Not the Devils Decade: Nicholas Kaldor in the 1930s

J. E. King*

Abstract:
At the beginning of the 1930s Nicholas Kaldor (1908-1986) was a
third-year undergraduate student at the London School of Economics. By the end of
the decade he was an established economic theorist with an international reputation
and a string of well-received publications to his name. In this paper I focus on
Kaldors four most important articles from this period: his 1934 paper on the nature
of equilibrium theorising; two pieces from 1939, on the compensation principle in
welfare economics and on money, finance and the consequences of speculative
behaviour; and the 1940 paper setting out his Keynesian model of the trade cycle.
All four articles, I argue, are of considerable continuing interest.

Introduction

The Anglo-Irish journalist Claude Cockburn gave his memoir of the 1930s the
title of The Devils Decade (Cockburn 1973). As a summary description of the
state of the world in the 1930s this is hard to gainsay, but it cannot be applied to
the life and career of the subject of this paper. Nicholas (Miklos) Kaldor was born
in Budapest on 12 May 1908 into a prosperous middle-class Jewish family.1 In
1925-27 he studied economics at the University of Berlin, his interest in the
subject having been aroused by his experience of the German hyperinflation
during a family holiday in Bavaria in 1923. Dissatisfied with the standard of
teaching in Berlin, Kaldor returned to Hungary before enrolling at the London
School of Economics in the autumn of 1927. Here he came under the influence
first of the idiosyncratic American Marshallian, Allyn Young, and then of Youngs
charismatic successor to the chair of Economics, Lionel Robbins (Kaldor 1986a, p.
12; 1986b, pp. 31-4, 38-9).2 On 1 January 1930 Kaldor was almost midway through
the third and final year of his degree. He was to pass with flying colours (except in
Statistics and Scientific Method, which was his worst subject); he was enrolled as a
research student in 1930, appointed to an Assistant Lectureship in 1932 and,
belatedly, promoted in 1938 to full Lecturer at the School.3
Kaldor was always a prolific writer. In Berlin, at the tender age of 19, he
had worked as a stringer for the Budapest press, a practice he continued for a
while after his move to London.4 By 1931 he was addressing dense three-page
missives to John Maynard Keynes querying details of the argument in the
Treatise on Money and eliciting a courteous, if rather frustrated, reply (Keynes
1987, pp. 238-42). He was also co-translator of Friedrich von Hayeks extended
critique of underconsumption theory (Hayek 1931). Kaldors own professional
publications began in 1932 with a 12-page article on the Austrian economic crisis
that Keynes had rejected for the Economic Journal but Kaldor managed to place
in the Harvard Business Review (1932a). Over the next eight years he published
22 journal articles, 8 substantial book reviews and a translated book.5 He was a
natural linguist, blessed with the rare ability to write perfect prose in his second
and third languages (he was effectively trilingual in Hungarian, German and
English, and seems to have had at least a good reading knowledge of French and

40 History of Economics Review

_______________________________________________________________________________

Italian). His written English was clear, concise and accurate, if not particularly
elegant (Dorfman 1961, p. 495), and his interests covered the entire gamut of
economic theory and to a more limited extent, in the 1930s of economic
philosophy and policy.
Four of Kaldors articles from this first decade of his academic career
stand out from the rest, and I shall devote a section to each of them. These are his
1934 paper on the nature of equilibrium theorising; two pieces from 1939, on the
compensation principle in welfare economics and on money, finance and the
consequences of speculative behaviour; and the 1940 paper setting out his
Keynesian model of the trade cycle. Of the rest, the Hayek translations are
important if only because they gave Kaldor the opportunity to familiarise himself
with the Austrians ideas. Hayek had been appointed in 1931 to the Tooke chair at
the LSE, partly on Kaldors initiative (Targetti 1992, p. 3), and together with
Robbins he was to dominate the teaching of economics there for almost twenty
years.
Although Kaldor was later to claim that the work of translation also
awakened him to many of Hayeks theoretical errors (Kaldor 1986a, p. 30 n. 7),
there is no indication of this in any of his own publications in the early 1930s,
which are essentially liberal, Austrian and (by implication) pre-Keynesian in
character. He criticises the German socialist Emil Lederer for claiming that
technical progress was responsible for a permanent increase in unemployment,
arguing that this is possible only if money wages are assumed to be rigid
downwards. Lederers case, he concludes, sounds very strange to people who were
brought up on the marginal productivity analysis (1932a, p. 190). Unemployment
is due not to technical change but to monopolistic interference with the price
system, in this case presumably by trade unions (ibid., p. 195).6 Reviewing
Carl Landauers Planned Economy and Market Economy, Kaldor invokes Ludwig
von Mises criticism of socialist economics:
Even if we assume that a free market for consumption goods can be
preserved, the methods of producing these goods will have to be
decided arbitrarily; as the Socialist producer cannot, even if he tried to,
find out the true displacement [i.e., opportunity] costs of the factors of
production. This problem, which emerged as soon as the conception of
real costs was abandoned, has so far proved insoluble. (1932b, p. 379)
It was his reading of the General Theory, and discussions with young LSE
colleagues like Maurice Allen, Abba Lerner and his then very close friend John
Hicks, that finally convinced Kaldor of the defects of Austrian macroeconomics. At
about the same time, in the mid-1930s, he adopted the moderate Fabian socialist
beliefs that he was to hold for the rest of his life.
In later life Kaldor read little and rarely reviewed books, but in the 1930s
he was a very active reviewer, in particular for the LSE house journal,
Economica. His knowledge of German made him an obvious choice to review
books in that language, and this allowed him to assimilate the latest developments
in economic thought in both Germany (before Hitler came to power in May 1933)
and Austria (where academic freedom survived largely intact until the Anschluss
in March 1938). In addition to his reviews of Landauer and Lederer, already
noted, he also reviewed Erich Schneiders Theory of Production, which failed to
convince him of the merits of a mathematical approach to economic analysis,7
and H. von Stackelbergs Market Form and Equilibrium, which impressed him
greatly (1936a, 1936b).

Nicholas Kaldor in the 1930s 41

_______________________________________________________________________________

There was another important Austrian connection. Kaldors original topic

for his PhD dissertation was Commercial policy of the Danubian states after the
war.8 Although he never submitted a thesis he did publish a substantial article on
the economic crisis in Austria, as already noted. Drawing a very long bow it is
just possible to detect in this paper a Minskyian theory of financial instability, in
which the operating losses of bank-owned industrial enterprises in Austria forced
them into a vicious circle of Ponzi finance, borrowing repeatedly to cover their
commitments from previous loans.9 For the most part, however, Kaldors
interpretation follows the conventional sound finance critique of Austrian
financial policy: both the government and the population were living beyond their
means, and the day of reckoning had eventually arrived.
In 1935 Kaldor was awarded a Rockefeller Scholarship; Oskar Lange and
Nicholas Georgescu-Roegen were the other European recipients that year. This
permitted him to travel widely in the United States in 1935-6. He visited
Columbia, Harvard, Chicago and the University of California, presenting a paper
on wage subsidies as a remedy for unemployment at the 1935 meeting of the
Econometric Society in New York (1936c, p. 271 n. 1) and also attending the
1936 meeting of the Society in Colorado Springs. Never again would Kaldor
spend so much time in the United States, or feel so close to prominent American
economists. One product of his Scholarship was a commission to write the 1937
Annual Survey of Economic Theory for the Societys journal, Econometrica
(1937a). For the most part this is an intense, respectful and rather uninteresting
account of the analytical differences between Hayek and the Chicago capital
theorist Frank Knight,10 but there is a sting in the tail. In the final two pages of
the survey Kaldor slides his stiletto into Austrian capital theory, silently and
without warning but with fatal effect:
The purpose of the investment period approach is to reduce the
production function to two variables, substituting waiting for the
services of all produced (or variable) factors, with interest as the price
of waiting. In this way and only in this way can capital as capital
be treated as a factor of production, commensurate with labour.
(1937b, 232; original stress)
This, however, can only be done if both the fixed factors and the final output were
themselves homogeneous.11 In general, Kaldor concludes, the degree of
roundaboutness is not a useful concept, even under static conditions. Once capital
accumulation comes under consideration, the Austrian theory is in even deeper
trouble: There can be no doubt that for an analysis of dynamic problems and
especially of the par excellence dynamic problem of the trade cycle the
investment-period concept could hardly be of any use (ibid., p. 233; cf. 1939a,
1940a). Twenty-five years later, irritated by Joan Robinsons apparent belief that
criticism of neoclassical capital theory had originated in Cambridge, he would
claim that it was, on the contrary, born at the LSE (King 1998, p. 428).
Kaldors early contacts with Cambridge economists, and their influence
over his thinking on macroeconomics, will be touched on in sections 4 and 5 below.
In the 1930s he also wrote frequently on microeconomics, and in particular on
another topic close to Cambridge hearts, the theory of the firm. His extended
review of Joan Robinsons Economics of Imperfect Competition, written before he
had met Robinson,12 is both complimentary and critical. She has neglected
oligopolistic interdependence between producers, Kaldor complains, and has little
to say about selling costs. Under the circumstances it is not surprising that her

