Bernhard Felderer • Stefan Homburg

* Macroeconomics
and New
Macroeconomics
Second Edition

With 97 Figures

Springer-Verlag
Berlin Heidelberg New York
London Paris Tokyo
Hong Kong Barcelona
Budapest
 Professor D r . BERNHARD FELDERER
Universität B o c h u m
Universitätsstr. 150
-"CA
D-Ì630 B o c h u m , F R G B l B L

Professor D r . STEFAN HO: A. E u C A U u Ï Ï H.

Universität B o n n
Adenauerallee 2 4 - 4 2 "UHIDAD ACADEMICA DE
D-5300 B o n n , F R G
Z cíaos PROFESIONAL
/8/7P- ^ f ^7727^1 Y M P O S G R A D O / ECH
"The present treatise is an attempt to present a modern version of
S " 0 N & M old doctrines with the aid of new work, and with reference to the new
problems, of our own age."
1 6 4 4 3
< , s i 3 fc5*Li!L
ALFRED M A R S H A L L
. 12398b
0 ' 0 ^ { ^ T r a n s l a t i o n of the German edition:

fSpringer-Lehrbuch), 5th corrected edition

SBN 3-540-53415-6 Springer-Verlag Berlin Heidelberg New York Tokyo
ISBN 0-387-53415-6 Springer-Verlag New York Berlin Heidelberg Tokyo

NOM! A
ISBN 3-540-55318-5 Springer-Verlag Berlin Heidelberg New York Tokyo
ISBN 0-387-55318-5 Springer-Verlag New York Heidelberg Berlin Tokyo

ISBN 3-540-18004-4 1st edition Springer-Verlag Berlin Heidelberg New York

ISBN 0-387-18004-4 1st edition Springer-Verlag New York Berlin Heidelberg

Library of Congress Cataloging-in-Publication Data. Felderer, B. (Berahard)

¡Makrookonomik una neue MaKrookonomik. English] Macroeconomics and new
macroeconomics / Bernhard Felderer, Stefan Homburg. - 2nd ed. p. cm.
Translation of: Makrookonomik und neue Makrookonomik. 5th corr. ed. Includes
bibliographical references and indexes.
ISBN 3-540-55318-5 (Berlin : acid-free paper). -
ISBN 0-387-55318-5 (New York : acid-free paper)
1. Macroeconomics. I. Homburg, Stefan, 1961- . II. Title. HB172.5.F4513
1992 339-dc20 92-15191 CIP

This work is subject to copyright. All rights are reserved, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in other
ways, and storage in data banks. Duplication of this publication or parts thereof is
only permitted under the provisions of the German Copyright Law of September 9,
1965, in its version of June 24, 1985, and a copyright fee must always be paid.
Violations fall under the prosecution act of the German Copyright Law.
© Springer-Verlag Berlin Heidelberg 1987, 1992
Printed in Germany
The use of registered names, trademarks, etc. in this publication does not imply, even
in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
Printing and Bookbinding: Konrad Triltsch, Wurzburg
2142/7130-543210 - Printed on acid-free paper
C h a p t e r III. National Income Accounting

Origin of GDP
United States 1990

Part Two Macroeconomics

The essence of an orthodoxy of any kind is to reduce the subtle
and sophisticated thoughts of great men to a set of simple princi-
ples and straightforward slogans that more mediocre brains can
Expenditure of GNP think they understand well enough to live by.
United States 1990
(HARRY G . JOHNSON)

private c o n s u m p t i o n

gross investment
149S

government consumpt.
20%

Distribution of GNP
United States 1990

employees' i n c o m e

indirect tax net

Figure 10
 Q 1 2 3 S 8 8

§ 17 Introduction to Part Two

For some fifty years macroeconomic discussion has been dominated by two competing
orthodoxies, called Classical-Neoclassical and Keynesian Theory respectively. While
these two approaches have been considerably elaborated since, their "cores" have
remained unchanged, and hence it is worth examining them more closely. Thereafter we
will be able to debate their implications for economic policy, and then pass on to more
recent developments in macroeconomics. But initially, we should specify the general
assumptions of our approach.
First, we will represent both Classical-Neoclassical and Keynesian theory in highly
condensed form. The interconnected strands of thought, characteristic of the historical
development of ideas, cannot be followed here because theory, not history, is at the
center of our interst. Hence we combine Classical and Neoclassical theory in spite of
the analytical differences between them since there are also important similarities. For
short, we will speak of the Classical theory from now on. As a result of this very
condensed exposition, every judgement of "Classicism" or "Keynesianism" is a judge-
ment of the central arguments only and cannot do justice to single authors.
Secondly, the theories are put in contemporary terms. For example, we will speak
of a "production function" in connection with Classical theory - although this mathe-
matical notion was perfectly unknown to the Classics. This procedure suits our purpose
because it helps in comparing the different theories; and it is legitimate because only the
form and not the essence of the original ideas is changed.
Thirdly, we limit our subject in several respects. We consider only

- a market economy with private property,

- where all markets are perfectly competitive,
- no external economic relations exist,
- and where, initially, there is no government.
These premises are, no doubt, very restrictive and we hope that they do not prevent
the reader f r o m reading further. In most cases it is suitable to start with the simplest
model; but the simplest is also the most abstract model and hence rather remote from
reality. But, with increasing knowledge and proficiency, the model can subsequently be
made more realistic; and this will happen in Chapter 5, for example, where government
is introduced in order to analyze the impacts of fiscal and monetary policy.
Nonetheless, the models are perfectly adequate to bring out the central ideas of
Classical-Neoclassical and Keynesian theory. What is more, they are quite difficult
enough.
Chapter IV. The Classical Theory

§ 18 The Classical Vision. Plan of the Chapter

The early economists were confronted by the theoretical problem that, in a market
economy, the individuals formulate their economic plans rather independently from one
another, and yet these plans are normally fulfilled. How could this be possible? The
answer was: The pricing mechanism brings about a coordination of those individual
economic plans. Due to this mechanism, the pattern of production tends to adjusts to
the pattern of demand.
If there is excess demand in a certain market, for example, the customers will
overbid one another to get the desired good, or the suppliers will be able to charge them
higher prices. Thus the production of the good becomes more profitable, which induces
some suppliers to produce more of it. At the same time some customers will be inclined
to buy other goods because this one now seems too expensive to them. Hence, the price
change is likely to result in a market clearance. But if this is true for any market - so
the early economists thought - it would be true for all markets, too.
With regard to the above mechanism, Classical and Neoclassical economists also
considered adjustment problems', indeed, they are the last to be accused of unrealistic lines
of reasoning. Nevertheless, the tenor of their analysis reads as follows: Ultimately, a
market economy tends towards a state of general equilibrium, i.e. clearance of all markets;
and in this sense, general equilibrium is the center of gravity of a market economy.
Classical and Neoclassical economists did not want to deny the possiblity of tempo-
rary crises: they only maintained that a convergence towards general equilibrium would
occur in the long run, even if this "long run" would take some five or even thirty years.
Furthermore, they considered general equilibrium an optimal state for society. They
asked: What could be better than an economic order which allows individuals to plan
freely and independently and yet typically fulfills their plans? So the early economists
recommended an economic order that guarantees economic freedom to the citizen: they
advocated the market directed economy.
Beside the analysis of the pricing mechanism that we call price or value theory, a
second achievement of the Classics consisted in discovering the veil of money. Viewed
naively, money seems to play the outstanding role in the economic process: Those who
possess plenty of it are considered rich; and nearly everyone tries to earn lots of money
and to spend as little as possible on any given quantity of goods. The Classics opposed
this common opinion by declaring: Money and wealth are entirely different. The social
wealth consists of the annual production or the existing stock of goods whereas
money is a mere medium of exchange. No one wants money for its own sake but only
because of the goods he can buy with it. Hence, money is a "veil" over the real events.
 34 Chapter IV. The Classical Theory § 19 Production Functions 35

The exaggeration of this view led to the macroeconomic dichotomy (bipartition), as In so far as we are abstracting from money, it is not the money price level P and
it is called today. This is the proposition that in a market economy the monetary and the nominal wage w that are crucial, but only their quotient, i.e. the real wage w/P.
the real magnitudes are entirely independent of one another. An increase in the quantity The real wage denotes the number of commodity units paid for one working hour.
of money, for instance, does not bring about any change in real wealth, but only a In the following, we will first analyze the behaviour of firms and households;
proportionate increase in prices. Accordingly, analysis was divided into value theory and subsequently we let them trade in the various markets; then we introduce money; and
monetary theory. Value theory deals with real magnitudes as well as with relative prices finally we sum up these isolated results by means of the Classical moclel.
whereas monetary theory deals with pure money prices. We mention this "dichotomy"
because it is decisive for what is to follow: To maintain the "Classical spirit", we will
first discuss the determination of real magnitudes only; afterwards we will supplement
this pure value theory with the quantity theory of money. §19 Production Functions
Two macroeconomic simplifications are basic to the following procedure. First, in
the economy considered there are only two kinds of individuals, called firms and In the firms sector, commodities are produced by means of factors of production.
households respectively, and these are combined into two sectors. The sectors each Clearly the technology of production exerts a strong influence on the behaviour of
pursue three economic activities: firms; so we first turn our attention to this. Classical theory distinguishes between
three factors of production, namely, labor, capital, and land.
The firms - produce commodities, The term labor here covers all work done by blue-collar and white-collar workers;
- demand labor power, and the work of the enterpreneur himself, however, was either neglected or conceived of
- undertake investments. as a special factor.
The households - consume commodities, Capital in general means either the entirety of producible means of production, or
- supply their capacity to work, and the financial funds to pay for them; here we use the term in its first sense. In the
- make savings.
theoretical analysis of production, inventories do not constitute capital because they
Our second assumption implies that only a single homogeneous commodity is cannot be used as a means of production.
considered whose quantity is designated as Y. Looked at ex post, Y is the real income Land, finally, covers all natural resources or, in other words, all non-producible
of the whole economy and, as only one commodity is supposed to exist, "real income" means of production. Land in the narrow sense belongs to this category, but so do
means the same as "the number of annually produced units of that commodity"; ore deposits, oil springs, and the like.
hence reai income is measured in commodity units. The relationship between input of factors of production and output can be de-
But in ex ante-analysis, three meanings of " Y " must be carefully distinguished. Y scribed by a production function:
can be interpreted as Y = F(N, K). (10)

