ECON10071 Chap.8 Inequality Constraints and Kuhn-Tucker
Inequality Constraints and Kuhn-Tucker
Inequality Constraints and Kuhn-Tucker allow an optimum to be interior or on a boundary. The conditions combine stationarity, primal feasibility, multiplier signs and complementary slackness. A slack constraint has a zero multiplier; a positive multiplier requires that the associated restriction bind. Candidate cases must be checked against the chosen sign convention.
Concavity and convex feasibility can turn a valid Kuhn-Tucker point into a global maximum.
What this chapter covers
- 01
Inequality orientation
- 02
Primal feasibility
- 03
Dual feasibility
- 04
Complementary slackness
- 05
Active-set cases
- 06
Multiplier comparative statics
Worked example · free
Solve a binding ceiling
- 3The unconstrained stationary point x=3 violates the ceiling, so test the binding case x=2.
- 3The objective is concave and the multiplier on 2-x is positive, so x=2 is the global maximum with value 8.
Key terms
- Primal feasibility
- Satisfaction of the original choice restrictions.
- Dual feasibility
- Satisfaction of the multiplier sign conditions for the chosen convention.
- Active constraint
- A restriction that binds at the candidate solution.
Inequality Constraints and Kuhn-Tucker FAQ
What does complementary slackness require?
For each inequality, the multiplier times its slack equals zero. Therefore a slack restriction must have a zero multiplier, while a positive multiplier can occur only when the corresponding restriction binds.
Why does the sign convention matter?
Writing a constraint as less than zero or greater than zero changes the multiplier sign required by the chosen Lagrangian. Mixing conventions can make an invalid candidate appear acceptable.
How should a multiplier be interpreted?
For a binding resource restriction, it gives the local change in optimised objective value from a small relaxation, using the units of objective per unit of resource and assuming the active set stays stable.
Exam move
Write the inequality orientation and multiplier sign at the top of the page. Solve the slack and binding cases separately, then audit feasibility, stationarity, sign and complementary slackness for each. Add concavity and convex-set evidence before claiming global optimality. Interpret a multiplier only after the active constraint and units are clear. Fix the sign convention before differentiating.
Rewrite each inequality in the form required by that convention, record the multiplier sign and include non-negativity constraints explicitly. A reversed constraint requires a corresponding change in the Lagrangian and dual sign rule. Keep the original restriction beside the transformed version so that slack can be translated back into the problem's language.
Many apparent Kuhn-Tucker solutions are merely sign-inconsistent candidates produced by mixing two conventions. Build cases from complementary slackness. For each constraint, either its slack is positive and its multiplier is zero, or it binds and its slack is zero; both may be zero in a degenerate case. Begin with the fully interior case, then examine plausible binding combinations.
In every case solve stationarity, calculate the choices and multipliers, and complete a four-part audit: primal feasibility, dual feasibility, complementary slackness and stationarity. Reject a row as soon as one item fails, but preserve the rejected calculation so that the active-set reasoning remains reviewable. Separate necessity from sufficiency.
A valid Kuhn-Tucker candidate still needs the relevant regularity and curvature conditions for a global conclusion. For maximisation, practise recognising a concave objective and a convex feasible set. Strict concavity can establish uniqueness. If those properties are unavailable, compare all feasible candidates or use a direct bound. Do not infer globality from a positive multiplier or from the fact that one constraint binds.
The multiplier belongs to the local sensitivity analysis after optimality has been established. Interpret shadow values with units and range. A multiplier measures the local change in the optimised objective from a small relaxation of its associated binding restriction under the chosen sign convention. A slack constraint has zero local value while the active set is unchanged.
Large relaxations can activate another constraint or make the current one slack, so a local multiplier should not be extrapolated mechanically. In timed practice, finish by naming the active set, objective value, sufficiency argument and the units of each nonzero multiplier. After solving a case table, vary the resource bound and identify the interval over which the same active set remains feasible.
Within that interval, compare the predicted value change from the multiplier with a newly solved optimum for a small adjustment. Then choose a larger change that activates another restriction and observe why the original shadow value no longer applies.
This exercise makes the local nature of comparative statics concrete and reinforces that multipliers are interpreted only after the underlying optimality conditions have been checked. Include a corner with a binding non-negativity restriction in the same set. Checking its multiplier and slack alongside the resource constraint prevents the active-set analysis from focusing only on the most visible inequality.
Working through Inequality Constraints and Kuhn-Tucker in ECON10071? Sia is AskSia’s AI Mathematics tutor — ask any ECON10071 Inequality Constraints and Kuhn-Tucker question and get a clear, step-by-step explanation grounded in how ECON10071 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.