Questions: Duality in Producer Theory

Question 1

A firm's cost function is c(w₁, w₂, y). According to Shephard's lemma, what does ∂c/∂w₁ give you?

• A. The marginal cost of producing one additional unit of output
• B. The conditional demand for input 1 — how much input 1 the cost-minimizing firm uses
• C. The elasticity of total cost with respect to the price of input 1
• D. The shadow price of the production constraint

Answer: (B)

Shephard's lemma states that differentiating the cost function with respect to an input price gives the conditional factor demand for that input: ∂c/∂w_i = x_i*(w, y). This is the central practical payoff of duality in producer theory — once you have the cost function, optimal input demands are obtained by simple differentiation, without re-solving the constrained optimization problem.

Question 2

An economist wants to study how a firm adjusts its input use when input prices change. She has estimated the firm's cost function from observed price and cost data, but has not specified a production function. Can she derive the firm's input demands?

• A. No — input demands require solving the optimization problem with an explicit production function
• B. No — the cost function only tells us total cost, not the composition of inputs
• C. Yes — by Shephard's lemma, differentiating the cost function with respect to each input price yields the conditional factor demands
• D. Yes — but only if the production function is Cobb-Douglas or another parametric form

Answer: (C)

Duality guarantees that the cost function encodes the same information as the production function. Shephard's lemma — ∂c/∂w_i = x_i* — gives conditional factor demands directly from the cost function, for any well-behaved technology. This is a general result, not restricted to specific functional forms. This is why empirical industrial organization regularly estimates cost functions: costs and prices are observable, and duality recovers the production-side information without specifying the production function.

Question 3

The production function and the cost function represent different aspects of a firm's technology, so they can seldom be derived from each other.

• T. True
• F. False

Answer: False

They are dual representations of exactly the same underlying technology. The production function asks: given inputs, what is the maximum output? The cost function asks: given prices and a target output, what is the minimum cost? Solving one optimization yields the other, and the envelope theorem (Shephard's lemma) provides the formal bridge. All information in the production function is encoded in the cost function and recoverable from it.

Question 4

Shephard's lemma states that differentiating the cost function with respect to an input price gives the conditional factor demand for that input — without requiring the original optimization problem to be re-solved.

• T. True
• F. False

Answer: True

This is the key result of duality in producer theory. The envelope theorem applied to the cost-minimization problem yields ∂c(w,y)/∂w_i = x_i*(w,y), where x_i* is the cost-minimizing demand for input i. Once you have the cost function, all comparative statics on input demands are available through differentiation — a major computational and conceptual simplification over repeatedly solving constrained optimization problems.

Question 5

Why is the cost-function approach often preferred over the production-function approach in empirical work on firms?

Model answer: In empirical settings, input prices and total costs are often directly observable in firm-level data, while the production function's input-output mapping may involve unobservable variables (effort, quality, managerial skill). Duality guarantees that any well-behaved cost function corresponds to a valid underlying technology, so estimating a cost function from price and cost data gives legitimate information about production possibilities. Shephard's lemma then yields conditional factor demands from simple differentiation, and comparative statics (how input use responds to price changes) follow without specifying or estimating the production function directly.

This reflects a broader principle in economics: dual representations are not just mathematical curiosities — they often align better with what is observable. The same logic underlies consumer theory duality: expenditure functions are estimated from household expenditure data when utility functions are unobservable. The mathematical equivalence of the dual problems means nothing is lost by working on whichever side is empirically tractable.