9. "The Propensity To Consume": I

0:00Chapter 9 The Propensity to Consume, Part 1 Digression on Mathematical Economics When we come in the general theory to the two chapters on the propensity to consume, we meet all our previous difficulties magnified. 1. A specialized and self-coined technical vocabulary to cover complex concepts, which, however, are never consistently adhered to. 2. Loose, unverified, unverifiable or meaningless statements. 3. A constant confusion or scrambling of cause and effect.

0:494. The same aversion to and derision of anything resembling individual thrift, prudence or forethought that was evident 16 years previously in the economic consequences of the peace. In the general theory, in brief, Keynes did not suddenly discover that the traditional economic virtues were really vices and vice versa. He had practically always thought so. All that he hit upon was a new rationalization for his old bias. The ultimate object of our analysis, he begins, is to discover what determines the volume of employment.

1:35So far we have established the preliminary conclusion that the volume of employment is determined by the point of intersection of the aggregate supply function with the aggregate demand function, page 89. Here we meet two special Keynesian technical terms, so we look back to remind ourselves what they mean. And on page 25, we find the so-called definition, Let z be the aggregate supply price of the output from employing n men. The relationship between z and n being written z equals phi n, which can be called the aggregate supply function. Similarly, let d be the proceeds which entrepreneurs expect to receive from on the employment of n men, the relationship between d and n being written d equals f n, which can be called the aggregate demand function.

2:41Here we have not so much two definitions as two mathematical equations, each of which expresses a complex relationship. If the volume of employment is determined by the point of intersection of the aggregate supply function with the aggregate demand function, then the volume of employment depends upon the relationship between two complex relationships. This ought to be difficult enough to keep in mind, but our troubles have only begun. For, if we go back to Keynes's definitions on page 25. We find that while the aggregate supply function depends upon the relationship between two actualities, supply price and number of men employed, the aggregate demand function depends upon the relationship between an expectation of proceeds and an actuality, number of men employed.

3:43Most logicians or mathematicians would doubtless agree that some actualities could be equated with other actualities, and some expectations with other expectations. But I doubt whether many would agree that expectations could be mixed up with actualities in the same mathematical goulash, or that the resulting equations could have any precise and if a mathematical equation is not precise, it is worse than worthless. It is a fraud. It gives our results a merely spurious precision. It gives an illusion of knowledge in place of the candid confession of ignorance, vagueness or uncertainty which is the beginning of wisdom.

4:36A short digression seems desirable at this point, not merely on Keynes's mathematical economics, but on mathematical economics in general. It is said in defense of mathematical economics, by, for example, Keynes's father, John Neville Keynes, in The Scope and Method of Political Economy, that That, Exact numerical premises are not always essential to the employment of mathematical methods, page 257. Mathematical methods, in other words, can still be useful in economics even when they do not deal with known or even determinable quantities, but are a means of clarifying merely hypothetical relationships.

5:27Francis Edgeworth, for example, in his Mathematical Physics, 1881, writes,

5:57To take a trivial instance, A is greater than B, and B is greater than C, therefore A is greater than C. Here is mathematical reasoning applicable to quantities which may not be All this is doubtless true, but the mathematical economists who make such points then tend to forget that out of a merely hypothetical equation or set of equations, they can never pull anything better than a merely hypothetical conclusion.

6:47As Whitehead remarks, The conclusion of no argument can be more certain than the assumptions from which it starts. If mathematicians cannot in some way or other determine the numerical values of their x's and y's, their equations are useless for applied or applicable economics. And Keynes's general theory professes to be a theory applicable to real situations. He does, in fact, constantly profess to apply it to real situations. But we may go much further in our criticism. Even a merely hypothetical equation may be worse than worthless A mathematical statement to be scientifically useful must, like a verbal statement, at least be verifiable, even when it is not verified.

7:54If I say, for example, and am not merely joking, that John's love of Alice varies in an exact and determinable relationship with Mary's love of John, I ought to be able to prove that this is so. I do not prove my statement, in fact, I do not make it a whit more plausible or scientific, if I write solemnly, let X equal Mary's love of John and Y equal John's love of Alice, then Y equals FX, and go on triumphantly from there. Yet, this is the kind of assertion constantly being made by mathematical economists, and especially by Keynes.

8:42The model was set by Augustin Cournot in his L'écheve sur le principe mathématique de la cellulite de la tisse published more than a century ago, in 1838. Let us admit therefore that the sales or the annual demand d is, for each article, a particular function, fp, of the price, p, of such article. And he went on to explain how a curve can be made to represent the function in question. It is from this that the famous Marshallian supply and demand curves later developed, and today's immense and bewildering body of mathematical economics.

