Kalecki's Principle of Increasing Risk and Keynesian Economics

Tracy Mott

• Publisher: Routledge
• Year: 2009
• Print ISBN: 9780415080392
• eBook ISBN: 9781134894772
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

Kalecki was one of an important generation of Cambridge economists. Here, Tracy Mott's impressive book examines the relationship of Kalecki's economics to different economic areas and its relationship to major alternative schools, such as Keynes and Marx.

Mott looks at Kalecki's 'principle of increasing risk' and how it gives the way in which the reproduction and expansion of wealth can bring a coherent unity to economic analysis. In so doing, it makes sense out of the fundamental conclusions of Keynesian economics on the underemployment of labour and capital.

1 Economic theory

Nearly seventy-five years later, it is probably safe to say that John Maynard Keynes’s (1964, [1936]) General Theory of Employment, Interest, and Money caused more problems for economics than it gave solutions. The blame for this can be laid at many feet, perhaps not the least of which are those of history herself. If, as Karl Marx (1970 [1859], p. 21) remarked, “Mankind … inevitably sets itself only such tasks as it is able to solve,” perhaps neither does it ever find a solution to its problems which does not then lead to new problems. However, as many have noted, part of the blame can be laid at Keynes’s feet for not “finishing” his work in the sense of giving us a complete-enough system of the vision of the economy underlying his theory. There were, though, good reasons for this: Keynes, after all, was not omniscient, though he did know that many would not understand no matter what he did.¹ I myself would like to put the biggest part of the blame on the economics profession for constantly refusing to hear sympathetically or to understand the revolutionary aspects of Keynes’s ideas.

But that also is a story that others have told. What I want to do in this book is to make a contribution toward adding the missing pieces to the conception of the capitalist economy necessary to make what is worthwhile in Keynes’s ideas part of a coherent whole. A lot of work along these lines has been done by those working in what has come to be called the “Post-Keynesian” approach, broadly conceived. My desire here is to use some of these ideas and to add some of my own to set up a general framework, but, more importantly, to present a conception that I believe can give an underlying unifying principle to the various suggestions and conclusions that the ideas proposed by Keynes really offer.

Opponents have often characterized any and all versions of “Keynesianism” as suffering from the defect of being “ad hoc.” That is, the conclusions of Keynesian arguments are said not to be derived from fundamental postulates about economic behavior, but rather to be based on appeals to “institutional,” or extra-economic, aspects of life. The call for adequate microfoundations of macroeconomics so the renewed emphasis on basing macro behavior on utility-maximization are examples of this complaint.

Though the call to unify economics should be correct in the sense that, if our explanations of microeconomic behavior and macroeconomic behavior were divergent, at the least we would like to have good reason for why this is and how the differing explanations were compatible with one another,² the principle of utility-maximizing behavior, in fact, does not explain what is required because in and of itself it implies nothing specific about economic behavior. Utility maximization can, perhaps, be given meaningful content and certainly can be made consistent with observed behavior. Its failings, however, are revealed precisely in that in order to give meaningful content to the principle of utility maximization, other information has to be supplied. The extra content needed to give the theory meaning bears no essential relation to the development of the theory. Other than as a way to make the theory “work,” i.e., correspond to observation, it has nothing to do with the content of the theory.³

For example, utility-maximizing behavior does not in and of itself imply that asset holders will hold diversified asset portfolios. To generate asset diversification, individuals must be risk-averse.⁴ Since we observe diversification, utility maximization “explains” this by means of the assumption of risk aversion. But risk aversion, however plausible, is not itself explained as part of the science. It is merely used to square observation of phenomena with the underlying fundamental principle of the science. Similarly, neither the labor supply schedule nor the saving schedule can be assured of having the “correct” positive slope without positing further conditions beyond only utility maximization to make the substitution effect from a higher return to labor or saving outweigh the income effect, which can work in either a positive or negative direction.

Utility maximization can “explain” any state of the world. This, pace Popperian falsificationism, need not be a weakness, but could be a strength of the theory. Or rather, it would be a strength except that the assumptions and conditions necessary to give utility maximization explanatory content are not themselves explained or explainable in terms of the utility theory. As with risk aversion, they are merely asserted because they “work,” and they appear to be “plausible.” When and if these intermediate assumptions are examined on their own terms, I believe we will find the theory in effect to negate itself and to point in some other direction to find adequate explanations.

