Click any slide to jump · press m or Esc to close
One person. A fixed income. Two goods. Out of this single decision we will construct the demand curve — one of the two blades of the scissors that determines every price you have ever paid.
You walk into a shop with a fixed amount of money. Many combinations of goods are affordable. Which one will you actually buy — and why that one?
Economists call this the problem of choice. The answer depends on exactly two things:
Your tastes and preferences — your likes and dislikes.
Handled by: utility, indifference curves
Your income and the prices of the goods.
Handled by: budget set, budget line
Best affordable bundle = the meeting point of what you want and what you can afford. Everything else is machinery for locating that point precisely.
Utility, Total Utility, Marginal Utility, the Law of Diminishing Marginal Utility — and a first derivation of the demand curve.
Preferences, indifference curves, MRS, the indifference map.
Budget set, budget line, and how it moves when income or price changes.
Putting the two halves together: the tangency condition.
Demand curve, law of demand, normal/inferior, substitutes/complements, shifts vs movements, market demand.
Measuring responsiveness, elasticity along a linear demand curve, and expenditure.
A consumption bundle (or simply a bundle) is any combination of quantities of the goods under consideration.
To keep every idea drawable on a flat page, we assume just two goods: bananas and mangoes.
x1 = quantity of bananasx2 = quantity of mangoes(x1, x2) = a bundlep1, p2 = their pricesM = the consumer's income(5, 10) = 5 bananas and 10 mangoes.
(10, 5) = 10 bananas and 5 mangoes.
Order matters. These are different bundles. Both quantities may be positive or zero.
Not because real consumers buy two things — but because two goods fit on a two-dimensional diagram. Every conclusion we reach generalises to many goods; we simply could not draw it. The same trick is used in Chapter 3, where the firm uses exactly two inputs, L and K.
Utility of a commodity is its want-satisfying capacity. The stronger the desire for a commodity, the greater the utility derived from it.
Utility is a subjective, ethically neutral idea. A cigarette has utility for a smoker even though it is harmful. Utility measures how much a person wants something, not whether they should want it.
Assumes utility can be measured in numbers, in imaginary units called utils. "This shirt gives me 50 utils."
Tools: Total Utility, Marginal Utility, Law of DMU.
Associated with Alfred Marshall.
Assumes utility can only be ranked, not measured. "I prefer this bundle to that one" — but by how much, we never say.
Tools: preferences, indifference curves, MRS.
Associated with Hicks and Allen.
The ordinal approach is more realistic — nobody genuinely says "that mango gave me 14.7 utils." But the cardinal approach is easier to learn and gives the same core conclusion. We study both, and they agree: the demand curve slopes downward.
Total Utility (TU) is the total satisfaction derived
from consuming a given quantity of a commodity. TUn is the total
utility from consuming n units.
Marginal Utility (MU) is the change in total utility resulting from the consumption of one additional unit of a commodity.
4 bananas give 28 utils; 5 bananas give 30 utils.
So MU of the 5th banana = TU5 − TU4 = 30 − 28 = 2 utils.
| Units consumed | Total Utility (TU) | Marginal Utility (MU) |
|---|---|---|
| 1 | 12 | 12 |
| 2 | 18 | 6 |
| 3 | 22 | 4 |
| 4 | 24 | 2 |
| 5 | 24 | 0 |
| 6 | 22 | −2 |
MU₃ = TU₃ − TU₂ = 22 − 18 = 4 ✓
TU₄ = MU₁+MU₂+MU₃+MU₄ = 12+6+4+2 = 24 ✓
| Phase | Marginal Utility | Total Utility | In Table 2.1 |
|---|---|---|---|
| 1 | MU is positive but falling | TU rises, at a diminishing rate | Units 1–4 |
| 2 | MU = 0 | TU is at its maximum (point of satiation) | Unit 5 |
| 3 | MU is negative | TU falls | Unit 6 |
The relationship between a total and its marginal is always this:
Marginal > 0 → Total rises · Marginal = 0 → Total is at its peak · Marginal < 0 → Total falls
You will meet the identical logic with Marginal Product and Total Product (Ch 3), with Marginal Cost and Total Cost (Ch 3), and with Marginal Revenue and Total Revenue (Ch 4). Master it here and you get three chapters at a discount.