42 History of Economics Review

_______________________________________________________________________________

book inevitably becomes a treatise on monopoly (1934c, p. 335). Although he

doubts whether monopsonistic exploitation of labour is still of practical
significance, given the strength of the British trade unions, he praises Robinsons
treatment of the labour market: Of all Mrs. Robinsons results, unquestionably
the most valuable are to be found in Books VII-IX, which deal with the extension
of the marginal-productivity theory of distribution to monopoloid situations
(ibid., p. 337). When she returned to Imperfect Competition, writing a new
introduction to the second edition in 1969, Robinson and Kaldor were barely on
speaking terms, and she did not refer to his review, but the principal weaknesses
that she now acknowledged were precisely those that had been identified by
Kaldor thirty-years earlier. She now admitted that her treatment of oligopoly
involved a shameless fudge, while she remained proud of her analysis of wages
under imperfect competition, what for me was the main point (Robinson 1969,
pp. vi, xii).
Kaldor also took on Robinsons great rival, the Harvard theorist of
monopolistic competition, Edward H. Chamberlin, in articles on the equilibrium
of the firm (1934b) and the relationship between market imperfection and excess
capacity (1935), and in a protracted reply to Chamberlins criticisms (1938b). The
Americans considered response can be found in his Towards a More General
Theory of Value (Chamberlin 1957, pp. 45-6, 73-9, 177-9, 187-8). Again there is
an almost Talmudic flavour to these pieces,13 and they have little abiding interest
for non-specialists.14 But this was the cutting edge of microeconomic theory in
the 1930s, and Kaldor had shown himself to be a force to reckon with in this area,
too.

Equilibrium (1934)

One by-product of his interest in the theory of the firm did prove to be of lasting
significance. This was A Classificatory Note on the Determinateness of
Equilibrium, published in the very first issue of the Review of Economic Studies 15
in February 1934, the very same month in which the Dollfuss putsch destroyed
democracy in Austria. Kaldors theme is the conditions necessary to make
equilibrium determinate: the conditions under which we can give a scientifically
precise description of the actual course of economic phenomena (1934a, p. 122).
He identifies three grounds for concern; in effect they are the now-familiar
questions of existence, uniqueness and stability. Kaldors treatment of the existence
problem is both original and idiosyncratic.16 He is concerned not with the properties
of mathematical systems his discussion is entirely non-mathematical but rather
with the problem of path-dependency:
For the mere fact that there is, in any given situation, at least one
system of prices, which, if established, would secure equilibrium, does
not imply that this particular set of prices will also be put into operation
immediately; and if any other set of prices is established, not only will
further price-changes become necessary, but the equilibrium system of
prices (i.e. that particular set of prices which does not necessitate
further changes in prices) will itself be a different one. It is not possible,
therefore, to determine the position of equilibrium from a given system
of data, since every successive step taken in order to reach equilibrium
will alter the conditions of equilibrium (the set of prices capable of
bringing it about) and thus change the final position unless the

Nicholas Kaldor in the 1930s 43

_______________________________________________________________________________

conditions are such that either (1) an equilibrium system of prices will
be established immediately, or (2) the set of prices actually established
leaves the conditions of equilibrium unaffected (in which case the final
position will be independent of the route followed). (ibid., p. 124;
original stress)
Path-dependency is always a possibility, Kaldor notes, unless any move from one
position of equilibrium to another is immediate (p. 126). Where such a move
requires time,
it is only by means of a theory of the path (a theory showing what
determines the actual path followed) that a causal-genetic approach can
arrive at generalisations concerning the nature of equilibrium and
such a theory has not hitherto been forthcoming, although the necessity
for it has frequently been emphasised by writers of the Austrian School.
(ibid., p. 128)
Here Kaldor cites a book by the Austrian theorist Hans Mayer and Appendix F of
Alfred Marshalls Principles (ibid., p. 128 n. 1, n. 2), together with classic texts by
William Stanley Jevons, Lon Walras and F. Y. Edgeworth and papers by his friend
John Hicks in both English- and German-language journals.17 Avoiding pathdependence, he continues, requires that tastes and obstacles [that is, technology],
on each day, for everybody, should be unaffected by the events of the previous day
(ibid., p. 128). Thus for equilibria to be determinate, or path-independent, there
must be no significant learning effects.18
Kaldors discussion of multiple equilibria includes the relatively familiar
cases of backward-rising supply curves and backward-falling demand curves
that had been recognised by Walras, Knut Wicksell, Marshall and Hicks (ibid.,
pp. 129-30). Significantly, Kaldor points out that non-uniqueness is a sufficient
condition though not a necessary condition for path-dependence:
multiple equilibrium will always be present whenever there are stages
of increasing returns to single industries, i.e. whenever there are
stages of diminishing technical marginal rates of substitution. In these
cases, therefore, the final situation will be indeterminate in the sense
that it will depend upon the direction which happens to be adopted
initially; though equilibrium may still be determinate on our definition
of the term, since all possible equilibrium positions may still be
deduced from the data of the initial situation. (ibid., pp. 131-2)
The first and second problems of equilibrium theorising are thus intimately
connected.
The third problem is whether in any given case equilibrium will be
definite or indefinite (i.e. whether it will be approximated to or not), and this
appears to depend on the velocities of adjustment of the factors operating in the
system (ibid., p. 133). Here Kaldor draws upon published work (all in German) by
his LSE colleague Paul Rosenstein-Rodan, by Henry Schultz and by Umberto
Ricci, and also cites an Italian paper by Mauro Fasiani. He distinguishes between
completely continuous and completely discontinuous adjustments, the latter case
being much easier to deal with. Here stability (or definiteness) of equilibrium will
depend on the relative elasticities of demand and supply; according to what may be
called the cobweb theorem of Professor Henry Schultz and Professor U. Ricci
(ibid., p. 134). This is illustrated in Figure 1, which reproduces Kaldors own
Figures 2 and 3.

44 History of Economics Review

_______________________________________________________________________________

Figure 1

Kaldor claimed to have originated the term cobweb theorem,19 and this is
confirmed by Pashigan (1987). The conclusions are in 2007 known to all students
of intermediate microeconomics, but in 1934 they were stated in English for the
very first time:
(i) If demand is elastic relatively to supply, the cobweb will be
contracting; equilibrium will be definite;
(ii) If supply is elastic relatively to demand, the cobweb will be
expanding; equilibrium will be indefinite;
(iii) If the elasticities of supply and demand are the same, there will be a
constant range of fluctuations. (Kaldor 1934a, p. 135)20
In the more difficult case of continuous adjustment, Kaldor concludes, stability is
governed by the relative speeds of adjustment of supply and demand rather than by
the relative elasticities.
Kaldor was always proud of this paper (1960, p. 4), and justifiably so. Its
most important contribution, however, is not the elucidation of the cobweb theorem
but the discussion of path-dependence. Again, this was not Kaldors discovery; it
can be traced back (like so much in the history of economic thought) to Alfred
Marshall and his doubts concerning the reversibility of the long-run supply curve, a
point which was emphasised by Allyn Young in his LSE lectures (Blitch 1990;
Sandilands 1990). But Kaldor brings out its microeconomic significance with
exceptional clarity, even if he fails to recognise just how subversive of neoclassical
economic theory it really is (see Kaldor 1972 for a belated acknowledgement of this
point; cf. Harcourt 1988). He does not, in 1934, indicate any awareness of the
implications of path-dependence for macroeconomic theory. Decades later he
would, however, invoke precisely this principle in his attack on the neoclassical
approach to economic growth (1975, 1985; cf. Setterfield 1998). By this time, of
course, similar arguments had come to figure prominently in Joan Robinsons
critique of mainstream theorising (Robinson 1974) and, under the fancy title of
hysteresis, were being used both by Post Keynesians and more orthodox
macroeconomists to attack monetarists claims about the so-called natural rate of
unemployment (Arestis and Sawyer 2004; cf. Cross 1987), while path-dependence
features prominently in the influential work of scholars like Paul David and
W. Brian Arthur (1994). All this was implicit in the work of the 25-year old
Nicholas Kaldor.