- the total quantity of planned production (Y s );

Here Y denotes the quantity of goods produced, N the number of working hoars,
- the total quantity of planned demand for goods (Yd); or
and K the size of the capital stock. In our macroeconomic model, capital consists of
- the sum of labor incomes, interest incomes, and profit incomes.
a homogeneous commodity that is identical to the produced commodity. Thus the
Our model covers four markets in total, namely output can be used either for consumption or investment; in the latter case, the stock
of capital increases.
- the commodity market where the supply of commodity Ys meets consumption Equation (10) assumes nothing but the existence of a maximum output correspond-
demand C and investment demand I;
ing to any combination of labor and capital. The functional relationship between
- the labor market where labor supply N s and labor demand Nd meet;
them depends on
— the capital market where capital supply S (savings) and capital demand I = ABS/P
meet (in fact, as explained later, I denotes both investment demand and capital - the given quantity and quality of land; and
demand); - the given level of technical and organizational knowledge.
and the money market which is described by the quantity theory of money. By equation (10) it is not meant that land and knowlegde exert no influence on
Since the first three markets are assumed to be perfectly competitive, in each of output, but only that these two are given and constant within the considered period.
them a single price is established, i.e. In the short run, the enterpreneurs can vary output by variations in labor and capital
input only.
- the price level P which denotes the money price of the homogeneous commodity; To obtain analytical results, the foregoing premises do not suffice; certain assump-
- the nominal wage w as the money price of labor; and
tions with respect to the kind of functional relationship must be made. We will assume
- the rate of interest i as the price of capital.
a Neoclassical production function:
34 Chapter IV. The Classical Theory § 19 Production Functions 36

inputs, the law of diminishing returns is unlikely to hold because the relative scarcity
of the factors remains unchanged. On the contrary, it would be supposed that a
doubling of every input results in a doubling of output.
In three-dimensional space, the Neoclassical production function has the follow-
ing shape:

Figure 11

This figure shows the relationship between the number of working hours and
output where the quantity of capital is taken as given. This Neoclassical production
function clearly exhibits the following features:

- The graph runs through the origin,

- its slope is always positive, Figure 12
- but decreases over the whole range.
What does this mean in economic terms? First, nothing can be produced without Obviously, the slopes of the surface of this graph with respect to the N- and K-axes
any effort; therefore, the graph runs through the origin. Second, more work yields are both positive and decreasing. They are called marginal productivities. For exam-
more output; therefore, the slope is always positive.
ple, the marginal productivity of labor is equal to the marginal product of the last
We now have to ask why the slope decreases over the whole range. Imagine a firm "small" unit of labor divided by exactly that "small" unit. In the limit of infinitesimal
with a certain amount of equipment where no work is done at all. This is point O in variations, the marginal productivities are identical to the partial derivatives of the
the figure above. An increase in labor input up to one hour will bring about an extra production function with respect to N and K.
yield of AYj, and point A is reached. A second hour of work will also increase output
The difference between marginal products and marginal productivities is that the
but by less than the first because the first hour is likely to have been used for the most
former represent certain quantities of output while the latter are ratios. In contrast to
urgent activity.
marginal products, marginal productivities do not increase if the additional output
At this level of abstraction - only two variable factors of production which are produced increases but only if additional output per unit of input increases.
both perfectly homogeneous - it is quite reasonable to suppose that both factors The Neoclassical production function has first partial derivatives1 that are strictly
exhibit decreasing returns: For if the quantity of some factor is increased gradually, positive: any extra input yields some extra output; furthermore, it has second partial
the quantity of the other being held constant, the former is rather "scarce" at first; derivatives that are strictly negative: the additional output per unit of additional input
and then gradually becomes more and more "abundant" in relation to the other. Thus decreases; and, finally, we suppose for simplicity that its cross derivatives vanish: the
the increase in output caused by a given increase in the variable factor is likely to marginal productivity of labor is independent of the quantity of capital, and conver-
diminish. This is the essence of the Neoclassical production function. sely.
The additional yield of the last unit of labor is called the marginal product of labor. To sum up:
Thus we can say that the Neoclassical production function implies diminishing mar-
ginal product of labor as the quantity of labor is increased. ^ > 0 and ^ > 0 . (ID
5N 0K
Thisline of thought can also be directly applied to the other factor, i.e. capital,
since it is appropriate to assume that the law of diminishing marginal product applies ^ < 0 and g < 0 . (12)
to this factor, too: Starting with a positive number of working hours and without any 8N 2 0K
capital in the sense of equipment, the first unit of capital will be put to the most urgent 02F 02F
.„. " (13)
utilization, the second unit to the next most urgent utilization, and so on; hence 0N0K. 0K0N
capital will also show a diminishing marginal product.
It is to be noted that the foregoing reasoning applies only to partial factor varia-
tions, i.e. to variations of a single input (or some inputs) keeping at least one other
1 Partial derivatives are introduced in the mathematical appendix. Assumption (13) is also
constant. In the case of a total factor variation, i.e. a proportionate increase of all
explained there, in subsection *4.5.
 §20 T h e F i r m s 39
38 Chapter IV. The Classical Theory

The representative firm is supposed to aim at maximizing its profit. This assump-
Some qualifications concerning the empirical validity of the Neoclassical produc-
tion does not suit reality perfectly; but it seems more reasonable than any simple
tion function must be made. First, this function implies perfect substitutability of the
alternative (e.g. revenue maximization). Planned nominal profits are expressed as
two factors of production: they can be combined in any arbitrary ratio. This is rather
money units per period and they consist of revenues minus labor and capital costs:
unrealistic if a specific capital equipment is given and a certain number of workers is
required to operate it. Here we speak of a limitational production function. However, it = P-Y! - wN" - i-B' (14)
the Neoclassical production function does not relate to such a situation but rather to profits = revenues — labor costs — capital costs
the planning stage in which it is perfectly reasonable to assume that the entrepreneur
can make a choice between different methods of producing a given output We thus In this formula, k represents annual nominal profits, P the commodity price, w the
recognize that the Neoclassical production function deals with a not too short time nominal stage rate, N d planned demand for labor, and i the rate of interest. Bs is the
horizon which enables the enterpreneur to decide on the method of production; and planned debt of the firm, that is the stock of bonds that are to be issued u p to the end
it neclects the time scale in which equipment is already installed and no choice of this of the period. Hence i - Bs represent the future interest payments.
kind can be made. Trying to maximize profits, the representative firm has to take into account its
Second, the Neoclassical production function contradicts the original law of di- production possibilities which are given by a Neoclassical production function:
minishing returns as put forward by TURGOT. TURCOT proceeded from an observation Y = F(N,K). (15)
in agriculture: If land is cultivated, the second hour of work potentially brings in a
higher yield than the first because the first hour does not allow an intense enough Changes in the physical stock of capital within a single period are called invest-
cultivation. Thus, TURGOT proposed returns to labor which increase at first; but he ment. If K denotes the optimal equipment and K 0 that equipment given a t the
also thought marginal productivity of labor would decrease after a certain optimal beginning of the period, investment amounts to
point was passed. To conclude, TURGOT'S production function showed a marginal
productivity of labor which, starting from the origin, first rose up to a certain point, I:=K-K0. (16)
and then declined in the familiar manner. Generally, the assumption of a Neoclassical where we assume absence of bottlenecks, so that the optimal stock of capital can be
production function is likely to be violated if there are special technical relationships. realized by means of a single investment decision. The representative firm finances its
In many cases, especially that of TURGOT, it is not an unwarranted simplification
investment by issuing bonds 1 :
however because it can be shown that, in perfect competition, firms always produce P • I = AB S . (17)
in the range of diminishing marginal productivity1. Hence, the range of increasing
marginal productivity is economically irrelevant and can be neglected accordingly This equation amounts to the assertion that the value of the investment, calculated
without loss of generality. as commodity price times the number of capital goods required, is wholly financed by
For the purpose of empirical investigation, special production functions have been borrowing; hence AB is the nominal change in the debt within the period. Of course,
constructed, such as the famous Cobb-Douglas production function. Such formulations it is also conceivable that investment is financed out of current profits; but we can
need not detain us here since our primary concern is theoretical rather than practical. interpret this as if the profits were first distributed and then given to the firm as
outside debt immediately afterwards. Here, we implicitly acknowledge the fact that
self-financing gives rise to (opportunity-) costs since the owners have t o forego other
way of using their funds This difficulty is avoided here since capital consists of
§ 2 0 T h e Firms
outside debt only.
Having considered the production function, we are now ready to turn to the behav- By rearranging terms in equation (17)
iour of firms. In a market economy, the behaviour of the productive sector depends AB S
crucially on the decisions of the various single enterpreneurs, and this obvious fact has 1 = ^ (18)
to be our point of departure. Hence, we take up MARSHALL'S famous notion of the
representative firm. it becomes clear that the demand for physical capital (equipment), I, is identical to the
The representative firm is an imaginary single firm which behaves, except in scale,
real demand for financial capital (funds), ABS/P.
exactly as the average of all firms, i.e. it works under normal conditions. If there are We now substitute the production function (15) and equations (16) and (17) into
100 firms, for example, which produce in total 1000 units of a certain commodity, the
the profit function (14) to get:
representative firm will produce precisely 10 units of it. Because we are searching for
qualitative results only, the difference of scale can be neglected, and we can speak of n = P • F(N, K.) - w • N d — i (B 0 + P (K - K 0 )). (19)
"the representative firm" and "the sector of firms", as well.
Note that every supply of bonds is equivalent to a demand for capital. T h e r e f o r e we use the
1 This is shown in the Mathematical Appendix, Subsection *3.9. subscript " s " referring to bonds rather than capital.
41 Chapter IV, The Classical Theory §22 TheLaborMarket49

