CONSUMPTION FUNCTION

Consumption means using up commodities. Function means the relationship between two
or more variables 1 . Consumption function means the technical relationship between
income and consumption, i.e., C = f (Y), Where C is the consumption and the dependent
variable Y is the income and the independent variable. There is direct relationship
between income and consumption.

CLASSICAL VIEWS ON CONSUMPTION FUNCTION

Classical argued that the level of consumption was determined by the rate of interest.
When rate of interest increases, saving also increases by reduction in current
consumption and when rate of interest decreases, saving also decreases and current
consumption increases. Therefore, there was an inverse relationship between rate of
interest and consumption.

KEYNESIAN CONSUMPTION FUNCTION

According to Keynes primarily the level of income determines consumption
expenditures. Keynes argued consumption is relatively unaffected by interest rate. They
said that Marginal Propensity to Consume (MPC) is positive and less than unity and that
APC declines as income increases.

PROPENSITY TO CONSUME
Men are disposed, as rule, and on the average, to increase their consumption as their
income increases but not by as much as the increase in their income, which is popularly
known as “Propensity to consume” or “Consumption function”.

PROPENSITY TO CONSUME SCHEDULE OR CONSUMPTION FUNCTION SCHEDULE

Propensity to consume schedule or Consumption function schedule is a schedule of the
various amounts of consumption expenditure corresponding to different levels of income.
A hypothetical consumption function schedule is given below.

TABLE 1
PROPENSITY TO CONSUME SCHEDULE

Income (Y) Consumption Saving (S) MPC MPS MPC + APC APS APC +
( in Rs.) (C) (in Rs.) (in Rs.) (ΔC/ΔY) (ΔS/ΔY) MPS (C/Y) (S/Y) APS
100 060 040 0.50 0.50 01 0.60 0.40 01
120 070 050 0.50 0.50 01 0.58 0.42 01
140 080 060 0.50 0.50 01 0.57 0.43 01
160 090 070 0.50 0.50 01 0.56 0.44 01
180 100 080 0.50 0.50 01 0.55 0.45 01
200 105 095 0.25 0.75 01 0.53 0.47 01
220 107 113 0.10 0.90 01 0.49 0.51 01
240 107 133 0.00 0.90 01 0.46 0.54 01

1
Variable means which varies with time or any other attributes.
By taking income in X-axis and consumption in Y-axis, we will get an upward sloping
curve as shown in figure 1. CC is the propensity to consume curve or curve of
consumption function, which rises from downward to upward indicating that as the
income increases consumption also increases but to a lesser degree than the increase in
income.
Figure 1 Figure 2
Propensity to consume APC Curve
110 .62

.60
100
.58

90 .56

.54
80
.52

.50
70
CONSUMPT

.48

60
.46

APC
.44
50
80 100 120 140 160 180 200 220 240 260
80 100 120 140 160 180 200 220 240 260

INCOME
INCOME

NOTE: Consumption, C, represents the amount of consumer expenditures made at a

given level of income, where as propensity to consume, C (Y), is a schedule of consumer
expenditures at various income levels.

AVERAGE PROPENSITY TO CONSUME [APC]

It may be defined as the ratio of consumption expenditure to any particular level of
income, i.e., APC = C / Y. Table 1, Shows that the APC declines as income increases
because the proportion of income spent on consumption decreases. Figure 2 shows the
APC Curve, which is a flatter curve than consumption curve in figure 1.

AVERAGE PROPENSITY TO SAVE [APS] FIGURE 3

It may be defined as the ratio of saving to
APS CURVE
any particular level of income, i.e., APS = .56

C/Y. Table 1, Shows that the APS .54

increases as income increases. Figure 3 .52

shows the APS Curve, which is an upward .50

.48
sloping curve. .46

.44

RELATIONSHIP BETWEEN APC AND APS .42

1. APC + APS = 1 .40

APS

.38
2. APC = 1 – APS 80 100 120 140 160 180 200 220 240 260

3. APS = 1 – APC INCOME

MARGINAL PROPENSITY TO CONSUME [MPC]