Nicholas Kaldor in the 1930s 45

_______________________________________________________________________________

Welfare (1939)

His second major contribution came in his short paper Welfare Propositions of
Economics and Interpersonal Comparisons of Utility, published in the September
1939 issue of the Economic Journal (1939d). Some background on the
development of welfare economics in the interwar years may help to put this brief
note in context. For conservative economists, utilitarianism had always seemed to
carry a dangerously egalitarian message. Since everyone accepted the law of
diminishing marginal utility, maximising utility for society as a whole appeared to
require the complete elimination of inequality in income and wealth, subject only to
qualifications about the need to avoid seriously damaging the incentives to work,
save and take risks.21 Since an extra dollar is worth so much more to a poor person
than to a rich one, the case for taking from the latter and giving to the former was
hard to deny (Do you really need 50 cents more than he does?, as the twenty-first
century charity advertisements would ask, under photographs of young African
famine victims). Even an old-fashioned liberal like A. C. Pigou was driven a long
way towards socialism by the force of this argument (Pigou 1937, chapter 2).
One way in which a utilitarian could escape from egalitarianism was by
means of a dogmatic denial of the very possibility that the utility levels of different
individuals could be compared in the first place. This defence of privilege required an
unconvincingly solipsistic approach to the problem of comparing the states of mind
of different individuals, but it was advocated with great energy and considerable
eloquence by Kaldors boss at the LSE, Lionel Robbins (1932, pp. 138-41), and it
caused some unhappiness among his less conservative colleagues. The Fabian
socialist Barbara Wootton, for example, made Robbins argument a centrepiece of
her bitter Lament for Economics, published in 1938. It was not just a question of rich
and poor: the inability to compare the utilities of different individuals destroyed the
economists ability to make any welfare judgements about resource allocation.
Nothing can get us past the difficulty that where one man works, and another uses
the product of his work, it is impossible to say with certainty whether or not the
product was really worth the making (Wootton 1938 pp. 14-15; original stress). It
was the inability to justify interpersonal comparisons of utility, more than any of the
other difficulties that she identified in contemporary economic theory, that led
Wootton to abandon the discipline in favour of sociology, which seemed to offer a
more coherent approach to social policy-making (King 2004). Hers was an extreme
case, but Wootton was by no means the only left-leaning economist to be worried by
Robbins position. Roy Harrod, though dismissive of her jeremiad (Harrod 1938, p.
384), agreed with her on this point: If the incomparability of utility to different
individuals is strictly pressed, not only are the prescriptions of the welfare school
[Pigou and other advocates of cardinal utility] ruled out, but all prescriptions
whatever. The economist as an adviser is stultified (ibid., p. 397).
Half a century later John Harsanyi dismissed Robbins position as
essentially fallacious, except where it was applied to artistic and cultural
judgements:
It seems to me that economists and philosophers influenced by logical
positivism have greatly exaggerated the difficulties that we face in
making interpersonal utility comparisons with respect to the utilities
and disutilities that people derive from ordinary commodities and, more
generally, from the ordinary pleasures and calamities of human life.
(Harsanyi 1987, p. 957)22

46 History of Economics Review

_______________________________________________________________________________

Those who agreed with Robbins, however, needed an alternative foundation for
their policy prescriptions. A solution was, in fact, at hand. In 1894 and 1896-7
Vilfredo Pareto, writing in Italian and French (respectively), had defined an optimal
allocation of resources as one in which it was not possible to make any one
individual better off without simultaneously making one or more other individuals
worse off.23 This principle of Pareto optimality (to use modern terminology) did
not require interpersonal comparisons of utility or, indeed, any notion of cardinal
utility at all; it relied upon peoples ordinal judgements as to whether a change from
one state of the world to another would make them better off or worse off, or leave
them indifferent. A Pareto improvement is thus a change that makes one or more
individuals better off without making someone else worse off. Again, in 2007 all
this is known to every intermediate student of microeconomics; in 1939, however,
Paretos work was yet to be translated into English, though it was available in
French (and, of course, in Italian).
Enter Nicholas Kaldor, by now a Fabian socialist, who may not have read
Woottons Lament but was certainly familiar with the arguments.24 He is in entire
agreement with Robbins, Kaldor confirms, that interpersonal comparisons of
utility are in principle impossible, but this does not entail that economics as a
science can say nothing about policy. Take a classic example: the repeal of the
Corn Laws, which benefited capitalist and workers at the expense of the
landlords.25 This was evidently not a Pareto improvement, but it was a good thing
nonetheless, since the British government could have compensated the landlords for
their loss of income while leaving everyone else better off than before. Kaldors
statement of the principle involved, and its implications for the assessment of
economic policy, is so lucid that I shall cite it at some length:
In this way, everybody is left as well off as before in his capacity as an
income recipient; while everybody is better off than before in his
capacity as a consumer. For there remains the benefit of lower corn
prices as a result of the repeal of the duty.
In all cases, therefore, where a certain policy leads to an increase in
physical productivity, and thus of aggregate real income, the
economists case for the policy is quite unaffected by the question of
the comparability of individual satisfactions; since in all cases it is
possible to make everybody better off than before, or at any rate to
make some people better off without making anybody worse off. There
is no need for the economist to prove as indeed he never could prove
that as a result of the adoption of a certain measure no-one in the
community is going to suffer. In order to establish his case, it is quite
sufficient for him to show that even if all those who suffer as a result
are fully compensated for their loss, the rest of the community will still
be better off than before. (Kaldor 1939d, p. 550)
It follows, Kaldor continues, that Pigou was correct to propose that welfare
economics be divided into two parts:
The first, and far the most important part, should include all those
propositions for increasing social welfare which relate to the increase in
aggregate production; all questions concerning the stimulation of
employment, the equalisation of social net products, and the
equalisation of prices with marginal costs, would fall under this
heading. Here the economist is on sure ground; the scientific status of
his prescriptions is unquestionable, provided that the basic postulate of

Nicholas Kaldor in the 1930s 47

_______________________________________________________________________________

economics, that each individual prefers more to less, a greater

satisfaction to a lesser one, is granted. In the second part, concerning
distribution, the economist should not be concerned with prescriptions
at all, but with the relative advantages of different ways of carrying out
certain political ends. For it is quite impossible to decide on economic
grounds what particular pattern of income distribution maximises social
welfare. (ibid., p. 551)
The economist is not qualified, as an economist, to make judgements about the
distribution of income, or to proclaim for or against a lesser or greater degree of
inequality. All that economics can, and should, do in this field, is to show, given
the pattern of income-distribution desired, which is the most convenient way of
bringing it about (ibid., p. 552).
Kaldor did not invent the compensation principle, which must have been
familiar to lawyers for centuries and in political discourse already had a long, if
chequered, past. In the 1790s, for example, the radical activist Thomas Spence
asserted the right of all citizens to a Basic Income as fair compensation for the
losses they had incurred as a result of the private appropriation of the land (King
and Marangos 2006). Surveying the economic literature, Chipman (1987) traces
the compensation principle back to Jules Dupuit (in 1844) and Marshall (in
1890), the latter making use in his Principles of what would later become known
as Pareto optimality. Pareto himself was the first to apply the principle to the
analysis of competitive general equilibrium, and his follower Enrico Barone
elaborated upon it in an important paper (again in Italian) on the economics of
socialism in 1908. The compensation principle was at the very least implicit in
their work. The concepts of Pareto optimality and of compensation were
rediscovered, apparently independently, by Kaldors friend (and at the time LSE
colleague) Abba Lerner in a 1934 paper on the measurement of monopoly power;
they were used four years later by Harold Hotelling (1938) to discuss the relative
merits of income taxes and excise taxes.
But, leaving Marshall aside,26 Kaldor was the first economist writing in
English to identify compensation as a general principle for judgements concerning
social welfare. His paper proved extremely influential, as he later noted with a
mixture of pride and exasperation:
This idea was subsequently taken up and developed by Professor Hicks
and, following him, the proposition was subjected to a searching
examination by a whole host of economists (including Scitovsky,
Baumol, Little, Samuelson, Arrow, Graaf, Reder, Dobb and many
others), until it became a veritable cause clbre under the flattering
title of the New Welfare Economics. On re-reading the original note
in the light of all this subsequent work (some of which, I must confess,
I found too tedious to read and some of which was plainly beyond my
comprehension), I still feel unrepentant in rejecting Professor Robbins
proposition that the impossibility of making inter-personal comparisons
of utility puts an effective bar to economics as a science saying
anything by way of prescription. (Kaldor 1960, p. 5)27
By the end of the century the compensation principle was widely used by moral
philosophers like Robert Nozick (Lacey 2001), apparently without any
recognition of Kaldors contribution. At all events, his short note must be counted
as one of the most influential three-page articles in the history of economic
theory.