The Law of Diminishing Marginal Utility states that as a consumer consumes more and more units of a commodity, the marginal utility derived from each successive unit goes on declining, while the consumption of other commodities is kept constant.
You are thirsty. The 1st glass of water is bliss. The 2nd is nice. The 3rd is fine. The 5th adds nothing. The 6th makes you uncomfortable — negative marginal utility.
Why? Because having obtained some of the commodity, the desire for still more of it becomes weaker.
Complete the schedule:
| Units | TU | MU |
|---|---|---|
| 1 | 20 | ? |
| 2 | ? | 14 |
| 3 | 42 | ? |
| 4 | ? | 4 |
| 5 | 46 | ? |
At which unit does the consumer reach the point of satiation?
| Units | TU | MU | Working |
|---|---|---|---|
| 1 | 20 | 20 | MU₁ = TU₁ − 0 |
| 2 | 34 | 14 | TU₂ = 20 + 14 |
| 3 | 42 | 8 | 42 − 34 |
| 4 | 46 | 4 | 42 + 4 |
| 5 | 46 | 0 | 46 − 46 |
Satiation at the 5th unit, where MU = 0 and TU is maximum (46). Note MU falls throughout — 20, 14, 8, 4, 0 — exactly as the Law of DMU predicts.
A consumer's total utility from bananas is rising, and her marginal utility is positive but falling.
A student concludes: "Since she still gets positive utility from every extra banana, a rational consumer should keep buying bananas until MU reaches zero — that is where total satisfaction is greatest."
(a) Is it true that TU is greatest where MU = 0?
(b) Is the student's advice correct? What has been left out?
(c) State the condition that actually determines how many bananas she buys.
(a) Yes — that part is correct. TU is maximum exactly where MU = 0. In Table 2.1 that is the 5th unit, where TU peaks at 24.
(b) The advice is wrong. What has been left out is the price.
Bananas are not free. Every banana bought uses up money that could have bought something else.
Consuming up to MU = 0 would be right only if bananas cost nothing. Since they have a price, each banana carries an opportunity cost — the satisfaction forgone from the mangoes that money could have bought. The student is maximising satisfaction from one good while ignoring the budget constraint and the other good entirely.
(c) The condition is the tangency condition:
She stops where the rate at which she is willing to trade mangoes for bananas equals the rate at which the market lets her. At that point MU of bananas is still strictly positive — she would happily eat more, but not at that price.
The general error is important beyond this question: maximising the satisfaction from a single good is not the same as maximising overall satisfaction. The consumer's problem is a problem of allocation across goods under a constraint — which is exactly why Chapter 2 needs both an indifference map and a budget line, not just one of them.
Note the structural echo of Chapter 1: a free good would be consumed to satiation; an economic good never is, precisely because it is scarce and carries an opportunity cost.
The Law of DMU already contains the Law of Demand. Here is the argument.
Demand for a commodity is the quantity that a consumer is willing to buy and able to afford, at a given price, in a given period of time.
Cardinal utility analysis is easy to teach — but its central assumption is false. In real life nobody expresses satisfaction in numbers. You have never said "this mango gave me 17 utils." Utility has no natural unit, and no two people's "utils" could be compared anyway.
But there is something you can honestly do: rank bundles.
"I prefer A to B" · "I prefer B to A" · "I am indifferent between A and B"
That is all the ordinal approach ever asks of a consumer — an ordering, never a measurement. And remarkably, it is enough to derive the entire theory.
Ordinal utility analysis assumes that a consumer can only rank consumption bundles in order of preference, not measure the utility from them in numbers.
A consumer's preferences are monotonic if and only if, between any two bundles, the consumer prefers the bundle which has more of at least one of the goods and no less of the other good as compared to the other bundle.
"More is better." If bundle A gives you more of something and no less of anything else, you must prefer A.
