The modern theory of income and employment, for which we may thank the genius of J. M.

Keynes (1936), is without

question the most important advance in economic analysis in the twentieth century. Keynes taught us to understand the
nature of depressions and radically changed our thinking about how to deal with them. Keynesian principles have been
widely applied in the policies of most industrialized countries, and governmental responsibility for the maintenance of full
employment is now generally accepted. In the United States the growing influence of the new economics led, in 1946, to
the passage of the Employment Act, which created the Council of Economic Advisers and committed the federal
government to promote maximum employment, production, and purchasing power.
In addition to its profound influence on economic policy, the modern theory of income and employment
has paved the way for important developments in many areas of economic analysis. The purpose of
the present essay is to provide a broad, simple outline of the theory. Many subtleties and extensions
have necessarily been glossed over or omitted completely. [The interested reader is urged to consult
the separate articles on topics treated only briefly in this article:Consumption Function; Employment
AND Unemployment; Fiscal Policy;Inflation AND Deflation; Interest; Investment; Liquidity
Preference; Monetary Policy; andMoney.]
Meaning of unemployment. It is important at the outset to be clear about the meaning of
unemployment. Full employment implies that all who wish to work at the existing level of wage rates are
employed. There are, to be sure, potential workers who choose not to seek employment because wage
rates are too low. But such persons are said to be voluntarily unemployed, and the theory of
employment is not concerned with them; rather, it is concerned with involuntary unemploymentwith a
situation in which there are workers who would be willing to take employment at prevailing wages but
cannot find it.
Pre-Keynesian economics generally held that involuntary unemployment could persist only as the
result of market imperfections. The excess supply of labor implied by involuntary unemployment would
cause wages to fall, and the combined effect of an increased demand for labor (owing to the fall in its
cost) and the voluntary withdrawal of some workers from the labor market (owing to the lower wage
rate) would be to clear the labor market and restore full employment, just as a fall in price in any
market would eliminate excess supply. On the other hand, Keynes and his followers argued that a
state of involuntary unemployment could persist and that many levels of employment could be thought
of as equilibrium levels. The path to full employment, moreover, lay not in the direction of wage rate
adjustments but rather in the direction of raising the total level of spending in the economy.

Structure of the theory

The theory of income and employment is an aggregative theory which lumps all markets for final
goods and services into a single product market, all financial markets into a single money market, and
all markets for labor services into a single labor market. Customarily the analysis proceeds with a
description of the properties of the individual markets and then links them together into an overall
picture.

Product market
The sum total of the production of final goods and services (defined as output that is not resold in any
form during the accounting period) when valued at market prices is the gross national product. The
deduction of a capital consumption allowance for the replacement of capital equipment that was used
up during the course of producing current output reduces this total to the net national product (NNP).
The figure that results when NNP is deflated by an index of prices of final products in order to obtain
constant dollar values we callreal NNP, which for present purposes we refer to simply
as income, denoted by the symbol Y.
Income and employment analysis begins by breaking income down into several components. The
typical approach is to divide the economy into sectors and to examine the determinants of spending

