CHAPTER 4

MODELS OF DEVELOPMENT PLANNING

• Aggregate planning models can be used to generate various scenarios ranging
from pessimistic to optimistic.

• Used to determine optimal accumulation paths far into the future and are
indicative of the potential growth path of the economy

• Almost all-economic planning models deal with causal forecasting or economic

growth determination, including Harrod-Domar Model.

• Planning models are useful for several reasons such as allow policymakers to
form quantitative estimates of the various trade-offs in preparing development
policies.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 1

• Planning models reflect the accounting regularities and conventions of
national income, balance of payments, income and expenditure balances of
the public sector (Taylor, 1979).

• Can be classified in several different categories like aggregate, main sector,

multi-sectoral, regional and project specific models (Chowdhury and
Kirkpatrick, 1994).

• They may be simulation models which use informal calibration procedures

or

• More traditional econometric models that are calibrated more formally,

using statistical theory, in turn based on the assumption of time-phased
structural stability.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 2

Models of development planning involves the following specific mentioned:
❖ Harrod-Domar Model of Development Planning
❖ The Fel‘dman Model of Development Planning
❖ The Mahalanobis Model of Development Planning
❖ The Leontief Input-Output Model of Development Planning
❖ The Linear Programming (Optimizing) Model of Development Planning
❖ Macro econometric Model of Development Planning

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 3

4.1 Harrod-Domar Model of development Planning

• The model simply states that economic growth can be thought of as result
of abstention from current consumption.

• For example, if a farmer wants to increase output more than he used to

produce, all he needs to do is increase the amount of seed by foregoing
current consumption.

• The act of increasing the stock of seed is investment while the total
accumulated level of seed is capital stock.

• Thus, capital stock is the machinery, equipment and the like which is used
to produce output.

• However, investment is the flow of output to increase or maintain capital

stock.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 4
 • Current output 𝑌𝑡 is the sum of what is consumed 𝐶𝑡 and saved 𝑆𝑡 :

𝑌𝑡 = 𝐶𝑡 + 𝑆𝑡
Assuming that what is saved is invested (𝐼𝑡 ), 𝐼𝑡 = 𝑆𝑡 and let a capital stock
have a path:
𝐾𝑡+1 = 𝐾𝑡 − 𝐷𝑡 + 𝐼𝑡

𝐾𝑡+1 = 𝐾𝑡 − 𝛿𝐾𝑡 + 𝐼𝑡,

where D is depreciation and 𝛿 is depreciation rate.

𝑆 𝐾𝑡
Define a saving rate as, 𝑠 = 𝑡,capital-output ratio,𝛼 = , and divide the first
𝑌𝑡 𝑌𝑡
by the second equation:
𝑠 𝑌𝑡+1 −𝑌𝑡 𝐾𝑡+1 −𝐾𝑡
= + = 𝑔 + 𝛿,
𝛼 𝑌𝑡 𝐾𝑡

where 𝑔 is growth rate of GDP(Y).

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 5
 – The Harrod-Domar equation relates growth of 𝑠/𝛼 = 𝑔 + 𝛿 for the
economy to answer the ability of the economy to save (𝑠) and the
capacity/efficiency of capital to produce (𝛼).

– For instance, to foster growth just increase s and/or increase efficiency of

capital (decrease 𝛼).

– For a given 𝛼 and domestic saving, one can see the resource gap (𝑆𝑡 − 𝐼𝑡 )
for a targeted level of 𝑔.

– This implies that the model determines the financial requirement from
abroad in the form of aid, loan, and foreign direct investment (FDI).

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 6

• The Harrod-Domar model can be extended to account for labor force
growth rate and growth rate of labor productivity(p) and population
growth.