It may be defined as the ratio of change in consumption to the change in income or as the
rate of change in the propensity to consume as income changes, i.e., MPC = ΔC/ΔY.
Table 1 shows that the MPC is always less than one. Generally, MPC decreases as
income increases that is less and less of an equivalent marginal income consumed.
Diagrammatically, the slope of consumption curve measures the MPC.
MARGINAL PROPENSITY TO SAVE [MPS]
It may be defined as the ratio of change in saving to the change in income or as the rate of
change in the propensity to save as income changes, i.e., MPS = ΔS/ΔY. Table 1 shows
that the MPS is always less than one. Generally, MPS increases as income increases that
is more and more of an equivalent marginal income saved.

RELATIONSHIP BETWEEN MPC AND MPS

1. The value of MPC / MPS lay between zero and one, i.e., 0 < MPC < 1 and
0 < MPS < 1
2. MPC + MPS = 11
3. MPC = 1 – MPS
4. MPS = 1 – MPC
5. The MPC declines as income increases and MPS increases as income increases.

RELATIONSHIP BETWEEN MPC AND APC

1. MPC refers to the marginal increase in consumption (ΔC) as a result of marginal

increase in income (ΔY) and APC means the ratio of total consumption to total
income (C/Y).
2. As income increases, the MPC as well as the APC both decline, but decline in the
MPC is more than the decline in the APC.
3. When the MPC is constant, the consumption function is linear, i.e., a straight-line
curve. The APC will also be constant only if the consumption function passes
through the origin. However, if it does not pass through the origin, the APC will
not be constant.
4. Sometimes the MPC and the APC may be equal. The post-Keynesian economists
have come to the conclusion that over the long run the MPC is equal to the APC
and approximate 0.9.
5. The MPC is higher in poor communities and lower in case of rich communities.

SIGNIFICANCE / IMPORTANCE OF THE MPC

1. Keynes is concerned primarily with the MPC for his analysis pertains to the short
run while the APC is useful in long run analysis.
2. The post-Keynesian economists have come to the conclusion that over the long
run the MPC is equal to the APC and approximate 0.9.
3. 0 < MPC < 1, which means that when income increases the whole of it, is not
spent on consumption.
4. The MPC is higher in poor communities and lower in case of rich communities which
means Under Developed Country’s (UDCs) MPC is higher than developed countries.
5. The higher the MPC, the higher is the multiplier value and vice-versa.
KEYNES’S PSYCHOLOGICAL LAW OF CONSUMPTION 2
Keynes’s Psychological Law of Consumption states that men are disposed, as rule, and on the
average, to increase their consumption as their income increases but not by as much as the
increase in their income, which is popularly known as “Propensity to consume” or
“Consumption function”.

Propositions of the law

The law has three related propositions:
1. When income increases, consumption expenditure also increases but by a smaller amount
2. The increased income will be divided in some proportion between consumption
expenditure and saving, i.e., ΔY = ΔC + ΔS.
3. Increased in income always leads to an increase in both consumption and saving.

Assumptions of the law

1. He presumes a constant psychological-institutional complex. It means consumption
depends on income alone and other factors like income distribution, taste, fashion,
population growth, price level etc remain constant.
2. There exist normal circumstances and not extra ordinary circumstances like war,
revolution, hyperinflation etc.
3. It assumes the existence of a laissez- faire capitalistic economy.

IMPORTANCE OF CONSUMPTION FUNCTION OR

IMPLICATIONS OF KEYNES’S PSYCHOLOGICAL LAW OF CONSUMPTION

1. Vital importance of investment: In short run the consumption function is assumed to be

stable. It is from the stability principle of propensity to consume that Keynes draws the
conclusion that employment can only be increased with an increase in investment.
2. Repudiation of Say’s law: Say’s law of market says that ‘Supply creates its own
demand’ means there is no over production. According to Keynes, 0 < MPC < 1, which
states that what ever produced, is not consumed, means general over production. So,
consumption function analyses of Keynes discard Say’s law.
3. Turning points of Trade cycle: Keynes doctrine of consumption function indicates that
a sever depression is arrested by the fact that consumption does not decrease as rapidly as
income decrease. Thus, the concept of consumption function gives an insight into the
theory of trade cycle too.
4. Underemployment equilibrium: MPC being less than one, consumers fail to spend on
consumption as much as increase in income, Hence, what we have in the economy is the
underemployment equilibrium.
5. State intervention: When consumption lags behind income and causes depression, state
has to intervene to encourage consumption or control it in case of inflation. Thus, state
intervention becomes an important implication of consumption function.

