have others undertaken to calculate it. Accurate data concerning credit terms are not available; and even if available, these data would not lend themselves readily to fine computations because of the relative stability of terms during the 1920's and early 1930's and the progressive liberalization between 1934 and 1937. Nevertheless, those engaged in selling consumers' durable goods are well aware that the size of their market is highly dependent upon the credit terms offered to purchasers. The reasons for this relationship are to be found in the structure of the market for consumers' durable goods and in the mechanics of family budgeting.

Let us look first at the structure of the market. If we exclude families earning less than $500 a year, a class which is practically, eliminated by virtue of its poverty from the market for new durable goods, the consumer market as a whole takes the form of a pyramid with large numbers of families earning low incomes at the bottom and with progressively smaller numbers of families earning progressively higher incomes as the top is approached. Although the incidence of purchases of consumers' durable goods increases with incomes, the bulk of the purchasers of new durable goods are nevertheless to be found in the modest income classes. Installment purchasers are even more heavily concentrated among relatively low-income classes.

Table 6 indicates the income structure of the market for automobiles in 1940. It shows the distribution of all families, of all new passenger automobile buyers, and of installment buyers by income classes. In constructing these estimates, the 1940 new passenger automobile sales in the domestic market were distributed in accordance with income figures for all families and for new-car purchasers in 1935-36, in the belief that the income distributions, both of all families and of automobile purchasers, has not changed significantly since that time. It was assumed that 65 percent of new passenger automobile sales were made on installment contracts. While accurate figures are not available, this assumption appears to be conservative if allowance is made for installment loans used specifically to purchase automobiles.

TABLE 6.-Estimated distribution of all buyers and installment credit buyers of new passenger automobiles by family income classes in 19401

[Numbers in thousands]

1 Figures from Consumer Incomes in the United States, National Resources Committee, 1938, p. 6. Although 6.7 million families had incomes of less than $500, they account for a negligible fraction of new-car purchases.

2 The figure for total domestic retail sales from Automobile Facts and Figures (1940 ed., p. 6) was distributed in accordance with the percentage figures for cash and installment sales by income classes given by Plummer and Young in Sales Finance Companies and Their Credit Practices. National Bureau of Economic Research, 1940, p. 101. This calculation required an assumption with respect to the division of total sales between cash and credit buyers. It was assumed that 65 percent of the sales were made on installment terms.

Figures for the income distribution of purchases of other kinds of consumers' durable goods are not available, but a recent analysis by the National Bureau of Economic Research of data from the family expenditures study makes it possible to indicate the income distribution of families that were purchasing certain types of consumers' durable goods on installment terms in 1935-36. These are given in table 7. A very large portion of the installment market for furniture and radios is within very low income classes. For refrigerators and "other electrical equipment" the concentration of installment sales in the low-income classes was only slightly less marked. It seems

probable that 80 percent of all refrigerators, washing machines, and automatic furnaces, 75 percent of all furniture, 70 percent of all suction cleaners and radios, and 75 percent of all other electrical equipment for which the selling price exceeds $20 are sold on installment terms. The addition of cash buyers, while increasing the percentages in the higher income levels, would probably not change substantially the total income-structure of the market.

It will be noted that the prices of the major types of consumers' durable goods are exceedingly high in relation to the incomes of the great, bulk of purchasers. For instance, the family incomes of 76 percent of new passenger automobile buyers are less than $3,000, while the cheapest cars cost in the neighborhood of $700. Thus, for those who receive the highest incomes among this large group of buyers, the purchase of a new car takes 23 percent of the annual family income. In the furniture field, almost three-fourths of the installment buyers and probably more than two-thirds of all buyers have incomes of less than $1,500 a year. For this group the purchase of a livingroom suite for $200 would take more than 13 percent of the highest annual income. Such substantial outlays obviously cannot be made from current cash incomes. They must be financed either by drawing down savings or by installment purchases.

Annual income class

TABLE 7.-Estimated distribution of installment buyers of certain types of consumers' durable goods by income classes, 1935–361

[Numbers in thousands]

Under $500

$500 to $999.

