs,recreation halls, institutions,
summer holidays, winter entertainments, sick andburial societies.
... By contributing to objects such as these, labour enjoys
thefullest liberty in managing its own institutions outside the
business, whilstmanagement is maintained in its proper place inside
the business. " (Lever,Econxymic Review, Jan. 1901, pp. 62-3.)
There can be little doubt that, under apolicy of this sort, wisely
carried out, a given sum of money will react moreforcibly on the
quality of workpeople than it would do if simply handed over tothem
in the form of coin.
-
OHAP. I WELFAEE AND ECONOMIC WELFAEE 11
hand, he will reflect that other satisfactions purchased bythe
richthose, for example, connected with literature andartare perhaps
ethically superior to satisfaction of primaryneeds, and certainly
superior to that part of the satisfactionof the poor which is
derived from excessive indulgence instimulants.
6. The discussion of the two preceding sections makesit plain
that anything in the nature of rigid inference fromeffects on
economic welfare to effects on total welfare is out ofthe question.
In certain oases the divergence between the twoeffects will be
insignificant, but in others it will be very wide.Nevertheless, I
submit that, in the absence of special knowledge,there is room for
a judgment of probability. When we haveascertained the effect of
any cause on economic welfare, wemay, unless, of course, we have
evidence to the contrary, regardthis effect as prolally equivalent
in direction, though not inmagnitude, to the effect on total
welfare ; and, when we haveascertained that the effect of one cause
is more favourable
than that of another cause to economic welfare, we may, on
thesame terms, conclude that the effect of this cause on
totalwelfare is probably more favourable. In short, there is
apresumption,what Professor Edgeworth calls an "
unverifiedprobability,"that. conclusions about the effect of an
economiccause upon economic welfare will hold good also of the
effecton total welfare. The burden of proof lies upon those whohold
that, in any particular case, this presumption should
beoverruled.
7. The above result suggests prima facie that economicscience,
when it shall have come to full development, is likelyto^furnish a
powerful guide to practice. Against the acceptanceof this
suggestion there remains, however, one considerableobstacle. When
the conclusion set out in the precedingsection is admitted to be
valid, a question may still be raisedas to its practical utility.
Granted, it may be said, that theeffects produced by economic
causes upon economic welfare areprobably, in some measure,
equivalent to those produced ontotal welfare, we have really gained
nothing. The effectsproduced upon economic welfare cannotthe
argument runs
be ascertained beforehand by those partial and limited
investi-
-
12 WEALTH AND WELFAEE pakt i
gations, which alone fall within the scope of economic
science.
The reason for this is that the effects upon economic
welfare
produced by any economic cause are likely to be modified
according to the character of the non-economic conditions,
which,
in one form or another, are always present, but which
economicscience is not adapted to investigate. The difficulty is
statedvery clearly by J. S. Mill in his Logic. The study of a part
ofthings, he points out, cannot in any case be expected to
yield
more than approximate results : " Whatever affects, in
anappreciable degree, any one element of the social state,
affects
through it all the other elements. . . . We can never
eitherunderstand in theory or command in practice the condition ofa
society in any one respect, without taking into considerationits
condition in all other respects. There is no socialphenomenon which
is not more or less influenced by everyother part of the condition
of the same society, and therefore,by every cause which is
influencing any other of the con-temporaneous social phenomena."^
In other words, theeffects of economic causes are certain to be
partially dependenton non-economic circumstances, in such wise that
the samecause will produce somewhat different economic effects
accord-ing to the general character of, say, the political or
religiousconditions that prevail. So far as this kind of
dependenceexists, it is obvious that causal propositions in
economics canonly be laid down subject to the condition that things
outsidethe economic sphere either remain constant or, at least,
donot vary beyond certain defined limits. Does this condi-tion
destroy the practical utility of our science ? I hold that,in
respect of nations in a reasonably stable state ofgeneral culture,
like those inhabiting Western Europe, themeasure in which the
condition fails is not, in general, largeenough to render the
results reached by economic inquiryother than reasonably good
approximations to truth. This isthe view taken by Mill. While fully
recognising " the para-mount ascendancy which the general state of
civilisation andsocial progress in any given society must exercise
over all thepartial and subordinate phenomena," he concludes that
theportion of social phenomena, in which the immediately
1 Logic, ii. p. 488.
-
CHAP. I WELFAEE AND ECONOMIC "WELFAEE 13
determining causes are principally those that act throughthe
desire fpr wealth, " do mainly depend, at least in the firstresort,
on one class of circumstances only." He adds that,"even when other
circumstances interfere, the ascertainmentof the effect due to the
one class of circumstances alone is asufficiently intricate and
difficult business to make it ex-pedient to perform it once for
all, and then allow for theeifect of the modifying circumstances;
especially as certainfixed combinations of the former are apt to
recur often, inconjunction with ever varying circumstances of the
latterclass." ^ I have nothing to add to this statement. If it
isaccepted, the difficulty discussed in the present section need
nolonger give us pause. It is not necessarily impracticable
toascertain by means of economic science the approximate effectsof
economic causes upon economic welfare. The bridge thathas been
bidlt in earlier sections between economic welfareand total welfare
need not, therefore, rust unused.