9:30Yet there is no proof whatever that even the most elementary of these functional economic equations represents a fact of the real world. There is no proof that demand is a particular function of the price of a particular article. We can, of course, assume such a relationship. We can draw a hypothetical demand curve, and derive from it a hypothetical functional relationship between demand and price. We can then point out that according to our hypothetical curve and hypothetical table, when the hypothetical price of our hypothetical commodity is x, the amount demanded will be y.

10:16When the price is X1, the amount demanded will be Y1. When the price is X2, the amount demanded will be Y2, etc. We can, in short, assert that a general hypothetical relationship implies specific hypothetical relationships. But of what practical use will all these deductions be? Is there any way in which we can apply them to the real world? When we ask this, we find that our mathematical equations are of very little use indeed. For whether our hypothetical demand curve corresponds to any real demand curve, we can never know. Our equation is useful only on the completely unreal assumption More unreal still, our equation is valid only for one state of demand, which cannot be assumed to exist for more than an instant of time.

11:28We can never tell whether a real change of price, between, say, today's closing price of wheat and yesterday's closing price of wheat, is the result of a change, or supposed change, in supply, or the result of a change in the state of demand, or demand curve. In short, we can draw all the beautiful supply and demand curves we like and cross them at the points that please us most. We can thus help to clarify ideas for college freshmen and even for ourselves, but we constantly run the danger of deceiving ourselves by our own diagrams, of giving ourselves the illusion that we know what we in fact do not know.

12:16For these supply and demand curves are merely analogies, metaphors, visual aids to thought, which should never be confused with realities. We never in fact do know what the present demand curve or demand schedule is for anything. We can only guess. Historical research or past statistics, and all statistics describe events in the past, may help entrepreneurs to do this guessing and reduce their range of error, but such statistics can never enable entrepreneurs to know any future relationships of price to demand or enable economists to predict it with confidence.

13:04Supply and demand curves and functional equations assume that an infinite number or an indefinitely A fairly large range of simultaneous relationships can be known, and even known in advance, but as the physicists would say, this is not an operational concept. All that we can know is that, say, in 1956 the average price of wheat was P and the supply was S, whereas in 1957 the average price of wheat was P minus P1 and the supply was S plus S1. But were the supply curves and the demand curves exactly the same in 1956 and 1957?

13:53Are the differences caused merely by the supply curves crossing the same demand curve at different points? This we can never know, and the obtainable market data will never tell us. Nor do we improve the situation from a theoretical or mathematical standpoint, when instead of average annual prices we take average monthly or weekly or daily prices, or prices from hour to hour or minute to minute. As a matter of fact, on the organized speculative markets, we do not find that when the price of a commodity or a share goes up, the amount sold invariably falls, or that when the price goes down, the amount sold invariably increases.

14:42Often, the price and the amount sold will both increase, or the price and the amount sold will both decline. A typical supply and demand curve exposition tacitly assumes, say, that a demand curve remains fixed while a supply curve moves up and down and crosses it at different points, which constitute the changing prices. But the truth is that the level and shape of the supply curve, and more particularly of the demand curve, are themselves constantly changing from hour to hour. If they could really be discovered and put on a motion picture film, we might find them writhing, vibrating, and jumping in a way to discourage even the cockiest mathematical economist.

15:34The stationary supply and demand curves of the textbooks, or the few alternatives shown, are grossly simplified static assumptions, and should never be taken for more than that. The attempt to put demand and price into a functional equation, and worse than this, the belief that in real life we could discover a set of actual numerical values to which such an equation would apply, is chimerical. A little mathematical economics here and there, to repeat, or occasional diagrams, may be useful for clarifying or generalizing some economic concepts, for making them more precise, or for protecting us against some errors that otherwise might not be recognized.

16:23But the great bulk of mathematical economics today is a manipulation of hypothetical abstractions that have no useful application to the real world. Keynes does not advance in the slightest beyond Cournot in setting up his own functions and his own formulas. In fact, as we shall see, he goes backward. His equations are not merely unverified and unverifiable, they are invalid or inadmissible in other ways. Let us begin, as an example, with is that entrepreneurs practically never think or act in the way Keynes implies.

17:32The entrepreneur usually begins by trying to determine what his net income will be from producing a certain quantity of a certain product and selling it at a certain price. Only when he has made this estimate does he decide how many men will be needed to turn out this product. How many men he hires or keeps, moreover, will also be determined heavily by the wage rates he is obliged to pay. Instead of thinking what his gross proceeds will be from hiring so and so many men, he decides how many men he will have to hire, or how many he can afford to hire at a given wage rate, to acquire a certain net income.

18:19His decision will also be governed, of course, by how much capital he has, or can borrow. But a Keynesian is never allowed to look at the matter the way an entrepreneur looks at it. Under threat of excommunication, he is not even permitted to hint that the amount of employment will have anything to do with wage rates. That unemployment might be primarily the result of excessive wage rates in relation to prices, or the Demand for Products is the very doctrine that Keynes started out to disprove and to ridicule. Thus, there is no reason to suppose, and there is the most serious reason to doubt, that the causal relationship is the one tacitly assumed by Keynes in the equation d equals f n.

