The following statement is made in the TNEC Report #3, "Who pays the taxes?" on page 13:

"Outlays for all commodities at a given income level, however, can be increased simultaneously only at the expense of saving, which is negligible in the income areas that account for most of the aggregate consumer expenditures. For this reason a general business-cost tax "cannot be passed forward to price without an increase of total consumer expenditure, which would require conditions of general economic expansion." (*And a sales tax.)

Under present conditions incomes are of course rising, so that there would be no difficulty in passing the tax forward. On the other hand, if it is true that much or the whole of an estimated increase in incomes of 10.9 billions in 1942 over 1941 went to the group with incomes over $2,500 a year, the purchasing power of large numbers of people has not increased, and sellers of commodities bought by these groups will experience a decline in demand. Two facts limit the significance of this interpretation (1) the prices of no important commodities have dropped, and (2) the distribution of increased income in actually very different from that suggested by the OPA figures alluded to above. Many in the lower groups have moved up, so that in fact the per capita income of all groups have increased.

Although the foregoing statements appear to dispose of the practical significance of this quotation, it raises one or two issues which used to be discussed as part of a general treatment of the relation of sales and business taxes to price inflation.

Suppose a sales tax to introduced when incomes are not increasing. Then, the TNEC Report states, consumption can only be maintained by decreasing saving. Those who have no savings are unable to maintain their previous volume of purchases. Consequently the prices of the goods they customarily buy must fall, and there is some tendency toward unemployment provided we abstract from any possible effects arising out of the expenditure by the Government of these sums.

This conclusion must be qualified, however. The report of the National Resources Committee (Consumer Expenditure, Table 8, page 48) states that as of 1935-36 consumers in the groups through $1,250 a year deceived to the extent of $1,533 millions. This figure takes account of changes in such various items as installment credits overdue rent, savings and other deposits, loans from individuals outside the immediate family, over-due taxes, possession of livestock and other family capital, charge accounts, insurance premiums, equity in life insurance policies, etc. Careful use must be made of any figures for a concept as complex as that of dissavings. Without going into a complete discussion at this point, we may indicate two sources of error in evaluation which seem to have found a place in the Colm Report.

First, the fact that the low income groups dissaved in any year (especially a year of substantial unemployment does not prove that these groups have no savings left, out of which they can maintain purchases in the event of a rise in cost of living. A family does not start to dissave when its resources are nil; it may dissave from bank deposits of say $1,000, and continue dissaving until its assets are negative. Consequently it is incorrect to imply that the 1935-36 dissaving groups (the groups through $1,250 a year) have exhausted their savings simply because they decreased them.

Second, if a group is shown to have dissaved there is good reason to believe that it can dissave further by failing to pay taxes, borrowing from individuals outside the family, unit, increasing rent arrears, etc. This point is strengthened by the fact that to some extent those in a given low income group who dissave in one year are in later years savers; but because of increased incomes occupy a higher group and therefore do not serve to offset dissavings in the low group to which they formerly belonged. To the extent that this is so, negative savings could be reported for a given low income group year after year.

It may be pointed out that there is will a further means by which low income receivers can maintain consumer expenditures when cost of living rises without any corresponding increase in incomes. This situation would be characteristic of depression and unemployment. Relief payments would augment the purchasing power of the low income groups.

Another statement by Colm, the significance of which is in doubt, is that saving is negligible for the income group that accounts for the greater part of aggregate consumer expenditures. The implication in that the bulk of consumer expenditures relevant to the sales tax or business taxes is made by income groups which are able to save little. First it must be noted that of an outlay for consumption of $5O billions in 1935-36, about half was made by the group with incomes above $1,750 a year. Although it to true that not a very large part of the savings of the group above $1,750 is made by those whose incomes do not greatly exceed this figure, if they save anything at all they are in a position to maintain consumption when cost of living rises provided their savings are in a form which is convertible into consumption. Although a considerable part of such savings will be represented by insurance, durable consumers' goods (like houses), etc., a relatively large part will be in the form of bank deposits, building and loan deposits, etc., and will therefore be available for conversion into consumption.

Second, from the point of view of a retail sales tax on consumer goods. though not from that of business taxes or sales taxes applied at earlier levels of production and distribution, exemptions are likely to include food, housing, medicine, and clothing. These are articles of consumption which bulk heavy in the budgets of the lowest groups. Outlay for consumption by the group through $1,250 in 1935-36 was $17.4 billions. Assuming a sales tax rate of 5 percent, and further assuming that one-half of the expenditure on consumption of these groups is subject to the tax (an assumption commonly made under State sales taxes), the total tax on these groups would be about $435 millions. In view of what has been said about the nature of dissaving in the low income groups, and the possibility of increasing it, it is difficult to see how the addition of $435 millions in taxes to an expenditure on consumption of over $17 billions involves a very considerable drop in the volume of commodities purchased. Business taxes, as well as sales taxes applied at earlier stages of production might have a rather more serious effect because personal exemptions cannot be given.

The following table, from Consumer Expenditures, Table 8, shows aggregate savings by the low income groups, which account for about one-half of the $50 billion consumer or expenditures in 1935-36:

_____________________________________________________________________
                      Outlay     
 Income group          for        Savings 
                   consumption   
_____________________________________________________________________
 
     0 -  500        2,817        -800 
   500 -  750        3,888        -382       
   750 - 1000        5,209        -254 
  1000 - 1250        5,487        - 97  Total negative savings: 1533    
  1250 - 1500        4,807          95
  1500 - 1750        4,278         196  Total positive savings:  291     

CONCLUSION: The Colm report exaggerates the effect of rises in cost of living due to sales taxes. Part of the reason for this is the fact that no attempt was made to analyze dissavings. It is suggested that further work be done on the problem of dissavings in the low income groups with a view to supplementing our knowledge of the economic effects of sales taxation.