. 48
( 126 .)


It has been said that economics is just systematized common sense. So let us now organ-
The consumption
ize and generalize what has been a completely intuitive discussion up to now. One thing
function shows the rela-
we have discovered is the apparently close relationship between consumer spending, C,
tionship between total
and disposable income, DI. Economists call this relationship the consumption function.
consumer expenditures and
A second fact we have gleaned from these figures is that the slope of the consumption
total disposable income in
the economy, holding all function is quite constant. We infer this constancy from the fact that the straight line
other determinants of con- drawn in Figure 4 comes so close to touching every point. If the slope of the consumption
sumer spending constant.
function had varied widely, we could not have done so well with a single straight line.7
Because of its importance in applications such as the tax cut, economists have given this
The marginal propensity
slope a special name”the marginal propensity to consume, or MPC for short. The MPC
to consume (MPC) is the
ratio of the change in con- tells us how much more consumers will spend if disposable income rises by $1.
sumption relative to the
Change in C
change in disposable
Change in DI that produces the change in C
income that produces the
change in consumption. On
The MPC is best illustrated by an example, and for this purpose we turn away from
a graph, it appears as the
U.S. data for a moment and look at consumption and income in a hypothetical country
slope of the consumption
whose data come in nice round numbers”which facilitates computation.

Figure 4 is limited to 17 years of data, so try fitting a single straight line to all of the data in Figure 3. You will

find that you can do rather well.

Copyright 2009 Cengage Learning, Inc. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.
Licensed to:
Chapter 8 161
Aggregate Demand and the Powerful Consumer

Columns (1) and (2) of Table 1 below show annual consumer expenditure and dispos-
able income, respectively, from 2002 to 2007. These two columns constitute the consump-
tion function, and they are plotted in Figure 5. Column (3) in the table shows the marginal
propensity to consume (MPC), which is the slope of the line in Figure 5; it is derived from
the first two columns. We can see that between 2004 and 2005, DI rose by $400 billion
(from $4,000 billion to $4,400 billion) while C rose by $300 billion (from $3,300 billion to
$3,600 billion). Thus, the MPC was
Change in C $300
MPC 5 5 0.75
Change in DI $400

As you can easily verify, the MPC between any other pair of years in Table 1 is also 0.75.
This relationship explains why the slope of the line in Figure 4 was so crucial in estimat-
ing the effect of a tax cut. This slope, which we found there to be 0.90, is simply the MPC
for the United States. The MPC tells us how much additional spending will be induced by
each dollar change in disposable income. For each $1 of tax cut, economists expect con-
sumption to rise by $1 times the marginal propensity to consume.
To estimate the initial effect of a tax cut on consumer spending, economists must first
estimate the MPC and then multiply the amount of the tax cut by the estimated MPC.8
Because they never know the true MPC with certainty, their prediction is always subject
F I GU R E 5
to some margin of error.
A Consumption

Consumption and Income in a C
Real Consumer Spending, C

Hypothetical Economy

(3) 3,900
(1) (2) Propensity
Consumption, Disposable to Consume,
C Income, DI
Year MPC
2002 $2,700 $3,200
2003 3,000 3,600
0.75 2,700
2004 3,300 4,000
2005 3,600 4,400
2006 3,900 4,800 0 3,200 3,600 4,000 4,400 4,800 5,200
2007 4,200 5,200
Real Disposable Income, DI
NOTE: Amounts are in billions of dollars.

Unfortunately for policy planners, the consumption function does not always stand still.
Recall from Chapter 4 the important distinction between a movement along a demand
curve and a shift of the curve. A demand curve depicts the relationship between quantity
demanded and one of its many determinants”price. Thus a change in price causes a
movement along the demand curve. But a change in any other factor that influences quan-
tity demanded causes a shift of the entire demand curve.
Because factors other than disposable income influence consumer spending, a similar
distinction is vital to understanding real-world consumption functions. Look back at the
definition of the consumption function in the margin of page 160. A change in disposable
income leads to a movement along the consumption function precisely because the consump-
tion function depicts the relationship between C and DI. Such movements, which are what
we have been considering so far, are indicated by the brick-colored arrows in Figure 6.

The word initial in this sentence is an important one. The next chapter will explain why the effects discussed in

this chapter are only the beginning of the story.

Copyright 2009 Cengage Learning, Inc. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.
Licensed to:
Part 2
162 The Macroeconomy: Aggregate Supply and Demand