Compare (10, 8) and (8, 6): the first has more of both.
A consumer with monotonic preferences must strictly prefer (10, 8) — she cannot be
indifferent.
Compare (6, 6) and (5, 6): same mangoes, one more banana.
(6, 6) must be preferred.
Monotonicity is what forces the indifference curve to slope downward, and what forces the consumer's optimum to lie on the budget line rather than inside it. It does a great deal of quiet work later in this chapter.
An indifference curve is the locus of all points representing those bundles among which the consumer is indifferent — every bundle on it yields the same level of satisfaction.
| Combination | Bananas | Mangoes |
|---|---|---|
| A | 1 | 15 |
| B | 2 | 12 |
| C | 3 | 10 |
| D | 4 | 9 |
This consumer is equally happy with 1 banana + 15 mangoes as she is with 4 bananas + 9 mangoes. She has no reason to prefer any one of A, B, C, D.
To gain each extra banana she gives up mangoes — but only 3, then 2, then 1. The sacrifice is shrinking. That is the key observation, and it has a name.
Marginal Rate of Substitution (MRS) is the rate at which a consumer is willing to sacrifice units of one good to obtain one additional unit of the other, so that her total satisfaction remains unchanged.
| Combination | Bananas (x₁) | Mangoes (x₂) | Δx₂ | Δx₁ | MRS |
|---|---|---|---|---|---|
| A | 1 | 15 | — | — | — |
| B | 2 | 12 | −3 | +1 | 3 : 1 |
| C | 3 | 10 | −2 | +1 | 2 : 1 |
| D | 4 | 9 | −1 | +1 | 1 : 1 |
MRS is the absolute value of the slope of the indifference curve at a point. Since the curve slopes downward, Δx₂/Δx₁ is negative; we take the magnitude and drop the minus sign. So MRS = 3 means "willing to give up 3 mangoes for 1 banana."
The Law of Diminishing Marginal Rate of Substitution states that as the consumer acquires more and more of one good, she is willing to give up smaller and smaller amounts of the other good for each additional unit.
Diminishing MRS ⇒ the indifference curve gets flatter as we move right ⇒ it is convex to the origin. This is the normal shape of an indifference curve.
Perfect substitutes are goods that can be used in place of each other and provide exactly the same level of utility to the consumer.
| Comb. | ₹5 notes | ₹5 coins | MRS |
|---|---|---|---|
| A | 1 | 8 | — |
| B | 2 | 7 | 1 : 1 |
| C | 3 | 6 | 1 : 1 |
| D | 4 | 5 | 1 : 1 |
A ₹5 coin and a ₹5 note are identical in value. The consumer will always trade exactly one for one, no matter how many she already holds. MRS does not diminish, so the IC does not bend.
An indifference map is a family of indifference curves representing a consumer's preferences over all possible bundles. Each curve corresponds to a different level of satisfaction.
Because preferences are monotonic, a curve lying farther from the origin gives higher satisfaction.
IC₃ ≻ IC₂ ≻ IC₁
The map tells us the order — IC₃ is better than IC₂ — but never by how much. That is precisely what "ordinal" means. There is no number attached to any curve.
An IC slopes downward from left to right: if Δx₁ > 0 then Δx₂ < 0.
Suppose the consumer gains a banana without giving up any mangoes. Then she has more of one good and no less of the other.
By monotonicity, she must now be better off — so she has moved to a higher indifference curve, not stayed on the same one.
Therefore, to remain on the same IC, every gain in bananas must be compensated by a loss of mangoes. Hence the curve slopes downward. ∎
An IC can never be horizontal, vertical, or upward-sloping — each of those would violate monotonicity.
| Comb. | Bananas | Mangoes |
|---|---|---|
| A | 1 | 10 |
| B | 2 | 10 |
| C | 3 | 10 |
A, B and C all contain the same 10 mangoes, but B has more bananas than A, and C has more bananas than B.
By monotonicity, C ≻ B ≻ A. Since they lie on different curves, the curve farther from the origin represents higher satisfaction. ∎
This is what turns the consumer's problem into a clean instruction: climb to the highest indifference curve your budget allows.
