Money, interest and capital accumulationin Karl Marx's economics: a monetary interpretation and some similaritiesto post-Keynesian approaches *

* This paper has evolved from a presentation at the session on ‘Marxian Economics’ at the 6th annual conference of the European Society for the History of Economic Thought (ESHET), 14 – 17 March 2002, Rethymnon, Crete, Greece. I would like to thank my discussant Costas Lapavitsas for helpful comments. The paper has also benefited from comments by Claudio Sardoni on an early version and from comments and suggestions by two anonymous referees. Many thanks also to them. Remaining errors are of course mine.

Abstract

Starting from Schumpeter's important distinction between ‘real analysis’ and ‘monetary analysis’, in this paper it is shown that major elements of Marx's economic theory fall in the camp of ‘monetary analysis’. This is true for Marx's theory of value, his rejection of Ricardo's interpretation of Say's Law, his treatment of the realization problem in the schemes of reproduction and his theories of credit and the rate of interest. The implications of this monetary interpretation for Marx's theory of distribution and growth display broad similarities to a monetary extension of a Kaleckian version of the post-Keynesian model, in which the equilibrium growth path is determined by the interaction between monetary and real variables.

Keywords:

• Money
• interest rate
• distribution
• capital accumulation
• Marx's economics
• post-Keynesian economics

Notes

* This paper has evolved from a presentation at the session on ‘Marxian Economics’ at the 6th annual conference of the European Society for the History of Economic Thought (ESHET), 14 – 17 March 2002, Rethymnon, Crete, Greece. I would like to thank my discussant Costas Lapavitsas for helpful comments. The paper has also benefited from comments by Claudio Sardoni on an early version and from comments and suggestions by two anonymous referees. Many thanks also to them. Remaining errors are of course mine.

1 See Shaikh (1978a, 1983a) for surveys on Marxian crisis theories. The early attempts by Fritsch (1968) and De Brunhoff (1976) to reconstruct Marx's theory of money and credit remained without major consequences for Marxian theories of accumulation and crisis.

2 See Marglin (1984) and Amadeo (1986a).

3 See Fan-Hung (1939), Alexander (1940), Kenway (1980), Dillard (1984), Foley (1986a), Sardoni (1986, 1987), Rogers (1989) and Rotheim (1991).

4 See Williams (1992, 2000), Lapavitsas (1994, 1997, 2000a,2000b), Matthews (1996), Evans (1997), Graziani (1997), Sardoni (1997a,b) Mollo (1999), Lapavitsas and Saad-Filho (2000) and Argitis (2001).

5 See Cottrell (1994) and Hewitson (1995) for surveys of post-Keynesian monetary theory and Kaldor (1970, 1982, 1985), Lavoie (1984, 1992: 149 – 216, 1996, 1999) and Moore (1989) for elaborations of the ‘horizontalist’ view in post-Keynesian monetary theory.

6 There is, however, no agreement on this point of view in Marxian economics. De Brunhoff (1976), Weeks (1981), Foley (1983, 1986b), Crotty (1987) and Evans (1997) hold that Marx's commodity theory of money is a correct, albeit historically restricted, starting point for the analysis of money and therefore not applicable to modern economies. Lapavitsas (2000a) and Lapavitsas and Saad-Filho (2000) do not only consider commodity money to be an appropriate starting point for Marx's theory of money and credit but also argue that anchoring the monetary system on a money commodity would stabilize capitalist reproduction also in modern times. For a more extensive treatment of some aspects of the Marxian debates see Hein (1997: 34 – 42).

7 In their ‘new solution’ to the transformation problem Foley (1982) and Lipietz (1982b) define the ‘value of money’ – also for money token – as the ratio of the expenditure of direct and indirect labour power to nominal value added: ‘The value of money expresses the social equivalence of money and labour time which is inherent in commodity production, and would be meaningfully defined even if money were an abstract unit of account’ (Foley 1982: 39). This concept, however, does not yet contain a determination of the price level.

8 In Marx's commodity money system, however, there seems to arise an effect of the quantity of money on the level of prices as soon as the money commodity in circulation is replaced by paper money (Marx 1867: 125 – 30). An increase in the supply of paper money should increase the level of prices measured in units of paper money, because a unit of paper money now represents less units of the money commodity in circulation. This quantity theory relation, however, can only be sustained, if the representative of the money commodity is used only for circulation purposes. But this need not be the case. According to Marx (1867: 130), the role of ‘money as money’, which includes the function of money as a store of value (hoard) may also be assumed by the money representative. Hence, there need not be a strict relationship between the quantity of paper money and the price level, because the amount of paper money in hoards may be variable and the quantity theory collapses.

9 Ricardo's version of Say's law differs from the neoclassical version because it is neither associated with full employment of labour nor is there an economic mechanism equating savings and investment. Ricardo's version of Say's law simply implies that savings and investment are identical (Garegnani 1978, 1979).

10 On Marx's rejection of Say's law in the formulation of Ricardo and the ‘possibility theory of crisis’ as opposed to a theory of the actual crisis, see more explicitly Kenway (1980), Sardoni (1987: 26 – 36, 1997b) and Hein (1997: 51 – 9).