Since we deal with perfect competition only, the representative firm acts as price ingly, a point will finally be reached where
taker: It considers the price, the wage rate, and the rate of interst given and adjusts 8F
output, labor and capital input so as to maximize its profits. As equation (19) P dN d = w • dN d , - (25)
0N
suggests, profit is indeed merely a function of two variables, i.e. labor and capital,
because the latter determine output uniquely. Thus, the conditions for maximum and this is the point of maximum profit with respect to labor! If the firm were to
profit 1 consist in setting the partial derivatives of the profit function tozero: increase its labor demand further, profits would decline because marginal cost is
constant and marginal revenue diminishes (due to diminishing marginal productivity).
0tt 9F i On the other hand, it would also be unwise to demand less labor than indicated by
= p w 0 (20
0N -aN- = ' > equation (25) since, before reaching this equality, (24) applies and profits can be
07t 0F ! increased by employing more labor. Therefore, the point where marginal cost equals
8K 0 K ~ l P = O
" <21> marginal revenue is optimal. Dividing (25) by dNd immediately yields equation (22)
*
which basically makes the same statement but with respect to infinitesimally small
Solution immediately yields: changes.
Condition for 0F The condition for the profit maximizing stock of capital can be explored similarly.
optimal labor demand: P
'0N = W>
~ (22) Any increase in the capital stock causes both

Condition for 0F - an increase in revenues of approximately P • 0F/0K • dK d ;

optimal capital demand: = L (23) - and an increase in cost by i • P • dK d .
0K
dK d represents the change in the capital stock, i.e. investment. The increase in
P • 0F/0N is called the marginal value productivity of labor as opposed to 0F/0N, costs is due to the fact that investment is financed by debt for which interst must be
the marginal physical productivity. It is the increase in monetary revenue caused by paid. Starting from zero and increasing capital gradually, marginal revenue will at
a "small" increase in labor - as opposed to the increase in physical output. According
first exceed marginal cost; but whereas the latter is constant due to given prices,
to equations (22) and (23), marginal value productivity of labor must equal the
marginal revenue declines. It declines until
nominal wage in equilibrium; and marginal value productivity of capital must equal
the rate of interest. P • — • dK d = i • P • dK d . (26)
What is the economic interpretation of these formulae? Let us turn to labor first. 0K
The effect of the employment of a small additional quantity of labor is two-fold: marginal revenue = marginal cost

- revenue rises approximately by P-6F/0N' • d N d (marginal revenue); This is the optimum condition for the stock of capital. Dividing both sides by dK d
- and costs increase by w • dN d (marginal cost)2. yields equation (23) which states the equivalent with respect to infinitesimal changes.
Using those maximum conditions we are ready to turn to the ultimate goal of our
From the standpoint of profit maximization, the employment of an additional model: to explore how the representative firm reacts to price changes. Again, we start
labor unit is sensible only if marginal revenue exceeds marginal cost: by considering labor, and divide equation (22) by the commodity price or price level:

0F
P • —- • dN" > w • d N d (24) ^ = (27)
0N 0N P
marginal revenue > marginal cost
Equation (27) states that the marginal physical product of labor must equal the
As long as this condition holds, the representative firm will find it advantageous real wage rate. How can we expect the firm to react when the real wage rate rises? The
to increase its labor demand, since revenue increases more than cost. Now the central firm will again choose labor input so as to equalize marginal productivity and the
idea is that marginal productivity will decline due to the assumption of a Neoclassical given real wage rate; because, as we saw before, only in this case profits will be
production function. The reader will recognize this immediately from Figure 11. maximized. If the real wage rises, marginal productivity of labor must also rise, and
Hence, if labor demand goes up, and marginal productivity of labor declines accord- this means that labor demand diminishes (see Fig. 13).
Therefore, as a rule, labor demand declines as the real wage rate goes up. The same
result is obtained mathematically if we differentiate (27) by means of the chain rule 1
1 Owing to our assumptions regarding the production function, these conditions are both
necessary and sufficient when we abstract from corner solutions. 1 Cf. the mathematical appendix, passim. We have to point out that in the simple case we need
2 The terms "marginal revenue" and "marginal costs" refer to one factor unit here. Instead, we not solve two simultaneous equations (20) and (21) because the cross derivatives of the
could also use them referring to one output unit, writing P • dy for marginal revenue and production function were assumed to vanish. In Subsection *4.5 of the Mathematical Appen-
w • SN/3Y • dY for marginal costs. dix we deal with the more tricky case of non-vanishing cross derivatives.
42 Chapter IV, The Classical Theory
§22 TheLaborMarket49

labor input labor input is

is low and high and marginal
marginal productivity is low
productivity
is high

N
Figure 14
The labor demand of firms

Figure 13

with respect to w/P (i.e., 6F/0N must be differentiated with respect to N first, and then
N with respect to w/P):
92F dN
0N 5 ' d W P ) _ 1
' (28)

dN 1
= < 0 (29)
^ d ( W P ) W "
2 Figure 15
0N The investment d e m a n d of firms
The negative impact of a rise in the real wage rate on labor demand follows
directly from assumptions (12) and (13). The impact of changes in the interest rate on Our description of the representative firm's decisions is now complete. We recog-
capital demand is derived similarly: nized that the firm - when price, wage, and interest are given - chooses its demand
for labor and capital so as to maximize its profits.
0 2 F dK dl /dK \
1 1 ; c f ( 1 6 ) (30)
In the following we will employ the very important assumption that the stock of
^ • d T d T df=
capital does not change within the current period, or, to put it differently, that one
dl _ 1 period is required for the installation of new capital goods. Under these circum-
<0. (31)
dï~ W stances, the current production depends only on the initial capital stock and on the
SK 2 level of employment, whereas the new capital goods have no impact before the next
period. The resulting model is not exactly the same as that we discussed above -
We can summarize the results of this section by the labor demandfunction and the because there we supposed new capital goods to increase production instantaneously
capital demand function of the representative firm: but it simplifies analysis considerably, and it is for this reason alone that we employ
it from now on. Changes in the stock of capital are rather difficult to handle, and that
N" = N d ( j ) (32) is why they are chiefly dealt with in growth theory 1 .
Due to our new assumption, we can rewrite the production function as
Y = f(N). (34)
I = I(i). (33)
because the only variable factor that influences current production is labor. Invest-
The following two figures illustrate that any rise in the real wage diminishes the ment, on the other hand, shows a demand effect within the current period, but not a
demand for labor; and any rise in the interst rate diminishes the demand for capital. capacity effect; capacity to produce cannot be enlarged before the following period.
In addition to these two functions, there exists a demand for financial capital or, According to (34), current production depends exclusively on labor; but the latter, in
equivalently, a supply of bonds. (Recall that, if someone offers a bond he demands
financial capital, i.e. money; thus demand for capital and supply of bonds are equiva-
lent.) This demand for financial capital is here identical to real investment since we
supposed the latter to be entirely financed by outside debt. Thus, "investment" has 1 Since there is no capacity effect in the first period, the p r o f i t function (14) must be interpreted
as to refer to two-period profit maximization. Strictly speaking, we should allow for expecta-
a two-fold meaning. tions. But that would m a k e the analysis harder without changing the qualitative results.
44 C h a p t e r IV, The Classical T h e o r y §22 The Labor Market 49

turn, is conditional on the real wage rate (cf. equation (32)). Combining these twdS§jp means of cardinal utility theory. Cardinal utility theory conceives of utility as a
functions, we can specify the firm's commodity supply function: quantity that is measureablerat least in principle, and marginal utility is amply that
utility received from the consumption of the last "small" unit of some good.
Y® = Y s The representatives of marginal utility theory proceeded from the empirical obser-
vation that the intensity of a stimulus declines when the cause of that stimulus (e.g.
the consumption of a certain good) is repeated. Or alternatively: marginal utility
If the real wage rate goes up, the demand for labor will decrease and so will the decreases as consumption increases. This principle is known today as Gossen's first
commodity supply. The latter does not, however, depend on the rate of interest. It is ; i | t law. According to it, the relationship between income (or consumption) and utility
true that the rate of interest will influence investment; but since we neglect the capacity exhibits the following shape:
effect of investment it will not have any impact on current production.
-M Utility

