Theory of Production - Microeconomics

1. Production Function

A production function gives the maximum technically feasible output obtainable from specified quantities of inputs, given the state of technology.

Q = f(L, K; A)

Q = output; L = labour; K = capital; A = technology or efficiency. Other inputs, such as land, materials and energy, may also be included.

This is a technical relationship. Input prices do not directly determine its shape; prices matter when the producer chooses an economical input combination.

Technical efficiency

Producing the maximum output from given inputs, or avoiding unnecessary inputs for a given output.

A point below the production frontier represents technical inefficiency.

Economic efficiency

Choosing the least-cost technically feasible input combination for a required output, given input prices.

A technically efficient method need not be the cheapest method.

Three important functional forms

Function	Interpretation	Input substitution
Q = A L^αK^β	Cobb–Douglas; α and β are output elasticities of the respective inputs.	Smooth substitution; elasticity of substitution = 1, for positive exponents.
Q = aL + bK	Linear function; inputs are perfect substitutes.	Straight isoquants; elasticity of substitution is infinite.
Q = min(L/a, K/b)	Fixed proportions; one unit of output requires a units of labour and b units of capital.	L-shaped isoquants; elasticity of substitution = 0.

Worked example: Cobb–Douglas production

If Q = 2L^0.5K^0.5, L = 9 and K = 16, then Q = 2 × 3 × 4 = 24 units.

If both inputs double, output becomes 48 units. This function has constant returns to scale because 0.5 + 0.5 = 1.

2. Short-Run and Long-Run Production

The distinction concerns input adjustability, not a fixed number of days, months or years.

Aspect	Short run	Long run
Inputs	At least one input is fixed.	All inputs can be varied.
Typical function	Q = f(L, K̄)	Q = f(L, K)
Central question	What happens when labour rises with the plant fixed?	What happens when the scale of all inputs changes?
Main law	Law of variable proportions.	Returns to scale.

Example: A bakery adds workers while its ovens remain fixed: a short-run adjustment. It builds a larger bakery and changes both ovens and workers: a long-run adjustment.

Exam distinction: The long run allows all inputs to change, but does not require technology to change. Returns to scale are normally evaluated with technology held constant.

3. Total, Average and Marginal Product

Total product (TP): total output produced by the variable input, holding other inputs fixed.
Average product (AP): output per unit of the variable input.
Marginal product (MP): additional output from an additional unit of the variable input.

AP_L = Q/L
MP_L = ΔQ/ΔL; for a differentiable function, MP_L = ∂Q/∂L

Read the production schedule

Workers (L)	TP	AP = TP/L	MP = change in TP
0	0	Undefined	—
1	10	10	10
2	24	12	14
3	42	14	18
4	56	14	14
5	65	13	9
6	66	11	1
7	63	9	−3

Interactive production laboratory

Select a labour level to read its products and see the output point on the TP curve.

Workers: 3

The AP–MP relationship

If MP > AP, AP rises.
If MP < AP, AP falls.
In the usual smooth case, MP = AP at the maximum of AP.
When MP is positive, TP rises—even if MP itself is falling.
In the usual smooth case, TP reaches its maximum where MP becomes zero and then turns negative.

d(AP)/dL = (MP − AP)/L

Memory aid: Think of MP as the newest student’s mark and AP as the class average. A new mark above the average raises it; a mark below the average lowers it.

A discrete table may show a plateau or skip the exact MP = 0 point. Exact boundary equalities refer to continuous curves.

4. Law of Variable Proportions

As successive units of a variable input are combined with fixed inputs, its marginal product eventually declines, assuming unchanged technology and comparable units of the variable input.

Early increases in MP can result from specialisation and better use of fixed equipment. Later, congestion and the limited fixed input cause diminishing marginal returns.

Stage	Boundary and behaviour	Meaning
I	Ends at maximum AP, where MP = AP. AP rises; MP may initially rise and then fall.	The fixed input is relatively underutilised.
II	From maximum AP to maximum TP. MP is positive but below AP; it reaches zero at the endpoint.	The economically relevant operating region under the usual assumptions.
III	MP is negative; TP falls as more variable input is used.	Extra units reduce output and add cost.

Frequent trap: Diminishing marginal returns begin when MP starts falling. This can happen within Stage I, before AP reaches its maximum. They do not mean that TP is falling.

Why is Stage II the relevant region?