Hence our assumption was false: two indifference curves can never intersect. ∎
Intersection would mean the consumer is simultaneously indifferent between two bundles and prefers one to the other. That is incoherent.
An indifference curve is normally convex to the origin — it is steep at the left and gets flatter as we move right.
Because of the Law of Diminishing MRS. Since MRS falls (3 → 2 → 1), the slope of the curve falls too, and a curve whose slope keeps flattening is convex to the origin.
| Feature | Rests on |
|---|---|
| 1. Slopes downward left to right | Monotonic preferences |
| 2. Higher IC = higher satisfaction | Monotonic preferences |
| 3. Two ICs never intersect | Monotonic preferences (proof by contradiction) |
| 4. Convex to the origin | Law of Diminishing MRS |
Notice that three of the four features come from a single assumption — monotonicity. If a question asks "why does an IC slope downward", the answer is never "because of diminishing MRS". Diminishing MRS explains the curvature, monotonicity explains the direction.
1. No. Bundle (10, 8) has more of both goods than (8, 6). Monotonicity requires her to strictly prefer (10, 8). She cannot be indifferent.
2. No — they are not monotonic. The bundles (5, 6) and (6, 6) have the same quantity of good 2, but (6, 6) has one more unit of good 1. Monotonicity would require (6, 6) to be strictly preferred. Since your friend is indifferent, monotonicity is violated.
3. (10, 10) ≻ (10, 9) ≻ (9, 9)
(10,10) beats (10,9): same good 1, more good 2. (10,9) beats (9,9): same good 2, more good 1. And (10,10) beats (9,9) on both counts.
A student draws an indifference map in which the curves are concave to the origin instead of convex, and insists this is legitimate because "the curves still slope downward and still never intersect".
(a) Which of the four features of an indifference curve does a concave curve violate,
and which does it satisfy?
(b) What would MRS be doing along such a curve?
(c) Show that if a consumer really had concave indifference curves, the tangency point would
give her the worst affordable bundle rather than the best.
(a) A concave curve satisfies features 1, 2 and 3 — it can still slope downward, higher curves can still mean higher utility, and two such curves still need not intersect. All three of those come from monotonicity, which concavity does not touch.
It violates feature 4, convexity — and therefore violates the Law of Diminishing MRS, which is the assumption feature 4 rests on.
(b) MRS would be increasing. As the consumer acquired more bananas, she would be willing to give up more and more mangoes for each additional banana — the opposite of diminishing MRS.
Economically that is very odd: it says the more bananas she has, the more desperate she becomes for another one, and the less she values the mangoes she still holds.
(c) The tangency would be a minimum, not a maximum.
With a convex curve, the indifference curve lies above the budget line on both sides of the tangency — so every other affordable bundle is on a lower curve. The tangency is the best.
With a concave curve, the indifference curve lies below the budget line on both sides — so moving away from the tangency in either direction reaches a higher curve. The tangency is the worst.
Such a consumer would push all the way to a corner — spending her entire income on bananas alone or mangoes alone, never a mixture. Concave preferences describe someone who wants extremes, not balance.
Why this matters: convexity is not decoration. It is precisely the assumption that makes the tangency condition identify a maximum, and it is why real consumers buy mixtures of goods rather than everything of one. This is the same logic you will meet in Chapter 4, where P = MC identifies a profit maximum only if MC is rising — a condition on curvature, not on the tangency itself.
So far, everything has been about what the consumer wants — utility, preferences, indifference curves. Not once have we mentioned money.
Wanting is free. Buying is not. We now build the second half: what she can afford.
Two things, and only two, determine affordability:
MA fixed amount of money to spend.
p₁ and p₂Given by the market — the consumer cannot influence them.
The consumer takes prices as given. She is a price taker. In Chapter 4 this same assumption is applied to the firm, and given a name: perfect competition. It has been quietly present since this slide.
