Kalecki on imperfect competition, inflation and money

06 Sawyer 245-261

30/01/01 9:37 am

Page 245

Cambridge Journal of Economics 2001, 25, 245–261

REVIEW ARTICLE Kalecki on imperfect competition, inflation and money Malcolm Sawyer* (Reviewing: Collected Works of Michal- Kalecki: Vol. VI, Studies in Applied Economics 1927–1941, edited by Jerzy Osiatyn´ski, Oxford, Clarendon Press, 1996; Collected Works of Michal- Kalecki: Vol. VII Studies in Applied Economics 1940–1967; Miscellanea, edited by Jerzy Osiatyn´ski, Oxford: Clarendon Press, 1997) This paper reviews the final two volumes of Kalecki’s Collected Works. It first discusses, based on the papers in Vol. VI, his contributions on cartels, on Nazi Germany, the development of indices of business fluctuations and of national income accounts, and the movement of prices and costs over the business cycle. Drawing on the papers reproduced in Vol. VII, this paper continues by discussing Kalecki’s writings on rationing, inflation, imperfect competition and unemployment, money and finance, the national debt and the welfare state, international economic arrangements, the post-war American economy and econometrics and methodology. Key words: Kalecki, Imperfect competition, Inflation, Money JEL classification: B31, P16

1. Introduction These two volumes of Kalecki’s Collected Works complete the publication of his Collected Works in English, which began in 1990 (and in Polish in the late 1970s), and appeared a couple of years before the centenary of Kalecki’s birth in 1899. For the editor Jerzy Osiatyn´ski, editorship has taken up over two decades of his life, with work on them starting in 1973 and completed in early 1995. Those two decades obviously saw many changes in Poland, and the editorial comments in many of the volumes reflect the conditions under which they were written. Jerzy Osiatyn´ski himself has also experienced fluctuations of fortune during his editorship, from the prior loss of his academic career in the anti-Semitic purges of the late 1960s, which also involved Kalecki, to a period as Minister of Finance of the Third (Post-Communist) Polish Republic in 1992 and 1993. The two volumes continue the style of the first five: the almost exhaustive reproduction Manuscript received 18 May 1998; final version received 27 March 2000. Address for correspondence: Malcolm Sawyer, Leeds University Business School, The University of Leeds, Leeds LS2 9JT, UK; email [email protected] *University of Leeds. I am grateful to Philip Arestis and Jan Toporowski for comments on an earlier draft.

© Cambridge Political Economy Society 2001

06 Sawyer 245-261

246

30/01/01 9:37 am

Page 246

Malcolm Sawyer

of Kalecki’s writings, organised by themes (and within themes generally in chronological order), with extensive notes by the editor on the various versions through which the paper went and the languages in which it appeared, and the reproduction of some significant material relating to the work of Kalecki by other authors. The papers in Vol. VI, with a few exceptions, have not previously been available in English, and the majority refer to economic conditions in pre-war Poland ; in contrast, most of the papers in Vol. VII have been previously available in English. Kalecki’s Collected Works are hereafter referred to as CW followed by the appropriate volume number. Kalecki’s major innovative contributions to economic theory were made in the 1930s and largely formulated while he was in Poland (before he came to Britain via Sweden in 1936). These contributions are the role of effective demand in the determination of the level of economic activity, the major role of fluctuations in investment in the movement of effective demand, the importance of the degree of monopoly in setting the price–cost margin, and thereby the distribution of income between profits and wages. It is significant that Kalecki’s work as an economist at the Institute for the Study of Business Cycles and Prices from 1929 to 1936 concentrated on the two areas of national income accounts and industrial developments with close links between those areas of empirical work and the major themes of Kalecki’s theoretical contributions. 2. Volume VI The papers in this volume are on applied economics and were largely written in Poland prior to 1936. Apart from a few papers concerned with some technical measurement issues, many of the papers relate to business conditions in Poland and the remainder concern international markets or economic conditions in specific countries. Part 1 of Volume VI has 41 papers, with an average length of five pages, under the general heading of business analyses 1927–1935, and these are divided into 12 papers on business conditions in commodity markets, seven on the structure and operation of trusts and cartels, and 22 on international economics relations. The papers on commodity markets (including five on rubber and three on cotton) are rather ephemeral, commenting on short-run trends in the commodity markets, and are not further discussed. Part 2 contains papers under the title of ‘Indices of business fluctuations’. The papers in Part 3 relate to Kalecki’s ‘estimates of investment, consumption and social income in Poland’ made in the first half of the 1930s, which anticipate the national income statistics which Richard Stone pioneered a decade later. Many of the papers are relatively slight, though how many pieces of economic journalism would be viewed as significant 70 years after they were written ? Their main interest arises from their general focus on industrial structure and behaviour and on business fluctuations, which form the empirical background against which Kalecki was formulating his theoretical insights. The page references in the next five sections refer to CWVI. 2.1 Cartels and trusts The six papers on cartels and trusts (a seventh is an obituary of Ivan Kreuger, who at one time had control of over 75% of the world production of matches) were written in the late 1920s and reflect the extensive role of international cartels in the inter-war period. The papers suggest that the market position of these cartels comes from the pursuit of power rather than for efficiency reasons (e.g., economies of scale) and point to the ability of the cartels to purchase (from the local state) a monopoly position. The coordination of production across countries by the cartels can be seen as a clear precedent for transnational