§21 The Households

•SB
As in the sector of firms, the behaviour of the household sector also depends on the
decisions of single units. We therefore consider a representative household that acts, ^ :
save in scale, like the households as a whole. Because we look for qualititative results | J
only, the terms "representative household" and "household sector" can be used
synonymously. The utility function
Within any period, the representative household receives labor income, interest ?!
income, and profit income:
The graph of this utility function shows that marginal utility of consumption -
p . Y = w - N s + i - B d + jt. (36)
though always positive - diminishes over the whole range. In contrast, labor was
As we know from national income accounting, P • Y, i.e. the price-level times the assumed to yield dissatifaction because it involves giving up some leisure. Since leisure
physical output of a period, is identical to total nominal income if depreciation is is treated like any other good, and therefore also shows diminishing marginal utility,
neglected. Or, to put it another way, profits are simply defined as the value of marginal disutility of labor must increase as labor supplied itself increases.
production minus labor and interest income by equation (14). Hence, the sum of The household receives additional real labor income
labor, interest, and profit income must be equal to the value of production ex postr4
But what holds ex post trivially, holds ex ante only if the individual plans are compat- dY = — • dN s . (38)
ible. Therefore, in ex ante analysis, we have to distinguish between planned supply ;* P
(Ys), planned demand (Yd), and income (Y). if it supplies dN s additional units of labor at the given real wage rate w/P. When total
In that we are abstracting from money for the time being, the representative labor supply and labor income are small, marginal utility of income will be high but
household divides its real income between consumption and savings: marginal disutility of labor will be low. Accordingly, the household will increase its
labor supply gradually until the marginal disutility of labor and marginal utility of
AB d
Y = C + S; S:= income are just in balance. This point is optimal to the household.
If the given real wage increases, marginal utility of labor income will rise too,
In this equation, C represents real consumption, and S real savings, i.e. real because now a greater quantity of goods is paid for every working hour, and labor
demand for bonds. (37) is called a budget constraint because it forces the household supply will be increased until the new optimum is reached. Hence, marginal utility
to balance income and expenditure. The concept of a budget constraint is important theory yields the tesult that high wages correspond to a high labor supply, and from
and should not be misunderstood: it does not state that some income is given to the this the Classical labor supply function can be derived:
household and the latter has only to decide how to spend it. On the contrary, incomes
and expenditures are planned simultaneously; by manipulating its labor and capital
supply, the household can decide on its labor and interest income. This point is very
important to Classical theory.
In order to choose a certain labor supply, the household weighs the advantages of This theory should be qualified a little in view of contemporary theory. From
income against the disadvantages of giving up leisure. The Classics described this by modern "indifference curve analysis", such a monotonic relationship between real
47 Chapter IV, The Classical Theory §22 The Labor Market 49

N
Figure 17
The savings (or capital supply) of households
wages and labor supply can not be inferred because the impact of an increase in the
real wage rate is two-fold. On the one hand, there is certainly a substitution effect:
When the wage rate goes up, leisure becomes dearer and this will induce the household rate of interest. This yields the Classical capital supply function which states that
to demand less of it, i.e. it will induce him to work more. But, on the other hand, there savings depend on the rate of interest only, and increase in line with rising interest:
is an income effect, too, which enables the household to demand both more goods and
leisure. Thus, the household can afford the now dearer leisure, and it is even possible S = S( i ). (40)
<+)
that more of it will be consumed than before, i.e. that labor supply will diminish. The
net effect of these two potentially opposing impacts cannot be determined theoreti- The Classical abstinence theory of interest is not to be taken too seriously. In order
cally; but empirical evidence over the last century suggests that, on the average, labor to explore the positive interest rates that occur in reality, we d o not need the assump-
supply diminishes due to rising real wages. Nevertheless, we will employ function (39) tion that savings mean a "sacrifice" to the household generally. Rather, it is perfectly
in the following analysis. adequate to assume that the households' capital supply is scarce in comparison with
If we now turn to the household's choice between consumption and savings, it has the firms' demand for it. If, at an interest rate equal to zero, the households would
to be stressed from the outset that this decision is not independent of the one above. wish to save an amount of $ 100.000, but the firms would like to get $ 2.000.000, the
The household does not decide on its income first, and on the utilization of income latter will overbid one another until an equilibrium interest rate of, say, 10 per cent
afterwards: but he makes these decisions simultaneously. It is only for pedagogical is established. It is even conceivable that capital supply would remain at $ 100.000, i.e.
purposes that we discuss these problems independently. that not a single household would be induced to save more, because substitution and
When the Classics spoke of "savings", they thought primarily of the investing income effects cancelled each other out. In this case, the economic function of the
entrepreneur, and "savings" and "investment" appeared almost identical to them. interest rate would merely consist in directing capital into its most efficient uses.
This view was quite appropriate in those days. But the Neoclassics - and we will In so far as interest receipts constitute a neglegible part of total income, (40) and
follow them in this respect - were well aware of the fact that the decisions to save and (37) suggest that the household will consume less when the rate of interest goes up.
to invest are entirely different. Firms undertake investments in order to make profits This is because total income remains almost the same; and because the household's
in the future; but it is primarily households that decide how much to save, and they aim to save more it must reduce consumption accordingly. Therefore, the Classical'
do this in order to distribute their income over time, to receive an interest, or to leave consumption function reads
a bequest. C = C(i). (41)
(-)
Tne abstinence theory of interest, put forward by WILLIAM NASSAU SENIOR, was
Our description of the household's behaviour is now complete, and we want to
one of the Classical explanations of savings and interest, and it was perhaps the most
important. According to this theory, the household generally prefers present to future summarize the results. According to Classical theory, the representative household
consumption (where future consumption is the same as savings) on purely subjective receives labor, interest, and profit income. It decides on labor and interest income by
grounds, as well as because of risk-aversion. It was argued that the household saved itself; only profits are given because they depend on the prevailing profitability of the
only if interest high enough to counteract the sacrifice and the risk of waiting was firms. (In the long run, however, the household can change its equity interests; and
offered to him. then, all components of total income are endogenous.)
For simplicity only \ve assumed that the choice between labor and leisure is
Considered with the aid of marginal utility theory, the household will increase its
independent of the rate of interest, and that the choice between present and future
savings up to the point where the marginal utility of the interest payment and the
consumption is independent of the real wage rate 1 . Owing to the special assumptions
marginal disutility of waiting just balance. A rise in the given rate of interest increases
the marginal utility of a certain level of savings and, hence, induces the household to 1 We admit freely that this is an inconsistent assumption: If the household increases its labor
save more. supply subsequent to a rise in the real wage rate, it must necessarily either spend or save more,
wfiich contradicts the premise t h a t both consumption and savings are independent of the real
Furthermore, we will assume for simplicity's sake that the decision to save is not wage rate. Allowing for the dependency, however, would involve no different results but
influenced by the real wage rate and that the decision to work is not influenced by the prevent any graphical demonstration of the model.
48 Chapter IV, The Classical Theory §22 The Labor Market 49

of marginal utility theory, labor supply and savings proved to be positively correlated
to the real wage and the interest rate, respectively, while consumption was seen to
decline as the rate of interest rises.
The representative household formulated its plans subject to the budget constraint
(37) which may be rewritten as
P - C + P - S = w - N s + i (B 0 + P • S) + 7i. (42)
On the left hand side of this equation, consumption and savings are the two possible
utilizations of income, whereas, on the right, there are the three sources of income.
B 0 + P • S is identical to B d : the initial stock of bonds, plus current savings, is
Figure 20
identical to the planned stock at the end of the period, and interest income stems from Disequilibrium in the labor market
the latter stock.

§ 22 The Labor Market In the Classical model - as in reality - , the price level P is exogenous to the labor
market so that only the nominal wage rate can changeT Hence the excessive real wage
We have already derived the labor demand of the firms and the labor supply of the rate (w/P) 0 is due to an excessive nominal wage rate. As seen from the figure above,
households. In this section, we combine these two constituents of the labor market in this causes an excess supply of labor, i.e. unemployment. The Classics presumed that
order to establish the equilibrium levels of employment and the real wage rate. From in such a case, the unemployed would be willing to work at a lower nominal wage rate,
the Classical point of view,/«// employment and the equilibrium real wage rate result and that they would underbid the employed. It was also considered possible that the
from equating labor demand and supply: employers could now enforce a reduction in wages.
Both possibilities entail a reduction in real wage rates, and - according to the
Classics - this reduction would continue until the equilibrium real wage rate (w/P)*
and full employment were regained.
Looked at mathematically, these are two simultaneous equations with two un- With regard to the converse case of a real wage below the equilibrium rate we can
knowns: N and w/P. Provided that a solution exists at all, it is unique given our argue analogously: The firms will want to get more labor than they can at the
assumptions. In a graphical Representation, the solution is determined at the point of prevailing wage rate; accordingly, some of them will offer higher wages to entice
intersection of the demand and supply schedules: workers away from other firms; and this will last until the equilibrium is re-
established. Obviously, the Classical adjustment process can only be expected to work
smoothly if there are no arrangements or cartels, like employers' federations and
unions.
The reader should notice that the theoretical notion of "full employment" is not
necessarily identical to that of labor market statistics. Full employmenfN* is con-
sistent with the fact that some members of the working population do not work; in
Figure 19, these are Nj - N* persons. But, and this is the central point, they volun-
tarily do not take up employment because the prevailing wage appears insufficient to
them.
Therefore, we will reserve the term unemployment for involuntary unemployment
Figure 19 only, and N* is called full employment equilibrium since no one is involuntarily
Equilibrium in the labor market unemployed. This terminology is suggested by a normative consideration: At N*, all
workers are acting in accordance to their preferences, and in this special sense the
The plans of firms and households are compatible only when the equilibrium real result can be called "optimal". On the other hand, a state can hardly be regarded as
wage rate (w/P)* prevails: in this case, every demand for labor is fulfilled, and every optimal in which the working population is forced to work as much as possible.
household can sell its labor supply. The results of Classical labor market theory can be summarized as follows. Long-
We now have to ask what happens when the real wage rate differs from its lasting involuntary unemployment is impossible since the wage rate can always adjust
equilibrium level; for our equilibrium analysis makes sense only if there is a tendency as to bring about an equilibrium; in equilibrium, the plans of both firms and house-
of the real wage rate to move towards the equilibrium rate. Let us, therefore, consider holds are fulfilled; and these hypotheses are only subject to the premise that there are
a situation where the real wage is too high: no institutional constraints that hinder a wage adjustment.
50 Chapter IV. The Classical Theory § 24 The Commodity Market 51

§23 The Capital Market

"Interest is commonly reckoned for Money ...but this is a mistake."
(NICHOLAS BARBON)

In the capital market, bonds are traded; they are demanded by the firms and supplied
by the households. We have already derived the firms' real demand for financial
capital, in § 20, which proved identical to their investment:

. AB S Figure 21
1(0 3 — , (44) Equilibrium in the capital market

The firms' capital demand was seen to be a declining function of the interest rate
because any increase in the interest rate increases capital costs. Similarly, in §21, we likely to accept a lower-return because it is better to receive this than none at all.
derived the households' saving function, i.e. their real supply offinancial capital, which
Therefore, the rate of interest will decline until equilibrium is regained where every
turned out to be an increasing function of the interest rate:
firm and every household can realize its plans and where, accordingly, there is no
reason for the interest rate to change. Since the capital market comes rather close to
AB D
S®: = -P-. (45) a perfect market in the theoretical sense, we can expect a fairly quick adjustment.
We have to stress that it is claims that are traded in the capital market - not
The Classics, as already indicated, conceived of investment and saving as almost commodities and not money either. Commodities are supplied and demanded only in
identical processes carried out by the same persons. To them, "saving" meant not only the commodity market (which we will be considering next), and money is not regarded
a sacrifice of consumption but also a demand for real capital goods; and hence it is as capital (which does not deny that it serves as a medium of exchange in the capital
hardly astonishing that the equality of savings and investment did not seem a severe market).
problem to them. This point is all the more important to Classical economics to the extent that it
The Neoclassics came closer to the present view as they clearly perceived the conflicts with common opinion. NICHOLAS BARBON, in 1690, was the first to suggest
difference between decisions to save (i.e. to postpone consumption) and decisions to this model of the capital market. He maintained that borrowers, if they take a loan,
invest (i.e. to undertake an enterprise). If these decisions are reached independently, do not want the money for itself but the goods it can procure and that the capital
there must be an adjustment mechanism that reconciles them. This adjustment mech- market can be conceptualized as the exchange of goods for claims (bonds). Therefore,
anism is the rate of interest. Equating investment and saving yields the equilibrium money does not enter into the determination of the interest rate: the latter is in-
condition fluenced only by real forces, i.e. marginal productivity of capital and marginal disutil-
I(i*) = S(i*) (46) ity of waiting.
Thus, any abstention from consumption on the part of households results in an
which determines the natural rate of interest i*. The natural (or equilibrium) rate of increase in the social stock of capital because this abstention is equivalent to saving,
interest entails the equality of savings and investment. It is equal to the equilibrium and savings become equal to investment by means of the interest rate. The results of
marginal productivity of capital because equation (23) this analysis are that, in the Classical model, claims (bonds) are supplied and de-
manded in the capital market, and the interest rate is expected to bring about a quick
9F
market clearance.
=
' 6K <47>

shows that the firms always adjust the capital stock in order to equate the marginal
productivity of capital to the rate of interest. The natural rate of interest is also equal §24 The Commodity Market
to the marginal disutility of waiting since the households increase savings up to this
point. Hence, the rate of interest can be said to equalize the marginal productivity of To conclude our non-monetary analysis, we discuss the commodity market in which
capital and the marginal disadvantage of waiting (see Fig. 21). commodity supply and investment and consumption demand meet. Equilibrium in
The actual rate of interest may, of course, depart from the equilibrium rate for the Classical sense means equality of commodity supply and demand:
some time - but not for too long. For if it is above i*, for instance, investment will
decrease whereas savings may increase; thus capital supply will be abundant and some s (48)
Y Q = C(i) + I(i).
households will not succeed in selling the capital they wish to supply. They are then

.Js,
52 Chapter IV. The Classical Theory §25 The Quantity Theory of Money 53

We now want to explain why a seperate analysis of this market is not necessary: demand occurs in one market, there must be at least one other market where excess
The two sectors of the economy, firms and households, make decisions on three supply exists because, otherwise, the sum of the excess demands would not vanish.
activities each. The firms decide on We can conclude from this that there must be an equilibrium in the commodity
market if the labor and capital markets are both cleared. Hence, the commodity
- the supply of commodities,
market can be neglected in our non-monetary analysis.
- the demand for labor, and
Note: W h e n studying the literature, one frequently encounters the opinion that
- the demand for financial and physical capital,
equality of savings and investment is in itself sufficient to ensure equilibrium in the
whereas the households decide on commodity market. The usual motivation of this runs as follows:
- the demand for consumption goods, Y = C+ I and Y = C+ S. (56)
- the supply of labor, and
From this it follows that S = I implies Y = Y . (57)
- the supply of capital, i.e. savings.
This, though a truism, holds perfectly only in ex post analysis. In ex ante analysis,
It can easily be seen that both sectors have two degrees of freedom. If the firms
however, the two " Y " do not mean the same, since the Y on the left hand side
choose some investment and labor demand, they have also committed themselves to
represents planned commodity demand, and that on the right, planned income of
a certain supply of commodities; and the households have implicitly decided on their
households. Equalization of S and I amounts to deducing that commodity demand
consumption once they have chosen certain amounts of labor supply and savings.
equals income - and this has hardly anything to do with equilibrium in the commodity
If the firms' demand for labor matches the households' supply of it, and if capital
market. A n equilibrium in the commodity market requires that commodity supply
demand and supply also coincide, then commodity supply must equal commodity
and demand match, and planned commodity supply, in turn, is entirely different from
demand, too. Or, to put it differently, simultaneous equilibrium in the labor and
households' planned income because it is planned by the firm sector.
capital markets implies equilibrium in the commodity market. This can be proved
By means of Walras's law we can infer that S = I implies an equilibrium in the
mathematically in the following manner.
commodity market on the condition that an equilibrium in the labor market prevails
Using the definition of profits (19) and replacing F ( N , K) and K. — K 0 with their
simultaneously 1 . It is hence a fallacy that S = I implies an equilibrium in the com-
respective equivalents Y and I yields:
modity market; and it is downright absurd to refer to the capital market as a commod-
7t = P - Y s — w • N d — i (B0 + P I ) . (49) ity market.
To conclude, let us discuss the possibilty of a demand gap in the commodity
The firms' financial constraint (17) stated that any investment must be financed market, i.e. unsufficient demand for commodities. At first sight, it would seem that
by an equivalent supply of bonds: this situation is likely to arise because the households do not spend all their income
ABS - which, in equilibrium, is equal to commodity supply - but save a certain amount.
I=-jT- (50) But, in equilibrium, these savings are equal to investment demand so that investment
fills the gap exactly. Hence, in the Classical model, a demand gap is utterly incon-
Furthermore, we retain the households' budget constraint (42):
ceivable.
s
P-C + P S = w N + i (B0 + PS) + k . (51)

Putting all terms of these equations on the left hand side and rewriting (50) as §25 The Quantity Theory of Money
P . I - P -1 = 0 yields
7 i - P Y s + w N d + i(Bo+PI) = 0 (52) "Money is a veil." (ARTHUR CECIL PIGOU)

P •I- P•I= 0 - (53) The Classical non-monetary analysis which we have been discussing u p to now was
P • C + P • S - w • N s - i (B0 + PS) - n = 0 . (54) complemented by the so-called quantity theory of money. Money, looked at from the
Classics' point of view, is chiefly commodity money: it consists of gold, silver, and
Finally, adding (52) to (54) and rearranging terms amounts to coins of precious metal. Bills of exchange and banknotes were not included in the
quantity of money for reasons that we will explain later. Money was supposed to serve
P(C + I — Y s ) + (1 - i)P(S - I) + w ( N d - N s ) = 0 (55) as
commodity m a r k e t capital market labor market
- a generally accepted means of payment and
According to (55), the excess demands in the three markets, each of them multi- - a general measure of value.
plied by the respective market price, sum to zero. This is called Walras' law. Walras's
law does not hold only in equilibrium but under any circumstances: If an excess 1 This will become only too obvious in Keynesian theory.
 54 Chapter IV. The Classical Theory §25 The Q u a n t i t y Theory of Money 55

