Supplementary material for

Chapter 9
9.8 Understanding aggregate Desired consumption and desired
demand and the multiplier in investment spending
terms of the Keynesian cross It is assumed that desired consumption, consisting
of expenditures consumers desire to make in order to
model (supplementary material, buy final goods and services, depends on consumers’
recommended for higher level) real income. There is a positive, causal relationship
between real income and consumption, shown in
This material is included for the interested reader Figure 1(a), where the vertical axis measures desired
as a continuation of Chapter 9, section 9.8 spending (the dependent variable), and the horizontal
(textbook page 264). axis measures real income (Y, the independent
John Maynard Keynes was a famous British variable).1 The higher the income, the greater is
economist who lived in the twentieth century, desired consumption spending. Note that the area
and whose work laid the foundations for modern between the two axes is cut by a line making a
macroeconomics. The model presented here is 45-degree angle with the horizontal axis. This line
attributed to him, though it is a simplification of his represents all points of equality between the variables
highly technical work. This model is very helpful measured on the two axes, and therefore equality
for understanding some important macroeconomic between desired spending and income. Therefore at
concepts, including the relationship between point b, desired consumption spending is exactly
aggregate demand and aggregate output (real GDP); equal to income; at a, desired consumption is greater
the Keynesian idea of less than full employment than income, while at c, desired consumption is less
equilibrium; and the multiplier effect. than income.
The C line, representing consumption spending,
Consumption and investment spending is called a consumption function, because C is a
We have defined both aggregate demand and real function of national income, Y. Since it is a straight
GDP to consist of C + I + G + (X − M), yet the two are ΔC
line, it has a constant slope, given by (the change
not the same. How is this possible? To understand ΔY
this, we must begin by making a distinction between in the dependent variable divided by the change in
actual expenditure, and desired expenditure. Aggregate the independent variable, between any two points2).
output, or real GDP, is measured by adding up all The slope of the C function has a special meaning. It
actual expenditures, C + I + G + (X−M), for the purchase is the marginal propensity to consume (MPC),
of output (in the expenditure approach). Aggregate representing the change in desired consumption that
demand, on the other hand, is concerned with all results when there is a change in income. For example,
desired aggregate expenditures, which adds up desired suppose that income increases by $1000 million and
C + I + G + (X − M) for the purchase of output. consumption spending increase by $750 million. The
With this distinction in mind, we begin with 750 3
slope, or MPC, is = .
a simple version of the model that includes only 1000 4
consumers and firms, and therefore only desired You may be wondering why the consumption
consumption (C) and desired investment (I) spending. function shown in Figure 1(a) has the particular shape

1 Note that income, the independent variable, is plotted on the horizontal axis, following correct mathematical convention; see ‘Quantitative techniques’
chapter, on the CD-ROM, page 11.
2 In the chapter ‘Quantitative techniques’ on the CD-ROM, page 18, the slope was defined as the change in the dependent variable divided by the change

in the independent variable, between any two points on a straight line. Here, because of the use of the correct mathematical practice of plotting the
dependent variable on the vertical axis, the slope is the vertical change divided by the horizontal change. This differs from the slope in demand and supply
functions because of the reversal of the axes (see ‘Quantitative techniques’ chapter, pages 19 and 21).

© Cambridge University Press 2012 Economics for the IB Diploma 1

 Supplementary material for Chapter 9
simply add desired investment spending to desired
(a) Consumption spending
consumption spending, in order to arrive at total
45° line desired spending, shown in Figure 1(b). You can
see that desired investment spending is a constant
amount for all income levels, because the C + I line is
desired spending (C)

C>Y c C parallel to the C line.

b Any kind of spending that is independent of
a
income (meaning it is not ‘caused’ by income) is
called autonomous spending. By contrast, any kind of
C<Y spending that is dependent on income (is ‘caused’
by income) is called induced spending. Therefore,
45° while consumption spending is induced, investment
0 YI spending is autonomous.3
income (Y)

(b) Consumption plus investment spending Equilibrium level of income and output
Now remember from the circular flow model that
45° line national income is equal to the value of aggregate
C+I
output, or real GDP. This means we can re-label the
desired spending (C + I)

=I horizontal axis of our diagram as in Figure 2(a).