06 Sawyer 245-261

30/01/01 9:37 am

Page 247

Kalecki on imperfect competition, inflation and money

247

corporations. While the international dimension of cartels is evident in Kalecki’s empirical work, this is not reflected in his theoretical writing on the degree of monopoly. Kalecki reported that cartelised industries accounted for 36·7% of total industrial output in 1930, government-owned monopolies (including spirits, tobacco, salt and matches) for 3·9% and non-cartelised industries for the remaining 59·4%. Five cartels (in coal mining, iron metallurgy, cotton-spinning, sugar-mills and jute-manufacturing) accounted for over twothirds of the output of the cartelised industries. The importance of cartels in Poland was by no means unique in the inter-war period, but Kalecki’s perceptions on the importance of cartels plays a rather peripheral role in his macroeconomic writings up to the mid-1930s. 2.2 Nazi Germany and the rise of Hitler Kalecki wrote a number of brief papers on the evolving situation in Germany in the early 1930s. The interest of the papers (apart from seeing how accurately Kalecki foresaw later developments in Nazi Germany), arises from there being a piece of typical Kaleckian analysis with a concern over the interplay of political forces and economic policies, the importance of aggregate demand for the level of economic activity, and a focus on the distribution of income A paper written in 1932 remarked on the support given by the ‘captains of industry’ to Hitlerism, which ‘came into being almost spontaneously: business spheres only contributed to its support and occasionally gave instructions on good manners’ (p. 162). That support for Hitlerism, Kalecki argued, was headed by heavy industry, which was subject to severe cyclical oscillations, leading to demands for protection. Under Hitlerism, ‘heavy industry would then have a privileged position. The part of Hitler’s ideology that advocates the abolition of “interest slavery” fits in beautifully with the high indebtedness of this industry; its concentration in syndicate structures in combination with the great influence the industry has with the Hitlerite leadership gives it the role of one of the main decision-making centres’ (p. 162). In another paper written in 1932, Kalecki foresaw the coming of Hitler to power in the following words: ‘If in the immediate future there are no decisive moves on the part of the German working class, the attainment by Hitler of official and decisive influence over the government of the Reich after the July elections seems certain’ (p. 180). Further, ‘it is unlikely that when in power, the Hitlerites will turn out to be very revolutionary in foreign policy; in particular, they are unlikely to decide on a war with Poland just now. However, with all the greater force will they unquestionably start fulfilling their true goal in domestic policy, i.e., to completely control and crush the German workers’ movement. Inflation will play a crucial role in this policy’ (p. 180). Inflation here refers to what could be termed reflation through the expansion of credit. This crucial role of inflation starts from the stimulus to the economy arising from the Reichsbank giving the government interest-free loans. Inflation in this sense was viewed as crucial for Hitler, to supply funds for maintaining the Hitlerite vanguard, and to increase employment. ‘It should be added here that the period before hyperinflation sets in may be relatively long in present conditions because of the considerable underemployment of capital equipment: the increase in demand caused by inflation will be easy to meet due to an increase in output, and hence at the beginning the increase in prices will not be too large’ (p. 181). The implementation of what Kalecki termed the inflationary programme would bring two benefits for the capitalists, ‘social peace’ and high profits. Problems would lie further ahead: ‘with hyperinflation a stabilisation crisis must occur, but today who would think so far ahead ! Today the salvation and strengthening of capitalism for several or a dozen or more years is a matter of primary importance’ (p. 182).

06 Sawyer 245-261

248

30/01/01 9:37 am

Page 248

Malcolm Sawyer

In papers written in 1934 and 1935, Kalecki saw the economic upswing in Germany as ‘greatly impaired by the developments in foreign trade’ (p. 192), as the trade balance worsened. He argued that the German authorities regarded a stop to the business upswing as out of the question, and devaluation was highly unlikely ‘for psychological and prestige reasons’. The use of an export premium and the suspension of capital transfers appeared more likely options to deal with the trade deficit. Kalecki argued that Hitlerism, as with other forms of autarky, would entail a reduction in living standards. He also argued (p. 194) that real wages tend to fall during the upswing of the business cycle, and this is presumably a reflection of rising unit costs. The upswing in Germany from 1932 to 1934 is reported by Kalecki as involving a 60% increase in industrial production, with the initial stimulus coming from public investment. The well-known two-way links between investment and profits whereby aggregate investment determines profits, and profitability influences future investment is reflected in Kalecki’s statement that the ‘increase in profits which occurred in 1933 as a result of financing public investment by credit caused a considerable rise in investment in 1934’ (p. 197). ‘The business upswing is always reflected in an increase in the degree of utilisation of equipment. The German example shows that this need not necessarily be associated with a commensurate rise in the standard of living of the broad masses of the population’ (p. 201). 2.3 Indices of business fluctuations Many of the papers in this section relate to a search for available statistics which can be used as indicators of broader trends: prices of semi-finished investment goods for the business cycle, sales of lime for activity in the construction industry, sales of thread as an indicator of textile business. Seasonal adjustment to unemployment figures by the Harvard method (seasonal fluctuations are assumed to be proportional to the non-seasonal volume) are criticised, and Kalecki argued that it would be more appropriate to apply the seasonal adjustment to the employment figures, from which the seasonally adjusted unemployment figures could be obtained. The final paper in this section (written jointly with Brian Tew in 1940) proposes ‘a new method of trend elimination’, which relates to ‘the case when a time-series y subject to trend is supposed at the same time to be a function of another time-series x’ (p. 340). These papers show Kalecki’s drive to devise ways around statistical impasses by using available statistics as proxies for unavailable ones. The papers reflect Kalecki’s view that ‘one of the main symptoms of the business cycle are fluctuations in fixed capital formation’ (p. 205)1 and the notion of ‘sticky prices’ engendered by imperfect competition as the process of cartelisation ‘has largely reduced or entirely eliminated the flexibility of these prices’ (p. 205). 2.4 National income accounts Empirical work on national income accounts was undertaken alongside Kalecki’s development of the theory of investment and the role of investment in the generation of the business cycle. The first paper (published in 1931) in Part 3 opens with the statement that ‘the volume of investment is one of the most important indicators in the study of business fluctuations. Investment activity generates changes in the business cycle and determines 1 A referee has indicated that a more accurate translation would be ‘one of the most fundamental symptoms of the business cycle are . . . ’ implying that such fluctuations are not just superficial manifestations, but are fundamental to the economic process.