As a generally accepted means of payment, money permits an indirect exchange of

commodities. This is an enormous advantage as should become evident by the follow- | money and the exchanging the money for something else, makes no difference in the
essential character of transactions. It is not with money that things are really pur-
ing example: Imagine a peasant who wishes to sell grain in exchange for tools, and chased. . . . There cannot, in short, be intrinsically a more insignificant thing, in the
an artisan who wants to sell tools in order to get some meat. No exchange at all can economy of society, than m o n e y . " 1
take place between these two, because the peasant's offer does not match the artisan's The interpretation of this passage must not be pushed t o o far, however, since M I L L
need. It will probably take a great deal of searching until they find someone whose conceded a short-run impact of money on real variables. Nevertheless, the central idea
wishes correspond to their own. is maintained: Money is, so to speak, the lubricant of the economic process; without
In a monetary economy things are much easier: The peasant can sell his grain to its aid exchange would be much more troublesome; but a single "drop" "of it is
another person first and then exchange the revenue for the tools. Finally, the artisan sufficient - the quantity makes no difference to the real variables. These thoughts gave
can use this money to buy his meat. Thus the advantages of indirect exchange rise to the notion of a " d i c h o t o m y " between the real and the monetary sectors of the
originate from the fact that there need not be a double coincidence of wants (i.e. one economy whereby the two have no influence on each other.
person demands what is supplied by the other, and vice versa) but only a single Hence, with respect to the monetary sector, only the price level is to be explained
coincidence - and, of course, an agreement on the price. This constitutes the transac- and this is achieved by the quantity theory of money. The Classics chose the following
tions motive for using money. point of departure: The price level, being the average price of all commodities, is the
Due to its function as general measure of value, money serves for reckoning the ratio at which money exchanges for the other commodities and can be determined by
values of all other commodities: these values are expressed in money terms. This yields equating demand and supply in a fictitious money market. Fictitious, because money
remarkable efficiency gains, too. In an economy where there are 1.000.000 different is not actually traded in a specific market, but is exchanged for commodities in every
goods - and that is not very many - , about 500 billion relative prices do exist which market. Yet this does not preclude us from mentally distinguishing a money market.
specify the pairwise exchange ratios of the goods. Thus, using a general measure of 5 The "money market" in this sense must not be confused with those markets for
value reduces the number of prices to a 500.000th of the original number (i.e. to short-term loans which are called money markets in business language.
1.000.000), and this considerably facilitates comparisons. A general measure of value In the following we suppose that money is not produced and supplied as other
need not physically exist: MONTESQUIEU, for instance, gave an account of African K• goods but that its quantity is given by the available quantity of precious metals or by
tribes who used a purely imagininary measure of value, so-called "makuten", for the policy of the central bank. In doing so we neglect the possibility of gold produc-
expressing the value of different goods. Contemporary examples of imaginary mea- tion or creation of money by modern banks. This amounts to assuming an exogenous
sures of value are the British guinea and the Special Drawing Rights of the Inter- money supply.
national Monetary Fund. In the following, we will speak of abstract prices if the If the money supply is a given stock, the demand for money must also be a stock.
measure of value is not identical to the means of payments; otherwise, we will speak According to the quantity theory of money, the demand for a cash stock arises from
of money prices. the time-lag between receipts and expenditures. When people receive their money
Money's function as a store of value was assigned to it by the Classical authors in income, they do not spend it instantaneously but need some time to decide on their
• a restricted sense only: In order to serve as a means of exchange, money must also be consumption and savings. This time-lag causes a permanent average stock of cash to
a temporary store of value; and this is the reason why it is always durable goods that be held by the individuals. Let us calculate this by means of a representative individual
are selected to serve as money. On the other hand, the Classics denied that money acts who holds every dollar for three months. Thus the average duration of holding cash
as a store of value proper because, in contrast to other stores of value, it yields no balances, k, amounts to
interest. Hence, as they argued, it would be unwise to hold more cash than was
required for transaction purposes thus incurring a loss of interest on the assets that k = three months = 1/4 year. (58)
could otherwise have been held. Consequently, the Classics considered hoarding an Henceforth we will call this variable the Cambridge k, for short, since it was first
irrational exception. introduced by the economists MARSHALL and PIGOU from Cambridge. Let us further
Now we are able to turn to the core of Classical monetary theory, i.e. the principle assume nominal national income of $ 100 annually:
o f t h e neutrality of money. According to this, money is a mere "veil" over real events;
it sometimes obscures them; but, apart from the efficiency gains described above, it P • Y = $ 100. (59)
does not influence the real events at all. The last great Classic, JOHN STUART MILL,
Now, what quantity of money will be called for? If each dollar were to be kept for
for instance, opens the second volume of his "Principles", after he has completed
the whole year (on average) then $ 100 would be necessary to match the demand for
more than 700 pages of real analysis, with the following words:
money because the nominal income is exactly $ 100. But if, on the other hand, each
"Having proceeded thus far in ascertaining the general laws of Value, without the
idea of Money (except occasionally for illustration) it is time that we should now
superadd this idea. . . . It must be evident however, that the mere introduction of a 1 MILL, J.SI. (1848) Principles of Political Economy. London: P a r k e r . Reprint Toronto 1965:
particular mode of exchanging things for one another, by first exchanging a thing for University Press, pp. 2 ff.

3 1
 52 C h a p t e r IV. T h e Classical Theory §25 The Q u a n t i t y Theory of Money 57

dollar is kept for a quarter of a year, only $ 25 are required. These $ 25 circulate four be an increased money demand, and consequently an increased money value, or price,
times a year and hence support an annual sales volume of $100. for things of all sorts. This increased value would do n o good to any one; would make
In general terms, we can calculate the nominal demand for money (L°) as the no difference, except that of having to reckon pounds, shillings, and pence in higher
product of nominal income and the Cambridge k: numbers... If the whole money ?I1 circulation was doubled, prices would be
doubled." 1
L" = k • P • Y . (60)
M I L L ' S Neoclassical succesors specified the crucial point more exactly: They as-
The demand for money must equal the exogeneous money supply in equilibrium: sumed, as M I L L did, that the cash balance of every individual is doubled instanta-
neously. Starting from an original equilibrium, this means that real cash balances are
M = k • P • Y. . (61)
now considered too high because prior to the increase they were considered just
This is called the Cambridge equation-, it is one way of expressing the quantity appropriate. To reduce their excess cash balances, individuals will increase their
theory of money. Another way was chosen by the American economist IRVING FISHER demand for commodities. This increased commodity demand, however, will not be
who considered the velocity of circulation instead of the Cambridge k. In the above accompanied by a higher commodity supply since the latter depends on real variables
example, the Cambrige k came to a quarter of a year and this, as we saw, meant that that have not changed. Our first result, therefore, is that there will be an excess
money circulated four times a year, i.e. the velocity of circulation amounted to four. demand in the commodity market which, over time, will bid up prices.
Thus the velocity of circulation is merely the reciprocal of the Cambridge k: When can we expect this increase in prices to stop? It would be a fallacy to suppose
that this would occur when all individuals have spent their surplus cash balances.
v - j . (62) Though a single individual unquestionably can reduce his cash balance, this is not
possible for the economy as a whole since one individual's expenditures are the
Substituting v for k in the Cambridge equation and rearranging terms yields the receipts of another. The price pressure hence is not eliminated by reducing nominal
quantity equation: cash balances. We unravel this problem as soon as we realize that the individuals do
M - v = P • Y. (63) not want to maintain a certain nominal but a certain real cash balance, i.e. command
over a certain quantity of goods. Therefore, prices will rise until real cash balances,
The Cambridge equation and the quantity equation are basically equivalent as v
M/P, have resumed their original level. When this happens, the purchasing power of
proved to be the reciprocal of k. But the former clearly indicates the individual
decisions to hold money whereas the latter refers to the somewhat mechanistical cash balances is just the same as before, and hence the latter match the individuals'
notion of a "given" velocity of circulation. Therefore, we will use the Cambridge preferences. In order to keep the original level of M / P , prices must clearly rise in
equation from now on; but we stress that the quantity equation could be employed proportion to the quantity of money.
equally well. We refer to the dependence of commodity expenditures on real cash balances as
Let us now have a closer look at the four variables of the Cambridge equation. the Cambridge-effect. The Cambridge effect is the link between changes in the quan-
The money supply (M), as stated above, is a given stock. The Cambridge k depends tity of money and changes in the price level; and it ensures that prices move in
on how much cash the individuals want to hold. We assume - and this assumption proportion to the quantity of money.
is crucial - that it is constant in the short run. Finally, real income (Y), as we The Neoclassical economist K N U T W I C K S E L L , offe of the most outstanding mone-
recognized above, is determined by the preferences of the households and the produc- tary theorists, confessed in his critique of the quantity theory of money 2 that the latter
tion technology. gives a logically valid explanation but can be attacked on empirical grounds. While
the space at our disposal does not permit a full account of W I C K S E L L ' S own contribu-
M = k • P • Y. (64)
tion (which, in fact, is similar to that of KEYNES) we would, at least, want to take note
Therefore, the price level is the only variable not given from elsewhere and this of his central criticisms:
immediately yields the conclusion of the quantity theorists: The quantity of money, First, W I C K S E L L doubts the constancy of the velocity of circulation (or the Cam-
k and Y being given, determines the price level uniquely and causally, and k and Y bridge k) which he believes to be one of "the most airy and less seizable variables of
themselves are independent of the quantity of money. For example, any doubling of the economy". But if the velocity of circulation is not constant at least in the short
the quantity of money must entail a doubling of prices. This is the quintessence of the run, then the impact of a change in the quantity of money on the price level cannot
quantity theory of money. be forecasted.
A doubling of the quantity of money must entail a doubling of the price level - Second, W I C K S E L L criticizes the quantity theorists' narrow definition of money,
O.K. But, as economomists, we should not trust in such "musts", but ought to ask pointing out that precious metals and coins can be substituted by notes, bills of
what is the adjustment process which brings a b o u t this doubling. On this score, let us
listen t o JOHN STUART MILL:
1 MILL, o p . c i t . p . 15 /, Q <C<\
"Let us suppose, therefore, that to every pound, or shilling, or penny in possession
2 WICKSELL, K. (1898) Geldzins und Güterpreise. Jena: Fischer. English translation ( i y b 5 )
of any one, another pound, or shilling, or penny were suddenly added. There would Interest and Prices. New York: Kelly
 58 Chapter IV. The Classical Theory §27 The Classical Model 59

exchange, or cheques. If a wider definition of the quantity of money is considered buyers. A general over-supply, or excess of all commodities over demand, so far as
appropriate, the latter can hardly be treated as exogenous since, for example, the demand consists in means of payments, is thus shown to be an impossibility". 1
individuals can decide freely on how many bills of exchange they want to draw. It may
Anyone who accepts all the arguments of Classical theory u p to this point must
also happen that an increase in the quantity of precious metals entails a decrease in
the quantity of bills of exchange such that the total quantity of money remains also accept Say's law. For what could be the cause of a general "glut" or a general
constant. " d e m a n d gap"?

