An Oversight in the Dominant Theory
of Interest
Harry Gunnison Brown
[Reprinted from the American Journal of Economics
and Sociology, 1959]
IN A RECENTLY PUBLISHED ARTICLE,[1] Professor Paul A. Samuelson
expressed the belief that "Bohm and Fisher have given us the
essential insights into the pure theory of interest."
Writing in Econometrica in 1948, the late Professor Joseph
Schumpeter of Harvard University characterized Fisher's book on The
Theory of Interest as "a wonderful performance, the peak of
achievement, so far as perfection within its own frame is concerned,
of the literature of interest." And Schumpeter went on to say of
Fisher's study that it is "an almost complete theory of the
capitalist process as a whole, with all the interdependences displayed
that exist between the rate of interest and all the other elements of
the economic system. And yet this interplay of innumerable factors is
powerfully marshalled around two pillars of explanation: Impatience
(time discount) and Investment Opportunity (marginal return over
cost)."[2] Schumpeter adds, in a footnote, that "Keynes himself
also accepted the time-discount factor, i.e., the whole of
Fisher's theory."
On the basis of such comments as these, it is perhaps not
unreasonable to conclude that Fisher's theory of interest is presently
the dominant one.
Certainly Irving Fisher must be ranked among those who have
contributed greatly to the theory of interest. And certainly he worked
out more fully than Bohm-Bawerk had done, the explanation of how the
preference for present income over future income affects the net (i.e.,
in excess of allowance for depreciation) marginal productivity of
capital, and how the productivity of capital affects individual rates
of preference for present income over future income.
Nevertheless, there is an aspect of Fisher's analysis which seems to
me and has long seemed to me to be incorrect.
I
FISHER CORRECTLY CONCLUDES that a high per cent net productivity of
capital - in his phraseology, a high "rate of return over cost"
- will operate to produce a high rate of interest. He correctly
concludes that this high productivity will tend to bring about a high
rate of preference for present income over future,
i.e., a high rate of "impatience." He insists,
however, that the high net productivity of capital does not have a
direct effect in raising the interest rate, but only an indirect
effect. It raises the rate of interest only by or through raising
men's rates of "impatience," - of preference for present
income over future income.
In order that the reader may form his own judgment regarding the
Fisher viewpoint, I shall quote several passages. In the first of
these, discussing critically a passage in Bohm-Bawerk's Positive
Theory of Capital, Fisher says[3] that "the only[4] way in
which the existence of long processes of production acts on interest
is by overendowing the future and under-endowing the present, thus
creating a 'scarcity value' of present goods."
In a second passage,[5] commenting on the gains that may be secured
from "a newly discovered method of exploiting capital," he
says: "The effect in raising interest comes merely[6] from the
shifting forward of the income stream, which leaves the immediate
income smaller than before, but compensates for this by a still
greater increase afterwards."
In a third passage, replying to a reference by H. J. Davenport to
equipment loans (loans to persons or corporations that are borrowing
for the purpose of securing equipment - capital instruments - to aid
in production), Fisher contends that[7] such loans "are made for
the purpose of securing large incomes in the future, and larger
incomes mean larger consumption. Production loans then are made only
in contemplation of future consumption. Hence, though loans for the
acquisition of intermediate goods do greatly preponderate in the loan
market, these loans have power to affect the interest rate only by[8]
changing the relative amount of future incomes compared to present
incomes."
If I have correctly interpreted these passages in Fisher, the views
expressed can be stated in five propositions, as follows:
1. A high net marginal productivity of capital - "marginal rate
of return over cost" - encourages investment in capital in order
to realize this gain.
2. Such investment in capital for the sake of larger future income,
involves sacrifice of present income.
3. This sacrifice of present income strengthens the desire for
present income which has thus become scarce and weakens, relatively,
the desire for future income which now promises to be larger, -
thereby raising the rate of preference for present income over future,
the "rate of impatience."
4. Because of this higher "rate of impatience," the rate of
interest rises.
5. The effect of the productivity of capital on interest is brought
about only via these successive steps.
However much of truth there is in the passages I have quoted, they do
not contain the whole truth. They leave out, in fact, an important
part of the truth. A high marginal productivity of capital - a high "rate
of return over cost" - has a direct effect on the interest rate,
apart from any indirect effect it produces on interest by first
changing the rate of preference for present income over future.
II
LET us SUPPOSE the net marginal productivity of capital (the yield
above depreciation) to be or to become 8 per cent a year. And let us
use the simplest, the least complicated, illustration we can. John
Deckleburg, a fisherman, is able to catch 1,000 fish a year, with
which, as best he can, he provides for himself and his family. If he
should be able to build a boat, he could thereafter catch enough more
fish per year to cover depreciation of the boat (
i.e., repay the cost of the boat during its life) and, in
addition, get 80 more fish - 8 per cent return above cost - per year.
