Hayek and the General Theory∗

∗ I thank the two anonymous referees for useful suggestions and criticisms. The usual disclaimer applies.

Abstract

Hayek did not review the General Theory, but he criticized it in Profits, Interest and Investment (1939) and in part IV of The Pure Theory of Capital (1941). First, he showed that only exceptionally does greater consumption favour investment and employment. Second, he rejected Keynes's liquidity preference and maintained that only in an ‘extreme case’ might it be said that Keynes's theory of the rate of interest is valid. Although he correctly identified the gist of Keynes's theoretical innovation, his criticisms were already implicitly answered in the General Theory.

Keywords:

• Hayek
• ‘General Theory’
• Keynes
• liquidity preference
• consumption
• investment

Notes

∗ I thank the two anonymous referees for useful suggestions and criticisms. The usual disclaimer applies.

1 Hayek 1995: 121 – 225. There are several accounts of this debate; see for example Dostaler 1990, 1991, Klausinger 1991, Cochran and Glahe 1994, Caldwell 1995: 21 – 31, Kurz 1995.

2 Hayek 1976: 219, 1995: 59 – 61, 240 – 1, 251 – 2, McCormick 1992: 170 – 1, Caldwell 1995: 40 – 5, 1998; Howson 2001.

3 As is shown in section 3 a brief but significant correspondence on analytical matters took place between 20 September and 20 October 1939, but it was not accompanied by a public debate.

4 Moss and Vaughn 1986: 557, McCormick 1992: 170, Cochran and Glahe 1994: 69 note 1, Burczak 2001: 61. Zappia 2001: 350 mentions this criticism.

5 Hayek 1929: 139 – 41, 187 – 8, 196 – 7. Hudson (1988: 181 – 3) stresses this role of money in Hayek's theory of the trade cycle: for Hayek money is ‘the causal factor which is imminent to the system itself’, whereas entrepreneurial expectations are the exogenous factor.

6 Hagemann and Trautwein (1998: 302, 306) observe that exogenous monetary changes take on importance in Hayek's later writings and they show that this ‘indicates a fundamental problem of construction in Hayek's attempt to wed the business cycle with general equilibrium theory in such a way that the long-run neutrality of money is preserved’.

7 This topic cannot be discussed here. Hayek keeps these concepts in 1932a, but he questions them in 1935a (mainly pp. 123 – 8) and in 1935b: 152. Keynes's criticism is focused on these concepts first in the 1932 correspondence (Hayek 1995: 164 – 73) and then in the General Theory (Keynes 1936: 60, 79 – 81).

8 The ‘Ricardo Effect’ asserts that ‘in conditions of full employment an increase in the demand for consumers’ goods will produce a decrease of investment, and vice versa’ (Hayek 1969: 274), i.e. there is a tendency of using less labour-saving methods of production as a consequence of the increase of consumers' goods prices and of a fall in real wages, and vice versa. As Moss and Vaughn 1986 and Birner 1999 show, Hayek explicitly introduces this mechanism in 1934/35. Hagemann and Trautwein (1998: 381 n.5) note that already in a footnote of the German edition of Hayek 1931 the cyclical changes in the production periods are compared to playing a concertina, and that in an appendix of the second English edition Hayek connects his theory of the trade cycle with Ricardo's doctrine of the conversion of circulating into fixed capital. For a synthetic exposition see Hayek 1942, 1969, Hagemann and Trautwein 1998. For a discussion of the Ricardo Effect in the context of the contemporary loanable funds theories see Maclachlan 1993: 146 – 54.

9 Hayek deals with this topic already in 1932b: 195, 1932c: 140 – 2, 1935a: 125 – 9, 1935b: 143.

10 Hicks 1967b criticizes Hayek's model in which the process is activated by a credit expansion, because, for it to work, some sort of assumption is necessary to justify a large lag of consumption behind wages. Hayek (1969: 277 – 82) replies to these criticisms. Hicks adds that Hayek's model is valid if it is interpreted as ‘an analysis – a very interesting analysis – of the adjustment of an economy to changes in the rate of genuine saving’ (210). Chapter XXV of The Pure Theory of Capital perfectly meets these requirements.

11 Here lies the fundamental difference with regard to the case of an increase in the demand for consumer goods. Because of this difference there is no symmetry between the two cases.

12 Hayek 1939: 5 – 6 italics added. See also pp. 3 and 42 note 1. Hayek indicates this as the ‘main difference’ between the 1939 work and older versions of his cycle theory. The other difference is that the analysis of the effects generated by an increase in the demand of consumers' goods on investment and employment is done by considering variations of real variables, such as the rate of profit and the real wage. On these analytical novelties see Haberler 1946: 481 – 91.