Third, W I C K S E L L doubted the crucial link between the quantity of money and the - Sudden hoarding of money is eliminated as a possible cause of inadequate commod-
price level, namely, the strength of the Cambridge-effect. Later on we"will encounter ity demand because hoarding was considered irrational by the Classics. Yet even if
a similar critique that suggests substituting the Cambridge-effect by another one (the we do admit this possiblity, it only implies a change in the circulating quantity of
Keynes-effect). money that will soon be offset by a corresponding change in prices.
To conclude. The prevailing monetary doctrine of the Classical age is the quantity - Real savings are still less suited for explaining a general glut since they are matched
theory of money. It implies a dichotomy between the real and the monetary sector of by an equivalent investment demand almost instantaneously.
an economy and states, essentially, that changes in the quantity of money cause Yet these two arguments are superficial. The deeper case for Say's law lies in the
proportional changes in the price level only, whereas all real variables remain con- obvious fact that nobody plans to produce something without simultaneously plann-
stant.
ing to demand something else. Because decisions to supply and demand take place at
For clarity of exposition, we have considered only the "pure core" of this theory; the same instant, it is impossible that demand and supply differ in total. This does not
and hence we want to add that the quantity theorists were indeed far from denying mean, of course, that they match in each market (this is the very meaning of Say's law
every real impact of monetary changes - though they were also far from stressing this that the newspapers usually assign to it; but it is completely absurd). On the contrary,
point 1 . As A L F R E D M A R S H A L L once put it: it was SAY himself who analyzed the disturbances which can take place in a single
"This so-called 'quantity theory of the value of money' is true in just the same way m a r k e t . A n d RICARDO states:
as it is true that the day's temperature varies with the length of the day, other things
"Too much of a particular commodity may be produced, of which there may be
being equal; but other things are seldom equal." 2
such a glut in the market; but this cannot be the case with respect to all commodi-
ties." 2
What is more, Say's law was not meant to deny the possibility of temporary crises
§ 26 Say's Law which could arise, for instance, from slow price adjustments. Rather, the Classics
"The fundamental things apply, as time goes by." (From "Casablanca") wanted to contest the possiblity of a general and lasting glut - a thesis proposed so
frequently in the popular (and even scientific) literature. They argued: Everybody
The law named after J E A N B A P H S T E SAY is one of the most famous propositions of supplies goods and services only in order to buy some other commodities; in the
Classical doctrine. One familiar definition of it reads: course of capital accumulation and rationalization, therefore, production will in-
Any supply creates its own demand. This is because every expansion of production crease only; but a general market saturation is downright inconceivable since no one
entails additional factor incomes that are used by their recipients for demanding already saturated would supply anything more. And some of them added that the
commodities. saturation of all individuals with respect to all their wants would not be an economic
However innocent a statement this may seem, vehement discussions have attended problem, but the solution of all economic problems.
to it up to the present. Let us try to gain a closer insight into this. In the first instance,
Say's law does not refer to the identity ex post of sales and purchases but to planned
magnitudes; it is not a mere truism. Second, Say's law is not an axiom but a theorem, §27 The Classical Model
i.e., a proposition which is not supposed true but can be derived from others. The
following passage from M I L L brings this home: In the preceding section we discussed the various individual markets of Classical
economics covering in the process all essential issues. In order to get a comprehensive
" . . . is i t . . . possible that there should be a deficiency of demand for all commodi-
impression of this doctrine, we are now ready to integrate these isolated elements into
ties, for want of the means of payment? Those who think so cannot have considered
an algebraic and geometric Classical model.
what it is which constitutes the means of payment for commodities. It is simply,
commodities. Each person's means of paying for the productions of other people The algebraic Classical model summarizes the labor demand and supply func-
consists of those which he himself possesses. All sellers are inevitable and ex vi termini tions, (32) and (39), the production function, (34), the savings and investment func-

1 MILL, op.cit. vol. 2, p.91. Italics added.

1 Cf. ESHAG, E.(1963) F r o m Marshall to Keynes. Oxford. Basil Blackwell 2 RICARDO. D. (1817) Principles of Political Economy and Taxation. Reprint of the 3rd edition
2 MARSHALL, A. (1926) Official Papers by Alfred Marshall. London: Macmillan, p.267 London 1924: Bell and Sons. p. 275.
 §27 The Classical Model 61
60 Chapter IV. T h e Classical Theory

p
tions, (40) and (33), and the Cambridge equation, (61). Supplementing this with the
identity w = (w/P) P yields the simultanous equations C, for Classics:

Nd | = S* = N ' ( - N* - (C.l) E
w,
/ w'
Y = f(N) (C.2)
S(i) = I(i) (C.3) y j J k V
P*
M = k • P •Y P* (C.4)
Y*
(C.5) X IB'
I
n
These are six simultaneous equations - since (C.l) contains two - which, due to N* ^
/r- •
our previous assumptions, determine the six indicated variables uniquely.
(C.l) represents the equilibrium condition of the labor market which yields full
employment, N*, as well as the equilibrium real wage rate, (w/P)*.
M
(C.2) is the production function. Given that we ruled out changes in the capital
stock, production depends solely on the variable employment. The commodity supply N
function could have been written down as well; but this formula indicates more clearly
that commodity supply directly follows from the employment level, N*.
(C.3), the market clearance condition of the capital market, determines the natural
interest rate, i*. Of course, the latter implies a certain amount of savings and invest-
ment.
(C.4) is the Cambridge equation. Since the quantity of money and its velocity of
circulation are assumed given, the equilibrium price level, P*, is implied by real
output, Y*.
(C.5), finally, is a purely formal identity. It states that a certain nominal wage rate,
w*, is implied by the equilibrium real wage rate, (w/P)*, and the equilibrium price
level, P*. The working of the model requires that this nominal wage rate is not given
Quadrant I reproduces Figure 19 of the labor market which has simply been
from outside.
The algebraic model, among other things, elucidates the Classical dichotomy. The inverted. The axes always represent positive values.
Quadrant II reproduces Figure 11 of the production function which has also been
real sector of the economy is covered by the equations (C.l) to (C.3) where all real
magnitudes are determined. From (C.4) and (C.5), the price level and the nominal inverted.
waj>e rate are derived as purely monetary variables which do not influence the real Quadrant III depicts the capital market.
sector. Quadrant IV must be explained in greater detail. T h e hyperbola there represents
Such an algebraic model serves very well to demonstrate the core of a doctrine. the Cambridge equation. This becomes clear if you recognize that the Cambridge
Recollecting the various equations facilitates recollecting the whole theory - but the equation can be converted to
danger exists that one develops too mechanistic an understanding of the doctrine. We "
want to stress, therefore, that the equations are not put as natural laws but merely give
a brief account of a few seemingly important relationships. where P is the dependent and Y the independent variable. The given magnitudes, M
Next we turn o u r attention to the geometric exposition. This usually has the and k, determine the position of this hyperbola and any real income (Y) yields a
advantage of depicting events simultaneously which, in reality, also take place simul- specific equilibrium price level (P).
taneously; in this respect they are similar to algebraic models but superior to verbal Quadrant V, finally, is the equivalent of equation (C.5). Since w/P and P are the
description. Assembling Figures 11, 19, 21, and supplementing them by two new two variables involved, we encounter a family of curves
graphs results in the total model (see F'ig. 22).
(66)
Any quadrant of Figure 22 corresponds to the accordingly numbered equation of P =
w/P
the algebraic model. It should be read in numerical order:
 62 Chapter IV. The Classical Theory §28 Digression: Walras* M o d e l 63

here with parameter w. In this quadrant, w/P and P are given from the labor market The subjective preferences of a household are described by a preference function
and the Cambridge equation, respectively. By means of these magnitudes, a particular
hyperbola is "singled out", such that it e w e r s the point ((w/P)*,P*). Uh(XH..-,x.,G); h — 1 ...H (67)
The Classical model is now described completely. In Chapter VI we will use both which it seeks to maximize. This means, the household chooses all its d e m a n d s a n d
the algebraic and geometric representations in order to analyze macroeconomic poli-
supplies of goods such that its utility attains a maximum. But it has to t a k e notice of
cies.
its budget constraint by which it is forced to keep expenditures and receipts in balance:

Z Pgxhg+¿7rhf = 0 (68)
§ 28 Digression: Walras' Model g=i r=i

"The Walrasian Auctioneer is a great myth; I emphasize both words." The budget constraint requires the (positive) receipts to match the (negative)
expenditures: i.e., the sum of both must vanish. 7thf denotes the profit i n c o m e house-
(JAMES TOBIN) hold h receives f r o m firm f. We take the profit shares as givens a n d define
H
Finally we want to cast a glance at the microeconomic model of general equilibrium as T. nhf.
first proposed by LEON WALRAS. Though our concern is with macroeconomics, this h= 1
digression is almost inevitable because Under certain conditions which cannot be adduced here, the optimization process
yields the supply and demand functions of the H households:
- in the first instance, (Neo-) Classical theory was primarily microeconomic in spirit;
x
thus taking note of its microeconomic formulation will deepen o u r understanding hg = x h g ( p 1 , . . . , p G ) ; h = 1...H; g = l...G. (69)
of this doctrine.
Any of these H times G functions depends on
- Second, Walras' model is the prototype of general'equUibrium theory which offers
a microeconomic total analysis of the economy. In this respect, it is both point of - individual preferences,
departure and reference point for various further developments that are to be - original stocks of durable goods, and
discussed in Part Three. - prices.