• Assume that aggregate growth models projects macro-variables:

K (t )  cY (t )
where K(t) is capital stock and Y(t) is output at time t as well as c is the
average and marginal capital-output ratio:

I (t )  K (t  1)  K (t )  K (t )  sY  S (t )
where I(t) is gross investment at time t, s is the savings rate, S is national
savings and  is the depreciation rate

• If 𝑔 is the targeted rate of output growth, then Y (t  1)  Y (t ) Y (t )

g 
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree
Y (t ) Y (7t )
 K cK ( K / Y )Y Y sY  K s
   ⇨g   
K K K Y K c
s
n  p  
c
⇨

where n is the labor force growth rate and p is the growth rate of labor
productivity and  are wage and profit incomes, s and sW are the marginal
propensities to save from wage income and profit
We can also have: 𝑠
= 𝑔∗ + 𝑛 + 𝑛𝑔∗ + 𝛿 ≅ 𝑔∗ + 𝑛 + 𝛿
𝑔

where 𝑔* and n are growth in per capita GDP and population, respectively.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 8
4.2 The Fel’dman Model of Development Planning

• It was a Neo-Marxian model of economic development, created independently

by a Soviet economist Grigory Fel’dman in 1928 and amended by an Indian
statistician Prasanta Chandra.

• Essence of the model is a shift in the pattern of industrial investment towards

building up a domestic consumption goods sector.

• The model argued that to reach a high standard in consumption, investment on

capital goods is firstly needed.

• A high enough capital goods in the long-run expands the capacity of

producing consumer goods. It is, therefore, continuing to follow “heavy-
industry-first” policies.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 9

It recognizes the significance of a number of factors, viz:

– The pattern of income distribution (Marxian thought )

– Increase in the effective utilization of the existing capital stock

– Distinction between degrees of effective utilization of the old and the

new capital stocks.

– Consideration of the rate of growth in terms of capital capacity as well

as in terms of the absorption of a growing labor supply.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 10

The following are assumptions of Fel’dman model of development planning:
– There is no government expenditure, only household consumption and firm
investment exist,

– Production is independent of consumption, there is no lags in the growth

process.

– Capital is the only limiting factor, among others. It follows the Marxian
division of the total output of an economy (W) into capital goods for both
producer and consumer goods, and all consumer goods including raw
materials for them:
W=C+V+S
where production of total output of an economy (W) is the sum of constant
capital(C), variable wages capital (V), and surplus value(S).

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 11

• The fraction of total investment allocated to capital goods is the key
variable in the model.

• The rate of investment is purely determined by the coefficient and stock of

capital.

• The fraction of total investment allocated on capital goods:

𝐼 = 𝐼1 + 𝐼2

where I, 𝐼1 and 𝐼2 = annual rate of total net investment and each net investment
allocated to the respective categories, V = the marginal capital coefficient or
accelerator for the whole economy; as well as 𝑉1 and 𝑉2 represent for the
respective category in year(t).

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 12

C = the annual output of consumer goods; Y = the annual net output/income
of the whole economy; α = the marginal propensity to consume; ά = the
marginal propensity to save; 𝑡0 , 𝐶0 and 𝑌0 are the respective initial magnitudes
of t, C, and Y; and 𝐼1 = 𝛾𝐼 is the annual net investment allocated to capital.

Thus, since only 𝐼1 increases the capacity of producing capital goods, then it
follows that:

𝑑𝐼 𝐼 𝛾𝐼
= =
𝑑𝑡 𝑉1 𝑉1
In time t, total investment will grow at an exponential rate:

𝛾𝐼
𝐼=𝑒 𝑉1

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 13

 – In other words, total investment will grow at a constant exponential rate
𝛾𝐼
of . It could be certain proportion.
𝑉1

– Similarly, the annual rate of net investment allocated to category 2, i.e.,

consumer goods which is given by 𝐼2 = 1 − 𝛾 𝐼.