But consumption also has other determinants, and a
Movements along change in any of them will shift the entire consumption
consumption function
function”as indicated by the blue arrows in Figure 6. Such
shifts account for many of the errors in forecasting consump-
Real Consumer Spending

tion. To summarize:
Any change in disposable income moves us along a given
consumption function. A change in any of the other determi-
nants of consumption shifts the entire consumption schedule
(see Figure 6).
Shifts of
consumption Because disposable income is far and away the main deter-
minant of consumer spending, the real-world data in
Figure 3 come close to lying along a straight line. But if you
use a ruler to draw such a line, you will find that it misses
a number of points badly. These deviations reflect the in-
Real Disposable Income
fluence of the “other determinants” just mentioned. Let us
see what some of them are.
F I GU R E 6
Shifts of the Wealth One factor affecting spending is consumers™ wealth, which is a source of
Consumption Function
purchasing power in addition to income. Wealth and income are different things. For
example, a wealthy retiree with a huge bank balance may earn little current income when
interest rates are low. But a high-flying investment banker who spends every penny of the
high income she earns will not accumulate much wealth.
To appreciate the importance of the distinction, think about two recent college gradu-
ates, each of whom earns $40,000 per year. If one of them has $100,000 in the bank and the
other has no assets at all, who do you think will spend more? Presumably the one with the
big bank account. The general point is that current income is not the only source of spend-
able funds; households can also finance spending by cashing in some of the wealth they
have previously accumulated.
One important implication of this analysis is that the stock market can exert a major in-
fluence on consumer spending. A stock market boom adds to wealth and thus raises the
consumption function, as depicted by the shift from C0 to C1 in Figure 6. That is what hap-
pened in the late 1990s, when the stock market soared and American consumers went on
a spending spree. Correspondingly, a collapse of stock prices, like the one that occurred in
2000“2002, should shift the consumption function down (see the shift from C0 to C2).
Using the same logic, as this book went to press, many economists were worrying that
falling house prices, which were making consumers less wealthy, would therefore make
them less willing to spend.

The Price Level Stocks are hardly the only form of wealth. People hold a great deal of
wealth in forms that are fixed in money terms. Bank accounts are the most obvious exam-
ple, but government and corporate bonds also have fixed face values in money terms. The
purchasing power of such money-fixed assets obviously declines whenever the price
A money-fixed asset is an
asset whose value is a fixed level rises, which means that the asset can buy less. For example, if the price level rises by
number of dollars. 10 percent, a $1,000 government bond will buy about 10 percent less than it could when
prices were lower. This is no trivial matter. Consumers in the United States hold money-
fixed assets worth well over $8 trillion, so that each 1 percent rise in the price level reduces
the purchasing power of consumer wealth by more than $80 billion, a tidy sum. This
process, of course, operates equally well in reverse, because a decline in the price level in-
creases the purchasing power of money-fixed assets.

The Real Interest Rate A higher real rate of interest raises the rewards for saving. For
this reason, many people believe it is “obvious” that higher real interest rates encourage
saving and therefore discourage spending. Surprisingly, however, statistical studies of
this relationship suggest otherwise. With very few exceptions, they show that interest
rates have negligible effects on consumption decisions in the United States and other

Copyright 2009 Cengage Learning, Inc. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.
Licensed to:
Chapter 8 163
Aggregate Demand and the Powerful Consumer

Using the Tax Code to Spur Saving
Compared to the citizens of virtu- get to keep the full 5 percent. Over
ally every other industrial nation, long periods of time, this seemingly
Americans save very little. Many small interest differential com-
policy makers consider this lack of pounds to make an enormous dif-
SOURCE: © Elizabeth Simpson/Taxi/Getty Images
saving to be a serious problem, so ference in returns. For example,
they have proposed numerous $100 invested for 20 years at
changes in the tax laws to increase 3.5 percent interest grows to $199.
incentives to save. In 2001, for ex- But at 5 percent, it grows to $265.
ample, Congress expanded Indi- Members of Congress who advocate
vidual Retirement Accounts (IRAs), tax incentives for saving argue that
which allow taxpayers to save tax- lower tax rates will therefore induce
free. In 2003, the taxation of divi- Americans to save more.
dends was reduced. Further tax This idea seems reasonable and
incentives for saving seem to be has many supporters. Unfortu-
proposed every year. nately, the evidence runs squarely
All of these tax changes are designed to increase the after-tax return against it. Economists have conducted many studies of the effect of
on saving. For example, if you put away money in a bank at a 5 percent higher rates of return on saving. With very few exceptions, they de-
rate of interest and your income is taxed at a 30 percent rate, your tect little or no impact. Although the evidence fails to support the
after-tax rate of return on saving is just 3.5 percent (70 percent “common-sense” solution to the undersaving problem, the debate
of 5 percent). But if the interest is earned tax-free, as in an IRA, you goes on. Many people, it seems, refuse to believe the evidence.

countries. Hence, in developing our model of the economy, we will assume that changes
in real interest rates do not shift the consumption function. (See the box “Using the Tax
Code to Spur Saving.”)

Future Income Expectations It is hardly earth-shattering to suggest that consumers™
expectations about their future incomes should affect how much they spend today. This
final determinant of consumer spending holds the key to resolving the puzzle posed at
the beginning of the chapter: Why did tax policy designed to boost consumer spending
apparently fail in 1975 and succeed only modestly in 2001?

To understand how expectations of future incomes affect current consumer
expenditures, consider the abbreviated life histories of three consumers given
in Table 2. (The reason for giving our three imaginary individuals such
odd names will be apparent shortly.) The consumer named “Constant”


. 48
( 126 .)