by Utsa Patnaik
Increasing Long-Run Supply Price
Large numbers of primary commodities are subject to increasing supply price, not just in the short run when production cannot in any case be easily augmented, but even in the long run. The size of the tropical and semi-tropical land that produces large numbers of crops for consumption purposes or for commercial use in the capitalist sector of the world is limited. Even though land-augmenting investment and technological change can increase the effective supply of land, it typically requires state action, which is ruled out by the conceptual representation of capitalism. (And in any case, there is the real-life opposition to state action by financial interests that insist on “sound finance” and minimal state expenditure of this sort.) Also, minerals like oil are subject to increasing long-run supply price in the absence of new discoveries.
Increasing long-run supply price can be accommodated only through a fall in the profit share or a fall in the wage share within the capitalist sector. David Ricardo believed that since the real wage rate was bounded from below by a certain subsistence level, which alone would ensure an adequate supply of labor, an increasing supply price of primary commodities (what he called “diminishing returns”) would squeeze profit share and hence the profit rate until it fell to zero in a stationary state, and accumulation would come to a halt.
Even with given money wages, an increasing supply price would still have to reduce profit margins and profit rates if the value of money is to be kept intact and accelerating inflation avoided. It is not enough, in other words, that money wages are not allowed to increase autonomously. Even if they are not, capitalists are still faced with the prospect of either reductions in profit margin or accelerating inflation. Even a reserve army of labor large enough to prevent autonomous increases in money wages is not adequate to rescue capitalists from reduced profit margins if the value of money is to be protected.
There is, however, a further point of importance here. With increasing supply price of primary commodities, even if there is a reduction in the capitalists’ profit margin and profit rate and hence no accelerating inflation in the final product price and no threat to the value of money because of inflation in such final product price, there would still be a direct threat to the value of money because of the inflation in primary commodity prices. Primary commodities, or goods whose prices are supposed to move in tandem with their prices, will replace money as a form of wealth-holding as people come to expect a secular increase in their money prices. And even with profit margin and profit rate falling in the primary commodity–using sector, there would still be a threat to the value of money, which would cease to be a form of wealth-holding, and therefore the system would cease to be money-using.
Therefore, as a money-using economy, capitalism must have recourse to some ways of protecting the value of money. However, these ways are incapable of being captured in the standard conceptual representation of capitalism that economic theory has conjured up over the years.
Summing Up
The specificity of a money-using economy, which capitalism preeminently is, is not captured in the conceptual representation of it as an isolated sector, consisting only of workers and capitalists, that much of economic theory has dealt with. If capitalism was only an isolated sector, then it would be stuck forever in a stationary state or a state of simple reproduction. And in the event of accumulation, it would not even be able to protect the value of money in the face of the increasing long-run supply price to which several of the primary commodities are subject. The fact that capitalism has not had such dire experiences is because its reality has been quite different from the conceptual representation of it in most strands of economic theory. But before examining this reality, we’ll take a brief look at some of these strands and see how, specifically, economic theorists have failed to take account of capitalism’s money-using character.
CHAPTER 2
Money in Some Theoretical Traditions
In the previous chapter we discussed that the conceptual representation of capitalism as an isolated sector consisting only of capitalists and workers cannot capture its character as a money-using economy. Yet much of economic theory has stuck to this representation, which is why its treatment of money has been flawed in various ways. In this chapter, we examine some major strands of economic theory to establish this point.
The Walrasian System
Let us take the Walrasian system first. In the simplest exposition of this system an auctioneer arrives at market-clearing prices, and all sales occur only at these “equilibrium” prices, and there is no need for money in transactions. Money is only a numeraire, in terms of which all prices are expressed, a role that any other good can play as well. There is nothing sui generis about money; indeed, there is no reason why some special thing called “money” needs to exist in such a world at all.
Ironically, the Walrasian system, with money being used for the circulation of commodities at a constant velocity, with a transaction demand for it that bears a constant ratio to the total value of the transactions carried out, has been the staple of much monetary theory. It has been the bedrock of the particular tradition called “monetarism,” so much so that Frank Hahn described monetarism as synonymous with the belief that the world conforms to a Walrasian equilibrium.1 But in grafting a transaction demand for money into a Walrasian universe, the question remains: In a world where transactions occur only at market-clearing prices, why should there ever be a transaction demand for money? The only possible answer can be that the bids by market participants, which the auctioneer has to consider, are affected by their possession of money; that is, the money balance held by an individual affects his or her demand function. The question is how can it do so in a manner compatible with the notion of a transaction demand for money?
