1
Unit-II: Cyclical Fluctuations in the Economy
Accelerator and Multiplier Principles and their Interactions: Business Cycle (BC) Models
These BC models have used the accelerator and multiplier (AM) principles, and the
interaction thereof, to explain the fluctuations found in business activities. The accelerator
expresses the tendency for investment to be induced by changes in the level of income or
production. In its simplest form it is the ratio of the amount of investment so induced to the
change of national income which induced it.
From a sophisticated viewpoint the accelerator may be regarded as a more complicated
description of the manner in which investment or disinvestment is influenced over time by
changes in the level of national income or of its components (such as consumption). The
underlying reasoning is that the change in national income leads to a proportional change in
investment.
On the other hand, the multiplier expresses the relationship between income and demand
(consumption), i.e, when the propensity to consume change with a change in the income
level, the demand will also change. With a change in the level of demand, investment
(autonomous) will adjust accordingly (could be in either directions) affecting employment.
This will change the level of income in the next period which is a multiple of the initial
change in income.
In his theory of business cycles, Keynes, with the help of the multiplier theory has shown the
effect of increase and decrease in investment on output and employment get magnified
when multiplier is working during either the upswing or downswing of a business cycle.
However, he did not explain the cyclical and cumulative nature of the fluctuations in
economic activity. This is because he did not lay any emphasis on utilising the accelerator in
the explanation of business cycles.
Later economists tried to utilise both the multiplier and the interaction principles to explain
economic phenomena including business cycles. Samuelson (Hansen’s student) was among
the pioneers, though the idea was already developed by others, including Hansen. He
showed that the interaction between the multiplier and accelerator caused cyclical
fluctuations in economic activities.
In 1939 Samuelson pointed out that the combination of the accelerator and the propensity
could, provided lags between income and consumption and investment were taken into
account, lead not only to steady growth but also to cycles or to a gradual approach to an
equilibrium value, depending upon the magnitude of the coefficients of acceleration and
consumption involved. Tinbergen also developed a model that was formally similar,
explaining the movement of non- wage income. Models similar to Samuelson's in
mathematical form but differing in substance had previously been developed by Frisch and
Kalecki, whose models do include the accelerator but not the propensity to consume.
Harrod's and Samuelson's models are distinguished from those of Tinbergen, Kalecki and
Frisch in that they contain both propensity to consume and accelerator, while the other
models have one or the other element, but not both. As compared with Harrod's approach,
2
Samuelson's model, through the introduction of lags, indicates that cyclical behavior can be
generated even if propensity and accelerator remain constant.
Samuelson’s Multiplier Accelerator Interaction model
In his model, Samuelson has described the way the multiplier and accelerator interact with
each other for generating income and increasing consumption and demand of investment.
He also describes how these two factors are responsible for creating economic fluctuations.
Samuelson used two concepts, namely, autonomous and induced investment, to explain his
model. Autonomous investment refers to the investment due to exogenous factors, such as
new product, production technique, and market.
On the other hand, induced investment refers to the increase in the investment of capital
goods produced due to increase in the demand of consumer goods. When autonomous
investment occurs in an economy, the income level also increases.
This brought the role of multiplier into account. The income level helps in determining the
marginal propensity to consume. If the income level increases, then the demand for
consumer goods also increases.
The supply of consumer goods should satisfy the demand for consumer goods. This is
possible when the production technique is capable to produce a large quantity of products
and services. This encourages organizations to invest more to develop advanced production
techniques and increase production for meeting consumer demand.
Therefore, the consumption affects the demand of investment. This is referred as derived or
induced investment. This marks the starting of the acceleration process, which results in
further increase in income level.
An increase in the income level would increase the demand of consumer goods. In this
manner, the multiplier and accelerator interact with each other and make the income grow
at a much higher rate than expected.
Autonomous investment leads to multiplier effect that result in derived investment. This is
called acceleration of investment. Derived investment would make the accelerator to come
into action. This is termed as multiplier-acceleration interaction.
Samuelson made certain assumptions for the explanation of business cycles.
● The production capacity is limited and consumption spending in the current period is
related to the change in income in the preceding period. The lag between
consumption and production is one year.
● Investment in the current period is related to the change in income with a lag of one
period.
● There is also an additional factor in the form of exogenous (government) spending,
which is constant.
● The model does not consider the determinants of this exogenous spending nor the
effects of it’s changes on the level of income.