Fundamentally, this is what Keynes did in the General Theory. The utility-maximizing, or neoclassical, theory held that there could be no involuntary unemployment because maximizing behavior implied that workers would supply labor to the point where the marginal disutility of labor equaled the real wage. Keynes saw that the attempt to engage in such behavior by offering work per units of a money wage would, other than by accident, inevitably fail. Every attempt to achieve desired employment by altering money wages in order to arrive at the right real wage would drag the price level with it, since the ratio of output prices to money wages costs tends to remain constant at least in the short run. (Keynes, 1964 [1936], pp. 4–22.)

In this case the extra condition necessary to make the neoclassical theory work would be that decentralized decision-making operates in the way that centralized decision-making does, i.e., that the “correct” money wage and price level be discovered as they would if imposed by an omniscient planner or “auctioneer.”⁵ This condition bears no necessary relation to the theory of a decentralized market system. And the explosion of the neoclassical labor market mechanism by pursuing its own logic not only destroys the old system, but also lays the ground for a new conception of the working of the capitalist economy. But here is where Keynes’s revolution only got part of the way.

Before we proceed with that problem, though, it is worth our while to note that Keynes’s method of exploding the neoclassical system by pursuing its internal logic is the same as Marx’s method of dealing with classical economics. Marx wrote,

Political Economy has indeed analyzed, however incompletely, value and its magnitude, and has discovered what lies beneath these forms. But it has never once asked the question why labour is represented by the value of its product and labour-time by the magnitude of that value.

(1967 [1867], p. 80)

By pursuing the latter question, Marx was able to present the conditions for capitalism’s origin as a mode of production and the conditions for its passing into some other form of economic arrangement. This is the dialectical method.

Once Keynes had realized that labor supply conditions could not determine the level of employment, he was led to develop arguments, begun in The Treatize on Money (Keynes, 1971 [1930]), about what did determine the level of employment by determining the level of aggregate demand. Keynes, however, never made a complete break with neoclassical value theory, though hinting at it. The process of grafting his system into neoclassical economics–the “neoclassical synthesis” of Paul Samuelson–thus began quite early. The major problem confronting the synthesis has always been that of “money illusion.” Given the neoclassical microfoundations, “Keynesian” unemployment could only result from the prevalence of incorrect prices. Thus, economic agents must be denying themselves opportunities for mutually beneficial gains because they are operating in terms of monetary magnitudes which do not correspond to the real quantities which they are seeking.

The problem of “Keynesian” economics then has been to devise ways whereby unemployment could be explained in basically neoclassical world. All the answers, including institutional wage rigidity, costly information, long-term implicit or explicit contracts, etc., are ways of generating unemployment according to pre-Keynesian ways of thinking, and many of these answers can be found in pre-Keynesian monetary economists like Irving Fisher or Knut Wicksell.

That Keynes was up to something different can be seen in his development of the other hallmarks of the General Theory: volatile investment determined by animal spirits, dependence of consumption on income, liquidity preference, and so forth. All of these were modified or attacked by the neoclassicals in the intervening years, on the grounds of some lack of compatibility between these ideas of Keynes and the neoclassical system. But in what way are these ideas opposed to neoclassicism by offering a different perspective?

Let us take a look at one such perspective–the Marxian conception. The first question is, of course, what this perspective offers as the unifying idea behind the working of the capitalist economy. Many would say the class struggle. Though there is a bit of class struggle in Keynes, I don’t believe that this will form a new unifying principle for Keynes’s economics. There are too many things going on in Keynes that bear no relation to conflicts between capitalists and workers. If we bring in the conflicts between the rentier class and the other members of society, we can relate to more of the General Theory. This gives Keynes more of a “classical” tinge, looking a bit like Ricardo or Malthus, but I don’t believe it enables his work to correspond exactly to the Marxian conception, at least not without further emendations.

There is a problem in Marx’s own work that we should take up at this point. David Levine (1977, pp. 153–157) has put it this way. In criticizing his classical predecessors Marx pointed out that classical political economy had seen the production of value by labor as the natural principle on which the market economy was based. Marx saw that markets and the determination of value by labor-time were rather a social phenomenon arising from the self-destruction and passing into a new form of the feudal mode of production. Thus, instead of being a freeing up of economic life in accordance with the order of nature, capitalism was merely another form of social organization with its own contradictions leading to further transformations. Yet Marx himself retained more than a trace of the tendency to subsume social determinations to natural. This took the form in his work of seeing the question of the social allocation of labor as the primordial social determination of the economy in whatever historical epoch. In capitalism, then, value and its production and realization and its reproduction were determined in accordance with this primordial social determination.