19:11Nor is there any reason to suppose that the equation expresses a truth. And there are too many factors, tangible and intangible, which entrepreneurs and consumers must take into account in their plans, which do not get into the equation. And there is no way of showing or knowing, even when an infinite number of other factors are assumed to remain unchanged or equal, that the functional relationship expressed in the equation actually exists. How could this functional relationship be proved? We have already seen that this was impossible, even in the simpler and more orthodox functional relationship postulated by Cournot.

19:59The Keynesian functional relationship cannot, in fact, be proved. It can only be arbitrarily and dogmatically asserted. And this is typical of practically all the Keynesian equations. The Fundamental Psychological Law Because of the foregoing digression on mathematical economics, it has seemed to take unconscionable space to analyze even a few Keynesian paragraphs. But we have really made far more progress than the distance so covered in Keynes's first chapter on the propensity to consume may imply. For once we have recognized the slipperiness, vagueness, and changeability of most of Keynes's basic terms and concepts, and his habit of begging the question by unproved mathematical equations and sheer assertion, we can begin to economize in our analysis.

21:01Still on the first page of the first propensity to consume chapter, we come to the assertion Note that the factors which govern two quantities, the sum which will be spent on consumption when employment is at a given level, and the sum which will be devoted to investment, are largely distinct, therefore we are told these two quantities will be discussed not only in separate chapters, but in separate books of the general theory, pages 89 through 90. But if we go back to Keynes's own equation that income equals consumption plus investment, or if we turn merely to common sense, we might conclude that every dollar of income spent on consumption goods must leave a dollar less to be spent on investment goods, while every dollar spent on investment must leave a dollar less to be spent on immediate consumption.

22:02How the factors governing two quantities related to each other as subtrahend and remainder could be largely distinct is puzzling, but we shall postpone consideration of this paradox to a later stage. We come at last to the famous Keynesian concept of the propensity to consume. We will therefore define what we shall call the propensity to consume as the functional A Real Relationship Between YW, A Given Level of Income in Terms of Wage Units and CW, The Expenditure on Consumption out of that level of income, page 90 Keynes creates several difficulties here as usual.

22:49is trying to define the relationship of consumption to income in real rather than in monetary terms. And therefore he talks in terms of wage units. But we have seen, page 64, that Keynes so defines the wage unit as to make it in fact a concept that can only be conceived in monetary terms. These wage units are added together exactly in proportion to monetary wages, because an hour of special labor remunerated at double ordinary rates will count as two units, page 41. We must forget wage units, therefore, in order to substitute what Keynes thought he was comparing, which is real consumption with real income.

23:40The difficulty to be dealt with is the misleading nature of the term itself, propensity to consume. Keynes's definition shows that he is not in fact dealing with a propensity in the dictionary sense, a natural or habitual inclination or tendency. He is dealing with a mathematical relationship. He is dealing with the fraction or percentage of its income that the community in fact spends on consumption goods at different levels of income, regardless of what its propensities or inclinations may be. If the community spends 90% or 9 tenths of its income on consumption, then its propensity to consume is 9 tenths or 9 over 10.

24:31But the term, it can now be seen, is doubly misleading. The propensity part of it is not a propensity, but a function or a fraction. And we are not even talking about the fraction of income that is spent, but only the fraction that is spent on consumption goods. If a man buys a capital good, this is not counted in his propensity to consume, Because this purchase is called an investment. If a doctor, however, buys a house, and this house is both his residence and his office, how is this term entered on the Keynesian ledger? As part of the doctor's propensity to consume, or as an investment?

25:16If you buy a power mower strictly to use on your own lawn, it is doubtless part of your propensity to consume. If you buy it to rent out to others, it is an investment. But what is it if you use it partly for your own lawn and partly to rent out to others? Such questions are enough to show that the line between consumption goods and capital goods, between consumption and investment, is not as clear and sharp as Keynes's elaborate theoretical division implies. But they indicate also that from a practical standpoint, it is irrelevant to the immediate total volume of employment, whether a given amount of money is spent on consumption or on capital goods, whether it is to be included under the propensity to consume or under investment.

26:14In fact, they raise the further question whether, from the standpoint of the immediate effect on the volume of employment, there is any difference between the dollars that go into consumption and the dollars that go into investment, and therefore, whether there is any good reason for dealing with each by two separate sets of equations, or for that matter, whether there is any good reason for the whole elaborate structure of Keynesian theory. Of course, it makes a great deal of difference to the direction or allocation of employment whether money is spent on consumption or on capital goods, but it also makes a good deal of difference to the direction or allocation of employment whether money is spent on one consumption good rather than another, say on houses rather than on automobiles, or on on Beef rather than on Television Machines.

