Questions for Review
1. First, Keynes conjectured that the marginal propensity to consume—the amount consumed
out of an additional dollar of income—is between zero and one. This means that
if an individual’s income increases by a dollar, both consumption and saving increase.
Second, Keynes conjectured that the ratio of consumption to income—called the
average propensity to consume—falls as income rises. This implies that the rich save a
higher proportion of their income than do the poor.
Third, Keynes conjectured that income is the primary determinant of consumption.
In particular, he believed that the interest rate does not have an important effect
on consumption.
A consumption function that satisfies these three conjectures is
C = C + cY.
C is a constant level of “autonomous consumption,” and Y is disposable income; c is the
marginal propensity to consume, and is between zero and one.
2. The evidence that was consistent with Keynes’s conjectures came from studies of household
data and short time-series. There were two observations from household data.
First, households with higher income consumed more and saved more, implying that
the marginal propensity to consume is between zero and one. Second, higher-income
households saved a larger fraction of their income than lower-income households,
implying that the average propensity to consume falls with income.
There were three additional observations from short time-series. First, in years
when aggregate income was low, both consumption and saving were low, implying that
the marginal propensity to consume is between zero and one. Second, in years with low
income, the ratio of consumption to income was high, implying that the average propensity
to consume falls as income rises. Third, the correlation between income and consumption
seemed so strong that no variables other than income seemed important in
explaining consumption.
The first piece of evidence against Keynes’s three conjectures came from the failure
of “secular stagnation” to occur after World War II. Based on the Keynesian consumption
function, some economists expected that as income increased over time, the
saving rate would also increase; they feared that there might not be enough profitable
investment projects to absorb this saving, and the economy might enter a long depression
of indefinite duration. This did not happen.
The second piece of evidence against Keynes’s conjectures came from studies of
long time-series of consumption and income. Simon Kuznets found that the ratio of consumption
to income was stable from decade to decade; that is, the average propensity to
consume did not seem to be falling over time as income increased.
3. Both the life-cycle and permanent-income hypotheses emphasize that an individual’s
time horizon is longer than a single year. Thus, consumption is not simply a function of
current income.
The life-cycle hypothesis stresses that income varies over a person’s life; saving
allows consumers to move income from those times in life when income is high to those
times when it is low. The life-cycle hypothesis predicts that consumption should depend
on both wealth and income, since these determine a person’s lifetime resources. Hence,
we expect the consumption function to look like
C = αW + βY.
175
C H A P T E R 17 Consumption
In the short run, with wealth fixed, we get a “conventional” Keynesian consumption
function. In the long run, wealth increases, so the short-run consumption function
shifts upward, as shown in Figure 17–1.
The permanent-income hypothesis also implies that people try to smooth consumption,
though its emphasis is slightly different. Rather than focusing on the pattern
of income over a lifetime, the permanent-income hypothesis emphasizes that people
experience random and temporary changes in their income from year to year. The
permanent-income hypothesis views current income as the sum of permanent income
Y
p
and transitory income Y
t
. Milton Friedman hypothesized that consumption should
depend primarily on permanent income:
C = αY
p
.
The permanent-income hypothesis explains the consumption puzzle by suggesting
that the standard Keynesian consumption function uses the wrong variable for income.
For example, if a household has high transitory income, it will not have higher consumption;
hence, if much of the variability in income is transitory, a researcher would
find that high-income households had, on average, a lower average propensity to consume.
This is also true in short time-series if much of the year-to-year variation in
income is transitory. In long time-series, however, variations in income are largely permanent;
therefore, consumers do not save any increases in income, but consume them
instead.
176 Answers to Textbook Questions and Problems
C
Y
Income
αw
αw
Consumption
'
Figure 17–1
4. Fisher’s model of consumption looks at how a consumer who lives two periods will
make consumption choices in order to be as well off as possible. Figure 17–2(A) shows
the effect of an increase in second-period income if the consumer does not face a binding
borrowing constraint. The budget constraint shifts outward, and the consumer increases
consumption in both the first and the second period. In Figure 17-2(A), Y1 is the first
period income and Y2 is second period income. In choosing to consume at point A or B,
the consumer is consuming more than their income in period 1 and less than their
income in period 2.
Figure 17–2(B) shows what happens if there is a binding borrowing constraint.