3
● There is no foreign trade in the economy. Government resort to government deficit
financing and government expenditure remained constant throughout, that is
spends a fixed amount per unit period (Gt = Ḡ)
With no foreign trade, the equilibrium would be achieved when
Yt = Ct + It + Ḡ ……(1)
Where, Yt = National income
Ct = Total consumption expenditure
It = Investment expenditure
Gt= Government expenditure
t = Time period
According to the assumption that consumption takes place after a gap of one year, the
consumption function would be represented as follows:
Ct = a Yt-1 …..(2)
Where, Yt-1 = Income for t-1 time period
a = ∆C/∆Y (multiplier propensity to consume and 0<a<1)
Investment and consumption has a time lag of one year; therefore, the investment function
can be expressed a follows:
It = b (Ct –Ct-1) ……. (3)
Where, b = capital/output ratio (helps in determination of acceleration)
By putting the value of Ct and It in the first equation of national income, we get
c = a Yt-1 + b (Ct –C
t-1) + Ḡ …… (4)
If Ct = a Yt-1, then Ct-1= a Yt-2. Putting the value of Ct-1 in equation (4), we get
Yt = a Yt-1 + b (a Yt-1 -a Yt-2) + Ḡ
Yt = a (1 + b) Yt-1 – a b Yt-2 + Ḡ …… (5)
With the help of preceding equation, the income level for past and future can be
determined if the values of a, b and income of two preceding years are given. It can be
depicted from the preceding equation that the changes in income level can be affected by
the values of a and b.
Equation (5) is a type of difference equations with constant coefficient and being of the
second order. The mathematical explanation need not be given here but the solutions will
depend on the roots which in turn depend upon the coefficients- a and b.
4
The unique fixed point is at
Ẏ= [1/(1-a)] Ḡ…..(6)
A numerical example:
We will try to analyse the movement of income innine periods, with four cases having
different values for the two coefficients. (a and b)and assuming Ḡ=1.
Time Gt Yt
Perio
Case I : Case II : Case III : Case IV :
d(t)
a=0.5, b=0 a=0.5, b=2 a=0.6, b=2 a=0.8, b=4
1 1 1 1 1 1
2 1 1.5 2.5 2.8 5
3 1 1.75 3.75 4.84 17.8
4 1 1.875 4.125 6.352 56.2
5 1 1.9375 3.4375 6.6256 169.84
6 1 1.9688 2.0313 5.3037 500.52
7 1 1.9844 0.9141 2.5959 1459.592
8 1 1.9922 -0.1172 -0.6918 4227.704
9 1 1.9961 0.2148 -3.3603 12241.122
The resulting Yt for
each case is depicted below, where on the X-axis we measure the time
period (t) and on the Y-axis, the income.
5
Case I: The income increase at a decreasing rate the change in income would be in one
direction only. This however is not realistic.
Fig 1:
Case II: The cycle has a more or less constant amplitude or fluctuation. Hence it can be
categorised as an un-damped cycle.
Fig 2:
Case III: The amplitude of the cycles becomes larger with time. Therefore we have explosive
oscillation or anti-damped cycle. The reverse is the damped or the dwindling fluctuation.
Fig 3:
6
Case IV: It represents one way explosion which increases exponentially. It can go in both
directions, that is, if it is increasing it will explode in the upward direction and if it is
decreasing it will explode in a downward direction. It is not a cycle and more importantly
not realistic.
Fig 4:
The exercise shows that the income is highly sensitive to the values of a and b, as we have
seen.
7
It may be shown that the possible values of a and b can be divided into four regions, with
each region giving different types of fluctuation. The different combinations of a and b are
plotted in a graph and corresponding to each point there is a model sequence of national
income through time (as in the numerical example depicted above). The type of fluctuation
in income for that particular period will depend on which region of the graph the point is
(point of combination of a and b).
We take four combinations of a and b (α and β in the chart) and shown them on a graph.
Fig 5:
8
There are four regions designated as A, B, C and D.
Region A (the value of b is very small): The infusion of government expenditure or
investment expenditure into the system will propel the national income in the upward
direction but at a decreasing rate till it finally reaches a new equilibrium.
Region B: The cycle will initially start with a large oscillation which gradually decreases.
However, if there is a constant and regular governmental spending or new investment
expenditure, the cycle do not die down but will be periodic, that is the oscillations will be
regular in nature.( as in fig2)
Region C: The income move upward or downward at an increasing rate, resulting in an
explosive cycle. It may be noted that in this region the mpc is higher compared to that in
region B, which makes the size of the multiplier very large.
Region D (large values of a and b): The constant government spending or an increase in
investment will result in an ever increasing income, which will eventually grow at a
compounding (exponential) rate. The reverse also holds- a small unit of disinvestment can
cause a sharp spiral in the downward movement. Thus, this region is highly unstable and
highly unlikely to take place.