He can build such a boat during a year - thus its cost of production
is the 1,000 fish he could otherwise catch during the year - he can
borrow 1,000 fish during the year to live on. For then he will not
have to spend the year catching fish and can devote the year to
building the boat. And because the boat will yield - or earn - 8 per
cent on its cost of production, it will pay him to borrow at any
interest rate below 8 per cent. It will pay him to borrow at 3 per
cent, 5 per cent, 7 per cent or 7.9 per cent. The fact that there is
an 8 per cent gain from using capital, i.e., from roundabout
production, makes him willing to offer interest to a lender. It gives
him a motive to bid against other potential borrowers.
Does not this 8 per cent net marginal productivity - an 8 per cent "rate
of return over cost" - motivate him directly? Surely we have here
a preference for more against less. And surely this preference for
more rather than less does not arise because preference for present
income over future has risen. The preference for more rather than less
is an influence in its own right and can act directly. It is not an
influence which can make itself felt only via first setting into
motion the other influence of "impatience."
It is, in my opinion, correct and more realistic to recognize that
our fisherman, John Deckleburg, could have his living from day to day
without borrowing; that he does not borrow in order to be able to
enjoy appreciably more fish - or other present income - this week or
this year; that, on the contrary, he borrows in order to be able to
build the boat instead of having to spend his time catching present
fish for present needs; that he borrows in order to be able to carry
on roundabout production; that, in short, he borrows chiefly, if not
solely, because he prefers more to less and not because he prefers
present income to future income or earlier income to later income.
This is the emphasis that the Fisher analysis - like the analysis of
Bohm-Bawerk earlier[9] - seems to lack.
It is the same if we consider the case of a potential lender. He also
can be motivated directly by a preference for more as against less,
just as certainly or as much as by a preference for present income
over future income. Suppose that he is able to produce more than he
and his family 'need to consume, and thus is able to save. Then this
excess producing, and saving, can take the form of productive capital
from which he can hope to enjoy a return. If he is unwilling to lend
to" another for 5 per cent or 7 per cent, this may be because, by
using his savings himself in the form of productive capital, he
believes he will be able to gain 8 pet cent. In that case, the reason
he does not appear on the supply side of the borrowing and lending
market is clearly that he prefers more to less. His reason for not
lending does not have to be that he prefers present income to future
income by 8 per cent. It can well be that he is influenced far more by
his opportunity to use his savings profitably himself than by any
desire to enjoy more present income at the expense of having less
future income.
Then how can it be said that the productivity - or the anticipated
productivity - of capital, affects his interest offer only by and
through first making his present income comparatively small and thus
increasing his preference for present income over future income?
A potential lender may be influenced by a preference for more as
against less, or by a preference for present income as against future
income, or by both. Preference for present income as against future
may affect the amount of saving, thus the amount of capital and
thereby the marginal productivity of capital. The productivity of
capital may affect the amount of saving and may thereby affect the
distribution of the saver's income between present and future and,
hence, the degree of his preference for present goods. But that
preference for more as against less can have no effect on the loan
market and the interest rate except through its effect on preference
for present income over future income, is simply not true.
Yet Fisher's study presents so carefully and thoroughly the various
interrelations involved in the matter of the interest rate, the "impatience"
rate and the net marginal productivity of capital, that one is tempted
to assume he realized clearly the direct effect of capital
productivity on the interest rate. It is his insistence, in the
passages quoted, that the net productivity of capital has only an
indirect effect, and the thought that students of Fisher's analysis
will - and presumably do - so interpret him and themselves accept this
view, that are the justification for these comments.
1 "An Exact
Consumption-Loan Model of Interest with or without the Social
Contrivance of Money," Journal of Political Economy, 66
(December, 1958), p. 467.
2 "Irving Fisher's Econometrics," Econometrica, 16
(July, 1941), pp. 225-6.
3 The Rate of Interest, New York, Macmillan, 1907, p. 72.
4 Italics are mine.
5 Ibid., p. 199.
6 Italics are mine.
7 The Theory of Interest, New York, Macmillan, 1930, pp.
433-54.
8 Italics are mine.
9 See my paper, "An Off-Line Switch in the Theory of Value and
Distribution," Am.Journal Econ. Social., Vol. 3, No. 4;
also reprinted in Some Disturbing Inhibitions and Fallacies in
Current Academic Economics, New York, Robert Schalkenbach
Foundation, 1910, Chapter 4. See also my Basic Principles of
Economics, 3rd ed., Columbia, Mo., Lucas Brothers, 1955, Chapter
XIII, especially pp. 326-5. This chapter grew out of an article
published in the Quarterly Journal of Economics, August, 1913,
entitled "The Marginal Productivity versus the Impatience Theory
of Interest."
|