06 Sawyer 245-261

30/01/01 9:37 am

Page 249


249

their intensity. Other areas of economic activity depend mainly on the volume of investment’ (p. 357). The papers on the measurement of investment and of consumption rehearse the steps and approximations which the authors had to make to arrive at their goal, given the scarcity of statistics. This work culminates in an estimate of ‘social income’ in 1929 for Poland, one of the first attempts at the measurement of national income. ‘The notion of income is strictly linked with that of productive activity: income is its result’ (p. 390). Gross accumulation (investment) is used rather than net, largely on the grounds that depreciation figures were not available. The measurement of income is based on market transactions: ‘it is clear that the estimates of the social income lose any real meaning with the inclusion of such components as, for example, the value of women’s work in the household, the valuation of income takes place through the market mechanism and only income that passes through the market can be evaluated objectively’ (p. 391), though the exception is made for the inclusion of own consumption of food by farmers. ‘Most public services should be treated . . . as a means of creating the conditions for the production of goods and services, i.e., as the cost of their production’, and apart from educational services were excluded from the estimates. 2.5 Prices, costs and the cycle The final paper in Part 3 considers the relationship between movements in prices, costs and industrial production in Poland during the years 1928–34 when production almost halved, falling in 1932 to 53·7% of its 1928 value, and recovering somewhat to reach 62·8% in 1934. Wage rates initially rose even though output was falling, then fell and continued to fall even when output recovered. Using a fixed weight approach, Kalecki (with Ludwik Landau) built up relevant price and cost indices on an annual basis. What is here termed an index of profit margins is calculated: this is a rather misleading term (at least in the English translation) for it appears to be a price index of value added (gross output minus raw materials and semi-finished products cf., Table 124).1 The results in Table 130 show a fairly close association between changes in costs (prime costs minus cost of raw materials) and changes in ‘profit margins’ (as defined). The authors concluded that ‘with output changes we associate not some specific direction of changes in prices and costs, but only direction of changes in their spread [this seem to mean ratio]—changes caused by business conditions, whether resulting from the mechanism of the business cycle, or taking place under the influence of external factors’ (p. 501, text in brackets added). However, a regression between ratio of price of value added to costs and output provides a high correlation: the implication of Figure 42 is that a halving of output would reduce the ratio by around 8%. Here, as elsewhere, Kalecki argued that cartels are able to maintain or even increase prices (relative to costs) in the face of a fall in demand, and this element of price rigidity has an influence on the course of the business cycle. 3. Volume VII The final volume is divided into four parts. The first part, of some 46 papers, has ‘studies of the British war economy’ written while Kalecki was at the Oxford University Institute of Statistics. At that time, the Bulletin of the Oxford University Institute of Statistics appeared every three weeks, providing one of the few non-governmental commentaries on economic . 1 Jan Toporowski has commented that in the original Polish this index is called ‘marza zarobkowa w przemys´le’, which literally translates as ‘the margin of earnings in industry’ (i.e., value added) rather than an ‘index of profit margins’.

06 Sawyer 245-261

250

30/01/01 9:37 am

Page 250

Malcolm Sawyer

policies and prospects. The second part has four papers on the post-war American economy, and the third part is a collection of often brief papers on specific economists and book reviews. The final part, rightly labelled Miscellanea, include ‘notes and papers for Hilary Minc’ (then First Deputy Prime Minister in the Polish government), calculations for ‘reinforced-concrete-constructions’ (written in the early 1930s: Kalecki had earned a living in the late 1920s by undertaking engineering calculations of reinforced concrete constructions, and in 1929 had taken out a patent, which is reproduced here), and ten mathematical papers (of which the best known to economists is the one on the Gibrat distribution). These mathematical papers are on the theory of numbers, with no indication of how Kalecki came to be interested in this topic or as to why Kalecki began writing mathematical papers when nearly 60 years old. These papers differ from the other technical papers in these two volumes, which clearly arose out of problems of economic analysis and investigation which Kalecki was seeking to resolve. The book concludes with seven annexes, 20 pages on the main dates and facts in Kalecki’s life and a bibliography of his publications (running to over 63 pages) with full details of translations and reprints. Page references hereafter refer to CWVII. The papers in Part 1 are divided into three sections, and this review will follow that division. 3.1 Rationing Kalecki wrote extensively on rationing and saw it having a positive economic role, in damping down inflationary pressures in cases where demand ran ahead of supply, and beneficial on income distribution grounds. The first five papers have the common theme of advocating general (rather than specific item) rationing with a common ceiling on expenditure. There could be some exceptions: Kalecki mentioned that, for example, travel could be directly rationed by the number of permitted trains. He argued that rationing was preferable to taxation, since the latter was ‘not much different from . . . price increases’ (p. 7) and ‘may curtail . . . expenditure in an unhelpful way by cutting, say, education and entertainment while the actual problem is the scarcity of food, clothing, etc.’ (pp. 7–8). Rationing can also be used to deal directly with the inflationary threats from high demand relative to supply. The strong egalitarian element in the general rationing scheme proposed no doubt appealed to Kalecki. 3.2 Inflation The second section is entitled ‘war financing and economic equilibrium’. While some of the papers deal with short-run prospects (e.g., one is entitled ‘the third quarter’), many provide insights into Kalecki’s analysis of some broader topics, and I focus on those. Kalecki did not present a comprehensive account of inflation, and his writings on inflation are scattered over a number of papers, some reprinted in CWI and CWII and others which appear here. In a paper written in 1941, Kalecki asked the question, ‘what is inflation ?’ He began by remarking that ‘no generally accepted definition of inflation actually exists’, whereas we would now respond that inflation is a sustained increase in prices. Kalecki’s discussion reflected some conflation of cause and effect. For example, he remarked that ‘inflation is sometimes identified with the existence of a large budget deficit’ but ‘this definition is not satisfactory’ (p. 83). He argued that budget deficits do not necessarily involve inflation and that a balanced budget is not a safeguard against inflation. He later indicated (p. 87) that a budget deficit is only inflationary in so far as it involves considerably increased