DETERMINANTS OR FACTORS AFFECTING CONSUMPTION FUNCTION

Subjective factors: The subjective factors which affect the propensity to consume adversely,
consists of psychological motives such as precautions, foresight, family affection, old age
security, improvement etc. Keynes doesn’t think that all the subjective motives impel one to
reduce consumption. The corresponding motives, which induce one to spend more on
consumption, are enjoyment, better standard of living, recreation, generosity, extravagance and

2
This is explained clearly in his book “ The General Theory of Employment, Interest and Money”
written in 1936.
ostentations. Similarly, there are motives the liquidity, initiative, enterprise, improvement in
techniques of production and financial prudence which induce governments, business companies
and corporations to curtail consumption and increase saving.

Objective factors:
1. Income: Income is the most important factor that influencing consumption. As income
rises or falls, consumption also rises or falls.
2. Distribution of income: The wide spread inequality in income lowers the over all
propensity to consume as the rich have already fulfilled most of their basic wants. A more
equal distribution of wealth will go to raise the propensity to consume.
3. Financial Policies of Corporation: If corporations and companies keep more reserves
and distribute less of their profits as dividends, it will lower the disposable income with
consumers. On the other hand, if more income is distributed in the form of dividend,
more will be spent on consumption.
4. Windfall gains or losses: Sudden and unexpected gains or losses in income affect
consumption positively or negatively.
5. Change in Wage rates / levels: If the wage level rises, the consumption shifted upward
and vice-versa.
6. Change in fiscal policy; Change in fiscal policy in the form of taxation and public
expenditure affect consumption function. Imposition of more taxes reduces the
disposable income of the consumers, which adversely affect consumption and vice-versa.
On the other hand, increase in public expenditure injects more money to the economy,
which ultimately reaches at the hands of the consumers, and more money boost
consumption expenditure.
7. Change in expectation: If people’s expectation change like they expect war in near
future, they will save more and consume less.
8. Change in rate of interest: If rate of interest increases people intended to save more and
consume less to get more benefits and vice-versa.
9. Holding liquid assets: Liquid assets (currency, bank deposits, shares and so on) and
change therein also affect the propensity to consume. When people have large amounts of
liquid assets, they show a tendency to spend more on consumption and vice-versa.
10. Pigou Effect: Prof. A. C. Pigou argued that that a general fall in prices induced by the
general wage cut will increase the real value of cash balances and other form of saving
thereby leading to a higher rate of consumption. This latter relationship (between the real
value of the liquid assets and consumption) has come to be known as the ‘Pigou Effect’ in
brief, means that the real value of money assets rises as a result of general wage cut and
prices. The rises in the real value of money assets shift the consumption function
upwards. It is also called ‘Real Value of Money Assets Effects’.

MEASURES TO STIMULATE CONSUMPTION FUNCTION

1. Income Redistribution
2. Increase social security
3. Increase in Wages
4. Easy credit facilities
5. Urbanization and industrialization
6. Advertising and propaganda
 ABSOLUTE INCOME HYPOTHESIS OF KEYNES
The Absolute Income Hypothesis of Keynes states that consumption expenditure (C) are
a function solely of current personal disposable income (Yd): C = C (Yd). The
determinants of consumption was detailed in the General Theory of Employment, Interest
and Money by Keynes who hypothesized that consumption would be functionally related
to income in the following way:
1. For any change in income the corresponding change in consumption would be in
the same direction but of a smaller magnitude. The MPC would be less than one:
0<MPC<1.
2. The MPC would be less than APC, i.e., MPC<APC.
3. The rate of change of MPC would be negative that is the slope of the consumption
function will become flatter as income rises.
While short run time series and cross section evidence on the form of consumption
function broadly support Keynes hypothesis, long run evidence contradicts it.
Figure 1
According to the Absolute Income Hypothesis,
the consumption income relationship in non- C
proportional, as shown in the figure 1. In the C3
figure 1, as income increases over time, C2
consumption follows the non-proportional
function shown by C1, but over the long run
the statistical evidence suggests that C1
consumption function follows the path of the
proportional function as shown by C3. The
advocates of the Absolute Income Hypothesis
argue that there are upward shift in the non-
proportional consumption function as shown O Y
by the shift from C1 to C2, caused by the INCOME CONSUMPTION RELATIONSHIP
factors other than income.