$1,000 to $1,499.

$1,500 to $1,999. $2,000 to $2,499. $2,500 to $2,999.

$3,000 to $3,999. $4,000 and over.

Estimates based upon figures obtained from Consumer Incomes in the United States, National ReSources Committee, 1938, p. 6; and from The Pattern of Consumer Debt, 1935-36, by Blanche Bernstein, National Bureau of Economic Research, 1940, p. 139. These figures undoubtedly understate the number of installment purchasers, but they probably do not seriously distort the income distribution of such purchasers.

A considerable proportion of the purchasers in these income classes have no savings whatsoever, and the savings of many others are in equities in homes and in insurance policies and securities that cannot be readily converted into cash. Still others, of course, have ample cashr savings from which to make such outlays. But the great majority of these are unwilling to draw down savings for what they believe to be a consumptive expenditure. These savings have generally been accumulated by dint of severe self-denial and are earmarked for specific purposes or as reserves against specific emergencies. Consequently, whether or not the situation is completely rational, millions of families with savings bank deposits drawing 2 percent interest or less are currently using installment credit services at much higher rates of charge.

These observations suggest not only that a large part of the market for consumers' durable goods is highly dependent upon the availability of installment credit, but also that the size of the effective market could be increased or decreased by changing the conditions under which installment purchases can be made. Because their family budgeting is done on a cash income and outgo basis, installment buyers tend to look upon down payments and periodic installments as measures of the cost of enjoying and eventually owning durable goods. Consequently, among installment buyers, demand for such goods is influenced far more importantly by changes in these factors than by changes in the total purchase price.

The elasticity of demand for consumers' durable goods with respect to installment credit terms cannot be accurately predicted. It would probably be influenced substantially by many collateral circumstances such as the age and physical condition of the stock of goods in consumers' hands, the distribution of incomes or of increments or decrements in incomes, and anticipations concerning the period during which restrictions would remain in effect and the likelihood of still further restrictions. Nevertheless, it seems possible to draw. certain rough conclusions concerning the extent of the changes in demand that would be induced by changes in installment terms.

It has already been pointed out that ability to buy durable goods on the installment plan depends in large part upon the relationship between the periodic payments required for such purchases and available margins of incomes beyond fixed costs. This relationship operates like the traditional scissors in agriculture. The effect upon the income margin available for the purchase of consumers' durable goods is quite similar, whether the margin is reduced by a reduction of incomes or reduced by an increase in the required installment payments.

It seems probable, therefore, that a given increase in the amounts required as monthly installments would have an effect upon the installment demand for consumers' durable goods similar to that of an equal decrease in monthly incomes. In order to translate this relationship into terms of installment-payment elasticity it is necessary to divide the income elasticity by a factor representing the ratio between the monthly income and the monthly installment payment. Among the overwhelming majority of installment buyers of automobiles, typical installment payments take 25 percent or more of the monthly income, and we may estimate installment-payment elasticity for this group roughly as 2.6 divided by 4, or 0.65. For most other major types of durable goods, installment payments take smaller portions of the monthly income, and the income elasticity is lower. However, large numbers of families make two or more simultaneous installment purchases, and it seems reasonable to conclude that the installment-payment elasticity of demand for consumers' durable goods as a whole is not far below the figure for automobiles alone.

There is still the factor of changes in down payments to be considered. While an increase in installment payments tends to eliminate from the effective market those who cannot meet the higher periodic payments out of current incomes, an increase in down payments tends to exclude those who cannot immediately raise the additional cash required to initiate the purchase. To some extent, those eliminated by an increase in down-payment requirements would already have been eliminated by the increase in periodic installments. But there are certainly very large numbers of potential buyers who would be able to meet an increased installment but not a larger down payment. The number of potential buyers able to meet progressively higher down payments probably decrease rapidly. Au installment-terms elasticity roughly approximating 1.0 would, therefore, appear to be a conservative estimate.