1 Logic, ii. pp. 490-91.
-
CHAPTEE II
ECONOMIC WELFARE AND THE NATIONAL DIVIDEND
S 1. In the preceding chapter some study has been made ofthe
effects produced by economic causes, immediately uponeconomic
welfare, and ultimately upon total welfare. Wehave now to notice
that few causes operate upon economicwelfare directly. In general,
they operate indirectly through
the magnitude of the national dividend, and through
itsdistribution between different persons and between
differentparts of time. The problem of this chapter is to
determinein broad outline the relations subsisting between changes
ineconomic welfare and changes in certain quantitative aspectsof
the dividend. Before this is attempted, however, some
closer study of the dividend itself must be attempted.
2. The ultimate elements of which the dividend is com-posed
consist of objective services, some of which are renderedthrough
commodities, while others are rendered direct. Inreckoning up these
elements, we must be careful not to countthe same thing twice
overnot to enter in our inventory, forexample, both the services of
the baker and the loaf thatthese services bake, or both the
railways of the country andthe value of railway shares. Under this
head difficulties ofsome importance need to be overcome. Their
nature may beillustrated from the recent official attempt, not,
indeed, tomake an inventory of the national dividend, but to
estimate,in part, its money value. In the Preliminary Eeportissued
in connection with the Census of Production, theDirector summarises
the method followed to avoid doublecounting as follows : " The
result of deducting the total cost
u
-
OH. II ECONOMIC WELFAEE AND THE DIVIDEND 15
of materials used and the amount paid to other firms for
workgiven out from the value of the gross output for any
oneindustry or group of industries is to give a figure which
may,for convenience, be called the ' net output ' of the
industry
or the group. This figure expresses completely and
withoutduplication the total amount, by which the value of
theproducts of the industry or group, taken as a unit, exceededthe
value of the materials purchased from outside; i.e. itrepresents
the value added to the raw materials in the courseof manufacture.
This sum constitutes for any industry thefund from which wages,
salaries, rents, rates, taxes, deprecia-tion and all other similar
charges, as well as profits, have tobe defrayed." ^ By adding
together the net products of all theindustries of the country,
along with the net products ofthe country's capital invested
abroad, we obtain a catalogue ofthe whole dividend expressed in
terms of money. If doublecounting is obviated by methods of this
kind, there is noobjection in principle to including the two sorts
of servicesdistinguished abovethose rendered through commodities
andthose rendered directunder the term " goods and services."When
this term is properly guarded and qualified, its meaningis
precisely the same as that which Professor Fisher andothers intend
to signify by the term " services." The choicebetween the two terms
is a matter, not, as Professor Fisherappears to suggest,^ of
principle, but of convenience. Per-sonally, while recognising the
awkwardness of wide departurefrom business usage, I am inclined to
prefer the shorter andsimpler term " services."
3. It is, again, a matter of convenience, and not ofprinciple,
to determine whether all or some, and, if some,what services or
goods and services shall be reckoned as partsof the national
dividend. As indicated at the beginning ofthe preceding chapter, I
propose here to follow Dr. Marshall'susage and to include only such
items as can be brought easilyinto relation with a money measure.
The test of easinessagain, I derive, still following Dr. Marshall,
from the practiceof the British Income Tax Commissioners. Hence, I
include
1 [Cd. 4896], p. 6.2 The Nature of Capital and Income, pp.
105-6.
-
16 WEALTH AND WELFAEE parti
everything that people buy with money income, together withsuch
services as a man obtains from a house owned andinhabited by
himself. But, " the services which a personrenders to himself and
those which he renders gratuitouslyto members of his family or
friends ; the benefits which hederives from using his own personal
goods [such as furnitureand clothes], or public property such as
toll-free bridges, arenot reckoned as parts of the national
dividend, but are left tobe accounted for separately." ^ Our
decision to use the termnational dividend in this restricted
significance must notbe forgotten when we proceed to argue that a
diminutionin the dividend, in general, implies a diminution in
economicwelfare.
4. Our next problem is of more substantial significance.Certain
services have been separated off by the two precedingparagraphs as
prima facie belonging to the national dividend.Among these
services, however, closer reflection exhibits twodistinct
groupsservices that are converted into psychic incomeforthwith, and
services that are devoted to the constructionof instruments for
yielding other services, which will be con-verted into psychic
income in the future. It is a questionof some difficulty to decide
whether both these groups arereally eligible, or whether the latter
ought to be excludedaltogether, or whether some part of both ought
to be excluded.The whole body of these services we may call, if we
will, thegross national dividend. Our problem then assumes the
form
:
" What part of the gross national dividend constitutes the
true,net, national dividend ?