1. With the increase in the accumulated wealth of households that has accompanied
the long run growth of income, households have tended to spend a larger fraction
of any given level of income, thereby contributing to an upward shift in the
consumption function.
2. Migration of population from rural to urban areas caused to shift consumption
function.
3. The percentage of older people in the population has increased over the long run
period. Because per capita consumption in this age group does not drop off as
rapidly as does per capita income, the consumption function tends to shift upward.
4. New consumer goods have been introduced at a rapid rate over this long run
period. As more and more of these goods become regarded as ‘essentials’ by the
typical household, the consumption function tends to shift upward.

Doubts about the adequacy of the Absolute Income Hypothesis arose because of its
apparent inability to reconcile budget data on saving with observed long run trends
 RELATIVE INCOME HYPOTHESIS (JAMES S. DUESENBERRY)

The Relative Income Hypothesis (RIH) of James S. Duesenberry is based on the rejection
of the two fundamental assumptions of the consumption theory of Keynes:
1. Every individual’s consumption behaviour is not independent but interdependent
of the behaviour of every other individual; and
2. Consumption relations are irreversible and not reversible in time.

The RIH states that consumption depends on two things:

1. Income relatives to that of other households or individuals; and
2. The level of income in the immediate preceding periods.

This two-part theory of Duesenberry explains the conflicting long run and short run cross
section evidence on the specification of the consumption function.

The first part suggests that the fraction of income consumed by a household, its APC, is
determined by its Relative Income, i.e., by its position in the income distribution
(Microanalysis). Therefore, if all incomes double its position in the income distribution is
unchanged and it will continue to consume the same proportion of income, its APC will
remain unchanged. It can, therefore, be shown that if the income distribution remains
unchanged in the long run the APC will remain constant as income increases. This aspect
of the theory explains the evidence on the form of the Long Run Consumption Function
(Macroanalysis).

In its focus on relative income, this hypothesis emphasizes the imitative or emulative
nature of consumption. A family with any given level of income will typically spend
more on consumption if it lives in a community in which that income is relatively low
than if it lives in a community in which that income is relatively high. This tendency
arises in part from the pressure on the family to “keep up with the Joneses” and in part
from the fact that as the family observes what seems to be the superior goods of other
families it will be tempted to spend as a result of what Jemes S. Duesenberry calls the
“Demonstration Effect”(Microanalysis).

The second part of the hypothesis explains the specification of the short run consumption
function. It suggests that households find it easier to adjust to rising incomes than falling
incomes. Therefore as income decline in the recession, household seeks to maintain their
standard of living and their consumption falls by less than their income. It follows that at
lower levels of income, during the recessionary phase of the trade cycle, the APC rises as
the evidence on the short run consumption function suggests. This is known as “Ratchet
Effect”.

Duesenberry combines his two related hypothesis in the following form:

Ct / Yt = a + b (Yh / Yt)
 Figure 2: Relative Income Hypothesis
Where Ct is the current level of C LRCF
expenditure, Yt is the current level of SRCF2
income, Yh is the highest level of income
previously earned, ‘a’ and ‘b’ are E2
numerical constants which relate income to C2 SRCF1
consumption. The above equation states
that when Yt increases relative to Yh, the C1 E0
APC decreases and increase in total
consumption is not proportional to the E1
increase in total income.
O Y
Y1 Y0 Y2

In figure 2, LRCF is the long run consumption function; SRCF1 and SRFC2 are the short
run consumption functions. Suppose income is at the peak level of OY0 where E0Y0 is
consumption. Suppose income fall to OY1, people are used to the standard of living at the
OY0 level of income, they will not reduce their consumption to E1Y1 level but to C1Y1,
moving backward along the SRCF1. If income continues to increase to OY2 level,
consumers will move upward along the LRCF curve from E0 to E2 on the new SRCF2. If
another recession occurs at OY2 level of income, consumption will decline along SRCF2.