– 𝐼2 being the source of increased capacity in consumer goods:

𝑑𝐶 𝐼2 1 − 𝛾 𝑉𝛾𝑡 𝛾𝑡
= = 𝑒 1 Since, I = 𝑒 𝑉1
𝑑𝑡 𝑉2 𝑉2

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 14

 • The annual rate of output of consumer goods is given by:

1 − 𝛾 𝛾𝑡 𝛾𝑡
𝐶 = 𝐶0 + . 𝑒 𝑉1 −1
𝛾 𝑉1

• The elements which determine the national income and the growth rate of
the economy are:
Y=I+C
• By substituting the values of I and C in the above equation, it gives:
𝜸𝒕 𝜸𝒕
1−𝛾 𝑽𝟏
𝒀=𝒆 𝑽𝟏 + 𝑪𝟎 + . 𝒆
𝑽𝟏 −𝟏
𝛾 𝑽𝟐

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 15

 𝛾𝑡 𝛾𝑡
1−𝛾 𝑉1
𝑌=𝑒 𝑉1 −1 + 1 + 𝐶0 + . 𝑒 𝑉1 −1
𝛾 𝑉2
𝛾𝑡 𝛾𝑡
1−𝛾 𝑉1
𝑌= 𝑒 𝑉1 −1 + 1 + 𝐶0 + . 𝑒 𝑉1 −1
𝛾 𝑉2
𝛾𝑡
1−𝛾 𝑉1
𝑌 = 1 + 𝐶0 + +1 𝑒 𝑉1 −1
𝛾 𝑉2

Assuming that 𝐼0 = 1, the equation becomes:

𝛾𝑡
1−𝛾 𝑉1
𝑌 = 𝐼0 + 𝐶0 + +1 𝑒 𝑉1 −1
𝛾 𝑉2

𝛾𝑡
1−𝛾 𝑉1
𝑌 = 𝑌0 + +1 𝑒 𝑉1 −1 , Since, 𝑌0 = 𝐼0 + 𝐶0
𝛾 𝑉2

𝛾𝑡
1−𝛾 𝑉1
𝑌 = 𝑌0 + + 𝑒 𝑉1 −1
𝛾 𝑉2

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 16

 – The fundamental equation shows that C and Y each represent a sum of
a constant and an exponential in t.

𝛾
– Their rates of growth will differ from since the values of C and Y
𝑉1
will be greater than the value of I.
𝛾𝑡
– With the going of time, the exponential will dominate the scenario
𝑒 𝑉1
𝛾
and the rates of growth of C and Y will gradually approach .
𝑉1

– But this may take quite a long time, but the following happens:

1 − 𝛾 𝑉1
𝐶0 = ,
𝛾 𝑉2
𝛾
the C and Y will grow at the rate of from the very beginning.
𝑉1
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 17
 – If the purpose of economic development is the maximization of investment
or national income at a point of time, when 𝑉1 greatly exceeds 𝑉2 and even
then for a short period of time.

– A high γ does not imply, however, any reduction in consumption. With

capital assets assumed to be permanent, even γ = 1 would merely freeze
consumption as its original level.

– If assets were subject to wear and tear, consumption would be slowly

reduced by failure to replace them.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 18

 Critics on Fel’dman model of development planning
• It is generally valid in a long term growth model to ignore some short- run
issues.

• It is not valid to assume that long term growth does not alter the
environment of the model.

• Fel’dman model ignores the impact of technological change on the rate of

growth.

• It has no place in a long run theory, where the available techniques of

production cannot be realistically “held constant”.

• The irreversibility assumption ‘capital goods are used by both sectors but,
once investment is made, they cannot be transferred from one sector to the
other’.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 19
 4.3 The Mahalanobis Model of Development Planning

• Mahalanobis, developed a single, two and four-sectors model that fit into
development planning of the Indian economy in 1953 & 1955.

• In the two-sector model, the entire net output of the economy could be
produced in the investment and consumer goods sector.

The Mahalanobis model holds based on the following assumptions:

• A closed economy, which consists of two sectors, consumption goods (C)
and capital goods (K).
• Capital goods are non-transferable once installed in any of the sector
• There is full capacity production with constant prices,
• Investment is determined by supply of capital goods.
• Capital is the only scarce factor.
• Production of capital goods is independent of consumer goods
production.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 20

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 21
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 22
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 23
 – Substitute different values for t (t= 1, 2, 3, . . .,) the solutions to the above
equation will be as follows:

– Similarly, by putting the value of t in above equation, it gives us the

following:

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 24

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 25
 𝑡
1 + 𝝀𝑘 𝛽𝑘 𝑡 − 1
𝑌𝑡 − 𝑌0 = 𝐼0 1 + 𝝀𝑘 𝛽𝑘 − 1 + 𝝀𝑐 𝛽𝑐 𝐼0
𝝀𝑘 𝛽𝑘

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 26

 𝑡
𝜆𝐶 𝛽𝐶
𝑌𝑡 − 𝑌0 = 𝐼0 1 + 𝜆𝑘 𝛽𝑘 −1 1+
𝜆𝑘 𝛽𝑘

𝑡
𝜆𝑘 𝛽𝑘 + 𝜆𝐶 𝛽𝐶
𝑌𝑡 − 𝑌0 = 𝐼0 1 + 𝜆𝑘 𝛽𝑘 −1 (𝑖𝑣)
𝜆𝑘 𝛽𝑘

Suppose 𝐼0 =𝛼0 𝑌0 and substituting it into (iv)

𝑡
𝜆𝑘 𝛽𝑘 + 𝜆𝐶 𝛽𝐶
𝑌𝑡 − 𝑌0 = 𝛼0 𝑌0 1 + 𝜆𝑘 𝛽𝑘 −1
𝜆𝑘 𝛽𝑘

𝑡
𝜆𝑘 𝛽𝑘 + 𝜆𝐶 𝛽𝐶
𝑌𝑡 = 𝛼0 𝑌0 1 + 𝜆𝑘 𝛽𝑘 −1 + 𝑌0
𝜆𝑘 𝛽𝑘

𝜆𝑘 𝛽𝑘 + 𝜆𝐶 𝛽𝐶
𝑌𝑡 = 𝑌0 1 + 𝛼0 ( 1 + 𝜆𝑘 𝛽𝑘 𝑡 −1) (𝑣)
𝜆𝑘 𝛽𝑘
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 27
where 𝛼0 is the rate of investment in the base year, 𝑌0 and 𝑌𝑡 are the gross
𝜆 𝛽 +𝜆 𝛽
national income on the base and at year t, respectively and 𝑘 𝑘 𝐶 𝐶 ratio is
𝜆𝑘 𝛽𝑘
the overall capital coefficient.

• Assume that 𝛽𝑘 and 𝛽𝑐 are given, then income growth will depend on the
rate of investment in the base year (𝛼0 ) and the policy instrument(𝜆𝑘 ).

• Assuming further that 𝛼0 to be constant, income growth will depend on 𝜆𝑘 .

• If 𝛽𝑐 > 𝛽𝑘 , implies that larger percentage investment on consumer goods

industries, the larger will be the income generated.

• The expression 1 + 𝜆𝑘 𝛽𝑘 𝑡 , shows that after a range of time, larger

investment in capital goods industries, will gener at larger income.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 28

 𝑡
• Initially a high value of 𝜆𝑘 increases the magnitude 1 + 𝜆𝑘 𝛽𝑘 and lower
𝜆 𝛽 +𝜆 𝛽
the overall capital coefficient 𝑘 𝑘 𝐶 𝐶.
𝜆𝑘 𝛽𝑘

• As time goes a higher value of 𝜆𝑘 would lead to higher income growth in

the long run.

• On the other hand, if 𝛽𝑐 = 𝛽𝑘 , then the reciprocal of the overall capital

𝜆𝑘 𝛽𝑘
coefficient, that is, = 𝜆𝑘 equals marginal rate of saving.
𝜆𝑘 𝛽𝑘 + 𝜆𝐶 𝛽𝑐

• The important policy implication of the model is that for a higher rate of
investment (𝜆𝑘 ), marginal rate of saving could also be higher.

• A higher rate of investment on capital goods in the short run would avail a
smaller volume of output for consumption while in the long run, it would
lead to a higher consumption growth rate.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 29
4.4 Leontief Input-Output Model of Development Planning

• The inter-industry or input-output approach pioneered by Leontief and first

implemented in the Soviet Union.

• It served a means by which consistent inter-sectoral plans could be drawn up.

• Input-output models were the first effectively to separate real from nominal
resources flows.