13 Hayek 1939: 35 – 7. In the meantime the prices of consumer goods decrease, so that the real wages will start rising, thereby setting the conditions for the upward movement to start again.

14 See the letter to J. Robinson, dated 6 March 1941 (Ingrao and Ranchetti, 2005: 403 – 4).

15 See, for instance, Hayek 1995: 247 – 8: ‘Paradoxical as this may sound, [Keynes] was neither a highly trained economist nor even centrally concerned with the development of economics as a science. In the last resort he did not even think much of economics as a science, tending to regard his superior capacity for providing theoretical justifications as a legitimate tool for persuading the public to pursue the policies which his intuition told him were required at the moment’.

16 One of the reasons McCormick (1992: 185) mentions for Hayek not having reviewed the General Theory is ‘the belief that investment depended on final demand whereas Keynes emphasized the importance of expectations’.

17 This correspondence consists of three letters by Keynes with as many replies by Hayek and it takes place between 20 September and 20 October 1939 (Ingrao 2005: 241 – 3).

18 On this aspect of the 1931/32 controversy between Hayek and Keynes see Tieben 1997: 110 – 21.

19 See Keynes's letter of 20 September 1939 and Hayek's answer of 24 September 1939 (Ingrao 2005: 241 – 2).

20 See Keynes's letter of 20 September 1939 (Ingrao 2005: 241 – 2).

21 See Keynes's letter of 16 October 1939 and Hayek's answer of 20 October 1939 (Ingrao 2005: 243).

22 Hayek's statement that Keynes ignore the changes in relative prices following changes in consumer goods demand is thus not correct.

23 Also Haberler (1946: 488 – 90) criticizes Hayek's confidence in the possibility of rapid changes in the relative utilization of capital and labour: ‘it seems to the present writer that professor Hayek tremendously overestimates the short-run possibility of substituting labour for capital and vice versa in response to changes in the rate of profit (or the rate of interest) [ … ] It would seem that factors other than changes in the profit rate are much more important in determining the volume of investment’.

24 Moss and Vaughn 1986: 358 – 9 justify this behaviour by noting that Hayek imagines a firm that is construed ‘as a “portfolio of investment projects”’ and an entrepreneur who manages ‘a number of on-line investment projects’. Entrepreneurs ‘calculate the rate of return on each separate project and then allocate money capital among the projects so as to equalize returns at the margin’.

25 Hayek (1941: 390) defines this function ‘the supply of investible funds at increasing rates of interest due to the release of money from idle balances’.

26 The function is vertical if the desire to hold money is perfectly inelastic, i.e. cash balances are rigidly fixed and independent of the money rate of interest; horizontal if the desire to hold money is perfectly elastic, i.e. enough cash balances will be released to keep the money rate of interest always constant.

27 Hayek (1941: 364) considers indifferently the expected rate of return or the productivity of investment, (or rate of profit). This stems from his assumption that the expectations on future yields depend on the behaviour of current yields.

28 In other words, the closer the function is to being vertical: see note 26.

29 Hayek (1941: 393 – 4) notes that ‘the rate of saving sets the limits to the amount of investment that can be successfully carried out’. It will affect the volume of investment via investment demand and it will operate indirectly on the money rate of interest or on the supply of investible funds.

30 In other words, when the function is horizontal: see note 26.

31 This does not happen only if the desire to hold money is perfectly inelastic. In this case the money rate of interest corresponds to the rate of profit determined by the equilibrium analysis.

32 Hayek (1941: 386 – 92) re-expresses this process in different terms than Hayek 1939, in as much as he uses the functions of investment demand and of investible funds supply.

33 Hayek 1941: 393, 407. The money supply is taken as given. If one assumes that it could vary, a further element preventing the money rate of interest from rising is added.

34 Connected to this criticism is another criticism that Hayek formulates against Keynes's economic policy (Hayek 1941: 407 – 10), which will become a recurring feature of his subsequent polemical writings. Hayek accuses Keynes of favouring a short-run monetary policy, because he sees only the immediate effects of money quantity variations, ‘completely disregarding the fact that what is best in the short run may be extremely detrimental in the long run, because the indirect and slower effects of the short-run policy of the present shape the conditions, and limit the freedom, of the short – run policy of to-morrow and the day after’.

35 Hawtrey (1941: 290) believes part IV is not sufficiently developed owing to war time pressures. Smithies (1941: 778) comments that ‘it almost gives the impression of having been included as an afterthought’. F. Lutz (1943: 302) deems it to be ‘excellent’, but he deals exclusively with the other parts of The Pure Theory of Capital, which contains Hayek's theory of interest in real terms. Steedman (1994: 7) considers part IV ‘hardly adequate to its topic’.