The consumer would like to borrow to increase first-period consumption but cannot. If
income increases in the second period, the consumer is unable to increase first-period
consumption. Therefore, the consumer continues to consume his or her entire income in
each period. That is, for those consumers who would like to borrow but cannot, consumption
depends only on current income.
Chapter 17 Consumption 177
C2
Y2
Y2
+ Δ Y2
Second-periodconsumption
I2
I1
C1
Y1
First-period consumption
B
A
Old
budget
constraint
New budget constraint
Figure 17–2A
New
budget
constraint
Old
budget
constraint
B
A
C2
Y2
+ ΔY2
Y2
Second-periodconsumption
Y1
C1
First-period consumption
I1
I2
Figure 17–2B
5. The permanent-income hypothesis implies that consumers try to smooth consumption
over time, so that current consumption is based on current expectations about lifetime
income. It follows that changes in consumption reflect “surprises” about lifetime
income. If consumers have rational expectations, then these surprises are unpredictable.
Hence, consumption changes are also unpredictable.
6. Section 17.6 included several examples of time-inconsistent behavior, in which consumers
alter their decisions simply because time passes. For example, a person may
legitimately want to lose weight, but decide to eat a large dinner today and eat a small
dinner tomorrow and thereafter. But the next day, they may once again make the same
choice—eating a large dinner that day while promising to eat less on following days.
Problems and Applications
1. Figure 17–3 shows the effect of an increase in the interest rate on a consumer who borrows
in the first period. The increase in the real interest rate causes the budget line to
rotate around the point (Y1, Y2), becoming steeper.
We can break the effect on consumption from this change into an income and substitution
effect. The income effect is the change in consumption that results from the
movement to a different indifference curve. Because the consumer is a borrower, the
increase in the interest rate makes the consumer worse off—that is, he or she cannot
achieve as high an indifference curve. If consumption in each period is a normal good,
this tends to reduce both C1 and C2.
The substitution effect is the change in consumption that results from the change
in the relative price of consumption in the two periods. The increase in the interest rate
makes second-period consumption relatively less expensive; this tends to make the consumer
choose more consumption in the second period and less consumption in the first
period.
On net, we find that for a borrower, first-period consumption falls unambiguously
when the real interest rate rises, since both the income and substitution effects push in
the same direction. Second-period consumption might rise or fall, depending on which
178 Answers to Textbook Questions and Problems
New budget
constraint
Old
budget
constraint
C2
Y2
ΔC2
Second-periodconsumption
I1
I2
C1
Y1
ΔC1
First-period consumption
B A
Figure 17–3
effect is stronger. In Figure 17–3, we show the case in which the substitution effect is
stronger than the income effect, so that C2 increases.
2. a. We can use Jill’s intertemporal budget constraint to solve for the interest rate:
C1 + = Y1 +
$100 + = $0 +
r = 10%.
Jill borrowed $100 for consumption in the first period and in the second period
used her $210 income to pay $110 on the loan (principal plus interest) and $100
for consumption.
b. The rise in interest rates leads Jack to consume less today and more tomorrow.
This is because of the substitution effect: it costs him more to consume today than
tomorrow, because of the higher opportunity cost in terms of forgone interest. This
is shown in Figure 17–4.
By revealed preference we know Jack is better off: at the new interest rate he
could still consume $100 in each period, so the only reason he would change his
consumption pattern is if the change makes him better off.
Chapter 17 Consumption 179
C2
1 + r
$100
1 + r
$210
1 + r
Y2
1 + r
B
A
100 191 C1
Consumption today
Consumptiontomorrow
100
210
C2
Figure 17–4
180 Answers to Textbook Questions and Problems
c. Jill consumes less today, while her consumption tomorrow can either rise or fall.
She faces both a substitution effect and income effect. Because consumption today
is more expensive, she substitutes out of it. Also, since all her income is in the second
period, the higher interest rate raises her cost of borrowing and, thus, lowers
her income. Assuming consumption in period one is a normal good, this provides
an additional incentive for lowering it. Her new consumption choice is at point B
in Figure 17–5.
We know Jill is worse off with the higher interest rates because she could have
consumed at point B before (by not spending all of her second-period money) but
chose not to because point A had higher utility.