• In the standard input-output model, there is no substitution of factors in the

production functions and final demand is exogenously determined.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 30

• A dynamic version of the input-output model accounted for the accumulation
of capital stock.

• But it is computationally clumsy and its linearity led to either a balanced

growth turnpike or explosive diversion there from.

• The model is used to analyze inter-industry relationship to understand the

interdependencies and complexities of the economy, is also known as “inter-
industry analysis.”

• It shows conditions for maintaining equilibrium between supply & demand.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 31

Before dealing with mathematical concepts it is important to note the basic
assumptions of the input-output analyses:

• No substitution takes place between the inputs to produce a given unit of

output and the input coefficients are constant.

• The linear input functions imply that the marginal input coefficients are
equal to the average.

• Joint products are ruled out, i.e. each industry produces only one
commodity and each commodity is produced by only one industry.

• External economies are ruled out and production is subject to the

operation of constant returns to scale.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 32

 If the total output of say 𝑋𝑖 of the 𝑖 𝑡ℎ industry be divided into various number
of industries, 1, 2, 3,…n, then it gives the balance equation:
𝑋𝑖 = 𝑥𝑖1 + 𝑥𝑖2 + ⋯ + 𝑥𝑖𝑛 + 𝐷𝑖 (1)

If an amount 𝑌𝑖 absorbed by the outside sector is also considered, then the

balance equation of the industry becomes:
𝑋𝑖 = 𝑥𝑖1 + 𝑥𝑖2 + ⋯ + 𝑥𝑖𝑛 + 𝐷𝑖 + 𝑌𝑖
This implies,
𝑛
𝑖=1 𝑥𝑖𝑗 +𝑌𝑖 = 𝑋𝑖 (2)

where 𝑌𝑖 is the sum of product outflows from the 𝑖 𝑡ℎ industry, to consumption,

investment and exports, net of imports.

Equation (2) shows the conditions of equilibrium between demand and

supply (𝑋𝑖 ) which illustrates the flows of outputs and inputs to and from one
industry to other industries and vice versa.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 33

• In the analysis of input-output, the system of equations (1) and (2) presents
the conditions of internal consistency of the plan.

• If these equations are not satisfied, there might be excess of some goods
and deficiency of others.

• As 𝑥𝑖2 represents the amount absorbed by industry 2 of the 𝑖 𝑡ℎ industry it

then follows that 𝑥𝑖𝑖 stands for the amount absorbed by the 𝑗𝑡ℎ industry of
𝑖 𝑡ℎ industry.

• Thus, the technical coefficient or input coefficient of the ith industry is

denoted by:
𝑎𝑖𝑗 = 𝑥𝑖𝑗 /𝑋𝑗 (3)
where 𝑥𝑖𝑗 is the flow from industry i to industry j, 𝑋𝑗 is the total output of
industry j and 𝑎𝑖𝑗 is constant which is called technical coefficient in the 𝑖 𝑡ℎ
industry.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 34
• It shows the number of units of one industry’s output that are required to
produce one unit of another industry’s output.

• Cross-multiplying the terms in equation (3) gives:

𝑎𝑖𝑗 . 𝑋𝑗 = 𝑥𝑖𝑗

• By substituting the value of 𝑥𝑖𝑗 in equation (2) and transposing the terms
gives the basic input-output system of equations in the form:
𝑛

𝑋𝑖 − 𝑎𝑖𝑗 𝑥𝑗 = 𝑌𝑖
𝑗=1
where n represents the number of sectors in the economy.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 35

 If n = 2, the two-sector economy has two linear equations that could be stated
symbolically in the form:

𝑥1 − 𝑎11 𝑥1 − 𝑎12 𝑥2 = 𝑌1

𝑥2 − 𝑎21 𝑥1 − 𝑎22 𝑥2 = 𝑌2

which can be represented in matrix notation as:

X − [A]X = Y ⇨ [I − A]X = Y

where matrix (I − A) is known as the Leontief Matrix.

• The above presentation, however, is an open static model of input-output

analysis.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 36
• The analysis can be extended to include many other sectors, like health,
education, communication, transportation, manufacturing, banking, foreign
trade and balance of payments, and so on, in the economy.