06 Sawyer 245-261

30/01/01 9:37 am

Page 251


251

expenditure, concentrated on industries whose supply cannot readily respond, with the consequent creation of bottlenecks. Any deficit generates enough savings to finance itself. However, if the deficit is financed by short-term Treasury bills sold to the banks, ‘the budget deficit is said to be financed by “credit creation” not by “genuine savings”, and it is this way of government borrowing which is held responsible for the evils of inflation’ (p. 84). Bank deposits increase in line with the increase in Treasury bills held by the banks. Kalecki would appear to assume here that the amount of money in existence (as bank deposits) corresponds to what people wish to hold. A greater stock of money does not itself lead to greater spending (though a higher level of spending would generally lead to a higher transactions demand for money), and an increase in Treasury bills simply results in the banking system having a larger and more liquid balance sheet. Kalecki considered the decision to ‘dissave’ to be the crucial one, but ‘the form in which past savings are held is of no importance’ and that ‘the obstacles in parting with a bond as compared with withdrawing a deposit seem to have been rather exaggerated’ (pp. 84–5). Thus, the stock of broad money is held for portfolio reasons and has no particular impact of spending decisions. Kalecki noted that ‘inflation is sometimes defined as a state in which a rising, effective (money) demand for goods is not met by a similar increase in supply’ (p. 85), and made the distinction between a one-off rise in the price level (which is ‘not “inflationary” in the proper sense’) and a self-generating spiral process (‘for which the term “inflation” is generally reserved’, p. 85). He gave examples of the former as a rise in prices arising from a devaluation, and price rises following an increase in wage-rates when there are unemployed resources. However, he saw inflation as arising from growing demand which cannot be met by growing supply ‘owing to scarcity of plant, labour, or raw materials’ (p. 85). The ‘characteristic of inflation . . . [is] a rise in price of consumers’ goods in relation to the relevant costs of labour and raw materials’ (p. 86). What is relevant for inflation are conditions of supply: ‘during inflation the steeply rising part of the supply curve comes into the picture’, hence unit costs and prices rise relative to wages and real wages decline. A general rise in prices comes through the rise of numerous individual prices. ‘It follows directly that to prevent “inflation in general” one must deal with “inflations” in particular groups of commodities, and this may be done only by rationing: to avoid inflation it is necessary to cut purchasing power in those sectors of the economy where it is directed on goods in short supply’ (p. 86). This was written under war-time conditions, and hence did not consider the possibilities of enhancing supply. There is a more general implication of this approach, namely that inflationary problems arise when demand runs ahead of capacity, even when there is unemployment of labour, for there may simply be inadequate capacity to support the full employment of labour. When enterprises operate with high levels of capacity, they are faced with rising costs: real wages fall, but money wages cannot catch up with prices to restore the initial real wage. In his work at the United Nations (UN) on issues of inflation, Kalecki argued that rapid price increases often occur in the upswings of the business cycle. As production starts to rise from a relatively low level, ‘the output of manufactured goods can be raised quickly because idle industrial capacity and unemployed labour are available for use. In the short run, however, the supply of raw materials and foodstuffs, especially agricultural products, is much less responsive to increases in demand. As a result, prices of primary products increase considerably. This leads to increases in the prices of finished products, the costs of living, and wages, which in turn tend to engender new price increases’. However, ‘the increases in prices and wages referred to above are not the result of the maintenance of a

06 Sawyer 245-261

252

30/01/01 9:37 am

Page 252

Malcolm Sawyer

high level of effective demand but rather a phenomenon connected with the rapid rise in this level’ (p. 573). This analysis of inflation focuses on the balance between demand and capacity, and does not presume that there is adequate capacity to sustain full employment. It stands in some contrast to the prevailing orthodoxy, which focuses attention on the labour market in the analysis of the inflationary process. Unemployment relative to the ‘natural rate of unemployment’ or the ‘non accelerating inflation rate of unemployment is seen as relevant for inflation, and little, if any, attention is paid to issues of capacity and of pricing behaviour. In contrast, Kalecki focused on the state of demand relative to productive capacity, and hence related inflation to the product market situation. In his writings on world economic conditions at the UN in the late 1940s and early 1950s, Kalecki appeared to acknowledge some form of trade-off between unemployment and inflation. ‘Kalecki was aware that the UN leadership would not accept a document expressing views that were contrary to the opinion of the ruling élites in the Western countries, and therefore he emphatically argued that the choice between the amount of employment and the rate of inflation was of a political nature; consequently the UN should not take a position on this question’ (p. 572). It is regrettable that Kalecki never provided a systematic treatment of inflation. However, a Kaleckian analysis of inflation can be pieced together which points to the role of inadequate capacity (relative to demand) as a source of inflationary pressures. 3.3 Share of wages and of profits There are a number of papers on the share of wages and the closely related issue of profit margins. These papers mainly focus on some statistical issues that raise little which is of continuing empirical or theoretical interest. 3.4 Imperfect competition and unemployment It is generally recognised that Kalecki and Keynes discovered the principle of effective demand and the key role of investment at around the same time in the early 1930s. Besides these overlaps, there is also a range of differences between the approach of Kalecki and that of Keynes, notably over the formulation of the demand for investment as in his review of Keynes (1936) (CWI, pp. 223–232). There are, however, two widely accepted views on the differences between Kalecki and Keynes which must be questioned in light of the material now available in the seven volumes of Kalecki’s Collected Works. The first is the idea that Kalecki’s explanation of the occurrence of unemployment relies on the assumption of imperfect competition in the product market, whereas Keynes’s explanation did not. It is clear that Keynes did not invoke imperfect competition in The General Theory, and that Kalecki’s analysis includes a role for the ‘degree of monopoly’. In his review of Keynes (1936), Kalecki presented a formulation of Keynes’s approach in terms of imperfect competition (described as the more general case), but did not remark on the industrial structure. Kalecki’s formulation of the principle of effective demand in 1933 (entitled ‘Three Systems’, CWI, pp. 201–19) did not specify the state of competition but did assume increasing marginal costs and price equal to marginal cost. He did, though, assume that although ‘existing unemployment does not exert any pressure on the [labour] market, we postulate that changes in unemployment cause a definite increase or fall in money wages, depending on the direction and volume of these changes’ (CWI, p. 215; emphasis in

06 Sawyer 245-261

30/01/01 9:37 am

Page 253


253

original). In another paper, also published in 1933, Kalecki ‘tacitly assumed that free competition rules unhindered in our economic system’ (CWI, p. 106) for his business cycle theory, and then considered in the last two pages ‘how cartels affect the business cycle’. Indeed, in response to criticisms of his ‘Three Systems’ paper, he acknowledged that ‘the charges of my critic regarding my assumptions about the progression of marginal costs boil down mainly to accusing me of assuming free competition. I admit committing this sin, which I share with the majority of economists who deal with the theory of money’ (CWI, p. 493). The assumption of ‘free competition’ is remarkable in light of Kalecki’s own work of the importance of cartels in Poland and elsewhere, his remarks on perfect competition as ‘a dangerous myth’ (CWII, p. 98, though this was published posthumously in 1971), and his later development of the ‘degree of monopoly’. But the assumption of ‘free competition’ is suggestive of the idea that Kalecki did not view imperfect competition as fundamental to the explanation of unemployment and of the role of aggregate demand. He did, though, see differences between a competitive system and a cartelised one: for example money ‘wage reductions do not cause any increase in production in the case of a competitive economy, in a fully cartelised system they lead, as a result of rigidity of prices, to a shrinkage of production and a rise in unemployment’ (CWI, p. 190). It is only in the second half of the 1930s when the associated assumptions of non-increasing average costs (up to capacity) and imperfect competition became a central feature of Kalecki’s analysis.