and the income receipts of each sector. A complete analysis would include a household sector, a
business sector, a government sector, and a foreign sector [see National Income AND Product
Accounts]. However, for present purposes it suffices to confine the analysis to the household and
business sectors. The portion of production that is purchased by households is called consumption,
C. The remainder of the nations output accrues to the business sector in the form of capital goods
(new plant and equipment) or of additions to the stocks of finished and unfinished goods. Denoting the
product retained by the business sector as net realized investment, Ir, we have the basic definition
Y C + Ir.
Provided that the business sector retains no earnings, each dollar of expenditure will be received by
households as income. Since households are free either to consume or to save their incomes, we also
have
Y C + S,
where S is the level of saving. When we equate the two expressions we obtain the fundamental
accounting identity
Ir = S.
We proceed with the theory of income determination by considering the consumption component first.
Although aggregate consumption spending is related to many variables, we confine ourselves to the
original proposition of Keynes that consumption is an increasing function of the level of income. In
Figure 1 the curve labeled C represents an imaginary aggregate consumption function. Consumption
rises as income rises, but not by as much. In the hypothetical situation of Figure 1 a rise in income of
200 from an initial level of 200 is accompanied by an increase in consumption of 150 to a level of 350.
The ratio of the rise in consumption to the rise in income, C /Y, is the marginal
propensity to consume, which is simply the slope of the consumption function; in this example it has a
value of .
Because saving is the difference between income and consumption, the level of saving can be
measured as the vertical distance between the 45 line and the consumption function, or the separate
S curve of Figure 1. The ratio of the change in saving to the corresponding change in income is
the marginal propensity to save. Since an extra dollar of income must be either spent or saved, the
sum of the marginal propensity to consume and the marginal propensity to save must be 1.
The consumption function is a schedule of intentions. It indicates what the level of consumption
spending will be at different levels of income. We may now introduce a schedule of business intended
investment spending, which we may either plot separately or add to the consumption function in order
to obtain theaggregate demand, C + I, schedule. The aggregate demand schedule shows the level of
total spending that will be forthcoming at different levels of income. In the present example it is
assumed that intended investment is 50 at all levels of income.
Care must be taken to distinguish between intended investment, I, and realized investment, Ir.
Realized investment consists of all output that is retained by the business sector; intended investment
is only that portion of output that the business sector actually intended to retain when production plans
were formulated. Realized investment will exceed intended investment if business overestimates the
level of potential sales and will fall short of intended investment if potential sales are underestimated.
The amount of this discrepancy is unintended investment, Iu, so by definition
Ir I + In .

The presence of unintended investment implies that sales forecasts have been mistaken. In order to
reduce unintended inventory accumulation, output will be cut back; in order to offset unintended
inventory depletion (negative unintended investment) due to underestimation of potential sales,
production will be increased. Intentions and realizations, then, will be equal only when unintended
investment is absent. Since by definition
Ir I + Iu S,
and since equilibrium requires Iu to be zero, we infer that equilibrium requires intended investment to
equal saving. Although realized investment must always equal saving, intended investment will equal
saving only when the product market is in equilibrium.
A glance at Figure 1 confirms that I = S at an income level of 400. This income level must necessarily
also be where the aggregate demand schedule cuts the 45 line. Any other income level would be
characterized by the presence of unintended investment. For example, if business produces $500
billion worth of output, consumption will be $425 billion and intended investment will be $50 billion.
Total demand of $475 billion therefore falls short of production of $500 billion, with the consequence
that unintended investment is $25 billion. Realized investment is $75 billion, and this does, to be sure,
equal the level of saving, but un-desired inventories will pile up as long as production is maintained at
$500 billion and thus business must revise its sales estimates and reduce the level of output.
Multiplier. Any change that raises the aggregate demand schedule will raise the level of income.
Moreover, the rise in the level of income will exceed the size of the shift in aggregate demand that
brings it about. Referring again to Figure 1, we may suppose for a moment that intended investment is
zero and observe that the equilibrium level of income is 200, since that is where saving is zero.
Assume next that intended investment rises from zero to 50 and remains there. The equilibrium level
of income jumps to 400an increase not of 50 but of 200. This multiplied increase in income results
from the fact that the increase in investment spending raises income, thereby inducing additional
consumption spending. It can be seen from Figure 1 that the rise in income from 200 to 400 is
composed of the increase in investment of 50 plus an increase in consumption of 150.
The ratio of the change in income to the change in investment, Y/I, is themultiplier, .and its
numerical value is the reciprocal of the marginal propensity to save. If intended investment rises by
one dollar, equilibrium will not be restored until saving has also risen by one dollar. If the marginal
propensity to save has a value of x, an income increase of 1/x dollars is required to raise saving by
one dollar. In the present example, when income rises by one dollar saving rises by 25 cents. Since
saving must rise by one dollar, the required increase in income is four dollars. [For an alternative and
more elaborate discussion of the multiplier, seeconsumption function.]
The multiplier permits us to calculate the effect on the level of income of an upward shift in the
aggregate demand schedule. The multiplier also tells us how much aggregate demand must be raised
to reach an income level which will bring full employment. In Figure 1, if the equilibrium level of
income is 400 and the full-employment level of income is 600, the required increase in income is 200.
Since the multiplier is 4, a policy that shifts the aggregate demand schedule up by 50 will raise income
to the full-employment level. The required shift in aggregate demand can also be found by measuring
the deflationary gap,which is defined as the deficiency of aggregate demand measured at the fullemployment level of income. Inspection of Figure 1 confirms that the magnitude of this gap is 50.
Before we add to this simplest of all models, it might be instructive to put the model into algebraic
form. In linear form the consumption function can be written
C = a + bY = 50 + .75Y,

