A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The consumer's problem has two dual formulations: (1) maximize utility subject to budget, yielding Marshallian demand and indirect utility v(p,m); (2) minimize expenditure for target utility, yielding Hicksian demand and expenditure function e(p,u). These problems are equivalent: v(p, e(p,u)) = u and e(p, v(p,m)) = m. All demand information is contained in either the indirect utility or expenditure function.
Work through the dual problems for a concrete utility function. Verify the identities linking indirect utility and expenditure. See how duality enables estimation: either form generates the same demand.
You already know the consumer's basic problem from utility theory: given prices and a budget, choose the bundle that maximizes utility. Duality reveals that this problem has a mirror image — minimize the expenditure needed to reach a target utility level — and the two problems contain exactly the same information about consumer behavior. Understanding duality means understanding that these are not two different theories of the consumer but two equivalent windows into the same underlying preferences.
The primal problem (utility maximization) starts with a budget m and asks: what is the best utility I can achieve? The solution gives you Marshallian demand x(p, m) — the quantities chosen as a function of prices and income — and the indirect utility function v(p, m) — the maximum utility achievable at those prices and income. The dual problem (expenditure minimization) starts with a target utility u and asks: what is the cheapest way to reach it? The solution gives you Hicksian demand h(p, u) — the quantities chosen as a function of prices and target utility — and the expenditure function e(p, u) — the minimum cost of reaching utility u.
The power of duality lies in the identities connecting these objects. If you solve the primal and plug the optimal utility into the dual, you get back your original budget: e(p, v(p, m)) = m. If you solve the dual and plug the minimum expenditure into the primal, you get back your target utility: v(p, e(p, u)) = u. These are not approximations — they are exact equalities that hold for any well-behaved preference relation. Similarly, Marshallian and Hicksian demands are related: h(p, u) = x(p, e(p, u)). At the optimum, the utility-maximizing and expenditure-minimizing bundles coincide.
Why does this matter in practice? Because the two formulations have different analytical strengths. Marshallian demand is what we observe — people shop with budgets, not utility targets. But Hicksian demand is what we need for welfare analysis, because it isolates the pure substitution effect of a price change by holding utility constant. The expenditure function, via Shephard's lemma, delivers Hicksian demands through simple differentiation: ∂e/∂p_i = h_i(p, u). Duality means you never need to solve the dual problem directly — you can derive everything from the indirect utility function using Roy's identity, or from the expenditure function using Shephard's lemma. The choice of which formulation to use depends on which is more convenient for the problem at hand, and duality guarantees the answers will always agree.