• In reality, most economic variables are dynamic that causes and effect, and
action and reaction do not occur immediately after one and other.

• Rather, it takes some time for certain economic activities to happen as a

result of some causes or actions.

• Thus, the analysis becomes dynamic when it is closed by the linking of

investment part of the final goods to output.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 37

• In the Leontief dynamic input-output model, the output of a given period is
supposed to go into stocks or capital goods, which in turn are distributed
among industries.

• The dynamic balance equation is of the form of:

𝑋1 𝑡 = 𝑥𝑖1 𝑡 + 𝑥𝑖2 𝑡 + ⋯ + 𝑥𝑖𝑛 𝑡 + 𝑆𝑖1 𝑡 + 𝑆𝑖2 𝑡 + ⋯ + 𝑆𝑖𝑛 𝑡

+𝐷𝑖 𝑡 + 𝑌𝑖 𝑡
• Multipliers from the external demand are simply the sum of direct, indirect
and induced effects that determine by how much the economy will change
because of a change in final demands.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 38

Direct effects: These may be thought of as the revenues, jobs, and wages that
a new business or expanding business brings into the local economy.

Indirect effects: Any business expansion or new entry into a market will lead
that business to make purchases from and/or sales to local firms.

– Because of new demand, local firms are likely to create some new jobs,
increase wages and revenues. All this, in turn, will have an additional
impact on the overall economy.

Induced effects: While direct and indirect effects measure the impacts of
business-to-business interactions, induced effects are specific to the
behavior of the labor force.

– What that means is, employees of the new business and the related
businesses will spend their earnings in the local economy to purchase
items such as food, transportation, housing, medical, etc..
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 39
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 40
Note: matrix transpose equals cofactor transpose

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 41

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 42
Note : if the matrix coefficients is a three by three, we can apply the following procedures.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 43

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 44
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 45
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 46
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 47
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 48
 as follows.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 49

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 50
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 51
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 52
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 53
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 54
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 55
Limitations of the Model
• National benchmark input-output models are constructed once every five
years and the data may not be available to the public for another three
years.

• Changes in the economic structure that occurred between the end of data
collection and end of calendar year are not recorded.

• Input-output models are also static. That is to say, an input-output model is

a snapshot of the economy at one point in time.

• Specific limitations to the input-output model’s accuracy include constant

coefficients, linearity, sector homogeneity and no capacity constraints

• Constant coefficients imply that advances in production technology, new

inventions, import substitution, changes in consumer patterns of demand,
and the change of relative prices do not alter.

• Despite these limitations, the input-output model is perhaps one of the

more powerful descriptive tools available to the regional analyst.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 56
4.5 The Linear Programming Model of Development Planning

• The model is essentially an extension and generalization of input-output

analysis that can be expressed as a Linear Programming (LP)form.

• Thus, the LP model differs in several important respects from the standard
input-output model on which it is based.

• In particular, production, imports and exports are variables whose level in

each sector are to be determined by an optimizing solution
.
• In addition, alternative activities and resource limitations are explicitly taken
into account.

• LP requires that all the mathematical functions in the model be linear

functions.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 57
• There is well known that an input-output model provides, for each
commodity- sector, which indicate that the total supply of and total demand
for commodities are equal.

• Linear programming (LP) is really a mathematical tool which is now being

increasingly used in economic analysis of development planning.

• Because it helps the planner to allocate resources optimally among

alternative uses within the specific constraints.

• At the micro level, the technique could be used to find out optimal and
efficient methods of production.

• Actually, LP can be regarded as a powerful and complementary tool which

can be used to analyze and solve the problems of choice of techniques on the
supply side and final demand.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 58
• It is important to emphasize that the LP technique helps to tackle the major
problems of investment planning, namely consistency between sectors; feasi-
bility of plans; and optimality in resource allocation.
Conditions and Generalizations to Linear Programming

The application of linear programming (LP) techniques to any problem rests on

certain conditions and generalizations. These are:

• There is a definite objective. It may be the maximization /the minimization of

costs.