48 History of Economics Review

_______________________________________________________________________________

Speculation (1939)

Kaldors article Speculation and Economic Stability, published in the Review of

Economic Studies in the month that the Second World War began, is easily the
longest, most densely written and most complex of his four seminal papers from the
1930s, and also arguably the best. It has by far the most convoluted publishing
history. I shall cite the version that Kaldor included in his 1960 collected essays
rather than the original, since that was his preferred text, but readers should be
warned that it differs significantly from the original.28
The intention of the paper is clearly expressed in the title. Kaldor defines
speculation as:
the purchase (or sale) of goods with a view to re-sale (re-purchase) at a
later date, where the motive behind such action is the expectation of a
change in the relevant price relatively to the ruling price and not a gain
accruing through their use, or any kind of transaction effected in them
or their transfer between different markets. (Kaldor 1980b, p. 17)
The term goods is slightly misleading, as it soon becomes apparent that financial
securities play a major part in Kaldors story. In fact the 1939 paper constitutes an
important if generally unheralded contribution to the theory of finance. It is
very much in the heterodox tradition,29 especially since, by 1940, Kaldor was
convinced by his critics, Christopher Dow (1940) and Ralph Hawtrey (1940), that
expectations cannot successfully be modelled in terms of a single representative
agent and that analysis of the forward market requires at the very least a distinction
between bulls and bears (Kaldor 1940d, p. 200).30 It is also a critique of the
General Theory, taking in not only Keynes chapter 17 discussion of own-rates of
interest but also extending to the theory of liquidity preference and the multiplier.
Finally, Kaldor touches on the theory of money and the analysis of monetary
policy, using arguments that are sometimes (contentiously) interpreted as
foreshadowing his post-1970 Post Keynesian views on endogenous money.
Objects of speculation, Kaldor notes, must be standardised, durable, and
valuable in proportion to bulk (1980b, p. 20). Only two classes of assets satisfy
these conditions: certain industrial raw materials, and financial assets (bonds and
shares) (ibid., p. 22). He sets out the necessary relationships between expected
price, current price and future price in a series of equations involving the rate of
interest, the yield, the carrying cost and the risk premium (ibid., pp. 23-5). This
draws heavily on Keynes treatment of these questions in his Treatise on Money,
and Kaldor was later to suggest that his own article provides a missing link or
rather, one of the missing links between Keynes ideas in the Treatise and in the
General Theory (Kaldor 1960b, p. 4). There follows a lengthy and involved
discussion of the influence of expectations on the volatility of prices, for both raw
materials and securities, and an equally long and demanding analysis of the effects
of speculation on the level of economic activity (Kaldor 1980b, pp. 31-54).
Writing in 1980, Kaldor described his article as an attempt
to generalise Keynes theory of the multiplier by demonstrating that it
results from the stabilising influence of speculative expectations on
prices which applies in all cases in which the elasticity of speculative
stocks is high [and] to show that Keynes theory of interest contains
two separate propositions. The first regards interest as the price to be
paid for parting with liquidity, and it arises on account of the
uncertainty of the future prices of non-liquid assets. The second

Nicholas Kaldor in the 1930s 49

_______________________________________________________________________________

concerns the dependence of the current rate of interest on the interest

rates expected in the future. While the first proposition provides an
explanation of why long-dated bonds should normally command a
higher yield than short-term paper, it is the second which explains why
the traditional theory of the working of the capital market was
inappropriate why, in other words, savings and investment are
brought into equality by movements in the level of incomes, far more
than by movements in interest rates. And this second effect will be the
more powerful the less is the uncertainty concerning the future, or the
greater the firmness with which the idea of a normal price is
embedded in the minds of professional speculators and dealers. (1980a,
p. xvii; original stress)
When he met Keynes at a Cambridge tea party a few weeks after the article
appeared, Kaldor was pleased to find that he had already read it, and said that
I might well be right that it is the price stabilising influence of the policies of
dealers and speculators, rather than the premium which the public requires for
parting with liquidity, which explains why an increase in the propensity to save is
not itself capable of generating more investment (ibid., p. xvii n. 2).

Figure 2

If Kaldors argument casts doubt on the theory of liquidity preference, it also

has significant implications for the operation of monetary policy. These are brought
out in section 4 of the article, which concludes that we must expect the bank rate
mechanism, as an instrument of economic policy, to become increasingly
ineffectual. While it is still available for dealing with an occasional boom, the bank
rate becomes more and more ineffective as a safeguard against the ravages of
deflation (1980b, p. 58). This passage, and the earlier discussion in an extended
footnote and the associated diagram, has often been seen as evidence that Kaldor was
already in 1939 an advocate of endogenous money. Figure 2 indicates a highly elastic
money supply curve (though not a horizontal one), and Kaldor comments that:
The elasticity of the supply of money in a modern banking system is
ensured partly by the open market operations of the central bank, partly
by the commercial banks not holding to a strict reserve ratio in the face
of fluctuations in the demand for loans, and partly it is the consequence
of the fact that under present banking practices a switch-over from
current deposits to savings deposits automatically reduces the amount
of deposit money in existence, and vice versa. (ibid., p. 39 n. 2)

50 History of Economics Review

_______________________________________________________________________________

This has something in common with both the accommodationist and the stucturalist
versions of the subsequent Post Keynesian theory of endogenous money (Pollin
1991; cf. Rochon 2000 for a contrary view).
To do full justice to Speculation and Economic Stability another full-length
paper would be required (see Sardoni 2006). As John Hicks told Kaldor, after the
publication of the 1960 version, it was the culmination of the Keynesian revolution
in theory. You ought to have had more credit for it (cited by Targetti and Thirlwall
1992, p. 4). Kaldor remained, justifiably, proud of the paper, which convincingly
analyses real-world capitalist markets where stocks dominate flows and where
expectations about the future course of prices are crucial in modelling both supply
and demand.31 But he never really followed it up, failing to develop the alternative
macroeconomic model, faithful to the spirit of Keynes but incorporating all his
criticisms, which it seemed to demand. His 1940 trade cycle model, for example,
which is described in the following section, owes absolutely nothing to his analysis of
speculation and economic stability and barely mentions money or finance.

Cycles (1940)

There is however a real sense in which Kaldor (1940c) can be said to contain the
first truly Keynesian model of the trade cycle, after an abortive first attempt two
years earlier (1938d). Keynes himself had provided only Notes on the Trade
Cycle, a chapter heading that accurately describes the contents. He attributes the
cycle very largely to fluctuations in the marginal efficiency of capital, themselves
explained by changes in the profit expectations of business people (sometimes
reinforced by the effects of sharp changes in stock market prices). The discussion is
informal and discursive throughout (Keynes 1936, chapter 22).
Kaldor sets out to formalise the argument, and in the process eliminates
much of its subtlety. In the Keynesian system the level of income is determined by
the relationship between saving and investment. If planned or intended saving and
investment differ, the equilibrium level of income will change, and so, Kaldor
suggests, it should be possible to construct a model of cyclical fluctuations in terms
of regular and systematic changes in the savings-investment relationship. Kaldors
own contribution is to recognise that cycles can be generated by requiring both saving
and investment to be non-linear functions of income, as illustrated in Figure 3
(Kaldor 1940c, Figure 5, p. 88).