CRITICISMS:
1. Increase in income along the full employment level does not always lead to
proportional increase in consumption.
2. The RIH assumes direct relationship between income and consumption but in
recession it is not correct.
3. The consumer behaviour is slowly reversible, instead of being truly irreversible.
4. Consumer preferences are interdependent towards only rich neighbour.
 PERMANENT INCOME HYPOTHESIS OF MILTON FRIEDMAN

The Permanent Income Hypothesis / Theory of consumption argues that consumption is

related not to current income but to a long term estimate of income, which Milton
Friedman, who introduced the theory, calls “permanent income”. Friedman provides a
simple example: consider a person who is paid or receives income only once a week, on
Fridays. We do not expect that person to consume only on Friday, with zero consumption
on the other days of the week. People prefer a smooth consumption flow rather than
plenty today and scarcity tomorrow or yesterday.

Permanent income may be thought of as the income an individual expects to derive from
his work and his holdings of wealth during his lifetime. More technically, each consumer
arrives at an approximation of his or her permanent income on the basis of his or her total
wealth, human and nonhuman. For some people, the wages and salaries from human
capital make up almost all of their income; for some others there is rental, interest and
dividend income derived from nonhuman wealth in the form of real property and
financial assets. Therefore, permanent income depends on the consumers ‘horizon and
foresightedness’.

Given this meaning of permanent income, a family’s measured or observed income (=

Permanent income + transitory income) in any particular year may be larger or smaller
than its permanent income. Friedman divides the family’s measured yearly income into
permanent and transitory components, so that its measured income is larger or smaller
than its permanent income, depending on the sum of positive or negative transitory
income components. Transitory income is the unexpected income or the windfall gains
(ex. bonus) or losses (ex. loss of income due to shut down of industry).

In the same way, Friedman divides measured consumption into permanent and
transitory components. A good purchased because of an attractive price or normal
purchases deferred due to unavailability of the goods are examples of positive and
negative transitory consumption 3 . As with measured income, a family’s measured
consumption in any particular period may be larger or smaller than its permanent
consumption.

ASSUMPTIONS:
1. There is no correlation between transitory and permanent income.
2. There is no correlation between transitory and permanent consumption.
3. There is no correlation between transitory consumption and transitory income.
4. Permanent income affects consumption systematically.

3
Friedman also defines consumption as spending on services and nondurable goods plus the depreciation
and interest cost on consumer durable goods. A net addition to the family’s stock of durable goods is
treated as saving. The purchase of consumer durables, the theory suggests, represents investment by the
household and only the flow of services from theses enters consumption. It is of course, extremely difficult
to obtain measures of the true value of these services.
Friedman’s basic argument that permanent consumption depends on permanent income.
The theory argues that permanent consumption (Cp) is a constant proportion of permanent
income (Yp), which depends only on the interest rate (r), the ratio of non-human wealth to
total wealth (human and non-human) (w), and tastes or utilities (u).
Cp = cYp; c = Cp / Yp
c = f (r, w, u)
Tastes are affected by factors such as age and family composition. The permanent
consumption of different families with the permanent income will, therefore, vary
according to their specific characteristics. However, if there is no reason to expect these
characteristics to vary with the level of income, it may be assumed that the average ratio
of consumption to permanent income for groups of families at different levels of
permanent income will be the same. The AP of families at all levels of family income is
held to be the same when APC = Cp / Yp.