3.5 Money and finance The second widely held view is that Keynes was also a monetary economist, while Kalecki largely ignored monetary factors. Kalecki did not produce any large-scale work on money and finance, and what he did write has been largely neglected. His Selected Essays (Kalecki, 1971) does not have any essays on money, and many models which adopted the epithet Kaleckian (e.g., Dutt, 1988) are ‘real models’ in the sense of not involving money is any essential way, though the background assumption is that credit has to be created to permit expansion of expenditure. From the full range of Kalecki’s writings, some important themes on money can be discerned, which show that Kalecki had some important insights on the nature of money. I see three themes in Kalecki’s writings on money and credit which are still highly pertinent and worthy of further development. First, Kalecki saw that the expansion of expenditure and of economic activity required the creation of credit. For example, ‘the financing of additional investment is effected by the so-called creation of purchasing power. The demand for bank credits increases, and these are granted by the banks’ (Kalecki, CWI, p. 190). But banks could respond by raising interest rates and that ‘the precondition for the upswing is that the rate of interest should not increase too much in response to an increased demand for credit’ (Kalecki, CWI, p. 191). The second is the idea that the amount of money in existence depends on the demand to hold a stock of money, and any ‘excess’ money is extinguished by the repayment of loans. In an article entitled ‘The “mysteries” of the money market’ (CWVII, pp. 64–7), Kalecki argued that the amount of bank deposits held depends on the rate of interest (on deposits) and the volume of transactions. But if those variables are stable, then ‘the funds returned to the [money] market by government expenditure will be used to increase current accounts only in so far as more is needed owing to a greater volume of transactions’ (p. 65) and under those circumstances ‘however great the rise in the floating debt, current

06 Sawyer 245-261

254

30/01/01 9:37 am

Page 254

Malcolm Sawyer

accounts will not increase’ (p. 65). Kalecki concluded that ‘if the increase in output and prices is not so great as to cause the current accounts to rise by an amount equal to the funds returned to the market by government expenditure and the short-term rate of interest is stable—the rest of these funds is likely to be invested directly in short loans to the discount market or bills’ (p. 65). In a similar vein, he argued that ‘if, for instance, banks buy more bills while the total volume of business transactions is unchanged, bank deposits will rise but the short-term rate of interest must fall. If the short-term rate of interest is unchanged . . . current accounts will increase more or less proportionately with the volume of transactions’ (p. 157), and this Kalecki sought to support by taking the wage bill as a proxy for the volume of transactions. This suggests that the stock of current account deposits (narrow money) is demand determined, but that the short-term interest rate may adjust to bring about a balance between the demand for money and the stock. The third theme concerns the appropriate definition of money. In the papers in CWVII, Kalecki rarely uses the term money, and more often refers to cash by which he meant notes and current accounts. The demand to hold ‘narrow money’ is clearly a transactions demand (cf., pp. 160–1, where the stock of current account is regressed on wage bill), whereas the demand for deposit accounts is viewed as a portfolio demand. Kalecki noted quite different trends in current accounts and deposit accounts, but ‘there is no reason to expect a similar pace in the movement of current and deposit accounts, for . . . with a given short-term rate of interest current accounts increase more or less proportionately with the volume of transactions while the rise in deposit accounts depends on the rate of accumulation of liquid reserves and the attractiveness of deposit accounts as compared with other relevant assets’ (p. 162). This distinction between narrow and broad money is significant for two related reasons. First, the two types of money are held for different types of motive (transactions and portfolio respectively), and, secondly, the demand for the different types of money will respond quite differently to interest rate changes. Insofar as narrow money yields a zero rate of interest, the demand for narrow money responds to the absolute level of interest rates, whereas the demand for broad money (minus narrow money) responds to interest rate differentials with other interest-bearing assets. The approach of Kalecki suggests that different definitions of money behave in quite different ways and that narrow money (or more accurately the creation of narrow money) is the one relevant for the expansion of effective demand. 3.6 Burden of the national debt In the paper ‘The burden of the national debt’ (written in 1943) (CWVII, pp. 163–7). Kalecki’s remarks that ‘it is frequently maintained that the burden of the debt has increased very much up till now and this, together with the future rise, will make it impossible to embark upon costly reforms’ (p. 163) has considerable overlap with views which present-day politicians and others frequently mouth. But Kalecki set out to show (i) that in spite of the great increase in the amount of the national debt in the course of the war up to the present, the future “burden” of the present debt is not likely to be higher than in 1938–39; and (ii) that it is possible to devise special taxes for financing the interest on the national debt which will render its increase harmless in the sense that it will have no repercussions on output and employment; that therefore it is not necessary to consider the reduction of the budget deficit as such to be an important problem of war finance; and that a regime under which budget deficits would be used whenever necessary to maintain full employment does not involve insuperable difficulties. (p. 163)