Figure 3

Nicholas Kaldor in the 1930s 51

_______________________________________________________________________________

The investment function is non-linear because, at low levels of income, excess

capacity is a strong deterrent to new investment and, at high levels of income,
construction costs rise rapidly and the difficulty of borrowing increases. The saving
function is non-linear because the average propensity to save rises with income and,
at very low levels of income, saving almost disappears (ibid., pp. 81-2). There are two
ways in which this differs from contemporary multiplier-accelerator models of the
trade cycle (for example, Harrod 1936; Steindl 1937; Samuelson 1939). First, since the
savings function is non-linear the marginal propensity to save, and thus also the
multiplier, varies over the cycle. Second, investment depends not on the rate of change
of income (or consumption) but on its level. A third difference from the Harrod model
is that there is no trend; in 1940 (though not in subsequent work) Kaldor is happy to
model the cycle in the context of a static and not a growing economy.
The mechanism is quite straightforward. In Figure 3 there are two stable
equilibria (A and B) and one unstable equilibrium (C). To generate cycles, Kaldor
makes the S and I functions shift over time as a result of changes in the capital
stock and in the level of income itself. This is illustrated in Figure 4 (ibid., Figure 6,
p. 84). When activity is high, as at B (Stage I), the S function shifts upwards and
the I function starts to fall, bringing B and C closer together (Stage II). When the
two functions become tangential, B and C are identical and equilibrium becomes
unstable downwards (since S exceeds I in both directions). In the ensuing slump,
equilibrium is restored at A. The level of activity is now low (Stage III), leading the
I function to shift upwards and the S function to fall. A and C now come closer
together (Stages IV and V), gradually increasing the level of activity until A and C
are identical and the economy becomes unstable upwards (Stage VI, where a boom
occurs and equilibrium is reached at B). Thereafter the curves gradually return to
the position shown in Stage I, and the cyclical movement is repeated (ibid., p. 85).

Figure 4

52 History of Economics Review

_______________________________________________________________________________

There are three necessary conditions (which together are sufficient) for
recurrent cyclical fluctuations. First, the I function must be steeper than the
S function at normal levels of activity. Second, the I function must be less steep
than the S function at extreme levels of activity (that is, in both boom and slump
conditions). Third, the level of investment at the upper turning-point must be large
enough for the I function to fall, relatively to the S function; and the converse must
be true at the lower turning-point (ibid., pp. 85-6). If the first condition is not met,
the economy will be stable, and there will be no trade cycle. If the second condition
is not met, the economy will be unstable downwards (and capitalism will collapse).
Kaldor concludes that the forces acting to bring a slump to an end are less reliable
than those that end a boom: the danger of chronic stagnation is greater than the
danger of a chronic boom (ibid., p. 87). All this is set out without any mathematics;
when the model was eventually subjected to formal analysis it proved to have
serious weaknesses.32 Apart from one brief reference to the difficulty of obtaining
finance as a constraint on investment in a boom there is also no monetary or
financial component; it is as if Speculation and Economic Stability had been
written by someone else altogether.33
What are the prospects for controlling the cycle through macroeconomic
policy? Keynes expressed strong doubts concerning the effectiveness of monetary
policy, since fluctuations in the marginal efficiency of capital were likely to swamp
any feasible changes in interest rates. I conclude that the duty of ordering the current
volume of investment cannot safely be left in private hands (Keynes 1936, p. 320).
Kaldor takes this for granted, arguing that public investment should be undertaken
early in the downturn, to prevent the movement from Stage II to Stage III. Once
Stage III is reached, a sharp contraction is unavoidable, he suggests, and the volume
of public investment required to move the system from Stage IV to Stage VI is much
larger than that needed to move it from V to VI. Thus just when the depression is
at its worst the difficulty of overcoming it is at its greatest; (Kaldor 1940c, p. 88).
On balance Kaldor is pessimistic:
The chances of evening out fluctuations by anti-cyclical public
investment appear to be remote. For if the policy is successful in
preventing the downward cumulative movement, it will also succeed in
keeping the level of private investment high; and for this very reason
the forces making for a down-turn will continue to accumulate, thus
making the need for continued public investment greater. (ibid., p. 88)
There is a strong hint here of Michal Kaleckis celebrated reference to the tragedy
of investment, which is that it causes crisis because it is useful (Kalecki 1939
[1990], p. 318). In an appendix Kaldor does compare his model with Kaleckis:
The drawback of such explanations is that the existence of an
undamped cycle can be shown only as a result of a happy coincidence,
of a particular constellation of the various time-lags and parameters
assumed
Moreover, with the theories of the Tinbergen-Kalecki type, the
amplitude of the cycle depends on the size of the initial shock. Here the
amplitude is determined by endogenous factors and the assumption of
initial shocks is itself unnecessary. (Kaldor 1940c, pp. 91-2)
These features of the 1940 Kaldor model were important. They opened up an
important new area of trade cycle research, in which regular endogenous
fluctuations could be generated without the need to rely on time-lags or erratic
shocks (Matthews 1959).

Nicholas Kaldor in the 1930s 53

_______________________________________________________________________________

However, the article received little critical attention on publication,

appearing as it did only months before the end of the phoney war and the start of
the Battle of Britain. Ten years later Kaldors old friend, John Hicks, referred
briefly to it in his own book on the trade cycle, claiming in a footnote that:
Mr. Kaldors model is not based on the acceleration principle
(investment depending on changes in output); it is based on an assumed
connexion between investment and the absolute level of output, a
relation which seems to me to be much less defensible. Nevertheless, it
must be said that the Kaldor theory is better than the assumptions on
which it is formally based; for a certain amount of what is really
acceleration is allowed to go on, as it were, behind the scenes. (Hicks
1950, p. 9 n. 2)
Kaldor did not stress his objections to the accelerator in the 1940 paper, but when
he published an extended review of Hicks in 1951 it was at the centre of his
critique: The real weakness of Mr. Hicks model consists in his use of the
acceleration principle, which is a crude and highly unsuitable tool for analysis
and also an obsolete one, that an economist of Mr. Hicks subtlety should long ago
have discarded (Kaldor 1951, p. 837). The capital-output ratio is not a technical
constant, he now maintained, but an economic variable, which depends on the
financial resources of firms and on entrepreneurial expectations (ibid., pp. 83941).34 Thus a non-linear relationship between investment and income, as in the
1940 Kaldor model, is much more plausible than the constant accelerator
coefficient used by Hicks.
The second half of Kaldors review is devoted to the connection between
the trend and the cycle (ibid., pp. 841-7). He returned to this question in a later
article, noting that Keynesian macroeconomics tends to preclude anything other
than a static model of the cycle:
Indeed, the development of trade-cycle theories that followed Keynes
General Theory has proved to be positively inimical to the idea that
cycle and dynamic growth are inherently connected For it has been
repeatedly (and in my view conclusively) shown that a few simple
additions to Keynes own model of a general equilibrium of production
in the economy will produce the result that this equilibrium will take
the form, not of a simple steady rate of production in time, but of a
rhythmical movement of constant amplitude and period in other
words, a perpetual oscillation around a stationary equilibrium position.
(Kaldor 1954, p. 54)
Keynes-inspired theories of growth, on the other hand, were not able to incorporate
cyclical fluctuations with any degree of success. They neglected the crucial factor:
human attitudes to risk-taking and money-making (ibid., p. 67). The same forces
therefore which produce violent booms and slumps, Kaldor concludes, will also
tend to produce a high trend-rate of progress; though the connection between the
two is far too complex to be reducible (at present) to a simple mechanical model
(ibid., p. 70).35
This was, perhaps, a rather veiled piece of self-criticism. Kaldors final
reflections on the 1940 model were more favourable. In the introduction to his
collected Essays on Economic Stability and Growth (Kaldor 1960b, pp. 9-10), he
notes that a non-linear investment function was used by both Hicks and Richard
Goodwin (1951) in their own cycle models. Kaldor reasserts his objections to the
accelerator principle and acknowledges that Goodwin, at least, later reformulated

54 History of Economics Review

_______________________________________________________________________________

his analysis in a more acceptable fashion (Goodwin 1955). When Kaldor returned
to questions of macroeconomic instability to explain the stagflation crisis of the
1970s he developed an entirely different method of analysis from that in the 1940
paper (Kaldor 1976), and at the end of the decade he claimed of his original model
only that it had not been rendered obsolete by subsequent work (Kaldor 1980,
p. xvi). This seems an entirely reasonable assessment.