Another basic argument of Friedman’s permanent income hypothesis is that the transitory
component of consumption is not correlated with the transitory component of income
This amounts to saying that in a period in which a family’s measured income contains a
transitory component, it does not reduce its consumption in response, nor under the
opposite circumstances, does it raise its consumption. Unexpected increases the decreases
in income, therefore, result in equivalent increases or decreases in saving; consumption is
unaffected by “wind fall” gains or losses. In other words, the MPC out of transitory, or
“wind fall”, income is held to be zero. By criticizing this argument of Friedman, H. S.
Houthakker says “the man who has a lucky day at the races does not buy his friends a
drink and the poor fellows whose wallet is stolen does not postpone the purchase of
new overcoat”.
Figure 3
In figure 3, Y is the measured income, C, Cp
Yp is the permanent income,
C is the measured consumption,
Cp is the permanent consumption. D Cp
According to PIH, APC out of permanent
B C
income is constant which is shown by Cp as
it starts from origin 4 . The C curve that
H A
intercepts the vertical axis above the origin
displays an APC that decreases at successively E F
higher levels of family income. G

Families in the upper income classes will tend

to have positive transitory income, where as
those in the lower income classes will tend to O Y1 Yp1 Y=Yp Yp2 Y2 Y, Yp
have negative transitory income. Families in The Relationship Between Measured
the average level will tend to have zero Consumption and Measured Income and between
transitory income, i.e., for families measured
Permanent Consumption and Permanent Income
income equals permanent income or Y = Yp.

4
Only a straight line through the origin displays the property of constant APC.
Let us examine any level of measured income in figure 3 other than the average level –
for example Y2. Because transitory income is correlated with measured income, families
at this measured income level will, generally, have permanent income below measured
income (positive transitory income). From the basic argument that permanent
consumption is proportional to permanent income, illustrated in figure 3 by the Cp curve,
it would follow that consumption at measured income of Y2 would have been that shown
by D on the Cp curve if Y2 had included no transitory income – that is, if it had been
permanent income. However, because there is a positive transitory component in Y2,
consumption is indicated by A rather than D. The fact that consumption is the amount
shown by A establishes that the average permanent income of families with measured
income Y2 is Yp2. This follows because at this level of permanent income, consumption
will be the amount indicated by point B on the Cp curve, which is an amount equal to A
on the C curve.

If we examine a level of measured income below the average – for example Y1 – the
relationship parallel those noted for the Y2 income level. However families at Y1 income
level, on the average, will have permanent income above measured income (negative
transitory income). Families with income of Y1 have measured consumption shown by
point E. as before, permanent consumption equals measured consumption, or equals E. If
Y1 had included no transitory income – that is, if it had permanent income – consumption
at Y1 would have been that shown by point G on the Cp curve. However, because there is
a negative component of transitory income in Y1, consumption is the larger amount
shown by E. As in the preceding paragraph, the fact that consumption is the amount
shown by E indicates that the average permanent income of the families with measured
income of Y1 is Yp1. at this level of permanent income , consumption will be the amount
indicated by point F on the Cp curve, which is an amount equal to E on the C curve.
 THE LIFE - CYCLE HYPOTHESIS OF FRANCO MODIGLIANI

This hypothesis developed by Franco Modigliani, Richard E. Brumberg and Albert Ando.
It states that individuals consume a constant proportion of the present value of their
lifetime income each period, i.e., Ct = kVt, where Ct is the current consumption by an
individual, k is the constant proportion of Vt consumed and Vt is the present value of the
lifetime income. Precisely what this proportion will be depends on each consumer’s taste
and preferences but provided the distribution of population by age and income is
relatively constant. Lifetime (Vt) income is the sum of the value of all property assets as
of that year (At) plus labour or non-property income for that year (YLt) plus the present or
year t value of expected labour or non-property income of future years (YeLt).

Vt = At + YLt + YeLt Therefore, Ct = k (At + YLt + YeLt)

Assumptions of the hypothesis:

1. There is no change in the price level during the life of the consumer.
2. The rate of interest remains stable.
3. The consumer does not inherit any assets and his net assets are the result of his
own saving.