where a is the level of consumption associated with a zero level of income and bis the marginal
propensity to consume. The saving function must be
S = Y - C = -a + (l - b)Y = -50 + .25Y.
where 1b is the marginal propensity to save. Since the equilibrium condition is
I = S,
we have
I = -a + (1-b)Y,
or
50 = 150 + 25Y.
Thus, we can solve for Y and obtain
The multiplier is
The next step in constructing the theory of income and employment is to drop the assumption that
intended investment is a constant somehow determined by factors outside the economic system.
Although many alternative theories of investment behavior have been developed, we take note here
only of the fact that ordinarily a business decides to borrow in order to expand its stock of productive
capacity or its inventory holdings only when the expected rate of return on the new investment
(the marginal efficiency of capital) is in excess of the cost of borrowing funds (the rate of
interest). Alternatively, if funds are available without borrowing, the firm must determine whether it is
more profitable to invest or to engage in some form of lendingfor example, the purchase of
government bonds. The higher the cost of borrowing (or the return on lending), the less inclined firms
will be to engage in investment spending. For this reason the level of intended investment is usually
regarded as a decreasing function of the rate of interest, i.
If the saving function is written in the general form
S = S(Y),
(read: saving is an increasing function of the level of income) and the investment demand function in
the form
I = I(i)
(read: intended investment is a decreasing function of the rate of interest), and if we equate intended
investment with saving, it becomes apparent that the equilibrium level of income depends upon the
rate of interest. To put it differently, we now have two variables, i and Y, but only one equation,
I(i) = S(Y),
and we must therefore find out how the rate of interest is determined.

Money market and the rate of interest

The rate of interest is the return on lending or the cost of borrowing money. It can also be thought of
as the return that wealth holders forgo when they hold money balances that could have been lent at
interest, and it therefore constitutes the cost of holding money balances. Given the quantity of money

held by wealth holders, the equilibrium rate of interest is that rate at which they feel no incentive to
convert money into other financial assets or vice versa. Such a situation is denoted as monetary
equilibrium; it obtains when the demand for money balances to hold equals the supply of money.
The simple theory of income and employment customarily treats the supply of money, usually defined
as the currency and demand deposits held by the nonbank public, as a policy variablei.e., its size is
determined by the central monetary-fiscal authority. Under fractional reserve banking, however,
money can, within limits, be created or destroyed by the commercial banking system. Money is
created, for example, when banks convert excess reserves into earning assets, since this involves
creation of additional deposits. Thus, the size of the money supply depends to some extent upon the
degree to which banks are willing to make this conversion. The uncertainty that arises from the
possibility of deposit withdrawal makes some holding of excess reserves desirable. However, banks
forgo earnings when they hold excess reserves. Since a rise in the rate of interest increases this cost,
such a rise is likely to be accompanied by an increase in bank lending and therefore in the money
supply. For this reason many modern writers treat the supply of money as an increasing function of
the rate of interest. A money supply function might be written in real terms as
M8/p = (R/p, i),
where R is the quantity of nominal bank reserves (determined by the monetary-fiscal authority), i is the
rate of interest, M8 is the nominal stock of money, andp is the level of prices.
Keyness theory of the demand for money was one of his most important contributions. He delineated
three motives for holding moneythe transactions, precautionary, and speculative (liquidity
preference) motives. The transactions and precautionary demands were recognized by traditional
theory. Keyness great insight was to add the liquidity preference motive and to recognize the
importance of the rate of interest in determining the demand for money.
The transactions demand for money arises from the necessity for economic units to hold certain levels
of money balances because money receipts and disbursements are not perfectly synchronized in
time. For example, an individual who receives $1,000 at the beginning of a month, makes
disbursements at a uniform rate throughout the month, and ends the month with a zero money
balance will have an average cash balance of $500. In the long run the magnitude of such average
balances depends upon the nature of institutional payments-practices and upon the growth of income
and wealth. In the short run the size of transactions balances depends upon the number and size of
the transactions that the individual makes. Keynes and earlier writers therefore regarded the
transactions demand for money as proportional to the level of income.
Precautionary balances are held in order to meet unforeseen contingencies and to take advantage of
fortuitous opportunities. The magnitude of these balances was also viewed as proportional to the level
of income.
Keynes held that the speculative demand for money arises out of fear that interest rates may rise in
the future and that wealth holders will therefore suffer a capital loss if they hold bonds instead of
money. The lower the rate of interest, the greater is the risk of such a capital loss and the lower is the
return on bond holdings (the cost of holding money). Consequently, the quantity of speculative
balances held is viewed as a decreasing function of the rate of interest.
The theory of liquidity preference has been attacked on a number of grounds (see Tobin 1958) which
we need not go into except to note that money balances can usually be converted into risk-less
savings accounts. Nevertheless, empirical research (e.g., Tobin 1947) discloses a strong correlation
between the rate of interest and the average length of time that money is held between transactions, a
circumstance which seems to indicate that the demand for money is a decreasing function of the