To buy x₁ bananas she spends p₁x₁. To buy x₂
mangoes she spends p₂x₂. Total spending must not exceed income:
Inequality (2.1) is the consumer's budget constraint.
The budget set is the collection of all bundles that the consumer can buy with her income at the prevailing market prices — that is, every bundle satisfying the budget constraint.
The budget set includes bundles costing less than M as well as bundles costing exactly M. It is everything she can afford, including options that leave money unspent.
Income M = ₹20. Both goods priced at ₹5, available only in whole units.
(0,0) (0,1) (0,2) (0,3) (0,4) (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (3,0) (3,1) (4,0)
Amber (on the line): (0,4), (1,3), (2,2), (3,1), (4,0) — these cost exactly ₹20. Her entire income is spent.
Grey (below the line): cost less than ₹20. Affordable, but money is left over.
(3,3) costs ₹30 and (4,5) costs ₹45. Not affordable — they lie above the line and are simply not available to her.
The budget line consists of all bundles which cost exactly equal to the consumer's income — the boundary of the budget set.
Rearranged into the familiar y = c + mx form of a straight line:
M / p₂
Bundle bought if she spends her entire income on mangoes.
M / p₁
Bundle bought if she spends her entire income on bananas.
− p₁ / p₂
The price ratio. Negative, so the line slopes downward.
Take two points on the budget line, (x₁, x₂) and
(x₁+Δx₁, x₂+Δx₂). Both cost exactly M:
p₁x₁ + p₂x₂ = M …(2.4)
p₁(x₁+Δx₁) + p₂(x₂+Δx₂) = M …(2.5)
Subtract (2.4) from (2.5):
p₁Δx₁ + p₂Δx₂ = 0 …(2.6)
Rearranging:
Suppose she is spending everything and wants one more banana. It costs
p₁. She must cut mango spending by p₁. With p₁
she could have bought p₁/p₂ mangoes.
So she must give up p₁/p₂ mangoes to gain one banana.
The absolute value of the slope of the budget line is the rate at which the consumer is able to substitute bananas for mangoes in the market.
MRS = the rate at which she is willing to substitute (her preferences).
p₁/p₂ = the rate at which she is able to substitute (the market).
Two rates. The whole of the next section is about what happens when they differ.
Income changes from M to M′; prices unchanged. The new line is
Only M changed. The slope −p₁/p₂ contains no M —
so the slope is unaffected. Only the intercepts move.
Both intercepts rise → parallel outward shift. More of both goods becomes affordable.
Both intercepts fall → parallel inward shift. The budget set shrinks.
Price of bananas changes from p₁ to p₁′; p₂ and
M unchanged:
Unchanged: the vertical intercept M/p₂ — if she buys only mangoes,
the price of bananas is irrelevant to her.
Changed: the slope, and the horizontal intercept M/p₁′.
Line becomes steeper; pivots inward. Fewer bananas affordable.
Line becomes flatter; pivots outward. More bananas affordable.
A change in p₂ works the same way, pivoting around the
horizontal intercept instead.
A consumer's income is ₹20. Prices are p₁ = ₹4 and p₂ = ₹5.
1. 4x₁ + 5x₂ = 20
2. Good 1: M/p₁ = 20/4 = 5 units. Good 2: M/p₂ = 20/5 = 4 units.
3. Slope = −p₁/p₂ = −4/5 = −0.8.
4. New line 4x₁ + 5x₂ = 40. Intercepts become 10 and 8
— both doubled. Slope still −4/5, so there is a parallel outward shift.
5. p₂ = 4, so 4x₁ + 4x₂ = 20. The good-1 intercept stays at
5; the good-2 intercept rises from 4 to 5. Slope changes from −4/5 to
−1. The line pivots outward around the horizontal intercept and becomes steeper.
6. No change at all. The new constraint is 8x₁ + 10x₂ ≤ 40. Divide
throughout by 2: 4x₁ + 5x₂ ≤ 20 — identical to the original. What matters
is real income (income relative to prices), not the numbers themselves.
(a) A consumer can just afford 6 units of good 1 and 8 units of good 2 when she spends her entire income. Prices are ₹6 and ₹8. What is her income?