T h e F r e n c h m a n LEON WALRAS w a s the first, in 1874, to p u t f o r w a r d a theory of Since preferences and original stocks are exogenous, only prices are explicitly
microeconomic general equilibrium 1 . Having at first been refuted or ignored, his noted in (69).
model was to become the "magna carta" of economics, as SCHUMPKTER put it. In the Let us turn to the firms, F in number. Here, again, we denote d e m a n d s by negative
following, we want to describe Walras' model in a rather simplifying manner so as to and supplies by positive signs. Every firm seeks to maximize its profits which can b e
bring out its essence. written in the familiar form "revenues minus costs" as
An economy piay consist of households and firms, H and F in number. They are
indicated by h and f, respectively. Furthermore, G different goods exist, indicated by V f = 1 ... F . (70)
g, which are traded in the same number of markets. The term " g o o d " is used in its t= i
broad sense and also covers services and claims. The model refers to an economy Trying to maximize profits, the firms are constrained by their p r o d u c t i o n functions:
without money.
x f g = x f g (x f „•••,*r G ); f = 1...F; g = l...G. (71)
Prices p , , p 2 , . . . ,p G of the G goods are expressed by an imaginary measure of
value: they are abstract prices. Thus, the relative price of goods i and j is reckoned as The production functions are indexed by g, bccause for any firm there are precisely
Pi/Pj- as many production functions as there are goods; and they are indexed by f, because
The quantity of a certain good, g, that is supplied or demanded by household h each firm's productivity will tend to be different from the others'. E q u a t i o n s (70) a n d
is labeled as x h g . We denote it by a negative sign if it is demanded and by a positive (71) yield, under certain conditions, the demand and supply functions of the firms:
sign if it is supplied by "the household. Take an example: If household no. 9 supplies
20 hours of a certain quality of labor which is good no. 6, we write x 9 6 = 20. But if xfg = x f g ( P l , . . . , p G ) ; f = 1 . . . F; g = l . . . G . (72)
it demands apples of some sort, referred to as good no. 4, we have x 9 4 = — 3.
Our description of individual behaviour is now complete, and we merely have t o
go a small step f u r t h e r to obtain the results of Walras' model. A general equilibrium
requires that d e m a n d and supply be equated in each market:

1 WALRAS. L. (1884) Elémcnls d'économie politique pure ou théorie de la richesse sociale.

Lausanne. English translation ( 1954) Elements of pure Economics. London: Allen and Unwin
E P , • Xh, + £ P, • xf„ = 0; g = 1... G. (73)
h=l r=i
 64 Chapter IV. The Classical Theory
§28 Digression: Walras' Model 65

T h i s equation says that the s u m of all individuals' (negative) demands for a certain
Walras' Law: T h e sum of the excess -demands in all (G) markets is equal to zero.
g o o d (g) m u s t be equal to total (positive) supply of this g o o d . Let us, for instance,
consider t h e market of good no. 5 ("potatoes"): Therefore, a n equilibrium in G - 1 markets implies that there is also an equilibrium
in the G t h m a r k e t .
Walras' law gives rise to the following problem: Since the s u m of the excess
Z P5-Xh5+ £ P5-*T5=0. (74)
h=l f=l demands is given (i.e. zero), we are left with G — 1 independent equilibrium conditions
and, therefore, the number of independent equations (G — 1) is less than the number
T h e sum of demands for p o t a t o e s must equal the sum of supplies in order to
establish an equilibrium in the p o t a t o e s m a r k e t In (73), this condition is applied to of unknowns (G).
all markets. Walras himself arrived at the result t h a t only G — 1 relative prices are determined
by the equilibrium conditions, but not the G abstract prices. He chose an arbitrary
E q u a t i o n s (69), (72), and (73) c a n be combined to yield W a l r a s ' model:
good as numéraire, i.e. as measure of value, and argued that the G — 1 independent
X
bS ~ X
hg ( P i , • - •, p c ) (W.i) equations a r e just sufficient to determine the G — 1 relative prices with respect to the
numéraire. T h u s , all G — 1 relative prices, in the sense of exchange ratios, are deter-
X X
hf = hf(Pl,.-.,PG) (W.2) mined.
» F
It is easily seen why the G abstract prices, sometimes called " m o n e y prices", are
£ P g • x h g + f Z p , • x,, = 0. (W.3)
not determined within the model. A doubling of all abstract prices has no impact on
Hence, Walras' model consists of utility functions or the budget constraints; and all relative prices, demands, and
supplies remain the same as before:
- H times G demand functions of the households,
- F times G demand functions of t h e firms, and x hg (P PG) = x h g (2 p ! , . . . , 2 p G ) (77)
- G equilibrium conditions, x
fg(Pi,---Pc) = x f g ( 2 p 1 , . . . , 2 p c ) . (78)
where " d e m a n d " is used as a generic term to refer to demand (denoted negatively) and Demands a n d supplies are said to be homogeneous of degree zero in abstract prices.
supply (denoted positively). The m o d e l determines Considered in economic terms, this is obvious because an alternative way of express-
- H times G demands of the households, ing values in no way alters real events in an economy without m o n e y . Indeed, this
- F times G supplies of the firms, a n d doubling of abstract prices is equivalent to the situation in which, f r o m a certain day
- G equilibrium prices. on, prices a r e n o t reckoned in dollars b u t in half-dollars. It is hardly conceivable that
- for this reason alone - demands or supplies should change."
T h u s , the model consists of j u s t as many equations as u n k n o w n s which is, of Finally, we want to reproduce the adjustment process of prices as put forward by
course, neither necessary nor sufficient for a unique solution b u t can be regarded as WALRAS. F o r this purpose, Walras developed the famous (or notorious) notion of the
a " h i n t " . However, adding all budget restrictions, (68), and all p r o f i t definitions, (70), auctioneer: A t the beginning of any period, the auctioneer cries out a certain trial price
yields vector. A f t e r doing so, he collects the demand and supply plans of households and
H / G F \ F / G firms, these plans being determined with respect to the announced price vector. The
E ZPg*hg +r=S l% J +E
h = i \g = i
S pgxrg- n
I = i \g«i
auctioneer c o m p a r e s demands and supplies in every market; he increases prices where
there is an excess demand; and lowers t h e m where there is excess supply. Afterwards,
a n d , a f t e r cancelling out the 7thg (all profits are distributed) and rearranging terms, this the new price vector is announced. N o w , the individuals formulate their plans again;
becomes and this interplay lasts until a price vector is found which establishes a general

X
Z ( Z PG' HG + E P„' 0 = 0. (76)
g=l\h=l f=l /

T h i s very important equation follows directly from the h o u s e h o l d s ' budget con-
straints a n d the firms' profit definitions. Thus, its validity is independent of equilib-
r i u m conditions, and the equation holds in equilibrium as well as in disequilibrium.
It is called Walras' law. Comparing (76) with (W.3) shows that (76) is just the sum of
the equilibrium conditions. This sum must vanish. Therefore, a n equilibrium in G — 1
m a r k e t s implies an equilibrium in all markets: if G — 1 terms in (76) are zero, and the
sum of all terms is zero, then the G t h term must vanish also.
 § 29 Conclusion 67
66 Chapter IV. T h e Classical Theory

equilibrium. Not before then are the transactions carried out. This "cybernetic" process Further Reading
is illustrated in Figure 23. ACKLEY,G . (1978) Macroeconomics: Theory and Policy. N e w Y o r k etc.: MACMILLAN
W A L R A S believed that this process of trial and error, he spoke of a tatdnnement MARSHALL, A. (1990) Principles of Economics. Reprint of t h e 8 t h e d . t . o n London 1952. M a c m i l -
(groping), would finally arrive at the general equilibrium. In short, he was the first to
MILL"J.SL (1848) Principles of Political Economy. L o n d o n : P a r k e r . Reprint T o r o n t o 1 9 6 5 :
give a mathematical account of S M I T H ' S invisible hand.
More recent research work in the field of general equilibrium theory, starting with R i c ^ Ä Principles of Political Economy a n d T a x a t i o n . Reprint of the 3rd e d i t i o n
a paper by A B R A H A M W A L D 1 , thoroughly investigated the problems of existence, L o n d o n 1924: Bell and Sons
uniqueness and stability of the general equilibrium because, as we already mentioned, SMITH, A. (1776) An Inquiry into the Nature and C a u s e s ol the Wealth of N a t . o n s , R e p r i n t
these issues can not be solved by counting equations and unknowns. However, it is O x f o r d 1976: Clarendon . ,
impossible to give even a brief account of this work here 2 . WALSH, V.Ch. and H. GRAM (1980) Classical and New- Classical Theories of G e n e r a l E q u i l i b -
rium. New York etc.: O x f o r d University Press

§29 Conclusion
Our discussion of Classical theory is now complete, and we have obtained some first
insight into the "vision" of the Classical economists. From the partial analyses of the
labor, capital, and commodity markets, the quantity theory of money, Say's law, and
the complete model, it has emerged that the Classical doctrine has no place for
involuntary unemployment, overinvestment, underconsumption, and general crises.
On the contrary, the "invisible hand", i.e. the princing mechanism, always results in
a general market clearance.
In view of the real depressions which arise so frequently, the reader may wonder
whether the Classical theory is really to be taken as a serious doctrine. With respect
to this, two remarks are in order.
First and foremost, it must be emphasized that the Classical and Neoclassical
authors were primarly concerned with the analysis of long periods. They only main-
tained that there is a strong tendency of market economies to converge towards the
general equilibrium; and that this equilibrium would finally be attained - even if the
adjustment process might last for many years. On the other hand, the Classics denied
a long-run tendency towards ever aggravating crises as M A R X and others had ex-
pected.
In the second place, the Classics were in fact well aware of real enonomic problems
but they traced them back to specific disturbances of the market process: to
monopolies, cartels, unions, and state intervention. Hence, it is altogether wrong to
think, for instance, that the Classical doctrine is refuted by the unemployment which
actually exists: The statement "Perfectly flexible wages entail steady full employ-
ment." Is not disproved by the fact that unemployment arises frequently - wages not
being perfectly flexible. We will return "to this later.
In our next chapter, we encounter a theory which not merely criticizes several
assumptions of Classical theory but turns some issues upside-down.

1 WALD, A. (1936) Ü b e r einige Gleichunessvsteme der mathematischen Ökonomie. Zeitschrift

für Nationalökonomie 7, pp. 6 3 7 - 6 7 0
2 Good introductions to general equilibrium theory a r c Q U I R K , J. and R. S A P O S N I K (1968)
Introduction to G e n e r a l Equilibrium Theory a n d Welfare Economics. New York etc.:
McGraw-Hill as well as A R R O W , K.J. and F . H . H A I I N (1971) General Competitive Analysis.
San Francisco: H o l d c n - D a y . We also recommend the articlc of W E I N T R A U B that sets aside the
more technical difficulties: WEINTRAUB. E.R. (1983) O n the Existence of a Competitive Equi-
librium: 1930-1954. J o u r n a l of Economic Literature 21, pp. 1 - 3 9