3. a. A consumer who consumes less than his income in period one is a saver and faces
an interest rate rs. His budget constraint is
C1 + C2/(1 + rs) = Y1 + Y2/(1 + rs).
b. A consumer who consumes more than income in period one is a borrower and faces
an interest rate rb. The budget constraint is
C1 + C2/(1 + rb) = Y1 + Y2/(1 + rb).
B
A
100 191 C1
Consumption today
Consumptiontomorrow
100
210
C2
Figure 17–5
Chapter 17 Consumption 181
c. Figure 17–6 shows the two budget constraints; they intersect at the point (Y1, Y2),
where the consumer is neither a borrower nor a lender. The shaded area represents
the combinations of first-period and second-period consumption that the consumer
can choose. To the left of the point (Y1, Y2), the interest rate is rb.
d. Figure 17–7 shows the three cases. Figure 17–7(A) shows the case of a saver for
whom the indifference curve is tangent to the budget constraint along the line segment
to the left of (Y1, Y2). Figure 17–7(B) shows the case of a borrower for whom
the indifference curve is tangent to the budget constraint along the line segment
to the right of (Y1, Y2). Finally, Figure 17–7(C) shows the case in which the consumer
is neither a borrower nor a lender: the highest indifference curve the consumer
can reach is the one that passes through the point (Y1, Y2).
C1
Y1
First-period consumption
Y2
C2
Second-periodconsumption
Interest rate
of rs
Interest rate
of rb
Figure 17–6
182 Answers to Textbook Questions and Problems
I
C2
C2
Y2
Second-periodconsumption
C1
First-period consumption
Y1
C1
Figure 17–7A
I
C2
C2
Y2
Second-periodconsumption
C1
C1
Y1
Figure 17–7B
I
C2
Y2
Second-periodconsumption
C1
Y1
First-period consumption
Figure 17–7C
Chapter 17 Consumption 183
e. If the consumer is a saver, then consumption in the first period depends on [Y1 +
Y2/(1 + rs)]—that is, income in both periods, Y1 and Y2, and the interest rate rs. If
the consumer is a borrower, then consumption in the first period depends on [Y1 +
Y2/(1 + rb)]—that is, income in both periods, Y1 and Y2, and the interest rate rb.
Note that borrowers discount future income more than savers.
If the consumer is neither a borrower nor a lender, then consumption in the
first period depends just on Y1.
4. The potency of fiscal policy to influence aggregate demand depends on the effect on consumption:
if consumption changes a lot, then fiscal policy will have a large multiplier.
If consumption changes only a little, then fiscal policy will have a small multiplier.
That is, the fiscal-policy multipliers are higher if the marginal propensity to consume is
higher.
a. Consider a two-period Fisher diagram. A temporary tax cut means an increase in
first-period disposable income Y1. Figure 17–8(A) shows the effect of this tax cut
on a consumer who does not face a binding borrowing constraint, whereas Figure
17–8(B) shows the effect of this tax cut on a consumer who is constrained.
The consumer with the constraint would have liked to get a loan to increase C1,
but could not. The temporary tax cut increases disposable income: as shown in the
figure, the consumer’s consumption rises by the full amount that taxes fall. The
consumer who is constrained thus increases first-period consumption C1 by more
A
B
A B
C2
Y2
C2
Y2
Second-periodconsumptionSecond-periodconsumption
Y1
+ Δ Y1
Y1
First-period consumption
First-period consumption
Y1
+ Δ Y1
I1
C1
I1
Y1
C1
I2
I2
Figure 17–8A
Figure 17–8B
184 Answers to Textbook Questions and Problems
than the consumer who is not constrained—that is, the marginal propensity to
consume is higher for a consumer who faces a borrowing constraint. Therefore, fiscal
policy is more potent with binding borrowing constraints than it is without
them.
b. Again, consider a two-period Fisher diagram. The announcement of a future tax
cut increases Y2. Figure 17–9(A) shows the effect of this tax cut on a consumer
who does not face a binding borrowing constraint, whereas Figure 17–9(B) shows
the effect of this tax cut on a consumer who is constrained.
The consumer who is not constrained immediately increases consumption C1. The
consumer who is constrained cannot increase C1, because disposable income has
not changed. Therefore, the announcement of a future tax cut has no effect on consumption
or aggregate demand if consumers face binding borrowing constraints:
fiscal policy is less potent.