06 Sawyer 245-261

30/01/01 9:37 am

Page 255


255

Kalecki proposed an annual capital tax levied at such a rate as to pay for the interest on the national debt incurred from a particular date, which he argued would mean that the increase in the national debt would not lead to any changes in output and employment. The essence of the argument is that the owners of capital as a whole will be receiving extra income (from the interest on the increase in the national debt) but paying an equivalent amount in capital tax, leaving their income, and hence consumption, unchanged. He asserted that fixed capital investment would not be affected, since all forms of capital are taxed and the comparative position of fixed investment remains unchanged. This is the weakest link in the argument since it assumes that fixed investment is rather akin to a portfolio decision, given the wealth to be allocated, rather than a decision dependent on profitability. Kalecki concluded that aggregate demand would remain unchanged, and that while the national debt rises, so does the tax revenue to pay the interest charges. He briefly reviewed some practical issues in the implementation of an annual capital tax, and sees income tax levied on unearned income as an alternative. Kalecki concluded that ‘if to maintain full employment . . . requires budget deficits there is no reason to be scared by the rising national debt’ (p. 167). 3.7 The welfare state and full employment The papers in the third section of Part 1 relate to post-war reconstruction. In the first, Kalecki discussed the ‘economic implications of the Beveridge Plan’, and his general approach is that the plan can be readily afforded and was anyway based on the pessimistic assumption of a 10% unemployment rate. He dismissed ‘objections to the Beveridge Plan on the ground that it will seriously impair the competitive position of British exports’ on the basis that costs and prices might rise by 1·5% (because of the increase in employers’ contributions to social insurance); such a rise in export prices ‘is of no importance’ and could anyway be offset by an appropriate devaluation. It was interesting in the light of current debates to read Kalecki noting the Beveridge Plan advocated the introduction of ‘compulsory attendance at a work or training centre after six months of unemployment’ (p. 225), which he saw as ‘a step backwards’, penalising those ‘who happen to be unemployed longer than six months by compelling them to do uncongenial work to earn their dole’. The next stage is typical Kalecki: in the face of a problem, seek a solution which involves progressive redistribution. He suggests that if the benefit to wage ratio is too high for those who would earn low wages, then raise low wages rather than reduce benefits. A later paper (pp. 238–44) discusses the ‘White Paper on Employment Policy’ (published in 1944 as Cmnd. 6527). Kalecki noted that this was the first time that an official document had acknowledged the responsibility of government for the prevention of large fluctuations in output and employment. This represented a ‘great advance on the creed that slumps are natural and even salutary which has hitherto prevailed, but it does not amount to a complete programme of full employment. For even if deep and lasting depressions are overcome by government intervention, it does not mean that unemployment will be reduced to a bare minimum’ (p. 238). The White Paper identified three ways to stabilise effective demand, but Kalecki was critical of each of them. The first was to keep the foreign balance in equilibrium, and Kalecki argued that, while there may have to be a foreign trade balance (because of the constraints on borrowing), equilibrium is not required for stabilising employment. The second was the reduction of the swings in private investment. Kalecki agreed on the inadequacies of using the rate of interest in this respect, and dismissed trying to use

06 Sawyer 245-261

256

30/01/01 9:37 am

Page 256

Malcolm Sawyer

persuasion of enterprises to smooth out their investment plans in the social interest. He believed that variations in profits tax offers one possibility and ‘a substantial contribution . . . from this “timing” of public investment’ (p. 241). The third way was the variation of social insurance contributions, where he doubted that such variations could in practice be made large enough to offset fluctuations in investment. Kalecki noted that the White Paper aimed for ‘high and stable employment’ rather than full employment and argued for the budget to be balanced over the long run. He chided the authors of the White Paper for referring to the ‘increasing burden of the debt’ (arising from the taxes to finance the interest payments), while admitting that the burden is a transfer problem. Kalecki noted that the White Paper saw no unfavourable effects if the debt increased in line with national income, yet failed to see that this would permit a budget deficit. It is perhaps surprising that Kalecki did not draw attention to the significance of the relationship between the (post-tax) rate of interest and the growth rate (of national income) for the stability of the debt to income ratio (cf., Lerner, 1943). He criticised the White Paper for its neglect of the redistribution of income through taxation and the acceptance of the prevailing distribution of income. He wrote elsewhere on the limits on attaining full employment through the stimulation of investment (Kalecki, 1945), and it is not surprising that he concluded that a programme for full employment ‘must be based either on a long-term budget deficit policy or on the redistribution of income’ (p. 244). In his paper on ‘Employment in the UK after transition’, he forecast that the transition period after the war during ‘which labour will be scarce and consequently many wartime controls will need to be maintained will be between six and seven years’ (p. 270), and in the out-turn the final rationing controls were lifted in 1953. Kalecki then produced a scenario for the post-transition period in which full employment could continue with a balanced budget (in effect Kalecki sets public expenditure at the level necessary to underpin full employment, which he found to be roughly equal to his postulated tax revenue). Kalecki did, however, foresee some ‘serious difficulties in the sphere of foreign trade, because it [the UK] will have to increase its volume of exports considerably over the pre-war level in order to obtain means for paying for the necessary imports (mainly because of the loss of foreign investments)’ (CWI, pp. 400–1). Kalecki’s estimates for the UK stand in some contrast to his estimates for the USA, and the two sets of estimates were compared in an article in the American Economic Review, 1945 (see CWI, pp. 387–401). In the American case, Kalecki foresaw the need for a significant budget deficit approaching 9% of GDP if full employment were to be secured resulting from a higher savings ratio in the USA. 3.8 International arrangements In a number of papers written during the war, Kalecki looked for ways to ensure high levels of demand to underpin full employment. In a couple of papers, a joint paper with E. F. Schumacher written in 1943 included in this volume, and another written in 1946 on multilaterialism included in CWI, Kalecki addressed the issue of avoiding a deflationary international economic environment. In the latter paper, Kalecki argued that ‘an ideal basis of multilaterialism . . . will be the maintenance throughout the world of a volume of domestic expenditure which in combination with foreign net expenditure financed by long-term loans, is adequate for securing full employment’ (CWI, p. 416). His paper with E. F. Schumacher on ‘international clearing and long-term lending’ was written as a commentary on the proposals of Keynes and of White for the post-war

06 Sawyer 245-261

30/01/01 9:37 am

Page 257


257

international financial system. They argued that there was ‘no merit in a general policy aiming at current account equilibrium for all countries’ (p. 226) but rather there was merit in what they termed unbalanced equilibrium by which they meant balance of current and capital account taken together. It would seem that they were implicitly assuming a fixed exchange rate regime. Kalecki and Schumacher saw the weakness of the British and American schemes as involving limited liquid reserves, so that ‘“disequilibrium” means that the surplus countries are draining off the reserves of deficit countries’ (p. 226). They also argued that ‘there is little reason to believe that any international machinery aiming at equilibrium as such would be able, in actual practice, to attain its end without “equilibrating downwards” ’ (p. 227). They sought a mechanism which would allow countries with surpluses to hoard them without endangering international liquidity, and that would mean that reserve funds were not limited and hence required some mechanism for the expansion of reserves. The proposals of Kalecki and Schumacher were aimed to allow any country to run an export surplus as large as they wished, which seemed to be on the basis that this would allow others to run trade deficits. Countries which needed a trade deficit for purposes of reconstruction and industrialisation should have their position safeguarded. In effect, industrialised countries should tend to run surpluses on trade account in order to lend to developing countries.They considered the possibility of the International Clearing Union only having the function of settling the outstanding balances between central banks, but not being concerned over the size of the balances to be settled. They raised the objection against this on the grounds that it could lead some countries to over-value their currency. Some countries would be in a stronger position to secure such an overvaluation (which Kalecki and Schumacher associate with a ‘high degree of internal planning’, p. 228). They concluded that ‘the uneven development of government control in the various countries might, therefore, make a system of unlimited overdraft facilities unworkable. . . . Competitive appreciations of national currencies are precisely as undesirable as competitive devaluations’ (p. 228). This consideration led them to propose a compromise solution which combined the clearing and the lending functions. The aim was to allow surplus countries to acquire whatever reserves they wished, while providing for countries needing an import surplus for purposes of reconstruction and industrialisation to do so. Further, the aim was to provide an international policy instrument to enable those countries which did not require an import surplus to maintain long-term balance in their current account. This approach involved some abandonment of the concept of equilibrium as an object of international policy. However, ‘current account equilibrium would be demanded from . . . those countries that have reached a sufficiently advanced stage of development to make this a rational policy’ (p. 229). No advanced country would be prevented from running an export surplus (and so I would read the previous sentence as saying that industrialised countries would have to avoid current account deficits rather than have to achieve current account balance). The export surpluses would be necessary to permit the non-industrialised countries to run current account deficits and to import capital. Kalecki and Schumacher believed that these objectives could be achieved through a modification of the British plan. Quotas would be established equivalent to 50% of foreign trade turnover (rather than the 75% proposed in the British plan), and there would be no fines on creditors and debtors (which had been proposed in the British plan). In addition to an International Clearing Union, there would be an International Investment Board which would decide the long-term loans to be granted to deficit countries.