11 See Sowell (1972) for an overview of the ‘general glut’-controversy.

12 The role of credit in economic crisis is explored in more detail by Marx in Capital, vol. III (Marx 1894: 476 – 519), where he shows that the credit system may exacerbate economic crisis. As Crotty (1986), Arnon (1994) and Pollin (1994) have observed, Marx's theory displays broad similarities to Minsky's ‘financial instability hypothesis’ (Minsky 1975, 1977). There is, however, a major difference between Marx's and Minsky's theories, because Minsky sees economic crisis caused in the financial sector, whereas Marx views economic crisis to be only exacerbated by financial relations. On Marx's and post-Keynesian financial crisis theories, see the more extensive treatment in Hein (1997: 252 – 76).

13 For a more extensive treatment of the schemes of reproduction see Kenway (1987) and Hein (1997: 136 – 55).

14 See Sweezy (1942) and Hein (1997: 148 – 55) for an extensive discussion.

15 See also De Brunhoff (1976: 60 – 72) and Foley (1986b: 86 – 9). Mollo (1999) and Lapavitsas (2000a), however, do not seem to be aware that hoarding and dis-hoarding of a money commodity can only play a limited role for money endogeneity in a growing economy.

16 We therefore disagree with Lapavitsas (1997, 2000b) and Lapavitsas and Saad-Filho (2000), who consider the credit system in Marx's theory to be mainly a mech- anism for the internal reallocation of idle funds among industrial and commercial capitalists.

17 In this respect, Marx agrees with the banking point of view in the ‘banking-currency-controversy’ (Lapavitsas 1994, Mollo 1999).

18 See also Pivetti (1987), Hein (1997: 63 – 9) and Argitis (2001) on the relation between profits of enterprise and interest in Marx's theory.

19 This objection can also be applied to Panico's (1980) attempt to reformulate Marx's approach in a production price model.

20 See Panico (1980, 1988), Pivetti (1987) and Argitis (2001) for similar results with regard to Marx's theory of the rate of interest and Sardoni (1997a) for some ambiguities in Marx's reasoning.

21 The compatibility of Marx's theory of interest, money and credit with the post-Keynesian ‘horizontalist’ approach does, however, not imply that Marx's approach agrees with post-Keynesian views on the origin and the role of money (Hein 2004b). In Marx's theory, money and credit are not considered to be means to overcome uncertainty as a historical characteristics of human life as such, as in Davidson (1994). Marx rather considers money and credit to be historically specific social institutions mediating the social division of labour, which itself cause uncertainty and instability as main characteristics of capitalist reproduction (see also Lapavitsas and Saad-Filho 2000).

22 It should be noted that the position sketched here differs from those post-Keynesian views that assume that a decreasing liquidity position of commercial banks and rising lender's and borrower's risk finally lead to rising interest rates when the quantity of credit is expanding in the accumulation process (Minsky 1986, Wray 1990, Palley 1996, Rousseas 1998). If an accommodating policy of the central bank is supposed, however, there will be no decreasing liquidity position of commercial banks when credit is expanding. If we further suppose that commercial banks only supply credit to creditworthy borrowers, there will also be no increasing borrower's or lender's risk when credit is increasing. For the economic system as a whole, increasing credit means increasing expenditures and hence increasing revenues from which credit can be repaid. There is, therefore, good reason to assume that the interest rate is the exogenous variable of the accumulation process and that the quantities of money and credit are endogenous variables. If interest rates are rising when the quantity of credit is expanding, this is due to restrictive monetary policies chosen by the central bank (Lavoie 1996).

23 See Lavoie (1995) for a comprehensive survey on the effects of the introduction of a monetary interest rate into the different variants of post-Keynesian theories of distribution and growth.

24 There may be, of course, feedbacks from accumulation to distribution in those models. See Marglin (1984) and Amadeo (1986a) for more specific formulations.

25 For the short run version of the profit-squeeze approach explaining trade cycles, see the seminal paper by Goodwin (1967), for the long run version explaining economic stagnation see Glyn and Sutcliffe (1972) and the more recent work in the ‘Social-Structure-of-Accumulation’ approach, i.e. Gordon et al. (1987). For the ‘falling-rate-of-profit-due-to-rising-organic-composition-of-capital’ theories, see Shaikh (1978a,1978b, 1983b, 1987) and Catephores (1989: 166 – 87). For a critique of the necessity of a falling rate of profit due to technical change in a model with prices of production instead of labour values see Van Parijs (1980).

26 See Marglin (1984) and Amadeo (1986a) for comparisons of the orthodox Marxian model with post-Keynesian models. For a more detailed discussion of post-Keynesian models of growth and distribution, see Lavoie (1992: 282 – 347) and Hein (2004a: 133 – 256).

27 Whereas Robinson (1962) only assumes full utilization of the capital stock on the equilibrium growth path, Kaldor (1957, 1961) also assumes full employment of labour.

28 To achieve this, the propensity to save out of profits of course has to exceed the propensity to save out of wages.

29 For a survey of monetary extensions of post-Keynesian theories of growth and distribution in the tradition of Kaldor and Robinson, see Lavoie (1995).

30 Dutt's (1989) ‘Marxian/post-Keynesian-model’ has the difference between the rates of profit and the rate of interest in the desired accumulation function and assumes actual accumulation to be determined by savings of capitalists and rentiers. The model assumes Say's law to hold and firms to produce at full capacity. It is, hence, not an appropriate starting point for a Marxian ‘monetary theory of accumulation’.

31 Bhaduri and Marglin (1990) derived different regimes of accumulation in a non-monetary Kaleckian aggregate demand model, in which investment is affected by effective demand and by costs of production.

32 See also Hein (1999) for such a model.

33 For a preliminary attempt towards such an analysis based on the present model, see Hein and Ochsen (2003).