This is why Marx, though realizing the problems with Say’s Law, based his conception of value and indeed his conception of the development of the capitalist economy on a fundamentally “full-realization” of the prices and profits producible by the employed labor. The effect of the capital accumulation process on growth and fluctuations and on the determination of value itself in the manner of the “Keynesian” type of effective demand problem was also downplayed in favor of the tendency for the rate of profit to fall due to a rising organic composition of capital.

If we turn to Marx’s insights about the capital accumulation process and build on it, however, we can develop a theory of growth, fluctuations, and value that overcomes these limitations and relates well to Keynes’s insights. This can be seen best by examining Michał Kalecki’s economics. Kalecki put profits at center stage. This is expressed most clearly in his “principle of increasing risk,” (Kalecki, 1990 [1937]b, pp. 285–293) which states that the fundamental limit on the expansion of any firm is the size of its own capital. Limits on growth due to diseconomies of scale Kalecki rejects on the grounds that any technical or engineering limitation can be overcome through replication of the optimal scale unit. Limits due to market size under less than perfect competition Kalecki accepts, but he argues that this is not enough to explain every limitation, e.g. why large and small firms are started at the same time in the same industry.

This idea has been appealed to in the study of business concentration (e.g. Josef Steindl, 1990 [1945]). Kalecki used it to justify the explanatory role of profits in his investment models (Kalecki, 1991 [1954]f, pp. 281–292). It has been seen to underlie various arguments within post-Keynesian theory about what determines the mark-up of prices over prime costs and the importance of the level of indebtedness to business fluctuations (e.g. Alfred Eichner, 1976; Adrian Wood, 1975; Mott, 1982).

Kalecki (1991 [1954]e, pp. 277–281) argued that the limit coming from the size of own capital arose because the more of one’s own wealth tied up in a particular fixed investment, the more danger one was exposed to in the event of failure and the more trouble one would be under in case of a sudden need for liquidity. Issuing debt compounds this problem by setting up an additional outflow of liquidity to be met.⁶ Issuing new equity avoids the fixed commitment of repayment, but dilutes the value of the investment to the original holders more than debt by letting new shareholders on an equal footing.⁷

When we tie this into Kalecki’s demonstration that spending out of profits determines profits, we have a fuller way of looking at Marx’s circuit of capital, M-C-M’. The willingness and ability to sink profits into productive capital in the hope of making more profits determines what profits will be there next period. That is, across firms the amount of profits determines what is available both in terms of own funds and ability to borrow, and in the aggregate the willingness to sink these funds determines the level of profits, which is then available for next period’s investments.

The willingness to sink funds into fixed capital for Kalecki is also of course a matter of the prospective profitability of investment. This in turn is largely a function of expected product demand relative to the capacity to produce. Productive capacity is the result of past investment minus depreciation, while product demand in the aggregate is given by the propensity to spend out of profits and wages times profits and wages, respectively. Consumption demand is the largest part of aggregate demand, and is taken to be mainly a matter of the level of wage income, since the propensity to consume out of profits is much lower than the propensity to consume out of wages.

Linking all of this with Keynes is straightforward. The important thing to note, though, is how Keynes’s ideas about investment, consumption, liquidity preference, and the like, which seem to be based on perhaps plausible but nevertheless “ad hoc” insights from the point of view of neoclassical economics can now become grounded in some fundamental principles of the income distribution and capital accumulation processes of a capitalist economic system. Relations which Keynes claimed to be dependent upon psychological considerations can be seen to be determined by income categories and their role in determining consumption and accumulation. The dependence of consumption on personal income, e.g. is due to the fact that the vast majority of household income consists of wages as payments for laboring rather than receipts from selling or renting property. If those who are dependent on laboring for their livelihood have little wealth other than their wage-income from which to finance consumption, the correlation between consumption and wages, in turn the major component of household income, must be close.⁸ The determination of investment by animal spirits and of interest by liquidity preference need to be subsumed into what we discussed earlier concerning the role of profits and profitability in M-C-M’. Discussions on these topics will follow in the subsequent chapters.