27:15At this point, Keynes digresses to discuss the Rate of Interest, but it will help the exposition here if we do not follow him in this digression but reserve criticism of his interest rate theories until we come to his chapters 13 and 14 wholly devoted to the Rate of Interest. The next point for us to pass on to, therefore, is Keynes's fundamental law upon which his confidence in the alleged propensity to consume function is based. The fundamental psychological law upon which we are entitled to depend with great confidence both a priori from our knowledge of human nature and from the detailed facts of experience is that men are disposed, as a rule and on the average, to increase their consumption as their income increases, but not by as much as the increase in their income.

28:15That is to say, if Cw is the amount of consumption and Yw is income, both measured in wage units, the function of Cw has the same sign as the function of Yw, but is smaller in amount, i.e. dcw over dyw is positive and less than unity. Page 96. Before starting to analyze this, it is important to emphasize the central role that this alleged consumption function plays in Keynesian economics, and the huge importance his admirers and disciples attach to it. Keynes's most notable contribution, writes Alvin H. Hansen, was his consumption function.

29:06The behavior patterns of the community are such that a gap exists, which gap widens absolutely as real income increases, between the amount the community wishes to consume and the output the community is capable of producing. But let us allow Keynes to explain a little further exactly what he means. This fundamental psychological law is both a secular and a cyclical law, and is apparently considered even more important as an explanation of business cycles than as a secular law. Continuing without gap from the quotation from Keynes above, this is especially the This is the case where we have short periods in view, as in the case of the so-called cyclical fluctuations of employment.

29:59Thus, a rising income will often be accompanied by increased saving, and a falling income by decreased saving. But apart from short period changes in the level of income, it is also obvious that a higher absolute level of income will tend, as a rule, to widen the gap between income and consumption. These reasons will lead, as a rule, to a greater proportion of income being saved as real income increases. But whether or not a greater proportion is saved, we take it as a fundamental psychological law of any modern community, that when its real income is increased, it will not increase its consumption by an equal absolute amount, so that a greater absolute amount must be saved, page 97.

30:52Now, how does Keynes try to establish this great discovery, this fundamental psychological law, on his mere say-so? Had he put this forward merely as a loose common-sense observation with no mathematical equations based on it Of course the consumption function is normally less than unity, page 96. This is merely a pretentious way of saying that a community cannot as a whole consume more than it produces, and that any community that has advanced beyond the most primitive and miserable state will save against contingencies at least part of what it produces and invest something in land improvement and tools to increase its future production.

31:53It is also true that richer people will, as a rule, save a greater proportion of their incomes than poorer people for the simple reason that they have more above their present bare minimum consumption requirements to save, and some saving is the course dictated by common prudence. But such truths in this loose general form have been known from time immemorial. They were embedded in Proverbs long before the birth of Adam Smith, and no one has hitherto called them a great economic discovery. But let us listen again, for a moment, to Hansen. Even a little reflection on the course of economic history is enough to disclose the unmistakable fact that consumption has risen, broadly conceived, more or less in proportion to the spectacular growth in productivity which the last 150 years have witnessed, whoever supposed otherwise.

32:56To this general knowledge, long and widely held, Keynes indeed added something very important, namely the precise formulation of the consumption income schedule, together with the concept of the marginal propensity to consume. And more significant still, he developed a theory in which this and other functions relevant to the determination of aggregate demand are integrated. In brief, Keynes took a loose truism that everybody knew and turned it into a double error, first by attributing a precision to it that it simply does not have, and secondly by making it the basis of a false theory. I have said that Keynes tried to establish his fundamental psychological law by mere assertion, by an ipsy dixit.

33:48There are two major ways in which an economic law, or any scientific law, might be established, deductively or inductively. In the deductive hypothetical method, a truth is established because it is a demonstrable implication of another truth already known or postulated. This is the kind of truth embodied in the ordinary laws of supply and demand. If the supply of an article remaining unchanged, the demand for it increases, the price will rise. If, with no increase in the demand for an article, the price is raised, less of it will be bought, etc. The conclusion follows inevitably from the premises, and from the very meaning of such terms as supply, demand and price, or if one wishes to be more precise and technical, of such terms as the curve of price and amount demanded, the curve of price and amount offered, etc.

34:53In economics, induction consists usually either of common observations or of statistics. Economic truths, like truths in almost all fields, are never established by pure induction or pure deduction, but by a combination of both methods. From hypothetical premises, only hypothetical conclusions can be drawn. For a conclusion to be true and to have practical application, the premises must be known, from common observation or otherwise to be true. In confirming a conclusion, statistics at best present nothing more than presumptive evidence, for they can never be complete.

35:39Now, though Keynes declares that his fundamental psychological law of the relation of consumption to income can be arrived at not only a priori, but, from the detailed facts of experience, He never deigns to offer the slightest statistical confirmation. We can only assume that this is because he has none to offer. When we do look at available statistics, we find in fact a prima facie refutation of his fundamental psychological law.