interest rate. This suggests that transactions and precautionary balances are also interest-elastic.
Individuals who accumulate cash balances in anticipation of a large future outlay may convert these
balances into short-term securities, which they liquidate when cash is needed. The greater the return
on such securities relative to the transactions costs which their purchase and sale entails, the greater
is the incentive to economize on money balances and the lower will be average transactions balance
held (Baumol 1952; Tobin 1956).
The distinctions between the three motives for holding money are regarded as arbitrary by
contemporary monetary theorists. Nevertheless, it is useful for expository purposes to maintain a
distinction between transactions (active) balances and speculative (idle) balances. In any case it is
clear that the demand for money is a function of both the rate of interest and the level of income, so
we may ignore the separate motives and write quite generally
Md/p = L(i, Y)
as the demand for money to hold in real terms.

Joint equilibrium of the two markets

It was seen earlier that product market equilibrium is established when intended investment equals
savingsi.e., when
Similarly, monetary equilibrium is established when the demand for and the supply of money are
equal. Accordingly, we have
as the condition for monetary equilibrium. If the real value of the stock of bank reserves is known, the
two equations determine the equilibrium level of income and the rate of interest.
In Figure 2 this general equilibrium solution is sketched out in a way originally presented by Hicks
(1937). The curve labeled ISo is a diagrammatic representation of equation (1). It represents all
combinations of interest rates and income levels that are consistent with equality of intended
investment and saving. The curve has a negative slope because a decline in the rate of interest raises
the level of investment and therefore also the level of saving and income.
The LMo curve depicts equation (2), representing all combinations of income levels and interest rates
that yield monetary equilibrium. The curve ordinarily has a positive slope because a rise in the level of
income raises the required quantity of transactions balances so that monetary equilibrium is usually
obtainable only if the rate of interest rises. The point of intersection of IS0 andLM0 defines the
equilibrium level of income, Y0, and rate of interest, i,0the values that are compatible with equilibrium
in both markets simultaneously.

Monetary versus fiscal policy

An increase in the money supply brought about by a central bank purchase of securities on the open
market shifts the LM curve to the right (for example, fromLM0 to LM, to LM1 in Figure 2) because for
any level of income, it is only at a lower rate of interest that wealth holders can be induced to hold the
additional money balances. As can be seen from Figure 2, the size of the increase in income that this
shift brings about depends upon the slope of the LM curve in the range in which it intersects
the IS curve.
The vertical range of the LM curve represents the classical view that in a state of monetary
equilibrium, money balances are never held except for transactions purposes. Thus, a rise in the
money supply must cause the interest rate to fall and investment and income to rise until income rises
by enough to absorb all of the additional money into transactions balances. If, therefore, the IS curve