– Every maximization problem has its dual problem, that of minimization.

If the primal pertains to maximization, the dual involves minimization
and vice versa.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 59
• There should be alternative production process for achieving the
objective. The concept of process or activity is the most important
in linear programming.

• There must be certain constraints or restraints of the problem. They are

restrictions pertaining to certain conditions of the problem, are also called
inequalities.

• There are choice variables, the various production processes or activities

so as to maximize or minimize the objective function and to satisfy all the
restraints.

• There are the feasible and optimal solutions. For instance, given the
income of the consumers and the prices of goods, feasible solutions are all
possible combinations of the goods that a consumer can feasibly buy.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 60

 Before dealing with the general formulation of Lp models, let's look at
the underlying assumptions:
– The decision making body is faced with certain constraints or
resource restrictions

– A limited number of alternative production processes avail

– Linear relations among the different variables which implies

constant proportionality between inputs and outputs with in a
process

– Input - output prices and coefficients are given and constant as

well as there are continuity and divisibility in products and factors

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 61

General formulation of Linear Programming Problems

• In an linear programming exercise, usually the objective is to maximize

or minimize some linear function of the variables given, given resource
constraints.

• The programmer seeks to obtain non-negative values of these variables

subject to the constraints and maximize/minimize the objective function.
More formally
𝑚𝑎𝑥 𝑉 = 𝑐1𝑥1+ . . . 𝑐𝑟𝑥𝑟

ai1 x1  ... air xr ,  ,  bi i  1, ..., m

xj  0 j  1, . . ., r
where aij, bi and Cj are given constants with m inequalities or equalities in r
variables, i.e.

27-Oct-24 . Prepared by Urgaia Rissa (Ph.D.) BA Degree 62

• The components of linear programming are decision variables, constraints,
data and objective functions

Characteristics of Linear Programming

The following are the five characteristics of the linear programming problem:
• Constraints: The limitations should be expressed in the mathematical
form, regarding the resource.

• Objective Function: In a problem, the objective function should be

specified in a quantitative way.

• Linearity: The relationship between two or more variables in the

function must be linear. It means that the degree of the variable is one.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 63

 • Finiteness: There should be finite and infinite input and output
numbers. In case, if the function has infinite factors, the optimal
solution is not feasible.

• Non-negativity: The variable value should be positive or zero. It

should not be a negative value.

• Decision Variables: The decision variable will decide the output. It

gives the ultimate solution of the problem. For any problem, the first
step is to identify the decision variables.

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 64

 Find the maximum value of the objective function given a simple case of two
variables given below:

𝑚𝑎𝑥 𝑉 = 5𝑥1 + 3𝑥2

S.t: 3 𝑥1 + 5𝑥2 ≤ 15
5 𝑥1 + 2𝑥2 ≤ 10
𝑥1 , 𝑥2 ≥ 0
Solution:
2 3 𝑥1 + 5𝑥2 = 15
−5 5 𝑥1 + 2𝑥2 = 10
𝑥1 = 20/19=1.053 and 𝑥2 = 2.368
The maximum value of the objective function will be :
𝑚𝑎𝑥 𝑉 = 5𝑥1 + 3𝑥2 =5(1.053) +3(2.368) = 12.37

❖ GAMS and other software programming can solve complex LP problems

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 65

 Calculate the maximal and minimal value of z = 5x + 3y for the following
constraints:
x + 2y ≤ 14
3x – y ≥ 0
x–y≤2
Solution:
The three inequalities indicate the constraints.

The optimization equation (z) = 5x + 3y. You have to find the (x,y) corner points
that give the largest and smallest values of z.
To begin with, first solve each inequality.
x + 2y ≤ 14 ⇒ y ≤ -(1/2)x + 7
3x – y ≥ 0 ⇒ y ≤ 3x
x–y≤2⇒y≥x–2
Here is the graph for the above equations.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 66
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 67
now pair the lines to form a system of linear equations to find the corner
points.
y = -(½) x + 7
y = 3x

Solving the above equations, we get the corner points as (2, 6)

y = -1/2 x + 7
y=x–2

Solving the above equations, we get the corner points as (6, 4)

y = 3x
y=x–2
Solving the above equations, we get the corner points as (-1, -3)

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 68

 For linear systems, the maximum and minimum values of the optimization
equation lie on the corners of the feasibility region.

