Papers on Welfare and Growth

Tibor Scitovsky

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781135033019
• Edition: 1
• Topic: Negocios y empresa
• Subtopic: Negocios en general

The volume is divided into three parts: A: Economic Growth and Related Problems (covering international trade and economic integration, including a comparative study between Europe and America)B: Theoretical Welfare Economics (welfare propositions in economics, profit maximization and its implications and the Theory of Tariffs)C: Practical Welfare Economics (the price of economic progress, equity and international payments).

A

Economic Growth and Related Problems

Chapter 1

A Study of Interest and Capital¹

1940

The aim of this paper is to study the determination and interdependence of the rates of interest yielded by different kinds of securities. The complexity of the subject makes it necessary to divide it into three parts. In Part I we make an attempt to determine the price and rate of output of securities in a general way; and as securities are a commodity of which large stocks are always extant, we shall begin with a short discussion on the general theory of supply and demand of commodities of which large stocks are usually held, and so derive the price of securities in analogy with ordinary supply and demand analysis. It will be seen that in this way we arrive at a theory of interest rates which is fundamentally a generalization of Mr Keynes’s liquidity preference theory for any number of securities. In Part II we shall analyse the several criteria according to which securities differ from each other. This will lead us to a more complete understanding of the interdependence of different rates of interest and to a new explanation of the relative movements of long- and short-term rates. In Part III we shall restate the theory in a slightly different form and compare it with the ‘real’ theory of interest and capital.

It is possible to conceive of the rate of interest as being determined by the supply and demand either of capital goods, or of securities, or of money. In this paper we shall always think of it as being determined in the market for securities, and to make the analogy with the supply and demand analysis of consumers’ commodities easier, we shall have to think in terms of security prices as well as in terms of yields. This may be a little inconvenient at first because of the inverse relationship between the price and yield of a security, but I hope that the advantage of this approach will outweigh the additional trouble. The ‘physical’ quantity of securities will be measured by their nominal value.

In Parts I and II we shall deal generally with any number of different kinds of securities. Money is just one of these. Securities differ from each other not because the capital goods they represent are physically heterogeneous, but according to their promised yield, life-time and risk of default. For goods must always be differentiated from the buyers’ point of view, and yield, life-time and riskiness are the only characteristics of a security that matter to the investor. For example, beet and cane sugar, or natural and artificial rubber, are identical commodities if consumers do not or cannot distinguish between them; similarly, two securities are homogeneous whatever the nature of the capital goods they stand for if they are equally safe and offer the same income for the same period.² The meaning of these characteristics will be discussed in detail in Part II. Until then we shall simply have to accept the fact that securities differ from each other in several ways, just as consumers’ goods differ in taste, colour, consistency and a hundred other respects.

We shall have to make two simplifying assumptions, to be retained throughout this article. The first is the exclusion of ordinary shares and all other non-fixed-interest-bearing securities. This is admittedly a limitation. I believe that the co-existence of stocks and shares has some very important and interesting consequences, but I also think that this problem is additional to ours, in the sense that its solution would supplement rather than distort our results, so that its discussion may be reserved for some future date. For the time being our analysis will be considerably simplified by the absence of all non-fixed-interest-bearing securities.

Our second assumption is rigid money wages within the short period, in the sense that money wages in any short period are independent of changes in the same short period. Dropping this assumption would lead to interesting and rather unexpected results,³ but the discussion of such problems is beyond the scope of this paper. Besides, I believe the assumption to be a realistic one.

I

In the supply and demand analysis of an ordinary consumers’ good, supply is represented by the flow of production, demand by the flow of consumers’ demand—both functions of price—and the quality of these two is said to determine the equilibrium price and flow of output. This is a legitimate approach to reality in the case of many commodities which cannot be stored or the stocks of which are small and independent of price—depending only on storage costs and the rate of turnover. When, however, the volume of stocks is a function of price and is large relatively to current production,⁴ then supply will no longer be equivalent to production, but will—at certain prices—be temporarily augmented by a reduction, or temporarily reduced by an accumulation, of stocks; and this factor may become so important relatively to current production as to render the above picture of the determination of short-period equilibrium incorrect and misleading.

