Theory of the Demand for Money: Marx and Keynes

There was a small exchange between David Glasner (here and here) and Scott Sumner (here) on the originality of Keynes' contribution to the theory of demand for money. Main content of the contribution is to establish the connection between the demand for money and the rate of interest as a determinant of the former.

In what Keynes called classical system, the interest rate is determined within the real sector by the two schedules of investment and saving with no consideration of money; this is the loanable funds theory. As opposed to this, in Keynes's liquidity preference theory, the money market figures significantly and the interest rate is determined by the interaction between the goods market and the money market as shown in the IS-LM framework.

That is, Keynes established the monetary aspect of the interest rate determination which was missing in the Cambridge school. David Glasner and Scott Sumner were contending on whether this is Keynes's original contribution or its essence was already there in Cambridge economists' writings.

In terms of originality of highlighting the monetary aspect of the interest rate determination, the honor shoud be attributed to Karl Marx much before Marshall's Cambridge school and Keynes.

In chapter 32 of Capital Vol.3, after suggesting that "The possibility of a high rate of interest of longer duration .... is given by the high rate of profit", Marx discusses the mechanism operating behind this causal relation:

"In the period when business revives after the crisis ...... loan capital [i.e. credit] is demanded in order to buy, and to transform the money capital into productive or commercial capital. And then it is demanded either by the industrial capitalist or by the merchant. The industrial capitalist invests it in means of production and labour-power. // The rising demand for labour-power can never be in itself a reason for a rising rate of interest, in so far as this is determined by the profit rate....."

"...the demand for variable capital may increase, and thus also the demand for money capital, while this in turn increases the rate of interest".

"...to say that the demand for money capital and hence the interest rate rises because the profit rate is high is not the same as saying that the demand for industrial capital rises and that this is why the interest rate is high".

In these, Marx is identifying two explanations of why a high profit rate leads to a high interest rate: First, an increase in profit rate leads to an increase in investment, which in turn leads to a rise in the demand for money and credit, which finally raises the rate of interest. Second explanation is simply that an increase of investment generates a rise in the rate of interest. It is easily seen that the latter is nothing but the loanable funds theory, or a theory of real interest, which Keynes tried to overcome and the former is something similar to Keynes's emphasis on the monetary aspect of the interest rate determination.

However, the difference between Marx's and Keynes's theory of the demand for money is more important than their similarity. In Keynes's case, or in the Hicks's representation of it, the demand for money is determined by the marginalist framework, i.e. by comparing the marginal utility of holding money and its marginal costs. On the contrary, in Marx the demand for money is modeled within a broader context of reproduction and accumulation of capital.

A Starting Point of Developing Marxian Theory of Money and Credit

See Palley’s words on Post Keynesian theory of endogenous money:

"PK endogenous money theory emphasizes that this linkage [connection the financial and real sectors] runs predominantly from credit to money to economic activity. The important feature is that credit is placed at the beginning of this sequence. This contrasts with conventional representations that place money first, as reflected in the standard textbook money multiplier story in which bank deposits are said to create loans" (Palley 2008, p.2).

The issue is this: Which one is logically prior, money or credit? This question has a long history of monetary debate of, following Schumpeter’s categorization, ‘monetary theory of credit vs. credit theory of money’ each corresponding to 'deposits make loans' and 'loans make deposits' thesis. Depending on an answer to the question, theoretical definition of money and bank will be drastically different.

In monetary theory of credit, money is usually defined as constituted of something real - hence real money or commodity money - while credit, as a substitute money, is created based on the real money. In this view bank merely intermediates money from surplus units to deficit ones; hence banks as financial intermediaries.

In credit theory of money, on the other hand, the essential task of bank is to create money through making loans. What is it then that bank loans out when money is created simultaneously with the lending? It is its own liabilities! In this approach money is essentially nothing but bank's liabilities. 

Now, where does Marx's monetary theory fit in? It is conventionally accepted that in Marx money is logically prior to credit; thus monetary theory of credit. For example, in the initial chapters of Capital Vol.1 Marx defines money essentially as commodity money with intrinsic value; a metallic money is assumed in the analysis of the circuit of capital and reproduction scheme in Capital Vol.2 and of transformation of value into price of production in Capital Vol.3; it is only in Part 4 and 5 of Capital Vol.3 where credit system is discussed. 

In particular, as Costas Lapavitsas has strongly emphasized in various places, it is shown in Capital Vol.2 that the credit system is structurally derived from the workings of the circuit of capital; more specifically, idle capitals regularly emerge due to the turnover times and will be lent out to other capitalists in search of profit opportunity; in this sense a pool of idle capitals is the foundation of credit system. 