36:27Here is a table taken from official statistics showing the disposable personal income in In the year 1944, income was $146.8 billion, savings were $36.9 billion, and savings as a percent of income was 25.2%.

37:03In the year 1945, income was $150.4 billion, savings were $28.7 billion, and savings as a percent of income were 19.1%. In the year 1946, income was $159.2 billion, savings were $12.6 billion, and savings as a percent of income was 7.9%. In the year 1947, income was $169 billion, savings were $4 billion, and savings as a percent of income were 2.4%.

37:51In the year 1948, income was $187.6 billion, savings were $10 billion, and savings as a percent of income were 5.3%. In the year 1949, income was $188.2 billion, savings were $7.6 billion, and savings as a percent of income were 4%. In the year 1950, income was $206.1 billion, savings were $12.1 billion, and savings as a percent of income was 5.9%.

38:39In the year 1951, income was $226.1 billion, savings were $17.7 billion, and savings as a percent of income was 7.8%. In the year 1952, income was $236.7 billion, savings were $18.4 billion, and savings as a percent of income was 7.8%. In the year 1953, income was $250.4 billion, savings were $19.8 billion, and savings as a percent of income was 7.9%.

39:28In the year 1954, income was $254.8 billion, savings were $18.3 billion, and savings as a percent of income was 7.2%. And in the year 1955, income was $269.4 billion, savings were $17.1 billion, and savings as a percent of income was 6.3%. Now let us see what these figures do to Keynes's alleged psychological law. Law. The events of 1955 were in themselves an emphatic contradiction.

40:15Disposable personal income increased by $14.6 billion, but savings fell by $1.2 billion. The total percentage of saving to disposable income fell from 7.2% to 6.3%. The same thing happened between 1953 and 1954. Disposable income went up $4.4 billion, savings down $1.5 billion. How many times Keynes's law was falsified in this 12-year period depends on how the law is interpreted. Total income went up each year as compared with the preceding year. If we take the amount of saving in 1944 as our base figure, therefore, the law was falsified in every one of the succeeding 11 years, for income was higher in each than in 1944, but savings were lower.

41:16If we interpret Keynes's law or consumption function, as he calls it, to mean merely that savings must rise in absolute amount when incomes rise, but only as compared with the preceding year, then Keynes's consumption function was falsified in six years and worked in only five. If we interpret the consumption function to mean that savings will rise as a percentage of income with every rise in income, then Keynes's law was falsified in seven of the eleven years and followed only in four. I may be accused of unfairness for including the enormous savings in 1944 and showing their decline in 1945, 1946 and 1947 in spite of substantial rises in disposable income in each year.

42:12It will be said that savings were heavy in 1944 and 1945 because these were war years and consumer goods were not available. This is indeed part of the answer, but this only underlines the fact that Keynes's law is no law, and that the relationship of spending and saving does not depend solely on total income changes, but on innumerable factors. Savings may depend less on what people earn today than on what they expect to earn tomorrow. Their spending this year may depend to a large extent on whether they expect the prices of and the things they want to buy to be higher or lower next year.

42:58People may buy on impulse. They may refrain from buying through loss of confidence, either in the general business future or in their own. These reasons for spending or not spending will be discussed more at length later. It is sufficient to note here that experience and statistics fail to support the consumption and Function, which Alvin H. Hansen regards as Keynes's most notable contribution. There are, I suppose, various ways of refusing to accept the kind of statistical refutation I have just presented. One can say that Keynes was not talking about the relation of money-income to money-savings, but of real incomes to real savings.

43:48Changing the foregoing income figures to figures in terms of constant dollars, e.g. 1956 prices, however, does not change the result much. Only in two years, 1946 and 1947, did the rise in money income cover a fall in real income. One can argue also, as Hansen apparently does, somewhat obscurely, that when the consumption function fails to work out statistically, it is because it has shifted. It still exists, and it is a major landmark in the history of economic doctrines. But if the consumption function is always shifting or can be upset completely by economic upheavals, then its existence can neither be proved nor disproved statistically, or in any other way, and it is worthless for cyclical or even secular prediction.

44:50Ambiguity of the Consumption Function Before we leave the consumption function, it may be as well to call attention to the uselessness and illegitimacy of the concept in other ways. The concept is highly ambiguous in what it tells us about the relationship of consumption and income. On page 116, I present a diagram to illustrate five main functional relationships that saving can bear to total income. Keynes could easily have clarified this point for himself and his readers by using a similar simple diagram, but though he peppers the general theory with functional and differential equations, he seems to have despised any simple clarifying device and uses only one diagram in the entire book.

45:45For reasons which I shall explain more fully later, it seems to me much more profitable to discuss the relation of savings to total income than the relation of consumption to total income. Therefore, I have labeled the lines in the diagram S1, S2, etc., rather This line, however, merely represents the dividing point between consumption and savings. The section below the line represents consumption, and the section above the line represents savings. The line yy' represents the line of total income.