(b) Two goods are each priced at ₹10; income is ₹40; goods come only in whole units. List all available bundles, and identify those costing exactly ₹40.
(a) The bundle (6, 8) costs exactly her income:
M = p₁x₁ + p₂x₂ = (6 × 6) + (8 × 8) = 36 + 64 = ₹100
(b) The constraint is 10x₁ + 10x₂ ≤ 40, i.e. x₁ + x₂ ≤ 4.
All 15 available bundles:
(0,0) (0,1) (0,2) (0,3) (0,4)
(1,0) (1,1) (1,2) (1,3)
(2,0) (2,1) (2,2)
(3,0) (3,1)
(4,0)
Costing exactly ₹40 (those with x₁ + x₂ = 4, lying on the
budget line): (0,4), (1,3), (2,2), (3,1), (4,0).
A consumer has income M and faces prices p₁ and p₂.
Inflation now raises both prices by 20%, and her employer raises her income by 20% as well.
(a) What happens to her budget line? Prove it algebraically.
(b) What happens to her optimal bundle?
(c) Now suppose instead that both prices rise 20% but her income rises only 10%. Describe the
change in the budget line precisely — is it a shift, a pivot, or something else?
(a) Nothing changes at all. The new constraint is:
1.2p₁·x₁ + 1.2p₂·x₂ ≤ 1.2M
Divide throughout by 1.2:
p₁x₁ + p₂x₂ ≤ M
Identical to the original. Both intercepts (M/p₁ and M/p₂) are unchanged, because the 1.2 cancels in each. The slope −p₁/p₂ is unchanged for the same reason.
(b) Her optimal bundle is exactly the same. The budget set has not moved and her preferences have not changed, so the same tangency point is still the best affordable bundle.
The consumer's choice depends on real income, not money income. What matters is income relative to prices. Scaling everything by the same factor leaves every real magnitude untouched.
This is why a pay rise that merely matches inflation makes nobody better off — a result you can now prove rather than merely assert.
(c) A parallel inward shift.
New constraint: 1.2p₁·x₁ + 1.2p₂·x₂ ≤ 1.1M. Divide by 1.2:
p₁x₁ + p₂x₂ ≤ (1.1/1.2)M = 0.9167M
This is the original prices with an income of about 91.7% of M. So:
Slope unchanged (both prices moved by the same proportion, so the ratio p₁/p₂ is untouched) · both intercepts fall by the same factor → a parallel inward shift.
Not a pivot. A pivot requires the relative price to change — one price moving differently from the other. Here both moved together, so only the real income fell. She is unambiguously worse off, and will end on a lower indifference curve.
The indifference map — what she wants. Higher curve = better.
The budget line — what she can afford. On or below it only.
A rational consumer knows her own preferences and always chooses, from the bundles available to her, the one that gives her maximum satisfaction.
Reach the highest possible indifference curve, given the budget set.
That single sentence is the whole of consumer theory. The next two slides just locate the point.
Take any point Z strictly below the budget line. Money is left unspent.
From Z, she can reach points on the budget line that have more of at least one good and no less of the other.
By monotonicity, every such point is strictly preferred to Z.
So Z cannot be the optimum.
Points above the budget line are simply not available — she cannot afford them.
The optimum bundle lies on the budget line: a rational consumer with monotonic preferences spends her entire income.
Higher satisfaction, but it lies entirely above the budget line. Unaffordable.
Points F and G are affordable, but they sit on a lower curve than E. Inferior.
Affordable and the highest reachable. E is the optimum — the bundle
(x₁*, x₂*).
At the optimum the budget line is tangent to an indifference curve. Tangency means the two curves have the same slope at that point:
The rate at which she is willing to trade mangoes for bananas equals the rate at which the market allows her to trade them. When the two rates agree, there is no trade left worth making — she is in equilibrium.
The IC must also be convex at that point (MRS must be diminishing). Tangency alone is not enough — on a concave curve the tangency point would be the worst point, not the best.