Conclusion

The outbreak of war marked the beginning of a new phase in Kaldors career. As a
British citizen he was not liable to internment but, as a recent ex-Hungarian, he
could not expect to play anything other than a menial role in the war effort. And so
he chose to remain in academia, moving to Cambridge with the residue of the
London School of Economics and turning away from theory to concentrate on
applied economics and issues of policy. The shift is already apparent in his short
but perceptive analysis of war finance, published in August 1939 in The Banker
(1939c). One can only speculate on what Kaldor might have achieved in economic
theory in the 1940s; his record in the devils decade suggests that he was capable of
great things.
For all its strength, Kaldors early work also points to a general and
continuing weakness in his work as a theorist: an unwillingness, or perhaps an
inability, to integrate his various insights into a coherent and systematic system. His
failure to integrate his ideas on money and finance with his trade cycle theory has
already been noted, and it is just one example of his inconsistency. There is some
justice in Joan Robinsons harsh assessment, made in private correspondence in
1952: The trouble with Nicky is that he never combs out his own head to make his
various ideas consistent with each other. Anything he has ever thought about is left
lying there and is apt to pop out however much some other idea has since made it
obsolete (King 1998, p. 416). Kaldor would have replied that consistency was
indeed the hobgoblin of little minds, but his failure to display it must have reduced
his influence in an economics profession that was increasingly giving priority to
rigour over relevance. In one sense, however, Kaldors grasshopper mind was a
source of strength; it allowed him to range over the whole of economics, being
original, provocative and invariably interesting.
________________________________
* Department of Economics and Finance, La Trobe University, Victoria 3086,
Australia. Email: [email protected]. I am grateful for comments from Geoff
Harcourt, Mike Howard, Michael McLure, two anonymous referees and participants
at the 2006 HETSA Conference in Ballarat. The usual disclaimer applies.

Notes
1
Biographical details can be found in Pasinetti (1986), Thirlwall (1987) and
Targetti (1992), and autobiographical material in Kaldor (1980, 1986a, 1986b).
2
On Young, see Blitch (1995) and Kaldors notes from his 1928-9 lectures at the
LSE (Blitch 1990; Sandilands 1990).
3
See Thirlwall (1987, chapter 1). The School Notes section of Economica
documents the progress of Kaldors career. In autumn 1930 he was awarded a Research
Studentship in Economics and Political Science to the value of 200 per year, plus fees.

Nicholas Kaldor in the 1930s 55

_______________________________________________________________________________

On 1 August 1931 he became Assistant in Economics and two years later he and two
colleagues were recognised as teachers of the University.
4
In later life Kaldor was a tireless writer of letters to The Times: several hundreds
of them altogether (Kaldor 1980, p. viii).
5
This was the English version of Hayeks Monetary Theory and the Trade Cycle,
translated by Kaldor and Honor Croome (Hayek 1933).
6
Kaldors review of Lederers Outline of Economic Theory was less critical, though
he rejected the German socialists theory of interest as an unsuccessful attempt to
synthesise Schumpeter and Marx (1932d, p. 481).
7
There is a deep underlying wisdom in the vagueness of Marshallian economics
which seems to escape altogether the precision of the mathematical mind (1936a, p. 97).
8
This is reported in the List of Theses in Economics and Allied Subjects in
Progress in Universities and Colleges in the British Commonwealth of Nations
published in the August 1931 issue of Economica (p. 373), with Kaldors probable date
of completion given as 1932. Twelve months later the details were unchanged. In the
final list to be published (August 1933, p. 366) the title had been broadened (to Studies
in the economic policy of Central European States since the war) and the completion
date had been deleted.
9
See especially 1932a, p. 29. But Kaldor never returned to this topic, and seems
always to have been unaware of, or uninterested in, Minskys ideas.
10 But see Cohen (2006) for a very different assessment of this article.
11 As an anonymous referee has pointed out, all this is straight Marshall.
12 For details of their first meeting, in April 1933, see King (1998, p. 412).
13 As noted by the reviewer of the first two volumes of Kaldors Collected Papers
(Dorfman 1961, p. 495). Strangely the same word (Talmudic) had been used by Joan
Robinson in correspondence 28 years earlier to describe his verbal exposition of
marginal productivity theory (King 1998, p. 412).
14 See however Targetti and Thirlwall (1989, pp. 2-3), who take a much more
favourable view of Kaldors writings in this area.
15 See Thirlwall (1987, p. 29) for details of Kaldors involvement with the Review of
Economic Studies, a journal which (as an anonymous referee has reminded me) formed
an important bridge between the young economists in Cambridge and at the LSE.
16 The term is admittedly anachronistic. Kaldor himself does not use it, instead
describing the three problems as being whether equilibrium is determinate or
indeterminate; unique or multiple; and definite or indefinite (1934a, p. 125).
17 He may also have drawn on Marshalls Pure Theory of Domestic Value
(Marshall 1930), privately printed for the author in 1879 and reprinted by the LSE in
1930 in its classical reprints series. (I owe this point to an anonymous referee.)
18 Again Kaldor himself does not use this term, but it is transparent in what he does
explicitly argue.
19 The name occurred to me in the course of an oral exposition of that theorem at
the L. S. E. seminar (Kaldor 1960, p. 4). On Robbinss celebrated graduate seminar,
see Kaldor (1986b, pp. 38-9); Galbraith (1981, p. 78); and McCormick (1992, p. 30).
20 Note that Kaldor also makes the standard undergraduate error by confusing
elasticity with slope!
21 These qualifications were later formalised by John Rawls (1971) in his maximin
principle.
22 It might be supposed that there is a substantial philosophical literature on the
possibility of making interpersonal comparisons. Interestingly enough, Harsanyi cites
no philosophical sources. In general, it may be said that moral philosophers recognise
the difficulty but refuse to take it seriously: because we often make rough and ready
comparisons between the effects of decisions on different persons, utilitarians continue

56 History of Economics Review

_______________________________________________________________________________

to suppose that interpersonal comparisons of utility are generally possible (Lyons

1992, p. 1265).
23 The relevant passage has been translated by John Chipman (1976 [1999], p. 178).
It is a remarkably turgid piece of prose; as a rhetorician, Kaldor beats Pareto hands
down. (I owe this reference to Michael McLure.)
24 During the War Kaldor worked with Wootton on William Beveridges technical
committee on the economics of full employment, but they do not seem to have been
close friends and I know of no evidence that they had met in peacetime (though as they
were both Fabian socialists living in central London the possibility cannot be ruled out).
25 As Chipman notes, this was an unfortunate example, since an optimal tariff would
have proved superior to both the pre-1846 tariff on corn imports and to free trade, as
Kaldor himself acknowledged in a later article. Thus free trade was only a second-best
solution (Chipman 1987, p. 526; cf. Kaldor 1940f). Kaldor continued to claim,
correctly, that the pre-1846 tariff had been the least advantageous of the three
alternatives (Kaldor 1960, pp. 5-6).
26 And Barone, whose paper was translated in 1935 as part of Hayeks anti-socialist
anthology Collectivist Economic Planning (Hayek 1935). The Italians contribution,
however, seems not to have struck home at the time (Chipman 1987, p. 525).
27 See Graaf (1957, pp. 84-90) for a full discussion of this literature.
28 First, a 16-page appendix on Keynes theory of the own-rates of interest was cut
from the original manuscript by the editor, Ursula Hicks, on grounds of length. This
was first published in 1960 (Kaldor 1980c). Second, Kaldor revised his analysis of the
relationship between expected price, current price and forward price in response to
criticism by Christopher Dow (1940) and Ralph Hawtrey (1940). Just to complicate
matters even further, his response to them (Kaldor 1940d) appeared in a Symposium
on the Theory of the Forward Market in the June 1940 issue of the Review,
immediately before their criticisms, which covered both his original article and his
response to theirs. Third, he made further significant changes to the text for the 1960
book edition. However, all the important issues were on paper, if not yet in print, by
1940. For this reason, and to keep the exposition reasonably uncluttered, I have decided
to treat the 1960 published text as canonical. All references to Kaldors preferred (1960)
version are to the 1980 reprint (Kaldor 1980c, pp. 17-74), except where it has been
necessary to cite the 1940 article (Kaldor 1940d).
29 On which see Stabile (2005) and Toporowski (2005). Mainstream finance theory
began much later with Markowitz (1952), at least according to the Nobel laureate
Merton Miller (1999).
30 Hawtrey complained that Kaldor still had not taken this point seriously enough
(Hawtrey 1940, p. 205).
31 I owe this important point to Geoff Harcourt.
32 The necessary and sufficient conditions specified by Kaldor are defective, and the
persistence of cycles depends crucially on parameter values, the speed of adjustment,
initial disturbances and the positions of the S and I functions (Chang and Smyth 1971).
In his reply, Kaldor acknowledged that a fourth condition was required, and had been
implicit in his own paper: movements along the I and S curves must proceed faster than
shifts in the curves due to capital accumulation. If this condition does not hold, no
Keynesian equilibrium is possible (Kaldor 1971, p. 45).
33 This was a common feature of the trade cycle models of the 1940s and early
1950s, as Hyman Minsky (1957) complained.
34 In reality, however, the rate of expansion of firms is confined by their financial
resources (quite independently of the behaviour of market rates of interest), which
means that they cannot take advantage of large investment opportunities as quickly as
of small ones (Kaldor 1951, p. 839). Kaldor links this point to Kaleckis principle of

Nicholas Kaldor in the 1930s 57

_______________________________________________________________________________

increasing risk (ibid., p. 841), and in 1951 stresses his affinities with Kalecki to a much
greater extent than he had done in 1940.
35 This is, in effect, a return to Keyness own emphasis on business expectations as
the driving force in the trade cycle (Keynes 1936, pp. 315-20).