C
We are now ready to put together all our
information and arrive at some important conclusions
using Figure 2(a), which is the same as Figure 1(b)
except that the C line (the consumption function) has
been left out. We can see immediately that at point
e, total desired spending of consumers and firms is
45°
0 YI Ye
exactly equal to national income or real GDP. This
means that the amount that consumers and firms
income (Y)
want (desire) to buy is exactly equal to the amount
that is actually produced. This is the equilibrium level of
Figure 1 The Keynesian cross model with consumption and investment output in this model, shown as Ye.
spending What happens when real GDP produced is less than
Ye? At Y1, the amount consumers and firms want to
it has. The answer has to do with desired saving. buy (point a) is greater than the quantity of real GDP
Income is equal to desired consumption plus desired produced (point b). At Y1, there is insufficient output
saving. Abbreviating saving as S, we can write Y = C + S. to satisfy consumers’ and firms’ spending desires. On
When income is low (lower than Y1 in the figure) and the other hand, if real GDP is greater than Ye, such as
C > Y, saving is negative, because income is too low to Y2, the amount that consumers and firms want to buy
provide consumers with enough money to buy their (point d) is less than real GDP produced (point c). Too
necessities (negative saving means that consumers are many goods and services are being produced relative
borrowing or else spending past savings from previous to what consumers and firms want to buy.
years). When incomes are higher (higher than Y1) and If the economy finds itself producing too much or
C < Y, saving is positive. too little real GDP relative to what buyers want, how is
What happens to desired saving; where does the equilibrium restored? At Y1, with too much spending
money that consumers want to save go? To answer relative to output produced, firms’ inventories (unsold
this question, we must recall the circular flow model stocks from production of previous years) are sold,
with leakages and injections, where we saw that thus providing firms with the signal to increase
saving, a leakage from the spending flow, is matched production, causing real GDP to increase to Ye. At Y2,
by investment, an injection into the spending flow. with too little spending relative to what is produced,
We will now add investment spending to our model, firms’ inventories increase, providing firms with the
which is assumed to be independent of income, and signal that they are producing too much, causing
so is a constant amount for all income levels. We them to cut back on production, so real GDP falls

3 Actually, only a portion of consumption spending is induced. When income is zero, there is a positive level of consumption spending (given by the
vertical-intercept of the consumption function); this is autonomous consumption. All consumption above this level is induced, as it is ‘caused’ by income.

© Cambridge University Press 2012 Economics for the IB Diploma 2

 Supplementary material for Chapter 9
Adding government and the
(a) Equilibrium in the Keynesian cross model with C and I foreign sector
45 degree line It is assumed that government spending (G), and
spending of foreigners for exports minus the spending
c C+I of domestic residents for imports (X − M or Xn for net
desired spending (C + I)

too much spending exports) are independent of national income, and are
C + I > GDP e d therefore autonomous. It is thus a simple matter to
too little spending add them into our model. We do so simply by
C + I < GDP
a adding G + (X − M) to the C + I function, as shown in
Figure 2(b).
b It may be noted that adding X, which is an
equilibrium GDP
Injections = leakages injection into spending, works to increase total desired
spending, while adding M, a leakage, works to decrease
45° it. Therefore, whether the addition of (X − M) will
0 Y1 Ye Y2 increase or decrease total desired spending depends on
income the relative size of exports and imports. If X > M, then
= real GDP (X − M) > 0, and desired spending increases. However
(Y) if X < M, then (X − M) < 0, and desired spending
decreases. It should also be noted that the addition of
(b) Equilibrium in the Keynesian cross model with C, I, G, and (X − M)
government spending G is actually more complicated
45° line than shown in Figure 2(b) because of the role of taxes,
which are ignored here for simplicity.
desired spending [C + I + G + (x – m)]

C + I + G + Xn > C + I + G + Xn
C + I + G + Xn, representing total desired spending,
real GDP
are referred to as aggregate expenditure. When
C + I + G + Xn < aggregate expenditure is equal to real GDP, the
real GDP economy is in equilibrium, as seen in Figure 2(b). At
this equilibrium, the sum of leakages is equal to the
C + I + G + Xn =
sum of injections, so that:
real GDP
S+T+M = I+G+X

It is easy to understand the meaning of this

45°
equilibrium if we refer to the principle we studied
in the circular flow model (see Chapter 8 of the
0 Ye income textbook, page 217): in any given time period, the
= real GDP
(Y) value of output produced by an economy is equal
to the total income that is generated in producing
that output, which is equal to the expenditures made
Figure 2 Equilibrium in the Keynesian cross model to purchase that output. When the economy is at
equilibrium in the Keynesian cross model, the value
to Ye. Through this mechanism involving changes in of output (real GDP) is exactly equal to the spending
inventories that provide signals to firms on whether that purchases that output. This is made possible
they should produce less or more, equilibrium real by the equality between injections and leakages. If
GDP is restored. the sums of injections and leakages are not equal to
We can also understand the meaning of Ye by thinking each other, the economy cannot be in equilibrium,
in terms of leakages and injections. Recall that saving because desired spending to buy output will be
is a leakage and investment is an injection into the different from the value of output actually produced.
income flow. Since saving is that part of income that
is not consumed, it follows that when total desired Relating aggregate expenditure
spending (C + I) is exactly equal to income, desired saving, to aggregate demand
or the leakage must be equal to desired investment,
or the injection. Therefore, Ye is the only level of real You may have noticed that in the discussion above,
GDP where desired saving is exactly equal to desired there was no mention of the price level. The reason
investment, so that the leakage equals the injection. is that in the Keynesian cross model, total desired

