Empirical Production functions:

Statistical Analysis
The Cobb – Douglas
&
Constant Elasticity of Substitution Production
function (CES Production Function)
 Cobb-Douglas Production Function
• Economists have examined several production functions and have used statistical
analysis to measure the relation between changes in physical inputs and outputs.
• One such statistical production function is this concept.
• It was proposed by Knut Wicksell (1851 - 1926), and tested against statistical
evidence by Charles Cobb and Paul Douglas in 1928
• In 1928 Charles Cobb and Paul Douglas published a study in which they modeled
the growth of the American economy during the period 1899 - 1922.
• They considered a simplified view of the economy in which production output is
determined by the amount of labor involved and the amount of capital invested.
– While there are many other factors affecting economic performance, their
model proved to be remarkably accurate.
• The Cobb-Douglas formula says that “Labour contributes about 75 per cent
increase in manufacturing production, while capital contributes only 25 per cent.”
 The formula is as follows: -
» Q = A L b K 1 –b

Where Q= total output

• L = Index of Employment of Labour in Manufacturing / Labour units,

• K = Index of employment of capital in manufacturing / Capital units,

• b and 1-b = exponents of elasticities of production / a parameters

• i.e., a and 1-a measure percentage response of output to percentage

change in labour and capital respectively.
Properties of Cobb – Douglas Production Function
1. Both L & K should be positive for Q to exist.
2. If we look the parameters we find that their sum [b + (1-b)] equal 1. This means
that the function in the original form assumes constant returns to scale.
– In the later version of the cobb-Douglas function, this condition was relaxed
and the functional form was rewritten as Q = A LαKβ
– where (α+β) could be greater than, equal to or less than 1.
• when (α+β) = 1 returns to scale are constant,
• when (α+β) > 1 returns to scale are increasing, and
• when (α+β) < 1 returns to scale are decreasing.
3. Another important feature of the function is that its parameters represent factor-
shares in output. In equation,
α = wage share / total income ; and β = rental share / total income.
4. We can also find the short-run relationship of inputs and output (e.g., marginal
product of Labour and marginal product of capital) with the help of this function.
In the two input case of equation,

– marginal product of Labour (MPL) = α (Q/L); and

– marginal product of Capital (MPK) = β (Q/K)

5. In its original form, this function has the elasticity of substitution as unity (Proof
given in the following section). This property has important policy implications for
formulation of an income policy.
Constant Elasticity of Substitution Production
function (CES Production Function)

• We know that under Cobb-Douglas function, Elasticity of substitution is assumed

equal to unity, but in “CES Production function” the elasticity of substitution is
constant and not necessarily equal to unity.
• The general for of CES function is, X = ϒ [ K C –α + (1-K) L –α] –ν/α

– X = output, C = capital input, L = labour input ----- 3 variables

– ϒ = constant, K= capital intensity factor coefficient and (1-K) = labour intensity coefficient,, and ν
= degree of returns to scale ---- 4 parameters
• the elasticity of substitution is derived with the help of the % α (known as the
substitution parameter).
• If  denotes elasticity of substitution, then α = { 1 – 1/  }
 Properties
• The value of  depends upon the value of α.
α = { 1 – 1/  }
• The MPL, C are always +ve if we assume constant
returns to scale (ν =1)
• The marginal product curves slope downwards, i.e.,
∂2X/ ∂L2 < 0 and ∂2X/ ∂C2 <0
• The marginal product of an input will increase when
other factor inputs increase.