06 Sawyer 245-261

258

30/01/01 9:37 am

Page 258

Malcolm Sawyer

Those deficit countries would be judged according to whether they were in process of industrialisation, readjustment etc. (‘A’ countries), or their deficit arose for some other reason (‘B’ countries), and presumably industrialised countries. ‘A’ countries would in effect be industrialising countries, and would have access to long-term loans. In effect, the loans would be paid into the country’s account, and the over-all limits on that account would apply to all countries. For ‘B’ countries, devaluation may be required, though it is recognised that the Marshall–Lerner conditions may not hold. But these proposals do not involve compelling surplus countries to adjust. However, borrowers from the International Investment Board could be directed to buy from the deficit ‘B’ countries. Kalecki and Schumacher also argued that the two separate agencies for current trade and for investment envisaged in the British and American plans should be fused into one. The key element of the Kalecki–Schumacher plan would be ‘an orderly supply of purchasing power to deficit countries through long-term lending by the International Investment Board’ (p. 231) with long-term lending separated from the ‘liquidity wishes of surplus countries’. These proposals were presented very briefly, and do not appear to have been further elaborated (apart from a prior paper by Schumacher, also published in the Bulletin of the Oxford University Institute of Statistics). The proposals raise the major question of whether there would ever be the sort of international political agreement necessary to operate such a system. But they highlight that a stable exchange rate system (whether involving fixed exchange rates, crawling pegs or a floating exchange rate system which is less volatile than the present one) requires an internationally agreed (or at least accepted) pattern of trade and lending. 3.9 The post-war American economy Four papers on the post-war American economy follow as Part 2. After a brief paper on the determinants of the cost of living, there follows one written in 1956 which compares the post-war economic situation (1955) with the pre-war one (1937): the years are deemed at comparable stages of the trade cycle. Over that 18-year period, national production had grown by 110%, personal consumption by 93%, and unemployment was down from 14·2% to 3·8%. The two broad conclusions of his analysis were: (i) a shift away from consumption (share down by six percentage points), some increase in the share of gross accumulation (investment plus export surplus), and substantially increased taxes on corporate profits, which Kalecki saw as directly financing armaments expenditure. ‘[T]he increase of the relative share of big business’s accumulation of the national product was absorbed by armaments (mainly through the tax on corporate profits and the export surplus, whose realization was also closely connected to the expenditures of the military-imperialist complex)’ (p. 281) (ii) the decline in unemployment (10·4% of the labour force) was largely matched by an increase in the armed forces and government employment (7% of the labour force). This paper illustrates Kalecki’s view on the significance of armaments expenditure for sustaining aggregate demand, and within his writings, support can be found for the notion that high levels of military expenditure underpinned the low levels of unemployment. The other two papers are more in the form of political journalism. In ‘The fascism of our times’, written in 1964, Kalecki points to the ‘fervent activity’ amongst groups such as the OAS in France, neo-Nazi groups in West Germany and supporters of Barry Goldwater in the USA whom he grouped together as ‘fascist groups’. But he noted that Goldwater-

06 Sawyer 245-261

30/01/01 9:37 am

Page 259


259

ism proclaimed a return to laissez-faire (in contrast to Nazism). He continued by arguing that ‘the majority of the ruling class does not like the idea of the fascists seizing power, but at the same time it does not wish to crush them. The fascism of our times is a dog on a leash; it can be unleashed at any time to achieve definite aims and even when on the leash serves to intimidate the potential opposition’ (p. 290). In the other paper (‘Vietnam and US big business’, published in 1967). Kalecki saw the Vietnam war as contributing to the weakening of the position of old big business, mainly located on the East Coast relative to the new business of the West and South. He saw the former group as ‘not being enthusiastic’ about the war in contrast with the latter group, and argued that the outcome of the war (and indeed ‘the fate of all mankind’) depended on ‘the fight between two competing groups within American big business’ (p. 297). 3.10 Econometrics and methodology Part 3 is entitled ‘On Political Economy and Economists’. It contains a number of very short (e.g., a review of Paul Baran’s Political Economy of Growth (1957) in just over a page, less than a page note on Joan Robinson), and a lengthy piece on Edward Lipi´nski, which the editor indicates was largely written by the co-author Tadeusz Kowalik. Most of these papers and those in the Miscellanea section contain little of any lasting interest, and are not further discussed. The Annexes reproduce some interchanges between Brian Reddaway and Kalecki over techniques of rationing, a paper on rationing of consumption and the problem of the black market in France. Annexe 5 reproduces some evaluations by mathematicians of Kalecki’s mathematical papers. Annexe 6, by the editor, discusses Kalecki’s work for the UN, and the conditions under which Kalecki resigned from the UN. Although there is a mention of the political atmosphere at the UN during the McCarthy period, and there are hints that this caused difficulties for Kalecki, this does not add anything substantial to what was already known, and tells us nothing of the ways in which the McCarthyite pressures were exerted. Annexe 7 is an extract from a UN document on inflationary and deflationary tendencies, largely drafted by Kalecki. The first two papers in Part 3 are of more interest, though each of them is characteristically laconic, and one wishes that the arguments had been fleshed out. In ‘Econometric model and historical materialism’, Kalecki in effect considered that econometric models are useful, provided that developments in productive relations and in the superstructure can be treated as constant. ‘In a more general case these functional relations [in the econometric model] alter under the impact of events in three other spheres of the system [those of natural resources, productive relations and the superstructure] and the economic development is then a much more complicated process than that presented by an econometric model as it reflects the evolution of the society in all the aspects’ (p. 301). Econometric models can then be used to capture surface relationships, but those surface relationships may well change as the superstructure changes. Further, and as an example, economic development causes changes in the productive relations which in turn affect the pattern of economic development. Kalecki concluded that econometric models can only be drawn upon when the two conditions of ‘no autonomous changes in the spheres [as defined above] other than strictly economic conditions or if any they do not affect significantly the pattern of economic development’ and ‘there is no significant feedback effect involved in the impact of economic development upon the other spheres of the system’ (p. 305). The clear implication of Kalecki’s arguments is that econometric models which assume constant parameter values are likely to be of rather limited validity, and only under the stringent conditions mentioned at the end of the previous paragraph.