Therefore, to find the optimum solution, you only need to plug these three
points in z = 3x + 4y

(2, 6) : z = 5(2) + 3(6) = 10 + 18 = 28

(6, 4): z = 5(6) + 3(4) = 30 + 12 = 42

(–1, –3): z = 5(-1) + 3(-3) = -5 -9 = -14

Hence, the maximum of z = 42 lies at (6, 4) and the minimum of z = -14 lies
at (-1, -3)

27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 69

4.6 Macro econometric Model of Development Planning
• The planning exercise and plan formulation of developing countries has
found basis in macro-econometric models recently.

• Application of such models takes the form of a simple Keynesian

framework of analysis:
Ct = α0 + α1Yt + α2rt + μt
It = b0 + b1Yt + b2rt + νt

Yt = Ct + It
Mt = a0 + a1Yt + a2rt + zt
wherein, Ct = consumer expenditure, It = capital formation, Yt = national
income, rt = interest rate, Mt = the exogenously supplied money, t = time
and μt, νt and zt error terms..
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 70
– Such a model is not dynamic, does not determine prices and ignores foreign
trade and also, a change in government taxes and spending in terms of
public policy is not assigned any role.
– However, Klein (1965) set out a more sophisticated version of the model
which is presented as follow:

Ct = α0 + α1 𝑌𝑡−𝑇𝑡 + α2Ct-1 + μ1t

𝑃𝑡

It = b0 + b1𝑌𝑡−1 + b2Kt-1 - b3rt-1 + μ2t

𝑃𝑡−1

Ft = c0 + c1 + c2 Ft-1 + c3 + μ3t

Et = d0 + d1Twt + d2 + μ4t

= Ct + It - Ft + Et + Gt
Tt = e0 + e1Yt + μ5t
It = Kt - Kt-1
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 71
 𝑌𝑡
= 𝑔0 +𝑔1 𝐿𝑡 + 𝑔2 𝐾𝑡 + 𝜇6𝑡
𝑃𝑡

𝑤 𝐿 𝑃𝑓𝑡
𝑃𝑡 = ℎ0 + ℎ1 𝑡 𝑡 + ℎ2 + 𝜇7𝑡
𝑌𝑡 𝑃𝑡 𝑃𝑡

𝑊𝑡 −𝑊𝑡−1 𝑁𝑡 −𝐿𝑡 𝑃𝑡 −𝑃𝑡−1

= 𝑗0 + 𝑗1 + 𝑗2 + 𝜇8𝑡
𝑊𝑡−1 𝑁𝑡 𝑃𝑡−1

𝑁𝑡 = 𝑘0 + 𝑘1 (𝑁𝑡 − 𝐿𝑡 ) + 𝑘2 𝑤𝑡 / 𝑃𝑡 + 𝜇9𝑡

𝑀𝑡 𝑌𝑡
= 𝑙0 + 𝑙1 + 𝑙2 𝑟𝑡 + 𝜇10𝑡
𝑃𝑡 𝑃𝑡

𝑃𝑒 = 𝑚0 + 𝑚1 P + 𝜇11𝑡
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 72
• In the above model; C=the real consumer expenditures, Y=national income
in current prices, T=taxes less transfer payments, p=general price index, I=
net real investment, K=real capital stock, r =interest rate, F=real imports,
E=real export,L=employment, w = wage rate, and N=labour supply are the
endogenous variables.

• The exogenous variables are p, import prices, is volume of world trade, G

is real government expenditures, and M is money supply.

• Practically, however, the planner will have to determine different types of

the mentioned variables to render such a model applicable to the special
problems of LDCs.

• The actual econometric techniques to be used depend upon initial

specifications of the equations.

• Subjective judgment of the planner in the light of the actual state of

information also determine the econometric model.
27-Oct-24 Prepared by Urgaia Rissa (Ph.D.) BA Degree 73