References
Arestis, P. and Sawyer, M. 2004. Re-examining Monetary and Fiscal Policy for the
21st Century. Cheltenham: Elgar.
Arthur, W. B. 1994. Increasing returns and Path dependence in the Economy. Ann
Arbor: University of Michigan Press.
Barone, E. 1908 [1935]. The Ministry of Production in the collectivist state, in
F. von Hayek (ed.), Collectivist Economic Planning. London: Routledge &
Kegan Paul, pp. 245-90.
Blitch, C. P. 1990. Kaldors lecture notes from Allyn Youngs London School of
Economics class, 1928-29, History of Political Economy, 22(3), Fall,
pp. 443-63.
Blitch, C. P. 1995. Allyn Young: the Peripatetic Economist. Basingstoke: Macmillan.
Chamberlin, E. H. 1957. Towards a More General Theory of Value. New York:
Oxford University Press.
Chang, W. W. and Smyth, D. J. 1971. The existence and persistence of cycles in a
non-linear model: Kaldors 1940 model re-examined, Review of
Economics Studies, 38(1), January, pp. 37-44.
Chipman, J. S. 1987. Compensation principle, in J. Eatwell, M. Milgate and
P. Newman (eds), The New Palgrave: A Dictionary of Economics, Volume 1.
Basingstoke: Macmillan, pp. 524-31.
Chipman, J. S. 1976 [1999] The Paretian heritage, Revue Europene des Sciences
Sociales, 14(37), 1976, pp. 65-173, reprinted in J. C. Wood and M. McLure
(eds), Vilfredo Pareto: Critical Assessments, Volume 2. London:
Routledge, 1999, pp. 157-288.
Cockburn, C. 1973. The Devils Decade. London: Sidgwick & Jackson.
Cohen, A. V. 2006. The Kaldor/Knight controversy. Is capital a distinct and
quantifiable factor of production?, European Journal of the History of
Economic Thought, 13(1), March, pp. 141-61.
Cross, R. (ed.) 1987. Unemployment, Hysteresis, and the Natural Rate Hypothesis.
Oxford: Blackwell.
Dorfman, R. 1960. Review of N. Kaldor, Essays on Economics Stability and
Growth and Essays on Value and Distribution, Journal of Political
Economy, 69(5), October, pp. 495-7.
Dow, J. C. R. 1940. Addenda to Mr. Kaldors Note, Review of Economic Studies,
7(3), June, pp. 201-2.
Galbraith, J. K. 1981. A Life in our Times: Memoirs. Boston: Houghton Mifflin.
Goodwin, R. M. 1951. The non-linear accelerator and the persistence of business
cycles, Econometrica, 19(1), January, pp. 1-17.
Goodwin, R. M. 1955. A model of cyclical growth, in E. Lundberg (ed.), The
Business Cycle in the Post-war World. London: Macmillan, pp. 203-21.
Graaf, J. de V. 1957. Theoretical Welfare Economics. Cambridge: Cambridge
University Press.
Harcourt, G. C. 1988. Nicholas Kaldor, 12 May 1908 30 September 1986,
Economica, 55(218), May, pp. 159-70.

58 History of Economics Review

_______________________________________________________________________________

Harrod, R. F. 1936. The Trade Cycle. London: Macmillan.

Harrod, R. F. 1938. Scope and method of economics, Economic Journal 48(191),
September, pp. 383-412.
Harsanyi, J. C. 1987. Interpersonal utility comparisons, in J. Eatwell, M. Milgate
and P. Newman (eds), The New Palgrave: A Dictionary of Economics,
Volume 2. Basingstoke: Macmillan, pp. 955-7.
Hawtrey, R. 1940. Mr. Kaldor on the forward market, Review of Economic
Studies, 7(3), June, pp. 202-5.
Hayek, F. A. von. 1931. The paradox of saving, Economica, 11(32), May,
pp. 125-69 (translated by N. Kaldor and G. Tugendhat).
Hayek, F. von. 1933. Monetary Theory and the Trade Cycle, translated by
N. Kaldor and H. M. Croome. London: Cape.
Hayek, F. von (ed.) 1935. Collectivist Economic Planning: Critical Studies on the
Possibilities of Socialism. London: Routledge & Kegan Paul.
Hotelling, H. 1938. The general welfare in relation to problems of taxation and of
railway and utility rates, Econometrica, 6(2), July, pp. 242-69.
Kaldor, N. 1932a. A case against technical progress?, Economica, 12(36), May,
pp. 180-96.
Kaldor, N. 1932b. Review of C. Landauer, Planwirtschaft und Verkehrswirtschaft,
(1931), Economic Journal, 43(166), June, pp. 276-81.
Kaldor, N. 1932c. The economic situation of Austria, Harvard Business Review,
11, October, pp. 23-34.
Kaldor, N. 1932d. Review of E. Lederer, Aufriss der konomischen Theorie,
(1931), Economica, 12(38), November, pp. 479-83.
Kaldor, N. 1934a. A classificatory note on the determinateness of equilibrium,
Review of Economic Studies, 1, February, pp. 122-36.
Kaldor, N. 1934b. The equilibrium of the firm, Economic Journal, 44(173),
March, pp. 60-76.
Kaldor, N. 1934c. Mrs. Robinsons Economics of Imperfect Competition,
Economica, n.s., 1(3), August, pp. 335-41.
Kaldor, N. 1935. Market imperfection and excess capacity, Economica, n.s., 2(5),
February, pp. 33-50.
Kaldor, N. 1936a. Review of E. Schneider, Theorie der Produktion, (1934),
Economica, n.s., 3(9), February, pp. 95-9.
Kaldor, N. 1936b. Review of H. von Stackelburg, Marktform und Gleichgewicht,
(1934), Economica, n.s., 3(10), May, pp. 277-30.
Kaldor, N. 1936c. Wage subsidies as a remedy for unemployment, Journal of
Political Economy, 44(6), December, pp. 721-42.
Kaldor, N. 1937a. Limitational factors and the elasticity of substitution, Review of
Economic Studies, 4, February, pp. 162-5.
Kaldor, N. 1937b. Annual survey of economic theory: the controversy on the
theory of capital, Econometrica, 5(1), July, pp. 201-33.
Kaldor, N. 1937c. Review of P. Kadosa, Die Preisgestaltung, (1935), Journal of
Political Economy, 45(5), October, pp. 707-8.
Kaldor, N. 1937d. Prof. Pigou on money wages in relation to unemployment,
Economic Journal, 47(188), December, pp. 745-53.
Kaldor, N. 1938a. On the theory of capital: a rejoinder to Professor Knight,
Econometrica, 6(1), April, pp. 163-76.