All these seem to suggest that Marx's monetary theory belongs to monetary theory of credit being incompatible with Post Keynesian type of endogenous money theory. According to this observation Marx's monetary theory would be rejected as invalid in the economy where credit money perform various functions of money.

However, this is not a proper assessment. In order to fully grasp Marx's theoretical system in its entirety we have to recognize that it is constituted of two different dimensions, i.e. method and theory. A very helpful place to look at for this is the chapter on simple reproduction in Capital Vol.2 where we find Marx's explanation why he assumed metallic money. It is worthwhile to quote.

"[The assumption of commodity money] is not made from mere considerations of method, although these are important enough, as demonstrated by the fact that Tooke and his school, as well as their opponents, were continually compelled in their controversies concerning the circulation of bank-notes to revert to the hypothesis of a purely metallic circulation. They were forced to do so post festum and did so very superficially, which was unavoidable, because the point of departure in their analysis thus played merely the role of an incidental point" (Capital Vol.2, p.290).

First of all, methodologically, commodity money is a right starting point and the analysis should then proceed to credit money. This is according to the method of presentation which captures in thought the logico-conceptual development of the object of analysis from the abstract to the concrete; it is important to note that this process does not necessarily corresponds to the actual historical development. Marx emphasizes the importance of this method by alluding to theoretical tumbling-blocks Tooke and both Banking School and Currency School had to face because they adopted a false starting point, i.e. credit money instead of purely metallic money, in the famous classical monetary debate.
Another reason than method is a theoretical consideration to emphasize that the capitalist economy systematically requires the credit system. Marx demonstrates this by showing that the supply of commodity money, due to its physical limit, cannot grow in step with the growth of the entire system. By starting with metallic money Marx was able to demonstrate effectively that extended reproduction of capital requires credit money which will bring the system beyond the metallic barrior. Marx's theory which emerges at this place states that in capitalist system money cannot but be a credit money, which is issuer's liabilities.  
 

From these considerations we have the following lessons on Marx's monetary theory: A distinction between method and theory is important; method of developing from the abstract (commodity money) to the concrete (credit money) on one hand and, on the other hand, theory of capitalist mode of production according to which credit system is a required condition for its extended reproduction. Once this is recognized we could see that both commodity money and credit money (as issuer's liabilities) perform an important role in Marx's monetary theory.

 
Recognizing these should be the starting point of any attempts to develop Marxian theory of money and credit.
 

Minsky and Marx: On Michael Roberts' Take on Minsky's FIH

Some Marxists tend to emphasize a distinction between Marx and Minsky dismissing the latter's theory of financial crisis. But I found, in many cases, this is grounded in a misunderstanding of Minsky, more or less. For example, Michael Roberts' piece on Steve Keen. (I would like to emphasize that the following is about a slight difference of opinion, on the specific issue on Minsky, I have with Michael, whose works I like to read.)

The very essence of Michael's understanding of Minsky seems to be this (if my understanding is correct): In Minsky, boom and bust in the financial sphere depend on the subjective and individual expectations of market participants and are independent of what's going on in the real; i.e. what's going on in the real rather depends on people's expectation. For Minskyians, Michael says, "profits depend on expectations and crises are the result of changed expectations by financial speculators".

I think this is an unfair rendition of Minsky. Even though Minsky himself is not crystal-clear on the related issue, I think interpreting his Financial Instability Hypothesis as a dynamics of the real and the financial interaction makes a lot of sense. Most importantly, profitability (and the resulting cash-flows) of firms is very crucial in the formation and frustration of expectation, which in turn leads to a corresponding fluctuation in the financial variables. In many Minskyian formal models, the future prospect, which affects debt level, is modeled as a function of profits relative to debt service payments.

In this sense, I think Michael's conclusion that "For Marxists, instability in the financial sector would not be enough to cause a major crisis if profitability is rising", also applies to Minskyians. In a sense, Minskyian concept of 'financial instability' cannot be conceived without in relation to profitability. For this concept describes the system's endogenous logic through which it gets financially fragile by overloading debts during good times with favorable profitability and ends up with collapses (or debt deflation) when the euphoric prospects are frustrated by unsatisfactory outcome of profit rate.

In many cases, there is a tendency for those Marxists who put Marx's TRPF and Minsky's FIH in opposition to perceive the financial sphere as secondary and peripheral despite their repeated rhetoric that money and finance are important in Marx's theory. Rather, they seem to be subject to an old dichotomy of the real vs. the financial. For instance, Michael contrasts Minskyians' focus on the financial sphere with Marxists' on the capitalist production process as the origin of the recent crisis.