Suppose the two goods have the same price, so the price ratio p₁/p₂ = 1. Compare it with her MRS.
She is willing to give up 2 mangoes for one more banana.
The market only asks for 1 mango for one more banana.
So she buys the banana but pays only 1 mango instead of the 2 she was prepared to give up. She therefore ends up with more of at least one good and no less of the other — a bundle she strictly prefers. She is better off.
⇒ She will keep buying bananas. Not an equilibrium.
She values a banana at only half a mango.
The market charges a full mango for it.
She is paying more than the banana is worth to her. Selling a banana back gains her a full mango, though she'd have accepted half. She is better off.
⇒ She will buy fewer bananas. Not an equilibrium.
Whenever MRS ≠ p₁/p₂, a profitable readjustment exists and the consumer moves. She stops moving only when MRS = p₁/p₂. That is why the optimum is at the tangency — and nowhere else.
A consumer buys bananas and mangoes. At her current bundle, MRS = 4, the price of a banana is ₹8 and the price of a mango is ₹4.
(a) Is she in equilibrium? (b) If not, what will she do — and why?
(a) The price ratio is p₁/p₂ = 8/4 = 2.
Since MRS (4) > price ratio (2), she is not in equilibrium.
(b) She is willing to sacrifice 4 mangoes for one extra banana, but the market only requires her to sacrifice 2. The banana is worth more to her than it costs. So she will buy more bananas and fewer mangoes.
As she does so, by the Law of Diminishing MRS her MRS falls (more bananas → each worth less; fewer mangoes → each worth more). She keeps adjusting until MRS = 2, at which point equilibrium is reached.
Two consumers, Anil and Bela, shop in the same market and face the same prices. Anil is far richer and buys much larger quantities of both goods. Their tastes are completely different.
(a) At their respective optimum bundles, is there anything that must be equal for
both of them? Prove it.
(b) Does this mean they derive the same satisfaction? Explain what the equality does and does
not tell us.
(c) Why is this result impossible to state in the language of cardinal utility?
(a) Yes — their MRS must be equal.
Each consumer optimises where MRS = p₁/p₂. They face the same prices, so the same price ratio. Therefore:
Despite different incomes, different quantities and different tastes, at the margin they value the two goods in exactly the same ratio.
(b) It certainly does not mean they are equally satisfied.
MRS is a rate of substitution at the margin — a slope. Equal slopes at their respective optima say nothing about the levels of satisfaction they have reached.
Anil is on a much higher indifference curve — his larger budget set lets him reach one Bela cannot afford. Their curves are tangent to different budget lines with the same slope.
Why the result is interesting anyway: it means no mutually beneficial trade is left between them. If Anil valued bananas more highly at the margin than Bela did, they could both gain by trading. Equal MRS means that opportunity is exhausted — which is a key ingredient of what makes a competitive market efficient.
(c) Because cardinal utility invites a comparison that is meaningless.
The cardinal approach would tempt you to say "Anil gets 500 utils and Bela 200, so Anil is 2.5 times happier". But utility has no natural unit, and there is no way to compare one person's utils with another's — an interpersonal comparison that no evidence can settle.
The ordinal approach never makes that claim. It only ever says each consumer ranks her own bundles, and the result above uses nothing more: it compares a slope — which is observable from choices — not a level of feeling. That is precisely why the ordinal approach is regarded as the more defensible of the two.
So far we found one optimum bundle for one set of prices and income.
Now we change one variable at a time and watch the optimum move. Changing the price of bananas traces out the demand curve — the single most important object in this book.
A consumer's demand function for a good gives the amount of the good she chooses at different levels of its price, when other things remain unchanged.
Held constant: the consumer's income, the prices of other goods, and her tastes and preferences. If any of those move, we are no longer on the same demand curve. This phrase is the difference between a movement and a shift.
The Law of Demand states that, other things remaining equal, there is a negative (inverse) relationship between the price of a commodity and its quantity demanded: when price rises, demand falls; when price falls, demand rises.