Nicholas Kaldor in the 1930s 59

_______________________________________________________________________________

Kaldor, N. 1938b. Professor Chamberlin on monopolistic and imperfect

competition, Quarterly Journal of Economics, 52(3), May, pp. 513-29.
Kaldor, N. 1938c. Mr. Hawtrey on short and long term investment, Economica,
n.s., 5(20), November, pp. 461-7.
Kaldor, N. 1938d. Stability and full employment, Economic Journal, 48(192),
December, pp. 642-57.
Kaldor, N. 1939a. Capital intensity and the trade cycle, Economica, n.s., 6(21),
February, pp. 41-66.
Kaldor, N. 1939b. Money wage cuts in relation to unemployment: a reply to Mr.
Somers, Review of Economic Studies, 6, June, pp. 232-5.
Kaldor, N. 1939c. Principles of emergency finance, The Banker, 51(163), August,
pp. 149-56.
Kaldor, N. 1939d. Welfare propositions of economics and interpersonal
comparisons of utility, Economic Journal 49(195), September, pp. 549-52.
Kaldor, N. 1939e. Speculation and economic stability, Review of Economic
Studies 7, October, pp. 1-27.
Kaldor, N. 1939f. Review of A. W. Marget, The Theory of Prices (1938), Economic
Journal, 49(195), September, pp. 495-8.
Kaldor, N. 1940a. The trade cycle and capital intensity: a reply, Economica, n.s.,
7(25), February, pp. 16-22.
Kaldor, N. 1940b. A comment on the above [H. M. Somers, Money wage cuts in
relation to unemployment: a rejoinder to Mr. Kaldor], Review of
Economic Studies, 7, February, p. 137.
Kaldor, N. 1940c. A model of the trade cycle, Economic Journal, 50(197), March,
pp. 78-92.
Kaldor, N. 1940d. A note on the theory of the forward market, Review of
Economic Studies, 7(3), June, pp. 196-201.
Kaldor, N. 1940e. Review of M. Abramovitz, An Approach to a Price Theory in a
Changing Economy (1939), Economic Journal, 50(188-189), JuneSeptember, pp. 278-80.
Kaldor, N. 1940f. A note on tariffs and the terms of trade, Economica, n.s. 7(28),
November, pp. 377-80.
Kaldor, N. 1951. Mr. Hicks on the trade cycle, Economic Journal, 61(244),
December, pp. 832-47.
Kaldor, N. 1954. The relation of economic growth and cyclical fluctuations,
Economic Journal, 64(253), March, pp. 53-71.
Kaldor, N. 1960a. Introduction to Kaldor, Essays on Value and Distribution,
London: Duckworth, second edition, 1980, pp. 1-9.
Kaldor, N. 1960b. Introduction to Kaldor, Essays on Economic Stability and
Growth. London: Duckworth, second edition, 1980, pp. 1-15.
Kaldor, N. 1971. Comment, Review of Economic Studies, 38(1), January, pp. 45-6.
Kaldor, N. 1972. The irrelevance of equilibrium economics, Economic Journal,
82(308), December, pp. 1237-55.
Kaldor, N. 1975. What is wrong with economic theory, Quarterly Journal of
Economics, 89(3), August, pp. 347-57.
Kaldor, N. 1976. Inflation and recession in the world economy, Economic
Journal, 86(344), December, pp. 703-14.

60 History of Economics Review

_______________________________________________________________________________

Kaldor, N. 1980a. General introduction to Collected Economic Papers, in Kaldor,

Essays on Value and Distribution. London: Duckworth, second edition,
pp. vii-xxxi.
Kaldor, N. 1980b. Speculation and economic stability, in Kaldor, Essays on
Economic Stability and Growth. London: Duckworth, second edition,
pp. 17-58.
Kaldor, N. 1980c. Keynes theory of the own-rates of interest, in Kaldor, Essays
on Economic Stability and Growth. London: Duckworth, second edition,
pp. 59-74.
Kaldor, N. 1985. Economics Without Equilibrium. Cardiff: University College of
Cardiff Press.
Kaldor, N. 1986a. Recollections of an economist, Banca Nazionale del Lavoro
Quarterly Review, 39(156), March, pp. 3-26.
Kaldor, N. 1986b. Ricordi di un economista, a cura di Maria Cristina Marcuzzo.
Milan: Garzanti.
Keynes, J. M. 1936. The General Theory of Employment, Interest and Money.
London: Macmillan.
Keynes, J. M. 1987. The Collected Writings of John Maynard Keynes. Volume XIII.
The General Theory and After. Part I: Preparation. London: Macmillan
and Cambridge University Press for the Royal Economic Society.
King, J. E. 1998. Your position is thoroughly orthodox and entirely wrong:
Nicholas Kaldor and Joan Robinson, 1933-1983, Journal of the History of
Economic Thought, 20(4), November, pp. 411-32.
King, J. E. 2004. Lament for economics, or how Barbara Wootton gave it all away
and became a sociologist, Research in the History of Economic Thought
and Methodology, 22C, pp. 301-21.
King, J. E. and Marangos, J. 2006. Two arguments for basic income: Thomas
Paine (1737-1809) and Thomas Spence (1750-1814), History of Economic
Ideas, XIV/1, pp. 55-71.
Lacey, A. R. 2001. Robert Nozick. Princeton: Princeton University Press.
Lerner, A. P. 1934. The concept of monopoly and the measurement of monopoly
power, Review of Economic Studies, 1, June, pp. 157-75.
Lyons, D. 1992. Utilitarianism, in L. C. Becker and C. B. Becker (eds),
Encyclopedia of Ethics. Chicago and London: St James Press, pp. 1261-8.
Marcuzzo, M. C. and Rosselli, A. (eds) 2005. Economists in Cambridge: A Study
through their Correspondence, 1907-1946. London: Routledge.
Markowitz, H. 1952. Portfolio selection, Journal of Finance, 7(1), March, pp. 7791.
Marshall, A. 1879 [1930]. The Pure Theory of Domestic Values. London: London
School of Economics and Political Science, reprints of economic classics.
Matthews, R. C. O. 1959. The Trade Cycle. London: Nisbet and Cambridge
University Press.
McCormick, B. J. 1992. Hayek and the Keynesian Avalanche. Hemel Hempstead:
Harvester Wheatsheaf.
Miller, M. P. 1999. The history of finance, Journal of Portfolio Management,
25(4), Summer, pp. 95-102.
Minsky, H. P. 1957. Monetary systems and accelerator models, American Economic
Review, 47(6), December, pp. 860-83.

Nicholas Kaldor in the 1930s 61

_______________________________________________________________________________

Pashigan, B. P. 1987. Cobweb theorem, in J. Eatwell, M. Milgate and P. Newman

(eds), The New Palgrave: A Dictionary of Economics, Volume 1.
Basingstoke: Macmillan, pp. 463-4.
Pasinetti, L. L. 1986. Nicholas Kaldor: an appreciation, Cambridge Journal of
Economics, 10(4), December, pp. 301-3.
Pigou, A. C. 1937. Socialism versus Capitalism. London: Macmillan.
Pollin, R. 1991. Two theories of money supply endogeneity: some empirical
evidence, Journal of Post Keynesian Economics, 13(3), Spring, pp. 366-96.
Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Belknap Press of Harvard
University Press.
Robbins, L. 1932. The Nature and Significance of Economic Science. London:
Macmillan.
Robinson, J. 1969. Preface to the second edition, in Robinson, The Economics of
Imperfect Competition. London: Macmillan, second edition, pp. v-xii.
Robinson, J. 1974. History versus equilibrium, Indian Economic Journal, 21(3),
January-March, pp. 202-13.
Rochon, L.-P. 2000. 1939-1958: was Kaldor an endogenist?, Metroeconomica,
51(2), pp. 191-220.
Samuelson, P. A. 1939. Interactions between the multiplier analysis and the
principle of acceleration, Review of Economics and Statistics, 21(2), May,
pp. 75-8.
Sandilands, R. J. (ed.) 1990. Nicholas Kaldors notes on Allyn Youngs LSE
lectures, 1927-29, Journal of Economic Studies, 17(3-4), pp. 18-114.
Sardoni, C. 2006. Kaldors Monetary Thought: a Contribution to a Modern Theory
of Money, University of Rome La Sapienza, mimeo.
Setterfield, M. 1998. History versus equilibrium: Nicholas Kaldor on historical
time and economic theory, Cambridge Journal of Economics, 22(5),
September, pp. 521-37.
Stabile, D. R. 2005. Forerunners of Modern Financial Economics: A Random Walk
in the History of Economic Thought. Cheltenham: Elgar.
Steindl, J. 1937. Der Konjunkturzyklus, Zeitschrift fr Nationalkonomie, 8,
pp. 229-237.
Targetti, F. 1992. Nicholas Kaldor: the Economics and Politics of Capitalism as a
Dynamic System. Oxford; Clarendon Press.
Thirlwall, A. P. 1987. Nicholas Kaldor. Brighton: Harvester.
Thirlwall, A. P. and Targetti, F. (eds) 1989. The Essential Kaldor. London:
Duckworth.
Toporowski, J. 2005. Theories of Financial Disturbance: An Examination of
Critical Theories of Finance from Adam Smith to the Present Day.
Cheltenham: Elgar.
Wootton, B. 1938. Lament for Economics. London: Allen & Unwin.