cuts the LM curve in the classical range, the level of income must rise in direct proportion to the
increase in the money supply.
The horizontal portion of the LM curve represents Keyness liquidity trap. Once the rate of interest falls
to a critically low level (i2, in Figure 2), wealth holders regard bonds and money as perfect substitutes,
and banks simply allow excess reserves to pile up. Consequently, if the IS curve cuts the LM curve in
the horizontal range, the open market purchase of securities by the central bank fails to lower the rate
of interest, since all the added money balances are simply held as speculative balances. As a result
the level of investment and the level of income fail to rise at all.
Whereas classical theory supports the effectiveness of monetary policy and Keynesian theory denies
it, exactly the opposite views emerge with respect to fiscal policy. An increase in government
purchases, for example, can be thought of as an upward shift in the investment demand schedule of
Figure 1. The consequence of this increase in aggregate demand is that the IS schedule shifts to the
right by an amount equal to the increase in government purchases times the multiplier. However,
whether the actual increase in income also equals the multiplier again depends upon the slope of
the LM curve.
The reader can visualize that a shift in the IS curve which cuts the LM curve in the liquidity trap range
will yield a fully multiplied increase in income whereas an intersection in the classical range will yield
no increase in income at all. In the classical case the attempt to increase total spending is frustrated
because a rise in the rate of interest cannot bring about an increase in transactions balances. The rise
in government purchases cannot generate a multiplier response, since any tendency for income to
rise will drive up the demand for money and raise the interest rate until private investment is reduced
by exactly the amount of the increase in government purchases. In the liquidity trap range, on the
other hand, rising income merely activates idle money balances without affecting the rate of interest,
and the level of private investment, therefore, does not decline.
In the classical view fiscal policy is incapable of raising the level of income, and fiscal changes merely
have the effect of redistributing national output between the private and the public sectors; reliance is
to be placed on monetary policy. In the Keynesian view, on the other hand, fiscal policy is the more
certain (and in some cases the only) method of raising the level of income.

Labor market and automatic adjustment

In the Keynesian view the equilibrium level of income that is jointly determined by product market and
money market equilibrium need not be the full-employment level. However, if an equilibrium level of
income (for example, Y0 in Figure 2) implies the presence of involuntary unemployment, will not
competition for jobs among workers cause wage rates to fall, and will this not cause the level of
employment and income to increase?
The theory of income and employment is a theory of the behavior of the economic system in the short
run. It therefore presupposes that the stock of capital equipment is fixed and views labor as the only
variable factor of production. From the theory of the competitive firm it can be deduced that labor will
be hired up to the point where the value of its marginal product just equals the wage rate. Expressed
in symbols, this condition is
= MP X p,
where is the money wage rate, MP is the marginal -physical product of labor,and p is the price of
output. Dividing through by p, we obtain
/p = MP,

where /p is the real wage rate. According to the law of diminishing returns, an increase in
employment will be accompanied by a decline in the mar ginal physical product of labor. A rise in
employment, therefore, cannot take place unless the real wage is reduced. Since the demand for
labor is a decreasing function of the real wage, we can write
Nd = D( /p)
as the demand for labor function.
In the classical view the quantity of labor supplied is normally an increasing function of the real wage
rate; it can be written
Ns = S( /p).
If there exists a labor market clearing mechanism such that Nd tends to equality with N8, these two
expressions can be equated to yield the equilibrium level of employment. Since involuntary
unemployment would be eliminated, this level of employment would be the full-employment level. The
full-employment level of income can then be determined by substituting the level of employment into
theproduction function,
Y = X(N),
which specifies the technical relation between factor inputs and the level of output.
Whether the level of income is determined by product market and money market equilibrium (the
Keynesian view) or by competition in the labor market (the classical view) depends upon whether a
labor market clearing mechanism operates so as to equate the demand for and the supply of labor. If
involuntary unemployment exists, the supply of labor exceeds the demand, and the restoration of full
employment requires that the real wage be reduced. Job competition among workers will reduce the
money wage rate, but whether the real wage rate will also fall depends on what happens to the price
level.
A fall in money wage rates leads business to expand the level of output and to increase the level of
employment. However, since the marginal propensity to consume is less than unity, only a fraction of
the additional output will be bought by consumers; unintended investment in inventories will therefore
take place, and output will tend to return to its original level. In Figure 1 it is apparent that a rise in
output of 100 from the original equilibrium level of 400 causes consumption to rise by 75, and the
remaining additional output of 25 represents unintended investment. Unintended investment will not
be eliminated until output falls back to 400. Since the equilibrium level of income will be the same as
before the wage cut, the level of employment must also return to its original level, and this implies that
the real wage will be restored to its original level as the result of a fall in the price level proportional to
the original fall in the money wage rate. It is clear, then, that the equilibrium level of income cannot
change unless the fall in money wage rates somehow succeeds in shifting the entire aggregate
demand schedule upward.
We have seen that the fall in money wage rates will induce a fall in the price level. This means that the
real value of the money supply, M8 /p, will increase, which in turn will cause the rate of interest to fall.
As a consequence the level of intended investment will rise, and so also will the levels of income and
employment. The argument can be visualized by reference to Figure 2. The increase in the real value
of the money supply shifts the LM curve from LMo toLM1, the rate of interest drops from i0 to i1, and the
level of income rises from Y0to Y1,.