06 Sawyer 245-261

260

30/01/01 9:37 am

Page 260

Malcolm Sawyer

Kalecki also raised a consideration which has more recently become associated with chaos theory, namely ‘whether the small random changes in the parameters lead to corresponding small changes in the economic variables in question or whether the effect is disproportionately large’ (p. 303). A paper of less than four pages entitled ‘Why economics is not an exact science?’ follows, which was Kalecki’s address on receipt of an honorary doctorate from Warsaw University. He concluded by stating that he saw his own role as having been to try to ‘overcome the imprecisions of political economy’ (p. 311). He noted that ‘in its most general aspects economics bears some resemblance to theoretical physics’ (p. 308), but lags far behind theoretical physics as a science. Kalecki did not say what he meant by science, though he did indicate that both economics and physics are ‘quantitative disciplines which, on the basis of general premises derived from the knowledge of the real phenomena, develop a deductive system which is then confronted with the external world’ (p. 308). Kalecki suggested that the difficulties of conducting experiments in economics has been a major obstacle for the development of economic analysis. But there are more significant sociological factors involved, which in the case of the survival of Say’s Law arose for two reasons: ‘the class interests of the capitalists and the apparent corroboration of the law by the experience of the individual’ (p. 309), namely that their income determines their expenditure without recognising that it is the expenditure of others which determines their income. He argued that Say’s Law ‘has been supplemented by a variety of contraptions resembling the devices used to prop up the Ptolomaic system’. The Ptolomaic doctrine had survived through its close connection with the feudal system and apparent agreement with everyday experience; similarly, Say’s Law ‘was a dogma buttressing the foundations of capitalism’ and bolstered by individual experience. A brief consideration of the theory of growth under socialism led Kalecki to the view that ‘we are still far from mastering even its most essential issues. Moreover, in our discussions we often fail, putting it mildly, to meet the highest standards of precision in formulation and argument’ (p. 311). Kalecki’s answer to the question of why economics is not an exact science appears to be a combination of the difficulties of conducting experiments, the role of ideological factors and a lack of precise and accurate formalisation. 4. Conclusions The final entry in the summary of Kalecki’s life states that ‘according to information received by Mrs Kalecki soon after his death, he was that year [1970] the leading candidate for the Nobel prize in economics’ (p. 605), which in the event was awarded to Paul Samuelson. One may speculate, given the subsequent track record of the Nobel prize committee, whether a heterodox economist would have stood any realistic chance of securing the Nobel prize, and if Kalecki had lived and been awarded it, whether the subsequent awards of the prize would have been more favourable to heterodox economists. There can though be little doubt that the contributions of Kalecki (and others such as Nicholas Kaldor and Joan Robinson) to economic analysis are far more important and long lasting than those of many who have recently been awarded the Nobel prize for economics. The papers in the earlier volumes of the Collected Works contain the original contributions of Kalecki which would have formed the basis of the case for the award of the Nobel prize. The papers in the two volumes under review can help us appreciate the background against which Kalecki was developing his ideas and to see him ‘in action’ as

06 Sawyer 245-261

30/01/01 9:37 am

Page 261


261

an applied economist, though many of the papers are rather ephemeral in nature. The interaction between Kalecki’s applied work and his theoretical work is clear from these papers. Many indicate the extent to which his theoretical ideas (especially on the role of investment and aggregate demand) had infused his applied work, prior to their formal appearance. The papers also show that Kalecki had interesting analyses of inflation and of money. Bibliography Baran, P. 1957. The Political Economy of Growth, New York, Monthly Review Press Dutt, A.K. 1988. Competition, monopoly power and the prices of production, Thames Papers in Political Economy, Autumn [reprinted in Arestis, P. and Kitromilides, Y. (eds), Theory and Policy in Political Economy, Aldershot: Edward Elgar, 1989] Kalecki, M. 1943. Political aspects of full employment, Political Quarterly, vol. 14, 322–31 Kalecki, M. 1945. Full employment by stimulating private investment? Oxford Economic Papers, no. 7 Kalecki, M. 1971, Selected Essays on the Dynamics of the Capitalist Economy, Cambridge, Cambridge University Press Kalecki, M. 1990. Capitalism: Business Cycles and Full Employment Collected Works, Vol. I, edited by Osiatyn´ski, J., Oxford, Clarendon Press Kalecki, M. 1991. Capitalism: Economic Dynamics, Collected Works, Vol. II, edited by Osiatyn´ski, J. Oxford, Clarendon Press Kalecki, M. 1996. Studies in Applied Economics 1927–1941, Collected Works, Vol. VI, edited by Osiatyn´ski, J., Oxford, Clarendon Press Kalecki, M. 1997. Studies in Applied Economics 1940–1967; Miscellanea, Collected Works, Vol. VII, edited by Osiatyn´ski, J., Oxford, Clarendon Press Keynes, J. M. 1936. The General Theory of Employment, Money and Interest, London, Macmillan Lerner, A. 1943. Functional finance and the Federal debt, Social Research, vol. 10, 38–51 [reprinted in Mueller, W. (ed.), Readings in Macroeconomics, pp. 353–360]