In the conventional production curve, Stage I permits improved average utilisation of the variable input. In Stage III, removing variable input raises output while saving its cost. The optimal positive-output choice is therefore normally in Stage II, with its exact position determined by output and input prices.

Under a competitive output market and a competitive labour market, an interior labour choice satisfies P × MP_L = w. With market power in the output market, use MR × MP_L = w. A firm can also choose to shut down.

5. Returns to Scale

Returns to scale describe the change in output when all inputs increase in the same proportion, with technology unchanged.

Compare f(tL, tK) with t f(L, K), for t > 1.

Type	Condition	If all inputs double
Increasing returns	f(tL,tK) > t f(L,K)	Output more than doubles.
Constant returns	f(tL,tK) = t f(L,K)	Output exactly doubles.
Decreasing returns	f(tL,tK) < t f(L,K)	Output less than doubles.

For Q = A L^αK^β, proportional input scaling multiplies output by t^α+β.

Experiment with Cobb–Douglas returns

Labour exponent α Capital exponent β Input multiplier t

Two separate questions: A Cobb–Douglas function can have diminishing marginal products for each input and increasing returns to scale. For example, α = β = 0.6: each exponent is below 1, but their sum is 1.2.

Returns to scale versus economies of scale: Returns to scale are technical output relationships. Economies of scale concern falling long-run average cost. Their connection depends on cost minimisation and input-price conditions; the terms should not be treated as identical definitions.

6. Isoquants and Isocosts

An isoquant shows combinations of inputs that produce the same output efficiently. An isocost line shows combinations costing the same total amount.

Isoquant

f(L,K) = Q̄

A higher isoquant represents greater output when inputs are productive.
Efficient isoquants do not intersect.
With positive marginal products, isoquants slope downward.
Standard smooth isoquants are convex when MRTS diminishes.

Straight and L-shaped isoquants are important special cases. Strict convexity is not a universal property of every production technology.

Isocost

C = wL + rK
K = C/r − (w/r)L

w = wage per unit of labour; r = rental price per unit of capital.

Labour intercept = C/w.
Capital intercept = C/r.
Slope, with L horizontal = −w/r.
Higher C gives a parallel outward shift if w and r are unchanged.

Marginal rate of technical substitution

Define MRTS_LK as the units of capital the producer can give up for one extra unit of labour while keeping output constant.

MRTS_LK = −dK/dL = MP_L/MP_K

Along an isoquant, dQ = MP_LdL + MP_KdK = 0. This gives its slope. For Cobb–Douglas:

MRTS_LK = (α/β)(K/L)

Interactive isocost explorer

Total cost C Wage w Capital price r

Try raising w while holding C and r fixed. The labour intercept falls; the capital intercept stays unchanged.

7. Producer Equilibrium

The producer’s input-choice problem can be expressed as:

Cost minimisation: achieve a target output at the lowest possible cost.
Output maximisation: obtain the highest possible output from a given input budget.

At a smooth interior optimum with the usual convexity conditions, the isoquant and isocost are tangent:

MP_L/MP_K = w/r
or MP_L/w = MP_K/r

The last form says that the extra output per rupee is equal across inputs. If MP_L/w exceeds MP_K/r, shifting expenditure toward labour can raise output.

Schematic: E represents the least-cost input combination for the illustrated output.

Worked example: equalising output per rupee

Suppose MP_L = 20, MP_K = 40, w = ₹10 and r = ₹20. Then MP_L/w = 2 and MP_K/r = 2. The tangency condition holds.

If the wage rises to ₹20 at this same input combination, labour gives only 1 extra output unit per rupee while capital gives 2. With substitutable inputs, the producer has an incentive to use relatively more capital.

Tangency is conditional: Perfect substitutes may give a corner solution; fixed proportions give a kink solution. Tangency alone is not a complete proof of a global optimum without suitable technology conditions.

Cobb–Douglas optimum with a fixed budget

For positive α and β, maximising Q = A L^αK^β subject to wL + rK = C gives:

L* = [α/(α+β)] C/w
K* = [β/(α+β)] C/r

Find the optimal input combination

Budget C Wage w Capital price r α β

This calculator sets A = 1 and treats inputs as divisible. It solves input allocation for a given budget, rather than choosing the profit-maximising output level.

8. Expansion Path

The expansion path joins cost-minimising input combinations for different output levels, holding input prices and technology constant.