However, if you understanding that money and finance are essential constituents of the totality of capital in Marx's theory, such contrast would be unnecessary. That is, for Marx, the concept of capitalist production encompasses both production and circulation spheres, and both the real and the financial sectors. And the circulation and the financial could possibly develop their own logic to disrupt the capitalist system; most typical way they do this is to distance themselves from the confines of the value space of the real, generating the systemic 'irrational' fetishism that profits appear to originate from money capital itself  (Marx showed this especially in ch.21, 30, 31, 32 of Capital Vol.3). And this is nothing but an overgrowth of financial sector which is doom to fall down.

If understood in this way, I think a better way for Marxists to appreciate Minskyian FIH is to recognize that Marx already had something similar in Das Kapital 200 years ago or so, rather than to dismiss it as foreign to Marx.

on marx's method 2

One more comment on Hegel before we move on to Marx:

There is a section titled "With What to Begin?" right after Introduction in Science of Logic - a.k.a Larger Logic - and there you can find Hegel's long discussion of the issue of 'starting point' in logic. I think one of the main lessons there is that 'the point of departure in reality' is for Phenomenology, and 'the logical starting point' is for Logic. (Patrick Murray (2000) understands the difference between Phenomenology and Logic in relation to the starting point in this way.)

Now turning to Marx:

As we already know so well, in Grundrisse Marx mentions two conceptual journeys, a descending one from the chaotic concrete to the abstract and a ascending one from the abstract to the concrete with many determinations; and he dubs the latter as an "obviously scientifically correct method," i.e. the abstract as an obviously scientifically correct starting point.

And as we remember, Nicolaus argues in his Preface that Marx changed his mind in Capital. InGrundrisse, the argument goes, Marx begins with an abstract category of 'production', but in Capital he abandons this method and starts with commodity. That is, for Nicolaus the category of commodity is something concrete. Of course, opposing view followed of Marquit (1977) and Carver (1980).

The controversy here is how to understand the nature of commodity as a theoretical category. On this respect, I think Chris Arthur's explanation is very interesting. He gives three criteria for the starting point: it should be i) simple, ii) historically specific, and iii) immediate. And he says that commodity satisfies the latter two but not the first; it is not simple enough since it can be further analyzed into use-value and value. What about 'value'? According to Arthur it satisfies the first two but not the third; it is not something that can be immediately grasped. Then what is the true starting point in Marx? Commodity or value?

Here we have Jairus Banaji (1979)'s wonderful solution to this puzzle: a concept of 'double starting-point' which Arthur accepts. According to Banaji, “the beginning is a movement between two points of departure.” The idea is that the commodity forms the analytic point of departure to arrive at the concept of value; and value as the ground of all further conceptual determinations (money, capital) forms the synthetic point of departure of Capital. I think this solution is also in line with Hegel's method having two distinctive starting-points.

Understood in this way, I think both of the two conceptual journeys described in Grundrisseconstitute Marx's method.

Finally, returning to our initial concern with the above discussions in mind, I think Marx's distinction between 'method of inquiry' and 'method of presentation' can be mapped into as follows:

*method of inquiry (Marx of Capital) = descending journey from the concrete to the abstract (Marx of Grundrisse) = analytical method (Hegel) = analytical shift from commodity to value (Banaji & Arthur)

-> commodity as a starting point

*method of presentation (Marx of Capital) = ascending journey from the abstract to the concrete (Marx of Grundrisse) = synthetic method (Hegel) = synthetic shift from value to commodity, money, and capital (Banaji & Arthur)

-> value as a starting point

on marx's method 1

'The starting point' and 'method of inquiry & presentation' theme are ones of the most important methodological issues in the initial chapters of CapitalVol.1 which have troubled Marxian scholarship so long. We could easily agree that questions of 'where to start' and 'how to proceed' are the two most grounding issues when we analyze something scientifically. So was the case with Marx whose object of analysis is capitalism.

I think there are two primary references in dealing with this subject; Hegel's Logic and the chapter on the method of political economy in Grundrisse. I think Marx either consciously borrowed or was heavily influenced by Hegel's method in Logic, and used the example of analyzing a given country starting with its population in showing how to apply Hegel's method in doing social science. (In this respect, I totally agree with Lenin in insisting that without reading Hegel's Logic one barely understands Marx's Capital, and that the latter is the best application of the former.)