Robert Clower sought to answer this question through this example:2 The market meets only once a week, and the sale proceeds of one particular week cannot be used entirely for buying goods in the same week. Only a part of such proceeds, a fixed part at that, can be used, and the remainder must be carried over to the next week, where demand consequently becomes dependent upon the money-stock in the possession of the buyers. The total value of transactions on any market day in such a case will clearly be constrained by the amount of money stock in the possession of the people, as monetarism postulates.3
However, this constraining role of money supply on the value of transactions, which restores some role to money even in a world where all markets clear simultaneously through an auctioneer, arises entirely because of the assumption that only a fixed ratio of current proceeds can be spent on purchases in the current week. Unfortunately, to assume such a fixed ratio is completely arbitrary. Monetarism and its assumption of a constant (income) velocity of circulation of money, collapses if this ratio became a variable. Besides, if it does become a variable there is no reason why it cannot reach 100 percent, that is, why the entire proceeds of a given day cannot be spent the same day, in which case the constraining role of money on transactions, and with it any role of money, would disappear altogether. In such a case we would be back to the Walrasian world with an auctioneer, a world where money is not needed.
Moreover, a variable time lag between sale and purchase by market participants gives rise to a deeper problem. If money is held for a variable period of time between sale and purchase, then the obverse of this period for which it is held, namely the velocity of circulation of money, also becomes a variable. But this immediately raises the possibility of a deficiency of aggregate demand for produced commodities, and Say’s Law, which denies such deficiency of aggregate demand, breaks down.
This is so because Say’s Law, by which there can never be a deficiency of aggregate demand as “supply creates its own demand,” amounts to saying the following. There can never be an ex ante excess supply of produced commodities because, for this to happen, there has to be something outside the circle of produced commodities for which there is a corresponding ex ante excess de
mand; and there is nothing outside this circle of produced commodities for which there can possibly be such an ex ante excess demand. Money, the only obvious entity that stands outside of the charmed circle of produced commodities, and one for which an ex ante excess demand could cause an ex ante excess supply for produced goods, thus nullifying Say’s Law, cannot play this spoiling role if the demand for money is a fixed ratio of the value of produced commodities.
In short, ex hypothesi, there can never be an ex ante excess demand for money if there is a fixed time lag between sale and purchase. Such an excess demand can arise only when the time lag between sale and purchase by a representative market participant becomes a variable. And if that happens and Say’s Law ceases to hold, then the Walrasian equilibrium, which assumes that all markets clear and that there is no involuntary unemployment (which would occur if there was an ex ante excess supply of produced goods), loses its relevance.
Summing this up, if there is no time lag between a market participant selling goods and buying goods, as is the case if all transactions occur at the same time at the market-clearing equilibrium prices arrived at by the auctioneer, then there is in effect no money in the economy. Money comes into the picture only when there is a time lag between a person’s sale and purchase. If this time lag is fixed, then one can perhaps tell some sort of a coherent story about the Walrasian equilibrium. But there is no earthly reason why this time lag should be constant. And if it is not, then the Walrasian equilibrium loses its relevance since there is no reason for Say’s Law to hold. In other words, giving money a role in the Walrasian system, a role that does not destroy the relevance of the system itself, can occur only under stringent and unrealistic assumptions.
The unrealism of this fixed–time lag assumption can be seen in a different way. During the time lag between sale and purchase in the C-M-C circuit, wealth is being held, no doubt fleetingly, in the form of money. But if wealth can be held fleetingly in the form of money, then there is absolutely no reason why it cannot be held in the form of money more than fleetingly. There can indeed be a whole range of circumstances that induce people to wish to defer their purchases, to hold their sale proceeds in the form of money, though this would nullify Say’s Law.
Say’s law and the Walrasian equilibrium assume that such a denouement, namely an ex ante excess supply of produced commodities, never happens, not that it may happen at certain times while other things happen at other times. It follows that the most “mainstream” strand of economic theory, which is the Walrasian one, is based on a conception that is not possible in a money-using economy.
The Ricardian System
Ricardo wrote much on money, including on monetary issues being debated in his time. But we shall be concerned here with discussing his monetary theory not with reference to his position in monetary debates of his time, but with reference to his overall economic theory.
Adam Smith had put forward his theory of price. which has often been referred to as an “adding up” theory of price,4 in which the price of corn determines the prices of all other commodities. When the corn price rises, money wages rise to keep the corn wage intact; hence the cost of production rises everywhere, and this gets passed on in the form of higher “natural prices” of all commodities.
This theory, however, which said that a “general enhancement of the price of all commodities” occurs “in consequence of that of labor,”5 made the level of money wages and money prices indeterminate, even in a world where the corn wage was a given. Now, such indeterminacy in a paper money or credit money world, where the level of money wages and prices could be anywhere but happened to be at some perch at any given time, could be understandable. But Smith was not talking about a paper or credit money world.