The reason for a general theory of stocks being as yet nonexistent seems to lie in the fact that the desire to hold stocks is usually a very complicated function of price, involving derivatives with respect to time and relative prices at different dates.⁵ For the following argument, however, we shall make the simple assumption that the form of this function is exactly similar to an ordinary demand function, i.e. that the desire to hold stocks is a diminishing function of price. This will be shown to be true of the demand for holding securities, and in the short run it also represents—at any rate approximately—speculative demand for commodity stocks whenever speculation is based on the notion of normality.

Let us consider an equilibrium situation where price is such that current production equals current consumption and the volume of stocks corresponds to the volume demanded by ‘stockholders’ at that price. Then assume a shift in consumers’ demand. In the absence of stocks there would be an immediate change in price, followed by the adjustment of production, and a new equilibrium would be established at the new point of intersection of the two curves. In the presence of stocks, however, the situation will be different, because the establishment of the new equilibrium will be retarded by the adjustment of the volume of stocks to the new price. So a diminished consumption demand (leftward or downward shift of the consumption curve) will be temporarily supplemented by the demand of stockholders who want to increase their stocks as price falls⁶—thereby checking the fall in price and the consequent adjustment of production.⁷ Similarly, an increased consumption demand will be temporarily satisfied—and the rise in price and production checked—by the dishoarding of stocks. The nature of this retardation can best be described by saying that the existence of stocks sets a limit to the time-rate at which price can change, the limit depending on the size of stocks and the price-elasticity of stockholding. For to each price change there corresponds a certain quantity of stocks which must be released (if price rises) or accumulated (if it falls) before the new equilibrium price can be established; and given the price change, this will be the greater the more elastic the demand for stockholding and the greater the total quantity of stocks. The time needed for the adjustment—and for the new equilibrium to be established—will be the longer the greater is this quantity in relation to the gap between current production and consumption opened up by the initial change in data (and not immediately closed by the requisite change in price), since—to put it crudely—stocks can only be released or accumulated through this gap. If this ratio is very large, because, for example, stocks are very large, then the adjustment of price will proceed so slowly that we can say the influence on price of primary changes in production or consumption is negligible in the short run, and it will be more correct to say that price depends on the quantity of stocks than that it is determined by the intersection of the production and consumption curves.

If the existence of stocks keeps the price of a commodity temporarily at a level where current production and consumption are not equal, this disparity will cause a change in the level of income which may—if the commodity is important and its stocks large—equate its production and consumption before price has had time to equate them. The way in which this happens has often been described and is known as the theory of the multiplier. A disparity in any particular industry between receipts and income paid out causes a similar but opposite disparity in the rest of the economy. This will disappoint (favourably or unfavourably) the expectations of entrepreneurs in those industries and induce them to change their rate of production and with it the rate at which they pay out income to the owners of the factors of production. This process is cumulative and will continue while the disequilibrium in the first industry subsists, i.e. until current demand is equated to current production, partly by the change in incomes causing a change in demand, partly—to the extent that stocks have adjusted themselves in the meantime—by the change in price.

For a further elucidation of the above argument let us take a specific example. Imagine a community which always spends all its income and has a margin of unemployed resources.⁸ We further assume that there is speculation in one of the staple commodities consumed by this society—let us call it wheat—and that the average value of speculative wheat stocks is large relatively to the value of all other speculative stocks taken together. Now imagine a shift in consumers’ demand away from wheat. The prices of all the goods which now are demanded more urgently will rise, causing an increased production and a higher level of employment in the industries affected. The price of wheat would fall and cause an offsetting diminution of production and employment in the wheat industry if it were not for the existence of speculation. But when the price of wheat begins to fall, people, who believe that the former price of wheat was ‘normal’ and therefore expect it to return after a while, will find it profitable to hoard wheat, thereby checking the fall in its price and production. So we get a net increase in employment in the community, and corresponding to it a rise in the level of incomes. This will increase consumers’ demand for most commodities. The increased demand for goods other than wheat will cause employment and incomes to rise yet further; the increased demand for wheat will check and ultimately stop the accumulation of wheat stocks. Short-period equilibrium will be re-established when incomes have risen sufficiently to make the consumption of wheat again equal to its production in spite of the change in tastes. This will probably happen at a wheat price somewhat below the original, corresponding to the now greater wheat stocks. It should be obvious that the argument for a shift in demand towards wheat is perfectly symmetrical.