36 Keynes 1936: 31 – 2, 292 – 4: ‘Thus the analysis of the propensity to consume, the definition of the marginal efficiency of capital and the theory of the rate of interest are the three main gaps in our existing knowledge which it will be necessary to fill. When this has been accomplished, we shall find that the theory of prices falls into its proper place as a matter which is subsidiary to our general theory’. For Keynes his theory is ‘a theory of value and distribution, not a separate theory of money’. It is worth recalling here that Hicks (1935) establishes a relationship between the theory of prices and production and monetary theory, which is opposite to the relationship proposed by Keynes. Hayek (1995: 57 – 9) affirms that Hicks' A Suggestion for Simplifying the Theory of Money ‘still seems to me more than the General Theory the most valuable result of the monetary discussions of the period’. On this topic, see Kohn 1986: 1202: ‘While Hicks considered the theory of money to be no more than a particular chapter in the theory of value, Keynes believed just the opposite: that the whole of the theory of value was just a special case of the monetary theory of production’. See also Townshend 1937: 159 – 61.

37 Keynes 1973b: 116 – 7. See also Townshend 1937: 159: capital assets prices ‘as well as money and monetary assets, have a value to hold for future exchange (i.e. for security or for speculation), causally independent of their value in present exchange, and determined by, and varying with, expectation’.

38 Keynes 1936: 137, 1973b: 106 – 7, 117, 213 – 4. Keynes had already clearly expressed these arguments in the Treatise on Money (1930: xxvii).

39 Keynes 1936: 137, 1973b: 118, 122. See also Leigh 1951, Chick 1987.

40 Keynes 1936: 168. More appropriately: ‘as to the complex of rates of interest for varying maturities which will rule at future dates’ (ibid.).

41 The methodological differences between Hayek and Keynes in treating uncertainty and expectations are the subject of many recent studies. See the references indicated in Carabelli and De Vecchi 2001.

42 For the notions of ‘uncertainty’, ‘state of confidence’ and ‘weight of an argument’ in the General Theory see Keynes 1936: chapter 12, and in particular pages 148 – 9, where reference to the Treatise on Probability is explicit. See also Carabelli 1988: 55 – 9 and McCormick 1992: 178 – 82. On the literature about Keynes's analysis of uncertainty see Dow 1995. On the relationship between changes in ‘weight’ or ‘confidence’ and changes in liquidity preference see Winslow 1995.

43 Keynes 1936: 169 – 70, 198 – 9. People who dissent from the predominant opinion and believe that the future rate will be above or below the rate ‘assumed by the market’ will sell or buy capital assets other than cash, thereby contributing to raising or lowering the interest rate with regard to the value assumed by the market.

44 Keynes 1973b: 230 italics added. ‘It is only to the extent that increased investment requires larger active balances that it reacts on the rate of interest – a reaction which can be prevented by increasing the quantity of money [ … ]. I have many pages on the theme that increasing investment involves increasing output and that this kicks back on the rate of interest by draining away more money into the active circulation’ (ibid.: 91). On this issue, see also Keynes 1936: 168 – 72, 203 – 4, 1973a: 522 – 3, 631, 1973b: 4, 11 – 13, 21 – 2, 115 – 6, 221, 223 – 6.

45 ‘If the liquidity preferences of the public (as distinct from the entrepreneurial investors) and of the banks are unchanged, an excess in the finance required by current ex ante output (it is no necessary to write “investment” since the same is true of any output which has to be planned ahead) over the finance released by current ex post output will lead to a rise in the rate of interest; and a decrease will lead to a fall. I should not have previously overlooked this point, since it is the coping-stone of the liquidity theory of the rate of interest’ (Keynes 1973b: 220).

46 Keynes 1973a: 522 – 3. See also 1936: 165, 184 – 5, 1973b: 202, Townshend 1937: 157 – 8.

47 Keynes 1973b: 222: ‘given the state of expectation of the public and the policy of the banks, the rate of interest is that rate at which the demand and supply of liquid resources are balanced. Saving does not come into the picture at all’. For the sake of completeness, note that Keynes does not exclude that variations of saving can affect the interest rate (Keynes 1936: 218 – 9), but this occurs ‘because in economics everything affects everything else’ (1936: 245 – 7, 1973b: 11).

48 Hayek does not manifest any interest for other two key topics of the General Theory: the expectations theory (key to understanding Keynes's entrepreneurs’ investment and liquidity preference theories) and the discussion of the ‘own rates of interest’ (a direct consequence of Sraffa's criticism of Hayek).

 This approach to the General Theory is shared by other commentators of Keynes, for example Hicks (1967a: 198 – 201) and Leijonhufvud (1981: 164 – 73). These scholars emphasize the importance of the ‘Wicksellian connection’, which is present in the Treatise on Money, do not challenge the notion of the natural rate of interest and, as a consequence, accept Keynes's liquidity preference theory with many reservations.