46:32The vertical line OY represents total income in the year of origin, the vertical line XY' total income in the final year of the diagram, the horizontal line OX the length of time in years over which the rise of income takes place. The various S-lines representing the dividing lines between savings and consumption all begin at the two-thirds point of the origin year income line O-Y. This assumes that in the initial year under consideration, saving is equal to one-half of consumption. This is, of course, a highly unrealistic assumption, because savings normally run, as we shall see, in the neighborhood of about one-tenth of total income.

47:24But the two-thirds point is taken in the origin year simply to make it easier for the eye to follow the various savings lines and not to crowd them into too narrow a space. For the same reason, the unrealistic assumption is made that the national income in the terminal year of the diagram is three times as great as in the origin year. Now the line SS1 represents what Keynes appears to be saying in his Psychological Law, namely that as total income increases, the proportion of saving to spending increases. In the long run, and in the absence of innumerable other disturbing factors, this is indeed what tends to happen.

48:12It is also, as I shall later show, what both the saving individual desires to happen, and what is most beneficial for the community. If this were what Keynes was saying in his consumption function, and if it were all that he was saying, it would have been true. would also have been a truth generally recognized not only prior to 1936, but prior to the birth of Hume and Adam Smith. Those who have more income than they must spend for the immediate necessities can afford to save something out of the rest. The more income they have above their immediate necessities, the greater proportion of it they can afford to, and the greater The same broad relation of savings to income applies both to individuals and to a whole community.

49:09But it applies as a loose rule of thumb generalization, and to the extent that it is true, it must remain one. The moment it is put into a mathematical equation, as Keynes attempts to put it, it becomes false. The mathematical precision is spurious. It is useless for practical application or short-run analysis, because the rise of income is only one among many factors, most of them intangible, that determine short-run changes in the volume of saving. And it is in any case astonishing, as I shall show, to regard the tendency toward increasing proportional saving when income increases as an ominous development that threatens to create secular unemployment and poverty.

49:59The line SS2 represents what could be meant by Keynes's psychological law. It represents approximately what does happen in reality over a series of years when the increase in income is not substantial. People tend to save about the same proportion of their income from year to year. This situation is illustrated by the diagram on page 119, which covers the years divided into quarters from 1951 to 1957 in the United States. The diagram is reproduced exactly without change of proportions from an official diagram published by the President's Council of Economic Advisers in the April 1958 issue of Economic Indicators.

50:49The chart shows that savings, while they fluctuated mildly over this period, showed no consistent tendency either to increase or decrease, but in general remained at the same proportion of total income, an average of about 7%. This is in line with what previous attempts to measure savings statistically have shown, though the proportion of savings to income depends on the particular ways in which savings and income are defined and measured. Thus, a study by Koznitz published in 1940 tended to show, in the words of Alvin H. Hansen, that, The percent of income saved and invested over the long run has been more or less constant at say around 12%.

51:41This would look on the surface like a stable situation and a healthy and progressive one. This saving and investment constantly increases the total amount of consumption and capital goods produced. The producers both of consumption and of capital goods could count, by and large, in such a The proportion of income saved remained substantially constant, but at higher absolute levels of If the gap in absolute terms between consumption and income widens as income increases, then aggregate demand will not be adequate to cover the aggregate supply price unless that gap is filled by an increase in investment.

52:59We shall later inquire whether or not Keynes and Hansen were unduly terrified by the prospect that saving might rise proportionately with income. On the same reasoning, Hansen should still be worried even if savings should be represented by the relation SS3, for though as income increased a smaller percentage of income would can be saved, it could still be greater in absolute amount than when income was lower. Only if the relationship were represented by the line SS4 might the souls of the Keynesians be at peace. Only then might the private enterprise system be allowed to take care of itself without the solicitous forced spending of the Keynesian bureaucrats to fill the gap.

53:50In fact, the Keynesians might be even more at peace if the relationship were represented by the line SS5, in which savings would grow smaller and smaller even an absolute amount as income increased. The Meaning of Saving But now a tiny doubt, at first no bigger than a man's hand, begins to grow and grow. If we accept Keynes's concept of saving, which is quite unrelated to his formal definition of saving in the general theory, can the Keynesian soul remain at peace as long as there is any saving at all? Or, to put it more accurately, as long as there is any saving of any amount that is not compensated or offset by an equal amount of investment? For though Keynes tells tells us in his formal definitional chapters that he rejects the new-fangled view that there can be saving without investment or investment without genuine saving, page 83.