As the target output changes, the producer moves to different isoquants. With fixed relative input prices, the relevant isocost lines have the same slope.

For Cobb–Douglas: K/L = βw/(αr)

The input ratio is constant, so its expansion path is a straight ray from the origin. More generally, homothetic technologies permit optimal input proportions to remain unchanged as output expands at fixed input prices; non-homothetic technologies can have curved expansion paths.

Trace an expansion path

Use Q = √(LK), w = ₹10 and r = ₹20. Optimal allocation always satisfies K = L/2.

Input budget: ₹120

Budget	L*	K*	Output √(LK)
₹80	4	2	2.83
₹120	6	3	4.24
₹240	12	6	8.49

Distinction: An expansion path tracks optimal input bundles across output levels. An isoquant holds output constant. An isocost holds expenditure constant.

9. Technical Progress and Production

Technical progress enables more output from the same inputs, or the same output with fewer inputs. It shifts the production function upward and a fixed-output isoquant inward.

A rise in output caused only by using more inputs is a movement under the existing technology, rather than proof of technical progress.

Form	Representation	Meaning
Hicks-neutral	Q = A(t)f(L,K)	At a fixed input ratio, both marginal products rise proportionately; MRTS is unchanged.
Harrod-neutral	Q = f(A_L(t)L,K)	Labour-augmenting progress: effective labour rises.
Solow-neutral	Q = f(L,A_K(t)K)	Capital-augmenting progress: effective capital rises.

Hicks classification of biased progress

Compare marginal products at the same K/L ratio, using MRTS_LK = MP_L/MP_K:

Labour-saving: MP_K rises relatively more, so MP_L/MP_K falls.
Capital-saving: MP_L rises relatively more, so MP_L/MP_K rises.
Hicks-neutral: the ratio remains unchanged.

Read the definition carefully: “Labour-augmenting” and “labour-saving” belong to different classifications and are not interchangeable labels. For Cobb–Douglas, input augmentation can be expressed as a multiplicative productivity change, so neutrality categories can overlap.

Productivity experiment

Hold L = 9 and K = 16 fixed in Q = A√(LK).

Technology A: 1.00

Growth-accounting connection

For Q = A L^αK^β with constant exponents, log differentiation gives:

g_Q = g_A + αg_L + βg_K

If output grows by 5%, labour by 2%, and capital by 4%, with α = 0.6 and β = 0.4, the approximate productivity-growth residual is:

g_A ≈ 5 − (0.6 × 2) − (0.4 × 4) = 2.2%

The equation is exact for instantaneous log growth rates. Substituting ordinary annual percentage changes is an approximation. A measured residual can reflect more than pure technological invention, including measurement and omitted factors.

10. Revision and Exam Traps

Question clue	Use this idea
One input changes; another is fixed	Variable proportions.
All inputs change proportionately	Returns to scale.
MP falls but remains positive	TP still rises, at a decreasing rate.
MP exceeds AP	AP rises.
Maximum AP on a standard smooth curve	MP = AP.
Maximum TP on a standard smooth curve	MP = 0, changing from positive to negative.
Least cost, smooth interior solution	MRTS = w/r.
Higher output from unchanged inputs	Technical progress or improved technical efficiency; distinguish the two from context.
Fixed-output isoquant moves inward	Improved technology.

Challenge 1: Can diminishing marginal products coexist with constant returns?

Yes. Q = √(LK) has diminishing marginal product of each input when the other is fixed. Doubling both inputs doubles output, so it has constant returns to scale.

Challenge 2: Calculate marginal products and MRTS

For Q = L^0.4K^0.6:

MP_L = 0.4L^−0.6K^0.6; MP_K = 0.6L^0.4K^−0.4.

MRTS_LK = (2/3)(K/L). With w/r = 1/2, tangency implies K/L = 3/4.

Challenge 3: Fixed proportions and redundant inputs

If Q = min(L/2,K/3), then L = 10 and K = 12 give Q = min(5,4) = 4.

Four output units require L = 8 and K = 12. The remaining 2 labour units are redundant. Adding only labour will not increase output while K stays at 12.

Knowledge-check progress: 0 of 13 attempted

These checks reveal explanations immediately. The final test below records your first answer to each question.

11. Final Practice Test

25 original UGC NET-style MCQs. Select one option to reveal the answer and explanation. Your first selection is recorded; you can revisit questions using the numbered buttons.

Progress is kept while this page remains open. Reloading the page starts a new session.