Investment and Saving in a Dynamic Context: The Contributions of Athanasios (Tom) Asimakopulos^*

* An earlier draft of the paper was presented at the 22nd FMM Conference, Berlin, 25–27 October 2018.

ABSTRACT

In the 1980s, Asimakopulos criticized both Kalecki and Keynes for the way they dealt with the problem of the investment multiplier. Kalecki's and Keynes's insufficient attention to the time dimension of the multiplier process led them to overlook some aspects of the relation between saving and investment and underestimate the importance of financing investment, especially with regard to the problem of the conversion of the firms' short-term loans into long-term loans. The paper looks at these problems by carrying out the analysis in a more formal way than Asimakopulos did. In a dynamic analytical context which takes explicit account of the time dimension of processes, the economy's propensity to save can affect investment through its effect on the long-term interest rate. Acknowledging this, however, does not imply the rejection of the view that investment ‘comes first’: it is not saving that determines investment, but the other way around.

KEYWORDS:

• Investment
• saving
• multiplier
• finance
• Asimakopulos's economics

JEL CLASSIFICATION:

• E10
• E12
• E43
• E44
• B3

Acknowledgements

The author would like to thank the participants in the session and, in particular, J. Bibow and A. Terzi, whose criticisms helped him improve my work. The author also want to thank G. Harcourt and P. Montalbano for their helpful comments and suggestions. Finally, the author would like to thank two anonymous referees for their comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

* An earlier draft of the paper was presented at the 22nd FMM Conference, Berlin, 25–27 October 2018.

1 Also Studart (1995) and Chick (1997) deal with some of the issues considered by Asimakopulos, but from a more balanced and less critical standpoint. Harcourt (1995) gives a thorough and careful analysis of Asimakopulos's many other contributions to Keynesian economics and considers also his 1983 article.

2 The debate was mainly concerned with Keynes's views rather than Kalecki's. Asimakopulos returned to consider Keynes's position on these issues in his book on The General Theory and accumulation (Asimakopulos 1991, pp. 109–15 in particular).

3 In The General Theory, Keynes's attention is focused on the final equilibrium produced by a larger investment, i.e., ‘on the logical theory of the multiplier which holds good continuously without time-lag, at all moments of time’ (Keynes 1973a [1936], p. 122).

4 Asimakopulos also points out that this is a necessary but not sufficient condition for restoring the initial liquidity positions. For the reasons why it is so, see Asimakopulos (1983, pp. 227–8).

5 Harcourt (1995, p. 13) points out that ‘Tom remained insistent that for post-Keynesian analysis to be operational, it must be done in terms of period analysis.’

6 For example, $S_{p_1}=s{I}_0$ denotes the amount of saving generated by the increase in investment $I_0$ in the first round of the multiplier process.

7 Firms are assumed to finance their investment entirely with external funds. If part of $I_0$ were financed with internal funds, this would not affect the analysis below in any significant way.

8 It could be assumed that firms convert their short-term debt at a sub-period $p_i$ with $i\ne 1$, but this would not significantly affect the main results of our analysis. In any case, the time interval between t = 0 and $p_i$ must be sufficiently short, otherwise the initial firms' debt should be regarded as a long-term liability.

9 We assume that the banks' lending decisions are independent of the amount of their reserves. On the relationship between banks' reserves and lending, see for example Werner (2014).

10 So that $I_0=b\left(r_{l,0}\right)$. We could consider a more general case in which investment at t = 0 is a function of the expected long-term interest rate at $p_1$, $E\left[r_{l,p_1}\right]$. Given the expected interest rate, which can be equal, larger or smaller than $r_{l,0}$, the analysis above would not change.

11 On Robertson's ‘step-by-step’ method see also Laidler (1995).

12 In particular, with respect to the process through which the initial liquidity positions are restored. See Robertson (1937, pp. 432–3), and Robertson (1938). See also Asimakopulos (1983, p. 228n) and Ingaro and Sardoni (2019, chapters 3 and 5).

13 See also Leijonhufvud (1981).

14 See also Keynes's considerations on the Swedish approach in some notes of his in 1937 (Keynes 1973b, p 183).

15 Also Lindahl (1954a, 1954b), another important exponent of the Swedish School, criticized Keynes's static method in The General Theory and for having abandoned the attempt to develop a dynamic approach in A Treatise on Money.

16 Asimakopulos (1971), inspired by Kalecki (1990), criticized also this aspect of Keynes's theory. See also Sardoni (1996) for a discussion of the topic.