54:57And though he himself insists in these chapters on the identity between saving and investment, page 84, the whole Keynesian theory of unemployment rests on the abandonment of these definitions Definitions and Concepts, and to return precisely to the now tacit definitions and concepts that he used in the treatise on money, and that he formally abandons in his definitional chapters of the general theory, with much regret for the confusions they have caused. Page 61. The real, unstated definition and concept that Keynes uses in his Unemployment Theory is that saving is the merely negative act of failing or refusing to spend money on either consumption or capital goods, directly or indirectly.

55:50To the extent that a community as a whole can have any such one-sided saving, or pure hoarding, or to the extent that it is even attempted, whether it can be universally achieved or not, then it must, other things unchanged, bring about unemployment. So Keynes did not even need his dubious fundamental psychological law, or his pretentious consumption function equations, to prove that saving, in the one-sided sense in which he thought of it, could cause unemployment. This brings us to still another puzzle. If Keynes had in the back of his mind this purely negative concept of saving, side by If I side with a positive concept of investment, why did he talk about a pure consumption function at all?

56:41Unemployment, even on his theory, is not caused by the amount that consumption falls short of income, but only by the amount that consumption and investment combined fall short of income. Suppose we were to choose definitions according to which saving and investment would equal each other in a position of equilibrium or in the long run, but in which for short transitional periods, savings could exceed new investment or investment could exceed real savings. If Keynes had held such a concept, and this concept is strongly implied in spite of explicit denials in much of what he wrote in The General Theory, Then it should have occurred to him that the relevant equations for his purposes did not concern the amount by which consumption alone fell short of income, but the amount by which consumption and investment together fell short of income.

57:39He would not have worried about the gap between consumption and income, but only about the far narrower gap between consumption plus investment and income. It was merely uninvested savings that disturbed him, not all savings. If he had built his functional equations on such concepts, his exposition might have been much clearer, and also, of course, his fallacies. Keynes's fundamental psychological law, as we have seen, is contradicted by experience. But even if the law were found to exist, it would prove very little about the future of Overall Employment. It would merely mean that there would be a tendency, and as I shall later show, a wholly desirable tendency, for a smaller percentage of the working force to be employed in turning out consumption goods, and a larger percentage in turning out capital goods. In fact, the so-called general theory rests on an arbitrary division, on a Verbal Trick Employment can only increase pari passu, Keynes contends on page 98, with an increase in investment, unless indeed there is a change in the propensity

58:58to consume. But this is like saying that our combined supply of ham and eggs can only increase with an increase in our supply of ham, unless indeed If, in fact, instead of dividing commodities into the two groups, consumption goods and investment goods, we were to divide them, whether consumption goods or investment goods, into two different groups, those whose names begin with the letters from A to M, and those whose names begin with the letters from N to Z, which we shall call the AM commodities and the NZ commodities respectively, Then, we could arrive at the following Keynesian conclusion.

59:45Employment can only increase pari-passu with an increase in the purchase of the AM commodities, unless, indeed, there is also an increase in the purchase of the NZ commodities to fill the gap. Brilliant revolutionary discovery! The New Economics! But we shall elaborate upon this when we come to Keynes's own elaboration of this basic point in his general theory. The Sinking Fund Bogie The next thing Keynes begins to worry about is sinking funds. Keynes, in fact, worries about practically everything that happens or can happen in a free enterprise system.

1:00:32His touching faith in the judgment and disinterestedness of government controllers is merely the other side of his distrust of the private businessman. Take a house which continues to be habitable until it is demolished or abandoned. If a certain sum is written off its value out of the annual rent paid by the tenants, which the landlord neither spends on upkeep nor regards as net income available for consumption, This provision constitutes a drag on employment all through the life of the house, suddenly made good in a lump when the house has to be rebuilt. Page 99 Thus, sinking funds, etc., are apt to withdraw spending power from the consumer long before the demand for expenditure on replacements, which such provisions are anticipating, comes into play, i.e., they diminish the current Page 100 Page 100 There are so many things wrong with the foregoing passage that it is difficult to know where to begin an analysis.

1:02:01Keynes lumps everything together and does not distinguish between a depreciation allowance or depreciation reserve, on the one hand, and an actual sinking fund, on the other. A depreciation allowance or reserve is merely a bookkeeping technique, a convenient accounting It's primary purpose is to give an entrepreneur a clearer idea of whether he is making a net profit or not, and if so, how much. A depreciation allowance has no necessary substance. An actual sinking fund is a fiscal technique and does have substance, but even that almost never exists in the form of unused cash.