Somewhat similar results can be obtained from another approach to the calculation of installment-terms elasticity. If we accept the thesis that the installment buyer looks upon the down payment and the amount of each periodic installment payment as the "price" of the goods, it seems possible to assume that the reaction of demand in the installment market to changes in down payments and installment payments would approximate the reaction of demand in the cash market to changes in total purchase price. From the actual data of automobile sales, Roos and Von Szeliski calculated the price elasticity of demand for automobiles at somewhat less than unity, but they admitted that these calculations were inconclusive in the absence of substantial price fluctuations. By other methods they arrived at priceelasticity figures ranging from 1 to 1.5, which appeared to be roughly confirmed by the effect of an excise tax upon the per capita automobile ownership in Canada.' Thus under these assumptions also an installment-terms elasticity of greater than unity appears to be a conservative estimate.

The extent of the impact of more stringent credit terms upon demand for consumers' durable goods will depend in the final analysis very largely upon consumers' anticipations. Our calculations of terms elasticity assume neutral anticipations, i. e., that consumers as a whole will expect neither more liberal nor

Since credit purchases decline more substantially than cash purchases, the income elasticity of installment demand is certainly higher than the income elasticity of total demand. However, this factor is offset by the fact that some installment buyers, particularly in the higher income classes, are not dependent upon this method of purchasing durable goods.

Dynamics of Automobile Demand, pp. 90-95.

more stringent terms in the near future. The tightening of credit terms means that throughout the installment market either cash savings or current expenditures must be sacrificed in some degree in order to meet the higher periodic payments. If consumers generally believe that the restrictions are temporary they will tend to avoid the inconvenience of the larger payments, and, under these circumstances, even a slight tightening of terms may have substantial effects upon demand. If, on the other hand, consumers generally believe that the current restrictions are only a beginning and that even more difficult conditions will be imposed in the near future, the limiting effects of more stringent terms upen demand may be substantially minimized.

Sociological reactions may be as important as anticipations in determining the actual installment terms elasticity. It is conceivable that the ability to buy a new automobile under the more stringent credit requirements might become the subject of family pride a symbol of financial worth and of the breadwinner's ability to provide. In this case, social pressures might compel sacrifices of consumption in other fields in order to demonstrate the family's capacity to meet the new terms. On the other hand, even more powerful social forces can be used to negate this possibility and to aid and abet the effects of more stringent terms. By skillful publicity, purchases of new automobiles and other durable goods could be made socially taboo during the period of defense emergency, and the use of old goods could be made to represent a demonstration of patriotism.

All signs point currently to a substantial further expansion of incomes during 1941. Some forecasts put the prospective increase of 7.5 billion dollars or 10 percent of the 1940 incomes. Assuming an income elasticity of 2, the demand for consumers' durable goods would increase by 20 percent above the high level of 1940. Figures for the first quarter of 1941 tend to confirm these expectations for the year as a whole. As has already been pointed out, this additional demand would compete directly with armaments industries for many materials and skills that are already scarce. The additional demand for automobiles alone would absorb approximately 2,000,000 additional tons of steel, 400,000 additional tons of iron, 7,000 additional tons of aluminum, 35,000 additional tons of copper, 75,000 additional tons of lead, 25,000 additional tons of zinc, and 3,000 additional tons of nickel.

By increasing down payments and requiring repayment over shorter periods of time, it seems possible to prevent the expansion of demand for consumers' durable goods. Assuming an installment terms elasticity of 1.0-or more properly -1.0-and assuming neutral anticipations, an increase of 30 percent in down payments and in installment payments would come close to stabilizing total demand, while larger increases in each factor could be expected to induce a decline. As has been pointed out, the impact of the more stringent terms may be increased substantially by appeals to patriotism and by making it clear that the restrictions are for the duration of the emergency only.

IV. SOCIAL OBJECTIONS REFUTED

Objections to this method of controlling demand for consumers' durable goods have been raised in some quarters on the ground of social policy. It has been contended that the impact of such a restriction would fall most heavily upon the lower income classes and that imposition of an excise tax upon specific goods would represent a more equitable method of restricting demand.

