Now both pieces are on the table: tastes (indifference curves) and limits (the budget line). The consumer picks the best bundle she can afford. This is the most-examined calculation in the subject, and there is one recipe: find where the indifference curve just touches the budget line.
She chooses the bundle she likes best among those she can afford. With more-is-better, the best bundle uses all her income, so it lies on the budget line (any bundle below it has a better one nearby). So the problem is: reach the highest indifference curve that still touches the budget line.
Written out: maximise u(x, y) subject to pxx + pyy = m. The answer, as a function of prices and income, is her demand (called Marshallian or uncompensated demand): x*(px, py, m) and y*(px, py, m).
Suppose she is on the budget line and an indifference curve cuts the line there. Then the curve’s slope differs from the line’s, and there are bundles on the line that lie above the curve: better bundles she can still afford. So she can only be at her best where the curve touches the line and does not cross it: the slopes are equal.
Read it as a rule of trade. If her MRS is 3, she would give up 3 units of y for one more x. If the market asks only 2 units of y for an x (px/py = 2), she gains by buying x. She keeps buying x, which lowers her MRS, until the two are equal. Solve this together with the budget equation to get both quantities.
Choose one of the exam utility functions. Slide the bundle along the budget line. The blue curve through the dot cuts the line almost everywhere; the dashed curve through the hollow dot is the highest one she can reach. Watch the three equalities at the end: MRS against the price ratio, marginal utility per rupee, and the multiplier λ.
u = x^0.5 y^0.5. Slide the bundle along the budget line. Its curve (blue) cuts the line until she is at the optimum, where the curve (dashed) only touches it.
Utility 30.619 (the best affordable is 35.355). Her MRS is 3/2; the price ratio px/py is 1/2.
Her MRS is above the price ratio: she would give up more y for an extra x than the market asks. Move right and buy more x.
On x: 0.6124. On y: 0.2041. Unequal, so a rupee moved from the lower to the higher would raise utility.
She spends 0.25 of income on x and 0.75 on y. At the optimum of a Cobb–Douglas consumer with weight a, the share on x is a = 0.5 whatever the prices.
The same condition comes out of the Lagrangian, which handles any number of goods:
Dividing the first two gives the tangency condition. Each also says λ = MUx/px = MUy/py, the equal marginal utility per rupee of Lesson 2. So λ is the marginal utility of income: the extra utility from one more rupee. For u = x1x2 maximised on x1 + x2 = 6, the best bundle is (3, 3), and λ = MUx1/p1 = x2/1 = 3 (Hyderabad 2020). A unit more of income would raise the maximum of u by about 3.
For u = xay1−a, the tangency condition and the budget give a result worth memorising:
The exponents, normalised to sum to one, are the budget shares. For u = xy they are ½ and ½: x* = m/2px and y* = m/2py (CUET 2023, JAM 2025, Hyderabad 2012, 2013, 2020). For u = x0.25y0.75 with prices 2 and 5 and income 100, she spends 25 on x and 75 on y: x* = 12.5 and y* = 15 (Hyderabad 2018). For u = xy² the shares are ⅓ and ⅔, so with prices 2 and 3 and income 9 she buys x = 1.5 and y = 2 (JNU 2011, 2015). The demand for y when U = a√(XY), M = 656 and Px = 8 is 656/2 ÷ Py = 328/Py (Hyderabad 2010).
The converse is how examiners test it: someone who spends fixed percentages of income on each good whatever the prices, such as 30% on housing and 25% on food, behaves as a Cobb–Douglas consumer (Hyderabad 2015). Only Cobb–Douglas has budget shares that stay constant at every price and income.
The same two steps work for any smooth, convex utility: set MRS = px/py, then put the answer in the budget.
UGC NET 2025 asked which of five statements describe the outcome of a consumer’s optimisation. Four are true and one is false:
| Statement | True? |
|---|---|
| The marginal utility per rupee is the same on every good | Yes |
| The MRS equals the price ratio | Yes |
| She reaches the highest indifference curve she can | Yes |
| The indifference curve is tangent to the budget line | Yes |
| She is indifferent between any two points on the budget line | No: only the optimum is on the highest curve; the others are on lower ones |
Tangency is necessary only for a smooth, convex, interior optimum. With kinks (perfect complements), straight lines (perfect substitutes) or curves that bow away from the origin, the best bundle can sit at a corner of the budget line. That is Lesson 6.
Definitions and one-step calculations: the shape of a curve, a budget line, an optimum for Cobb–Douglas tastes. Most CUET, UGC NET, Hyderabad and early JNU questions sit here.
Statements to judge, matching lists, and demands to compute with unusual tastes, up to JAM, GATE and JNU level.
This topic’s hardest questions, from the DSE entrance, are worked through in Lesson 17: Min, max and kinked utilities, and deals with three goods.
These are questions from the TestStop bank. A free account answers 10 a day, each with its worked solution; All-Access opens every set with no daily limit.