In this rather lengthy section, the author examines the effects of changes in interest rates. He notes that they can be handled, similarly to changes in prices, by dividing the effects into separate income effects and substitution effects. Because a rise in the rate of interest will lower the discounted prices of future purchases as compared with present purchases, such a rise "will cause a general substitution all along the line" from present purchases to less distant future purchases, to more distant future purchases. In other words, the substitution effect will cause "a general postponement of expenditure." The author does note, however, that "there is plenty of opportunity for all sorts of cross-effects, and all sorts of complementarity to muddle things up."
Regarding income effects, the net result of a rise in interest rates will depend on how the change affects the discounted values of the planned series of expenditures (including the amount that is planned to be left over at the end of the planning period) versus the discounted value of the planned stream of receipts. For a rise in interest rates, both of these capitalized values will be reduced, but it is not immediately clear which one will be reduced more. The author notes that this question is "formally identical" with the question addressed in the context of income examining "the relative movement of the capitalized values of two streams (previously of the same capitalized value), when the rate of interest changes." In that context, the relative changes in capitalized values of these streams depended on the average periods of these streams (weighted by the discounted values of the various payment amounts). A rise in the interest rate will make the individual better off if the average period of his stream of receipts is less than the average period of his expenditure stream.
When the period of expenditures is greater than that of receipts, the individual, in effect, "plans to spend less than he receives in the near future, to 'spend' more than he receives in the remoter future" (recalling that the capital sum to be accumulated at the end of the planning period is considered part of spending). Such a person, whom the author describes as "planning to be a lender," is made better off when interest rates rise. Because these individuals are made better off, they may then decide to consume more. Thus, "the income effect and the substitution effect go in opposite directions" for such persons, and "either may be dominant. We cannot say whether their present expenditure will be increased or decreased by a rise in the rate of interest."
The author goes on to discuss the nature of these results, and he explains that they arise "from the same cause as in the effect of changes in wages on the supply of labour, or of changes in the price for one commodity on the demand for another." But he notes that "the most important thing which emerges is the way in which this indecisiveness depends upon the assumption that the individual 'plans to be a lender.'"
In what the author calls "the contrary case," someone whose average period of expenditure is less than the average period of receipts will be made worse off by a rise in interest rates. For such a person, the income effect and substitution effect both work in the same direction, namely, to reduce current expenditure in response to an increase in the rate of interest.
These individuals, whom the author describes as people who "plan to be borrowers," include entrepreneurs who are undertaking investments (as well as "spendthrifts" whom he dismisses from further consideration).
In the remainder of the section, the author considers the implications of lenders' and borrowers' income effects for the supply and demand sides of the market for securities.
While those persons who plan to be lenders have an income effect increasing their present expenditure when the rate of interest rises, those who plan to be borrowers have an income effect reducing it. If these income effects cancel out, then there is nothing left but the two substitution effects, each of which tends to reduce current expenditure.
The author then asks the question, "Are the income effects likely to cancel out?" He notes that there is "one broad reason" to expect that they will tend to, but that this tendency "is subject to two sorts of exceptions."
The broad reason to expect them to cancel out is that current lending and current borrowing must always be equal when the market for securities is at equilibrium. But this isn't sufficient for concluding that the aggregate income effects among borrowers and lenders will exactly balance out. And this gets at the first of the two sorts of exceptions: namely, the possible inconsistency between the planned quantity of borrowing/lending and the actual current borrowing/lending.
[P]lanned borrowing and lending, being mainly inside people's heads (and not very definite even there), are not matched on the market. There may be an excess on one side or on the other; though, if there is, it spells inconsistency between plans, and consequent potential disequalibrium.
The second of the two exceptions, which the author deems "doubtless" more important, relates to the possible relative speeds with which borrowers and lenders adjust their expenditure to new conditions.
If borrowers are quicker to adapt themselves than lenders (I should judge that in practice this is probably the case), the income effect on the borrowers' side is likely to be stronger than ... on the lenders' side. This would make the net income effect work in the same direction as the total substitution effect, and reinforce the conclusion that, for the market as a whole, a rise in the rate of interest will reduce current expenditure, a fall in the rate of interest increase it.