Installment buying is not by any means confined to low-income classes. Nevertheless, the incidence of installment buying increases with progressively lower income classes and, even more important, the impact of restrictions upon credit terms and upon ability to buy increases with progressively lower incomes. It must be conceded, therefore, that restriction of installment credit terms is to some extent a discriminative method of curtailing demand. If such a restriction were to be the only method of making resources available for armaments production, the objections on social grounds might be valid. But this is not the case. For excellent reasons of social policy and of practicability of collection, increased taxes will fall most heavily on the social classes that are least affected by the restriction of consumer credit terms. It seems probable that the impact of tightened installment credit terms upon effective demand for durable goods in the lower income classes will be no greater than that of increased taxes upon high income classes.

Moreover, restriction of credit terms is not a tax. Those who would be restrained from buying new durable goods could spend their money for other things or save it for subsequent cash purchases. In the long run the purchasing power of the lower income classes and their share of the national income would be increased rather than diminished by virtue of the saving in finance charges and other credit costs. The effect of the restriction of terms is to defer or to divert expenditures, not to decrease the long-run volume of expenditures of the social classes affected.

Also, curtailment of demand for new consumers' durable goods does not mean that consumption of these goods must be sacrificed. Stocks of such goods in the hands of consumers of all social classes are presently higher than ever before in our history. These goods are normally capable of use over a considerable period of time and their useful life may be even further extended if there are incentives to repair them rather than to junk and replace them. We learned during the depression of 1930-33 that production of new automobiles can be violently curtailed for a considerable period without reducing the number of cars in use. These use-values of durable goods in consumers' hands represent a national reserve which can and should be called upon in time of emergency. Of all producers' and consumers' goods, production of consumers' durable goods can probably be curtailed with the least social hardship. Apart from the necessities of defense, it would seem possible to find positive justification for restriction of installment credit terms on the ground of general social and economic policy. The total indebtedness of families of small means has risen precipitously. Shiny new cars, refrigerators, and radios have taken precedence over decent housing conditions, reserves for rainy days, and even adequate nourishment in the scale of values of many families. The immediately prospective period of unequaled prosperity would seem to be the most desirable time to put a brake upon the discounting of future incomes, to restore family expenditures to a cash basis, and to encourage the accumulation of reserves for emergencies and for purchases of family capital goods.

If consumer credit is allowed to expand from its already very high level, its liquidation during the inevitable economic aftermath of the present defensestimulated prosperity is likely to have terrible social and economic consequences. To the extent that purchases of new durable goods can be postponed and present stocks in the hands of consumers can be drawn upon, a reserve demand will be created which could be called upon to fill the gap when defense expenditures are curtailed. To the extent that consumer credit can be liquidated during the present emergency, its subsequent expansion could be relied upon to provide a business stimulant whenever it may be needed.

These comments do not deny the desirability of a special excise tax on consumers' durable goods that compete with defense weapons. On the contrary, this writer has long urged that curtailment of demand for such goods be undertaken both by restricting installment terms and by excise taxes. The objections that can be raised against restriction of credit terms alone are equally applicable to excise taxes alone. For instance, for many people automobiles are necessary means of getting to work, and a tax upon such necessities, which have no relation to ability to pay, is highly inequitable. By combining both methods, however, the discriminative characteristics of each can be minimized. Moreover, the two methods tend to reinforce each other. 'The effects of a given increase in downpayment requirements and a given limitation of the period of payment are increased by the addition of a tax to the purchase price. On the other hand, without a restriction of installment credit terms, the effects of a given tax could and undoubtedly would be minimized in the installment market by an extension of the period of payment.

V. CONSUMER CREDIT AND INFLATION

It has been indicated that compulsory increases in down payments and periodic installments would curtail the demand for consumers' durable goods. To some extent, demand would merely be diverted to other goods. But there would also be a curtailment of the total volume of effective demand.

The outstanding amount of installment credit would be reduced not only because fewer consumers would be able to meet the more stringent conditions for installment purchases, but also because new installment contracts would have smaller initial credit balances and would be paid off more rapidly than previous credit commitments. Unless funds that would otherwise be saved should be used