The hypothesis argues that the general pattern of lifetime earning is such that income
level would reveal a disproportionately large number of people in middle age at the upper
end of the income distribution and a disproportionately large number of young and old
people at the lower end. Young and old households have a high average propensity
consume, indeed, they have often engaged in dissaving. Frequently they either borrow
agaisnts their future income, as in case of young, or run down their life savings as in the
case of the old. In contrast middle-aged household are either paying back earlier debts or
saving for old age and therefore have a low average propensity to consume. As a result
low-income households reveal high APC and vice versa; these facts explain the form of
consumption function by cross sectional analysis.
Figure 4
In long run this hypothesis suggests
household consume a constant proportion
of the present value of their lifetime
income and the long run consumption
function is suggested to capture this effect.

Finally, since actual labour income will

rise and fall around this long run average
the APC will found to vary inversely with
income over the length of the business
cycle as suggested by short run
consumption function.

In figure 4 the Y curve is a profile of The Life Cycle Hypothesis of

lifetime income stream of a typical person. Consumption
Measuring time from the year in which he begins full time employment, his income in
each following year rises until it reaches a peak in his middle or late working years;
thereafter it declines. The C curve shows his lifetime consumption stream, here drawn to
show a gradually increasing level of consumption from year to year. Assuming that he
plans zero bequests, he will seek to make the present value of his consumption over the
life cycle equal to the present value of his income. The shaded areas at the left of the
figure show that the individual’s consumption exceeds his income in the early years of
his working life – he is a dissaver, or net borrower. In the middle years, his income
exceeds his consumption – he is a saver. He not only repays the debts earlier incurred but
aquire assets on which he earns interest. Finally, in the late years indicated by the shaded
areas at the right of the figure 4, his consumption again exceeds his income – he is again
a dissaver. However, here the dissaving is financed not by borrowing but out of the
savings accumulated during the middle years.
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

THE LIFE-CYCLE-PERMANENT-INCOME THEORY OF CONSUMPTION AND SAVING

Modern consumption theory emphasizes lifetime decision-making. Originally, the life cycle hypothesis
emphasized choices about how to maintain a stable standard of living in the face of changes in income over
the course of life, while permanent income hypothesis focused on forecasting the level of income available
to a consumer over a lifetime. Today, these two theories have largely merged. The life cycle hypothesis is
essentially a permanent wealth hypothesis rather than a permanent income hypothesis.

A numerical example illustrates the theory: suppose that a person starts life at age 20, plans to work until
65, and will die at 80 and that annual labour income, Y, is $30,000. Lifetime resources are annual income
times year of working life (WL = 65 – 20 = 45) – in this example, $30,000 x 45 = $1,350,000. Spreading
lifetime resources over the number of year of life (YL = 80 – 20 = 60) allows for annual consumption of C
= $1,350,000 / 60 = $22,500. The general formula is

C = (WL/YL) Y
Where C is the annual consumption, WL is the year of working life, YL is the year of life remaining and Y
is the annual labour income, and (WL/YL) is the MPC.

Continuing with the numerical example, we can compute MPC by considering variation in the income
stream. Suppose income were to rise permanently by $3,000 per year. The extra $3,000 times 45 working
year spread over 60 years of life would increase annual consumption by $3,000 x (45/60) = $2,250. In other
words, the MPC out of permanent income would be WL/YL = 45/60 = 0.75. In contrast, suppose income
were to rise by $3,000 but only in one year. The extra 43,000 spread over 60 years would increase annual
consumption by $3,000 x (1/60) = $50. In other words, the MPC out of transitory income would be 1/YL =
1/60 ≈ 0.017. While the exact examples are slightly contrived, the clear message is that the MPC out of
permanent income is large and the MPC out of transitory income is very small fairly close to zero.

The life cycle theory implies that the MPC out of wealth should equal the MPC out of transitory income
and, therefore, be small. The reasoning is that spending out of wealth, like spending out of transitory
income, is spread out over the remaining year of life.

Note that the MPC out of permanent income, WL/YL, changes with age. In the text example, the MPC out
of permanent income at age 20 is 45/60. As a person ages, both the number of working years and the
number of life decline. By age 50, for example, the MPC would have declined to 15/30. The MPC out of
transitory income would rise from 1/60 to 1/30 at age 50.

Microanalysis: Theory
Macroanalysis: Empirical evidences taking long run, short run and cross sectional data.