5. In this question, we look at how income growth affects the pattern of consumption and
wealth accumulation over a person’s lifetime. For simplicity, we assume that the interest
rate is zero and that the consumer wants as smooth a consumption path as possible.
a. Figure 17–10 shows the case in which the consumer can borrow. Income increases
during the consumer’s lifetime until retirement, when it falls to zero.
B
A
B
A
I1
C1
Y1
Y2
C2
Y2
+ Δ Y2
Second-periodconsumptionSecond-periodconsumption
C2
Y2
+ Δ Y2
Y2
C1
Y1
I2
New budget
constraint
First-period consumption
I2
I1
New budget
constraint
Figure 17–9A
Figure 17–9B
Desired consumption is level over the lifetime. Until year T1, consumption is
greater than income, so the consumer borrows. After T1, consumption is less than
income, so the consumer saves. This means that until T1, wealth is negative and
falling. After T1, wealth begins to increase; after T2, all borrowing is repaid, so
wealth becomes positive. Wealth accumulation continues until retirement, when
the consumer dissaves all wealth to finance consumption.
b. Figure 17–11 shows the case in which a borrowing constraint prevents the consumer
from having negative wealth. Before T1′, the consumer would like to be borrowing,
as in part (a), but cannot. Therefore, income is consumed and is neither saved nor
borrowed. After T1′, the consumer begins to save for retirement, and lifetime consumption
remains constant at C′. In Figure 17-11 the consumption path has consumption
rising up to time T1′, and then consumption remains constant at C′.
Chapter 17 Consumption 185
Wealth
Income
Saving
Dissaving
ConsumptionC
Levelsofincome,consumption,andwealth
End of
life
Retirement
begins
T1
T2
Figure 17–10
Wealth
Income
Saving
Consumption
C
Levelsofincome,consumption,andwealth
End of
life
Retirement
begins
C'
T1
'
Figure 17–11
186 Answers to Textbook Questions and Problems
Note that C′ is greater than C, and T1′ is greater than T1, as shown. This is
because in part (b), the consumer has lower consumption in the first part of life, so
there are more resources left when there is no constraint—consumption will be
higher. The case where people are borrowing constrained in their early working
years is more realistic since it is difficult, if not impossible, to borrow against
expected future income.
6. The life-cycle model predicts that an important source of saving is that people save
while they work to finance consumption after they retire. That is, the young save, and
the old dissave. If the fraction of the population that is elderly will increase over the
next 20 years, the life-cycle model predicts that as these elderly retire, they will begin
to dissave their accumulated wealth in order to finance their retirement consumption:
thus, the national saving rate should fall over the next 20 years.
7. In this chapter, we discussed two explanations for why the elderly do not dissave as
rapidly as the life-cycle model predicts. First, because of the possibility of unpredictable
and costly events, they may keep some precautionary saving as a buffer in case they
live longer than expected or have large medical bills. Second, they may want to leave
bequests to their children, relatives, or charities, so again, they do not dissave all of
their wealth during retirement.
If the elderly who do not have children dissave at the same rate as the elderly who
do have children, this seems to imply that the reason for low dissaving is the precautionary
motive; the bequest motive is presumably stronger for people who have children
than for those who don’t.
An alternative interpretation is that perhaps having children does not increase
desired saving. For example, having children raises the bequest motive, but it may also
lower the precautionary motive: you can rely on your children in case of financial emergency.
Perhaps the two effects on saving cancel each other.
8. If you are a fully rational and time-consistent consumer, you would certainly prefer the
saving account that lets you take the money out on demand. After all, you get the same
return on that account, but in unexpected circumstances (e.g., if you suffer an unexpected,
temporary decline in income), you can use the funds in the account to finance
your consumption. This is the kind of consumer in the intertemporal models of Irving
Fisher, Franco Modigliani, and Milton Friedman.
By contrast, if you face the “pull of instant gratification,” you may prefer the
account that requires a 30-day notification before withdrawals. In this way, you precommit
yourself to not using the funds to satisfy a desire for instant gratification. This
precommitment offers a way to overcome the time-inconsistency problem. That is,
some people would like to save more, but at any particular moment, they face such a
strong desire for instant gratification that they always choose to consume rather than
save. This is the type of consumer in David Laibson’s theory.