All this, of course, will happen only if prices in those other industries are flexible. If they were as rigid as the price of wheat (e.g. because of the existence of equally large stocks), then the disequilibrium between receipts and outgoings in the wheat industry caused by the adjustment of its stocks would merely cause an exactly offsetting adjustment of stocks in all the other industries, without inducing entrepreneurs to any action. Thus the degree to which the equality between the current demand and current production of any particular commodity is brought about by a change in the level of income rather than by a change in its price, depends on the value of its stocks relatively to the value of all other stocks taken together. The greater the relative value of its stocks, the more sticky will be its price, and the more quickly will the level of income adjust itself through changes in the activity of other industries.

The reader may object here that speculators will not keep stocks at their unnaturally high (or low) level indefinitely; in fact, that speculators’ demand for stocks cannot be represented by a simple diminishing function of price. He will be perfectly right. The change in the level of income described above is only a temporary effect of the shift in consumers’ tastes, and the whole argument may be taken as an illustration of the harmful destabilizing nature of speculation. But it is also an illustration by analogy of the permanent effect on the level of income of changes in the propensity to consume. For it will be shown on p. 20 that unlike commodity stocks, the stocks of securities are simple decreasing functions of price, so that as regards the latter it will be true to say that they determine security prices while the level of income equates the flow of saving (‘current consumption’) to the flow of investment. This is the more true because securities are the commodity with large stocks par excellence. Another respect in which securities differ from other goods is that the current demand for them represents not consumption, but what may be called normal additions to stock. It will become apparent that this fact cannot affect the argument, but it will make the exposition easier if we assume for the moment that new securities are bought out of new savings for an entirely different purpose from that which induces people to hold the stock of ‘secondhand’ securities. (In Part III the same argument will be repeated without this analytical dodge.) In this way we can split up the individual’s economic problem into two parts: firstly he has to decide in what forms to hold his already accumulated stock of wealth, secondly he has to allocate his flow of income between the flows of different kinds of consumers’ goods and securities.

The second of these, to be dealt with later, is the familiar problem of the subjective theory of demand and is generally considered to be solved. The first is exactly analogous to the second. Each security promises a certain yield, distributed over time in a certain way, and each promise inspires a certain degree of confidence.⁹ The individual aims at the highest possible average yield, subject to the limitations of his wealth and preference for safety and liquidity. At each constellation of market prices, which to him are given, there exists an optimum selection of securities which, in his opinion, involves the exact amount of risk and illiquidity he is willing to bear and at the same time maximizes his average yield.

From the community’s point of view the quantity and not the price of each security is given in the short period, and these, together with the individuals’ preference scales and the level of income determine security prices.¹⁰ Just as in the general theory of demand for consumers’ goods, it can be shown that, neglecting capital-gain and capital-loss effects (these correspond to the income effects of the theory of consumers’ demand), the price of each security is a diminishing function of its quantity, while an increase in the quantity of another security will lower or raise its price according as the securities are substitutes or complements.

The effects of variations in the quantity of money need special consideration. They can be analysed most easily if we think of the price of a security as the ratio of its marginal utility to the marginal utility of money, the numéraire. Then an increase in the quantity of money will lower the marginal utility of money, i.e. lower the value of the denominator in the expression for price, and raise or lower the value of the numerator according as the security in question is a complement or substitute of money. So an increase in the quantity of money will cause all security prices to rise, those of complements most and those of substitutes least. In the limiting case the price of a perfect substitute for money would remain unchanged, because both the denominator and numerator of the expression for its price would fall in the same proportion.

This is little more than a generalization of Mr Keynes’s liquidity preference theory of the rate of interest. In the special case when there are only securities of one kind besides money, the rate of interest is an index of their relative prices which is a function of their relative quantities. The rate of interest, therefore, will rise equally whether the quantity of securities is increased relatively to that of money, or the quantity of money is diminished relatively to that of securities. For short-period analysis it is better to think of the rate of interest as a function of the quantity of money, simply because a change in the quantity of money is easily conceivable within the short period while a significant change in the quantity of other securities is impossible by definition.¹¹

When we distinguish between different kinds of securities, the crude picture of the rate of interest as the demand price of money breaks down because of the multiplicity of interest rates. But it still remains true that the quantity of each kind of capital and h...