It is also important to note that this requires a reappraisal of Marx as well as of Keynes. The emphasis here is on the capital accumulation process and on the role of income distribution in determining that process. But we are not concerned with the determination of full-realization values or prices of production. Rather, we are concerned with a state of affairs in which such a kind of full-realization values cannot be determined. One could appeal to the notion of “monopoly capital” or imperfect competition or to the disequilibrium state as the short-run fluctuations “market prices” around the long-run benchmark “natural prices”. I do not wish to take refuge in such a stance. For one thing, the idea of value-determination appropriate to capitalism should not, I believe, embrace a distinction between some things called “perfect” and “imperfect” competition. Capitalist competition, as in Kalecki, is a matter of “degrees” of competition. The importance of profits to the accumulation process, as in the classical economists and Marx, implies that competition with positive profits is not “imperfect.” As competition is a process of uneven development, however, equal rates of profit across industries, as in the classical economists and Marx, are unlikely.⁹ Thus, the long-run prices of production, as in Piero Sraffa (1960), are ruled out even as a “benchmark” of some sort. The formula “Sraffa plus effective demand” embraced by some (e.g. Nell, 1984) as the correct model of the capitalist economy to meld Marx’s and Keynes’s insight embodies a contradiction. Once effective demand is let in, Sraffa’s prices are out.

This is not just because of the problem that a shortfall in effective demand will not allow these “full-realization” prices to be fully “realized.” It has to do with something more fundamental which lies behind the realization problem, the nature of competition, and the determination of prices. This is the notion of an adequate conception of capital. The notion of capital necessary for the prices of production model is that of circulating capital. Where fixed capital is discussed in Sraffa, it is treated as circulating capital by applying a depreciation principle to determine value used up and treating the surviving value as a joint product. With only circulating capital in essence, equal rates of profit can be posited due to the mobility of capital between sectors which this conception allows. But this conception reduces the prices of production model to David Ricardo’s model where capital is “corn” and effective demand problems need not exist because production can always be readjusted to meet any changes in demand. It is only where capital is not so malleable in physical or value terms that the failure to “realize” value can cause a “crisis.” (See Levine, 1977, pp. 115–152; 1980a; 1980b.)

Kalecki’s ideas express this quite clearly. The degree of willingness to sink liquid capital into a fixed form, given the availability of liquid capital, is what determines the rate of investment. This degree of willingness is of course dependent upon the expected rate of profit on investment. But the expected rate of profit is mainly dependent in its turn upon the present degree of the utilization of fixed capital more than any other factor. And the availability of liquid capital, moreover, is itself dependent upon the expected rate of profit as well as the amount of profits realized by and retained in the firm last period.

This is “the principle of increasing risk.” The willingness to become illiquid, to sink liquid capital into fixed capital, decreases with the amount of one’s own funds so invested. Thus, the availability of liquid capital in the aggregate is not solely a function of the “pure” supply of money and credit as provided by the central bank and automatically multiplied by the reciprocal of the required reserve ratio. It is also affected by the level of own capital or unencumbered funds, i.e., retained profits plus depreciation allowances. “For to everyone who has will more be given, and he will have abundance; but from him who has not, even what he has will be taken away.” (The Gospel According to St. Matthew 25: 29.) In more common parlance, “It is easier to get a loan the less you need it.”

The Keynesian problem of effective demand turns out then to revolve around the tension between fixed and liquid capital in the following manner. If demand for a firm’s products falls so that the firm is no longer utilizing sufficiently the fixed capital it already has in place, the firm cannot melt down its capital and convert it into a new type of capital to produce goods for which there is a high demand. Since demand for its products has fallen, the value of these fixed assets has fallen and so the firm cannot convert them into liquid assets at much more than a small percentage of its original investment.

So, a drop in effective demand because investment is not considered sufficiently profitable in several or all industries...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Foreword
• Preface
• 1 Economic theory
• 2 Prices, profits, and costs
• 3 Real and money wages
• 4 The theory of value
• 5 Investment spending
• 6 Consumption spending
• 7 Taxation
• 8 Macroeconomic cycles and growth
• 9 Interest rates, inflation, and monetary policy
• 10 Financial institutions and financial markets
• 11 Economic policy and political economy
• Notes
• Bibliography