Underemployment equilibrium. There were two reasons, in Keyness view, why this interestinvestment mechanism might fail to work. First, there is the possibility of the liquidity trap, which, if
present, would mean that the increase in the real value of the money supply could not lower the rate
of interest. Second, there is the strong possibility that under depressed economic conditions
investment would be insensitive to changes in the rate of interest. The first possibility can be
visualized by further reference to Figure 2. If the IS curve cuts the LM curve in the horizontal (liquidity
trap) range, the increase in the real value of the money supply would have no effect on the rate of
interest or on the level of income. The second possibility can be visualized if we recognize that
the IScurve would be vertical if investment were totally insensitive to changes in the rate of interest. If
it is true, as is implied by the Keynesian analysis, that a fall in money wage rates will not raise the
level of income, either because of the presence of the liquidity trap or because investment is interestinelastic, then the labor market clearing mechanism fails to operate, and the existing level of income
can be thought of as an equilibrium level.
The most elegant challenge to the Keynesian doctrine of underemployment equilibrium came from
Pigou (1943), who suggested that if consumption were a function not only of the level of income but
also of the level of wealth, a fall in the price level would increase the real value of the stock of
currency and government debt held by the private sector, and this increase in wealth would cause the
consumption function to shift upward. As long as the labor market remained uncleared, wages and
prices would continue to fall, wealth would continue to increase, and the consumption function would
continue to shift upward until the full-employment level of income was reached.
The theoretical issue cannot be said to have been resolved. A fall in wage rates may either raise or
lower the level of income. The corrective interest-investmentand Pigou effect mechanisms discussed
above may, for example, be offset by the generation of adverse expectations. If wage reductions take
place in a sluggish, piecemeal manner, entrepreneurs in industries where wages have not yet fallen
will anticipate cost reductions by reducing output and employment and by selling from inventory. And if
consumers expect the price level to fall, post-ponable consumption expenditures will be reduced.
Moreover, a wage reduction will redistribute income in favor of profit earners, whose marginal
propensities to consume may be lower than those of wage earners, and this could cause the
aggregate consumption function to shift downward.
For practical purposes it is sufficient to recognize that in advanced economies wages and prices tend
to be downwardly rigid. It would be difficult to implement a national wage reduction policy, and, as
Keynes recognized, there is little that can be accomplished by such a policy that cannot also be
accomplished by a relatively painless expansionary monetary policy. Output and employment cannot
be raised without an increase in aggregate demand. Such an increase in demand might conceivably
be brought about by wage reduction, but a policy of wage reduction would be inefficient and might not
work at all. If it did work, the desired effects might take an intolerably long time to materialize. And the
policy would certainly be inferior to a policy of direct demand expansion through monetary-fiscal
measures.
Money illusion. In recognizing the institutional facts of life, Keynes broke away from the classical
theory of labor supply. Instead of assuming that the supply of labor depends on the real wage, he
assumed that labor is subject to money illusioni.e., that the quantity of labor supplied responds to
changes in money wage rates but not to changes in the price level and that the supply of labor is
therefore a function of the money wage rate rather than of a real wage rate. In Figure 3, 0 is the
historically given money wage rate, and P0 is the ruling price level. At money wage w workers will
offer anywhere from zero toN* units of labor. Thus, at w0/Po the labor supply curve is a horizontal line
up to N*.Although the money wage rate cannot be made to fall, it will rise when all who are willing to
work at w are employed and additional workers are desired. Consequently, the labor supply curve
bends up sharply once N* is reached. The labor

demand schedule cuts the supply schedule at No, and the distance N*N0therefore measures
involuntary unemployment.
Since the money wage rate is assumed to be rigid downward, full employment can be restored only
through a fall in the real wage brought about by an increase in aggregate demand and a rise in the
price level. If such a rise in the price level takes place, the entire labor supply schedule shifts
downward, and involuntary unemployment is eliminated. Thus, at real wage 0/Pi the labor demand
curve cuts the supply curve at N*, where all who are willing to work at the new real wage are
employed. Even though the real wage has fallen, the assumption of money illusion on the part of
workers implies that the same quantity of labor will be supplied at the new real wage as at the old.