"
5. To this problem two principal solutions have beengiven by Dr.
Marshall and Professor Fisher respectively.Dr. Marshall writes :
"The labour and capital of the country,acting on its natural
resources, produce annually a certainnet aggregate of commodities,
material and immaterial, in-cluding services of all kinds. This is
the true net annualincome or revenue of the country, or the
national dividend." ^
But, he adds elsewhere : " If we look chiefly at the incomeof a
country, we must allow for the depreciation of the sources
' Marshall, Principles of Economics, Gt\ ed. p. 524.2 Ibid. p.
523.
-
OH. II ECONOMIC WELFAEE AND THE DIVIDEND 17
from which it is derived." ^ In other words, Dr. Marshall'sview
is that the true net national dividend comprises thewhole of the
gross dividend, minus such part as would suf&ceto maintain the
country's capital intact. It may, therefore,according to
circumstances, be either greater or less than thesum of services
devoted to immediate consumption during theyear ; if new capital is
created, it will be greater than that sum,whereas, if renewals and
repairs are neglected, it will be less.Professor Fisher, on the
other hand, placing in the forefrontof his argument the proposition
that savings are in no circum-stances income, claims unequivocally
to identify the nationaldividend with those services, and those
only, that enter directlyinto consumption. According to him. Dr.
Marshall's nationaldividend represents, not the dividend that
actually is realised,but the dividend that would be realised if the
country's capitalwere maintained. These two things are, indeed,
materiallyequivalent in the special case where resources are set
aside,whether in renewals of obsolescent plant or in
depreciationfunds, to an extent that exactly offsets the
depreciation incapital that occurs. In practice, however, it is
extremelyrare for the two things to be materially equivalent, and
it isimpossible for them to be analytically equivalent.
6. There can be no question that, in point of
logicalcompactness. Dr. Marshall's definition labours under a
seriousdisadvantage. In a thorough-going stationary state there
isno ambiguity or difficulty about the conception of what
isrequired to maintain capital intact. If a particular sort
ofmachinery wears out in ten yearsTaussig's estimate for theaverage
life of machinery in a cotton mill ^it is obvious thatthe net
national dividend over ten years falls short of thegross dividend
by the value of this machinery. Again, in sofar as the growth of
any sort of crop wastes the productivepowers of the soil, the net
dividend falls short of the grossdividend by the cost of returning
to the soil those chemicalingredients that have been removed.^
Again, when minerals
' Principles of Economics, p. 80.* Quarterly Journal of
Economics, 1908, p. 342.^ Professor Carver writes of the United
States : " Taking the country over,
it is probable that, other things equal, if the farmers had been
compelled tobuy fertilisers to maintain the fertility of their soil
without depletion, the whole
C
-
18 WEALTH AND WELFAEE pakt i
are dug out of the ground, a deduction should be made equal
to the excess of the value, which the minerals used during
the year had in their original situation (theoretically
represented by the royalties paid on their working), over
the value which whatever is left of them possesses to the
country after they have been used. If "using" means
exporting in exchange for imports that are not used as
capital,
this latter value is zero. If, on the other hand, it means
working up into some enduring instrument, the value of the
mineral after use will be greater than the value it had in
the
mine, and, in order to obtain the net from the gross
dividend,
we shall need to add, and not to subtract, something.All this is
simple enough. When, however, we are con-
cerned, not with an imaginary stationary state, but with the
condition of affairs that actually exists, the mere
maintenance
of the physical efficiency of our plant is no longer
obviously
equivalent to the maintenance of our capital intact.
Machinery
that has become obsolete because of the development of im-
proved forms is not really left intact, however excellent
its
physical condition ; and the same thing is true of machineryfor
whose products popular taste has declined. If, however,
in deference to these considerations, we decide to make
anallowance for cases of obsolescence, we are exposed to the
retort
that this concession logically implies the recognition of
the
value, and not the physical efficiency, of instrumental goods
asthe object which is to be maintained intact. But, it is
thenurged, the value of instrumental goods, being the present
value
of the services which they are expected to render in the
future,necessarily varies with variations in the rate of
discount.Is it really a rational procedure so to define the net
national
dividend that its relation to the gross dividend shall
depend on a circumstance of this kind ? No doubt, we canand do
protest vigorously against the stretching of our conces-sion upon
this logical Procrustes' bed. We assert that capitalis maintained
intact in our sense, when physical depreciation
industry would have become bankrupt. . . . The average farmer
had never(up to about 1887) counted the partial exhaustion of the
soil as a part of thecost of his crop " (Sketch of American
Agriculture, p. 70). Against thiscapital loss has to be put,
however, the capital gain due to the fact of occupationand
settlement of the land.
-
CH. II ECONOMIC WELFAEE AND THE DIVIDEND 19
and depreciation through obsolescence have been made
good,whether the rate of discount has varied or not. This is
adefinite position, and one that renders the conception of thenet
national dividend, which is linked up with it, reasonablyprecise.
Nevertheless, we are bound to confess that this con-ception is in
the nature of a compromise and does not presentthose clear-cut
logical facets, with which the rival conceptionof Professor Fisher
confronts us.
7. The issue between these two conceptions of thedividend
cannot, however, be