He was talking about a commodity money world, which was relevant for his time, and which was clear from the use he made of Hume’s ingenious theorem on gold flows to refute the mercantilist claim that a nation grew richer by amassing precious metals.6 He argued based on Hume that if a country accumulated more gold, then its price level would rise (based on the Quantity theory of money), which would make its commodities uncompetitive, resulting in a current account deficit, which would be settled only through its gold flowing out. And in a commodity money world, Smith’s adding up theory could not hold. The value of money in such a world, where the production of the money commodity got augmented if it became more profitable, had to be linked to the conditions of production.
This was the starting point of Ricardo’s theory.7 In a commodity money world (or where paper money is convertible to the money commodity), money became part of the charmed circle of produced commodities, with the wage rate and the profit rate being identical between the money and non-money sectors (in a situation of free mobility of capital and labor across sectors). The equilibrium value of money, which constitutes the “center of gravity” of the “market price” of money, is given by its price of production, that is, the relative price vis-à-vis non-money commodities at which the profit and wage rates in the money sector are equal to those in the non-money sector (though Ricardo generally used relative labor values as an approximation for this equilibrium value).
It followed from this that when money wages rose, the rate of profit fell. Indeed. instead of all money prices rising as Adam Smith had suggested, if the money commodity was produced with labor alone, and all other commodifies required some means of production, then a rise in money wages would mean a fall in all money prices of all non-money commodities, which was the exact opposite of what Smith had suggested.
Ricardo’s monetary theory, however, had no role for any wealth demand for money, which also explains why Ricardo was an adherent of Say’s Law. His carrying over of the entire array of concepts—“market price,” “natural price,” and “center of gravity”—to his theory of money, which was the consequence of analyzing money as a produced commodity that was used only as a means of circulation, made him a monetarist in the short run though not in the long run.
This is because if perchance there was an increase in the supply of money, then the market price of money would fall, meaning that the market prices of all non-money commodities in terms of money would rise, exactly in accordance with what monetarism postulates. But this would mean a lowering of the rate of profit of the money-producing sector compared to the other sectors, and hence a shift of capital and labor away from producing money into producing other goods, which would lower the supply of money and thereby ensure once more that prices of production prevail and that the supply of money is no greater or no less than what is required for circulating commodities at those prices of production.
Therefore, in the short run, commodity prices depend upon the supply of money. In the long run, the supply of money adjusts to the demand, with commodity prices being equal to the prices of production. In Joan Robinson’s telling phrase, the Quantity equation MV= PQ is read by Ricardo from the left to the right in the short run, which is what monetarists do, but from the right to the left in the long run, which is of course contrary to monetarism.8
But a constant income velocity of circulation of money was central to Ricardo’s theory, both his theory of market prices of commodities in terms of money and his theory of prices of production, as indeed it was for Hume. James Mill, who was Ricardo’s interpreter and popularizer, tried to justify this assumption of a constant income velocity of circulation by suggesting that even though the different components of the money stock made different numbers of circuits in the process of circulation, there had to be an average number among them, a justification against which Marx had rightly remarked that an average did not mean a constant.9 And once this assumption of constancy of the income velocity of circulation is dropped, we have to reckon with a wealth demand for money, which makes a belief in Say’s Law logically untenable.
The Ricardian theory of money, like that of Walras, ignored the basic property of a money-using economy. Ricardo’s conceptual representation of capitalism consisted, in the
core of his theoretical discussion, of an isolated sector with capitalists and workers, and in analyzing it he abstracted, not surprisingly, from the central features of a money-using economy.
Exactly the same can be said of the Marshallian system, which we looked at in chapter 1 through the Cambridge equation. Though money in this system, as in the Walrasian, is not a produced good, so that there is no “natural price” of money and no need to distinguish between that and its “market price,” the assumption of a constant k (or a constant income velocity of circulation of money) rules out any wealth demand for money and hence any possibility of a deficiency of aggregate demand for produced goods. And whether it be in the Walrasian or the Ricardian or the Marshallian theories, ruling out any wealth demand for money and recognizing only a transaction demand for it, is logically untenable. This is because if money is held “fleetingly,” then there is no earthly reason why it cannot be held more than “fleetingly.”
The Keynesian System
The Keynesian system, like the Marxian one, rightly recognizes the wealth demand for money, because of which it rejects both monetarism and Say’s Law. According to Keynes, the value of money in any period is given not by the demand for and supply of money but by the level of the money-wage rate, that is, by the fact that the exchange ratio between a unit of money and a unit of one particular commodity, “labor-power,” to borrow Marx’s terminology, which is used directly or indirectly in the production of all produced commodities, is given.