© Cambridge University Press 2012 Economics for the IB Diploma 3

 Supplementary material for Chapter 9
spending and real GDP are compared for a single given appearing on the horizontal axis, shows various
price level. The Keynesian cross model does not show quantities of aggregate output that can be produced.
changes in the price level. The question that arises is If we were to measure this output for a particular
what happens if the price level changes? year, we would add up the actual spending of the four
Consider Figure 3, showing an aggregate groups of buyers (C + I + G + (X − M)) in order to obtain
expenditure curve, AE1, for a price level 1, with the value of aggregate output. Aggregate demand,
equilibrium real GDP Ye1. Suppose there is an increase on the other hand, shows the amount of aggregate
in the price level, from price level 1 to price level 2. output that the four groups of buyers want to buy at
This will appear in the figure as a downward shift of each possible price level, when that output generates just
the entire aggregate expenditure function, to AE2. enough spending allowing them to buy that output.
With AE2, the equilibrium level of real GDP falls from
Ye1 to Ye2. If the price level increases further to price The Keynesian cross model and
level 3, the AE function shifts further downward to output gaps
AE3, and equilibrium GDP falls to Ye3. Output gaps were noted in our discussion of the
Why does an increase in the price level cause a fall business cycle in Chapter 8 (page 232), as well as in
in aggregate expenditures? There are three reasons for Chapter 9 (pages 244 and 252). They arise whenever
this: actual output differs from potential output. We can
• A higher price level means a decrease in the real now see how output gaps arise in the context of the
value of wealth; consumers feel worse off and cut Keynesian cross model, shown in Figure 4. In both
back on their desired level of spending. parts, Yp represents potential output, which is full
employment output. Ye represents the equilibrium
• A higher price level means an increase in the
level of output, determined by the level of aggregate
demand for money, resulting in higher interest
expenditures, AEe. Note that the equilibrium level of
rates that cause the desired spending of consumers
output need not be equal to potential output.
and firms to decrease due to the higher cost of
borrowing.
• A higher price level means exports become more
expensive to foreigners, while imports become less
expensive to domestic buyers, causing Xn to fall. (a) Aggregate expenditure with varying price levels
45° line
Note that these reasons are exactly what accounts AE1 = [C + I + G + Xn]
a
desired spending [C + I + G + Xn]

for the downward-sloping aggregate demand curve. In price level 1

fact, by varying the price level in the Keynesian cross AE2 = [C + I + G + Xn]
b price level 2
model, we are deriving the aggregate demand curve. This AE3 = [C + I + G + Xn]
price level 3
can be seen in Figure 3(b), which derives an aggregate
c
demand curve from the corresponding equilibrium
points of the Keynesian cross model. In Figure 3(a),
at price level 1, equilibrium is at point a, which
corresponds to point a of the aggregate demand curve 45°
in Figure 3(b). As the price level falls, equilibrium 0 Ye3 Ye2 Ye1
moves to point b, and then to point c, resulting in the real GDP (Y)
corresponding points of the aggregate demand curve.
We thus arrive at an important conclusion: Price (b) The aggregate
level
demand curve
price level 3 c
Each point on the aggregate demand curve
corresponds to a particular price level where the price level 2 b
amount of output buyers want to buy is just that
output that generates the spending needed to buy price level 1 a
the output (recall the circular flow model).
AD

We can now return to our initial question: how 0 Ye3 Ye2 Ye1
is it that aggregate demand and real GDP, two real GDP (Y)
very different concepts, are both defined in terms
of C + I + G + (X − M)? The answer is that real GDP, Figure 3 Relating aggregate demand to aggregate expenditure