1:02:46It may be invested in long-term bonds of other corporations, or at the very least in short-term government securities. All but a modest percentage of it will exist, in short, in the form of investments. Or it may have been used to retire outstanding indebtedness. When a corporation decides whether or not to abandon an old factory to sell its equipment as scrap If a plant is to be scrapped or to erect an entirely new plant, its decision will probably be based solely on considerations of profitability. If continuing to work the old plant or old equipment results in a loss, or in a smaller net profit than a new plant would yield, or if a new plant, regardless of whether or not the old plant were scrapped, would promise to yield a profit, Keynes' illustration of a landlord who writes off a certain sum from the value of his house out of the annual rent, but neither spends that sum on upkeep nor regards it as net income

1:04:07some available for consumption but then suddenly rebuilds and apparently exactly duplicates the house in a lump when the house has to be rebuilt is so completely unrealistic that it hardly deserves serious discussion. No landlord with a grain of sense leaves his depreciation write-off as idle cash during the life of the house. The depreciation allowance in the first place may not even exist as a tangible sum. Depreciation allowances do not come into existence as tangible sums simply because they are deducted on the books. After deduction for depreciation, the landlord of a house or the owner of any other business may show a loss.

1:04:56If the loss is as great as the depreciation reserve, then there is no tangible reserve left to be either spent, invested or hoarded. If the loss is greater than the depreciation reserve, but if, for the sake of argument, the depreciation allowance does exist as a tangible fund, then the landlord may either spend it on his own consumption, he is under no legal compulsion to maintain the value In any case, he will not leave the money idle, not drawing interest. But putting aside all these previous objections to Keynes's theories about sinking funds, what an individual landlord or individual corporation does is not important for the and the state of employment or industrial activity of the community as a whole.

1:05:58What is important is only what the actions of landlords and corporations add up to collectively. In the spectacle of landlords and corporations collectively not putting up any new houses or factories for twenty or fifty years, say, not from 1910 to 1959, and then suddenly putting Building them all up in one year, say, 1960, is so absurd that one wonders how it could be seriously held for five consecutive minutes of thought. What happens is that each year a certain number of houses, office buildings, factories, machines, bridges and roads are being built or replaced.

1:06:43They are being put up or installed regardless of the state of sinking funds, and regardless are all mainly of the prospects of future profit, but even if the year in which a given structure were replaced were strictly dependent upon the year in which it was built, the replacement year would nevertheless differ with each plant or house, depending on when it was built. The individual sinking fund periods collectively considered overlap. Next year, say, Corporation A is lending out its depreciation reserve to help build the new factory of Corporation B. Next year, Corporation B will be lending out its depreciation reserve money to build the new factory of Corporation C, or even of Corporation A.

1:07:34And so around the circle. But Keynes takes his whole preposterous theory so seriously that he virtually attributes the 1929 depression to it. In the United States, for example, by 1929 the rapid capital expansion of the previous five years had led cumulatively to the setting up of sinking funds and depreciation allowances in respective plant which did not need replacement on so huge a scale that an enormous volume The aim of entirely new investment was required merely to absorb these financial provisions and it became almost hopeless to find still more new investment on a sufficient scale to provide for such new saving as a wealthy community and full employment would be disposed to set aside.

1:08:26This factor alone was probably sufficient to cause a slump. 100. There have been some pretty silly explanations of the 1929 collapse, but it remained for Keynes to attribute it to the way in which corporations kept their books. Their new investment or lack of it, their actual amount of physical replacement in any given year, had in fact nothing to do with these accounting technicalities. It was determined by the actual physical need for replacement, or rather, more accurately, by the outlook, as it seemed to the corporation officers or directors, for profits from the replacement or new investment.

1:09:11A depreciation allowance may prove in practice to be either too great or too small. But entrepreneurs are not guided in their present decisions by their past expectations The drop in new investment was chiefly the consequence of the 1929 slump, not the cause. But this strange theory about sinking funds enables Keynes to indulge himself once more in his favorite sport of ridiculing financial prudence and sound finance. These it turns out are the great enemies of employment. All prudence will be liable to diminish aggregate demand and thus impair well-being.

1:09:59The greater, moreover, the consumption for which we have provided in advance, the more difficult it is to find something further to provide for in advance, and the greater our dependence on present consumption as a source of demand. Yet the larger our incomes, the greater, unfortunately, is the margin between our incomes and our consumption. So, failing some novel expedient there is, as we shall see, no answer to the riddle except that there must be sufficient unemployment to keep us so poor that our consumption falls short of our income by no more than the equivalent of the physical provision for future consumption which it pays to produce today. Page 105 In a Nutshell Here is the general theory in a nutshell, with its transvaluation of all values.

1:10:55The great virtue is consumption, extravagance, improvidence. The great vice is saving, thrift, financial prudence. We shall reserve to a later point an exposition of why Keynes's riddle is a riddle of his own imagination, not of the free enterprise economic system. And we may anticipate one of our chief criticisms here by calling attention to Keynes's crude and naive conceptions of consumption and investment purely in quantitative terms, whereas every civilized human being in his actual consumption and use of capital equipment concedes them just as much in qualitative terms. There are definite limits, of course, to the quantitative use or consumption of food, clothing, Housing and Capital Equipment, but there are no assignable limits to possible improvements in the quality of capital equipment and in the products and services that it can help to produce.