Notice that the conceptual proceed of 'Being - Essence -Concept' in Hegel's Logic corresponds that of 'chaotic concrete - thin abstract - concrete with many determinations' in Grundrisse. We could also compare logical developments in Capital such as 'commodity - money - capital' or 'exchange value (immediately perceived) - value (as underlying essence) - exchange value (conceived in thought as form of appearance of value)' ('homology thesis', an argument that there is an one-to-one correspondence between Hegel's system and Marx's system. Of course the concrete correspondence is different from scholars to scholars. For example some argue that the category of 'capital' in Marx corresponds to that of Essence, not of Notion, in Hegel, excluding Notion from homology.)

First on Hegel's Logic: Hegel starts his Shorter Logic with mentioning how difficult it is to begin in doing philosophy. Here's what he says.

"But with the rise of this thinking study of things [i.e. philosophy], it soon becomes evident that thought will be satisfied with nothing short of showing the necessity of its facts, of demonstrating the existence of its objects, as well as their nature and qualities. Our original acquaintance with them is thus discovered to be inadequate. We can assume nothing and assert nothing dogmatically; nor can we accept the assertions and assumptions of others. And yet we must make a beginning: and a beginning, as primary and underived, makes an assumption, or rather is an assumption. It seems as if it were impossible to make a beginning at all." (Hegel's Shorter Logic §1)

Yet, we have to start anyway and the first mode of knowledge we form in our mind is concrete and empirical image established through empirical observation. However, Hegel criticizes empirical science, empirical knowledge for its shallowness and being contained in sense-perception, and argues that knowledge should be pursued beyond. But, he does not reject it all together but incorporates it within the totality of history of philosophy as an element constituting the latter. (This is a superior aspect of Hegel's 'wholistic' systematic philosophy.)

On the other hand, however, Hegel's system of Logic starts not with something empirical, i.e. something concrete-complex but with the category of Being, which is something abstract-simple, and which can be obtained (from the initial empirical perception) through some sort of thought process, i.e. reflection, and proceeds towards more concrete-complex categories, Essence and Notion. Then what is Hegel's true starting point; the concrete-complex empirically given or the abstract-simple?

Before answering this question, notice that there are two movements working in Hegel's system: First from the empirical concrete-complex to abstract-simple, and the second from the abstract-simple to the concrete-complex. Hegel calls the first movement the Analytical Method and the second the Synthetic Method (for definition of each refer to Hegel's Shorter Logic §227and §228).

From this we could easily figure out that there should be two different starting points in Hegelian logic for two different Methods; 'the point of departure in reality' for the Analytical Method and 'logical starting point' for the Synthetic Method.

A Rise in Gold Price and Marx's Theory of Money

Last week saw the gold price skyrocketing over $1,250 per ounce, the highest price ever. Since the collapse of the Bretton Woods system, the connection between money and gold has officially disappeared. However, the demand for gold as a storage of value continued to exist; especially so in times of crisis. And we have observed the price of gold soaring during the last couple of years marked by consecutive blows of financial crisis and sovereign crisis with tumbling euro and stagnating dollar. Does this lend support to Marx's commodity money theory, as presented in the first three chapters, to any meaningful extent?

As is well known, there are largely two responses within Marxian camp to Schumpeterian critique of Marx's money theory as metallist and thus outdated:

1) To dismiss the current sovereign money regime as a superficial, temporary and unstable phenomenon (e.g. Claus Germer 1999, Anitra Nelson, John Weeks 2012).

2) To reject the traditional Marxian commodity money theory and modify Marx in a very similar way with Post Keynesian Chartalism and hence in this approach an emphasis is given to the combination of money theory and state theory (Duncan Foley's 1983, 1987 pieces are the most prominent examples).

3) To argue that within Marx's theory, it is not a logical necessity that money needs to have intrinsic value despite Marx's assertion; non-commodity - e.g. fiat money anchored on state power or credit money backed by issuer's creditworthiness - can possibly function as money. Consequently, the post gold exchange regime does not pose a threat to Marx's theory (e.g. Michael Williams 2000, Fred Moseley 2010 etc.).

One of the important theoretical concerns in this debate is whether money itself necessarily has to have an intrinsic value or not within Marx's theoretical system. I think the recent soaring of gold price gives us some insights to this question in relation to the stability of the value of money. First, an institutional environment, where valueless non-commodity performs essential functions of money such as measure of value and storage of value, cannot avoid experiencing a disruption in the value of money now and then, especially in times of crisis. Second, if the circulation of non-commodity money relies upon the state whose credibility depends on its military and financial power and thus is subject to fluctuation, then something like the recent rise in the demand for gold at the expense of government-printed money or government bond may be repeated in a much more violent way anytime in the future.

Jun 15 2010