© Cambridge University Press 2012 Economics for the IB Diploma 4

 Supplementary material for Chapter 9
In part a, where Ye is less than Yp, there is an output
gap called a recessionary gap (or deflationary
gap). At Ye, there is unemployment and recession. (a) Recessionary (deflationary) gap
Actual equilibrium output is less than potential 45° line

desired spending [C + I + G + Xn]

output, and unemployment is greater than the
natural rate of unemployment. For potential output
to be achieved, it would be necessary for aggregate AEp
expenditures to increase to the level of the dotted line, recessionary gap
AEe
represented by AEp.
In part (b), Ye is greater than Yp, and here there is
an output gap called an inflationary gap. Actual
equilibrium output is greater than potential output,
and unemployment is less than the natural rate of 45°
0 Ye Yp
unemployment. For potential output to be achieved, it
would be necessary to have lower aggregate expenditures real GDP (Y)
at the level of AEp, shown by the dotted line.
However, there is nothing to guarantee that (b) Inflationary gap
aggregate expenditures will be at the level of AEp in 45° line

desired spending [C + I + G + Xn]

both parts (a) and (b). This leads to the following AEe
important conclusion (which you are already familiar inflationary gap
AEp
with from Chapter 9).

In the Keynesian cross model, equilibrium can

occur at any level of output, and there is nothing
to guarantee that the equilibrium level of output
will be the same as potential output. Recessionary
(deflationary) gaps or inflationary gaps may persist 45°
0 Yp Ye
indefinitely.
real GDP (Y)

Figure 4 Output gaps in the Keynesian cross model

The Keynesian multiplier
The Keynesian multiplier can be very conveniently into two parts: that due to an increase in autonomous
illustrated by use of the Keynesian cross model. investment spending and that due to induced
According to the Keynesian multiplier effect, consumption spending.
whenever there is an autonomous change in any of Also, note that the increase in real GDP from Ye1
the components of aggregate expenditure (C, I, G or to Ye2 in Figure 5 is the result of the full multiplier
X − M), a larger change in real GDP results. Consider effect, which can only occur when the price level is
Figure 5, showing an initial aggregate expenditure constant (see page 263 in the textbook). Since the
function AE1, with equilibrium real GDP at Ye1. Keynesian cross model presupposes a constant price
Suppose there is an increase in investment spending level, this means that an increase in autonomous
of ΔI. This will cause aggregate expenditures to shift spending and the subsequent increases in induced
upward to AE2 (= AE1 + ΔI), with the new equilibrium spending will be felt in their entirety as increased
real GDP at Ye2. As you can see in the figure, the output. This can be seen in Figure 9.18 (page 263)
increase in real GDP, shown by Y2 − Y1 is larger than ΔI. as the AD shift that takes place entirely within the
The multiplier effect has caused real GDP to horizontal part of the AS curve, which represents a
increase by more than the increase in the component constant price level.
of aggregate expenditure. The increase in real GDP, or
the difference between Y2 and Y1, consists of two parts: The multiplier is a Keynesian concept showing
one is an increase due to the increase in autonomous the power of increased spending to induce larger
investment expenditure equal to ΔI; and the rest is an increases in real GDP when the economy is in
increase due to induced consumption expenditure. recession and is not experiencing upward pressures
This idea corresponds to Figure 9.17 (page 262), where on the price level.
the total increase in aggregate demand is broken down

© Cambridge University Press 2012 Economics for the IB Diploma 5

 Supplementary material for Chapter 9
the multiplier (see page 262). Now that you know
that the MPC is the slope of the aggregate expenditure
desired spending [C + I + G + Xn] 45° line function,4 you can see why this is so. The larger the
slope of the aggregate expenditure function, the steeper
AE2 the aggregate expenditure function, and the larger is the
= ΔI
AE1 multiplier and increase in real GDP given a change in
autonomous expenditure. The smaller the slope, the flatter
the aggregate expenditure function, and the smaller the
multiplier and the increase in real GDP given a change in
autonomous expenditure. To convince yourself that this
is so, draw an aggregate expenditure function, AE1, that
45°
0 is parallel to the horizontal axis; this has a slope = MPC
Ye1 Ye2
= 0. Now assume an increase in investment spending
real GDP (Y)
equal to ΔI, and draw a new aggregate expenditure
due to due to induced function, AE2 = AE1 + ΔI; this will also be parallel to the
antenomous ΔI C spending horizontal axis. You will see that the resulting change
in real GDP will be equal to ΔI. The reason is
1
Figure 5 Illustrating the Keynesian multiplier that when the MPC = 0, the multiplier of is
1 − MPC
It is now also possible to use Figure 5 to understand equal to 1. Therefore, the change in real GDP is equal
the relationship between the size of the multiplier and to the change in investment spending, and induced
the MPC. As you know, the larger the MPC, the larger consumption spending is zero.

4Actually, the MPC is the slope of the consumption function, as we saw earlier. However, since the aggregate expenditure function is a parallel upward
shift of the consumption function, the slope does not change.

© Cambridge University Press 2012 Economics for the IB Diploma 6