Synthesis of Keynesian and classical models

The Keynesian and classical models may now be sharply delineated. Although these models are
static systems in which the equilibrium values of the variables are simultaneously determined, it is
nevertheless instructive to think of the determination of the equilibrium values as running in a definite
sequence.
In the Keynesian system, product market equilibrium can be written more generally as
I(i, Y) = S(i,Y)
to reflect the traditional view that a rise in the rate of interest induces individuals to increase the level
of saving and to incorporate the notion that the level of investment depends upon the levelof income
as well as upon the rate of interest. When the monetary equilibrium condition,
(R/p,i) = L(i,Y),
is added and the level of real bank reserves is assumed to be known, the two relations jointly
determine the level of income and the rate of interest, as well as the levels of consumption, saving,
and intended investment. The level of income being known, the production function,
Y = X(N),
defines the level of employment. The demand for labor, together with the historically given rigid money
wage rate, wo, provides the equation needed to determine the price level,
N = Nd = D(wo/P).
In the classical system the sequence can be thought of as beginning with the labor market, where the
real wage and the level of employment are determined by labor market equilibrium,
D(w/p) = S(w/p).
The full-employment level of employment, N*, is now known, and the full-employment level of income
follows from the production function; since the level of income is known, the product market
equilibrium equation collapses to
I(i) = S(i),
which implies the classical proposition that the investment and saving functions serve only to
determine the rate of interest and the way aggregate output is divided between consumption and
investment. Finally, since both the level of income and the rate of interest are known, the monetary
equilibrium condition can be collapsed to

R/p = constant,
which implies that the price level is entirely determined by the nominal stock of bank reserves. In the
absence of fractional reserve banking, the preceding expression can be replaced by the more familiar
M/p = constant,
which reflects the traditional notion that given a fixed level of income, the level of prices must be
directly proportional to the quantity of money.
The substantial differences between the ways in which the equilibrium values of the variables are
determined rest upon whether the equilibrium level of income can differ from the full-employment
level. This is illustrated in Figure 4, where it is assumed that the original LM curve is LMo and the fullemployment level of income is Y*. If the IS
curve cuts LM in the horizontal range at less than full employment (curve ISo),the rate of interest is the
liquidity trap rate which cannot be changed by a shift in the IS curve within the horizontal range of
the LM curve. The IS curve, in turn, determines the equilibrium level of income which cannot be
affected by changes in the money market (shifts in the LM curve).
If the IS curve (curve 7S,) cuts the LM curve at less than full employment but in a range where
the LM curve has a positive slope, the equilibrium level of income, Y1, and rate of interest, i1, are jointly
determined by conditions in the product and money markets, and both can be affected by changes in
either market.
Finally, suppose that the IS curve (curve IS2) cuts the LM curve at the full-employment level of income
and interest rate i2. Suppose next that the money supply is increased and shifts the LM curve
to LM,. Joint product market and money market equilibrium would now obtain at interest rate i 3 and
income level Yi. But Y3 is in excess of the full-employment level of income and is therefore not
attainable. Excess demand for goods and services must now exist. As a consequence, the price level
rises, and the real value of the money supply decreasesthe LM curve shifts back to the left. Since
the LM curve must shift back exactly to its original position if equilibrium is to be restored, the increase
in the money supply only produces a proportional increase in the price level and leaves unaffected the
other equilibrium values of the variables. The only possible equilibrium interest rate in this classical
case is the natural rate of interest, i2. It is found at the point of intersection of the IS curve and the
vertical line at the full-employment level of income, and it exactly equates intended investment with
the full-employment level of saving.
Thomas Dernburg
[See also the biography ofKeynes, John Maynard].

BIBLIOGRAPHY
works cited
Baumol, William J. 1952 The Transactions Demand for Cash: An Inventory Theoretic
Approach. Quarterly Journal of Economics 66:545-556.
Hicks, John R. 1937 Mr. Keynes and the Classics: A Suggested Interpretation.Econometrica 5:147159.
Keynes, John Maynard 1936 The General Theory of Employment, Interest and Money. London:
Macmillan. A paperback edition was published in 1965 by Harcourt.