In mathematics the independent variable goes on the horizontal axis. In economics we do the opposite for the demand curve: price (the independent variable) is on the vertical axis and quantity on the horizontal.
A long-standing convention in economics, usually credited to Alfred Marshall (though Cournot drew it this way earlier). Everyone follows it; just be aware you are reading the diagram "sideways".
The demand curve is the graphical representation of the demand function. It gives the quantity demanded by the consumer at each price.
The downward slope has two separate causes. Both push in the same direction.
When bananas become cheaper, they become relatively cheaper than mangoes. The consumer substitutes bananas for mangoes to get the same satisfaction more cheaply.
Demand for bananas rises.
When bananas become cheaper, the same money income now buys more — her real income (purchasing power) rises. She can afford more of things generally.
Demand for bananas rises (for a normal good).
For a normal good both effects work in the same direction, so a fall in price definitely raises quantity demanded. The demand curve is negatively sloped.
Substitution effect: the change in optimal quantity when the price changes and
income is adjusted so that she can just afford the old bundle.
Income effect: the change in optimal quantity caused by the change in
purchasing power that the price change brings about.
If d(p) = 20 − 2p: at p = 0, demand is 20; each ₹1 rise cuts demand
by 2; demand hits zero at p = 20/2 = ₹10.
Now hold price constant and change income instead.
A normal good is one whose demand increases when the consumer's income increases, and decreases when income decreases.
Demand moves in the same direction as income.
Examples: branded clothing, milk, restaurant meals, cars.
An inferior good is one whose demand decreases when the consumer's income increases, and increases when income decreases.
Demand moves in the opposite direction to income.
Examples: coarse cereals, low-quality food items, second-hand clothes.
A good is inferior only relative to a consumer's income level. The same good can be normal at low incomes and inferior at higher incomes.
At very low income, more money means more coarse cereal. Beyond a point, more money means she switches to better cereals, and coarse cereal demand falls. The good crossed over from normal to inferior.
For a strongly inferior good, a rise in purchasing power makes the consumer buy less. So when the price falls:
The two effects now oppose each other.
A Giffen good is an inferior good for which the income effect is stronger than the substitution effect, so that demand is positively related to its price — demand rises when price rises.
Now hold price and income constant, and change the price of a related good.
Substitute goods can be used in place of each other to satisfy the same want.
Examples: tea and coffee; Coke and Pepsi; rice and wheat.
Rule: demand for a good moves in the same direction as the price of its substitute.
Price of coffee ↑ → people switch → demand for tea ↑
Complementary goods are consumed together to satisfy a want.
Examples: tea and sugar; shoes and socks; pen and ink; car and petrol.
Rule: demand for a good moves in the opposite direction to the price of its complement.
Price of sugar ↑ → tea becomes costlier to enjoy → demand for tea ↓
Substitutes move with the related price. Complements move against it. Both words start with the direction they take: Substitute → Same, Complement → Contrary.
Caused only by a change in the price of the good itself.
Downward movement (price ↓) = extension of demand.
Upward movement (price ↑) = contraction of demand.
Caused by a change in anything other than the good's own price.
Rightward shift = increase in demand.
Leftward shift = decrease in demand.
| Change in… | Rightward shift (increase) if… | Leftward shift (decrease) if… |
|---|---|---|
| Income — normal good | Income rises | Income falls |
| Income — inferior good | Income falls | Income rises |
| Price of a substitute | Substitute's price rises | Substitute's price falls |
| Price of a complement | Complement's price falls | Complement's price rises |
| Tastes & preferences | Change in favour of the good | Change against the good |
Summer arrives and preferences turn towards ice cream. At every price, more ice cream is demanded. The whole curve moves right.
A study reveals cold drinks may harm health. Preferences turn against them. At every price, less is demanded. The curve moves left.
Ask: did the price of this good itself change? If yes → movement along the curve. If no → shift of the curve. Nothing else matters.
For the market for tea, state whether there is a movement along the demand curve or a shift — and in which direction:
Only item 1 involved the price of tea itself — and only item 1 is a movement. Everything else shifts the curve.
The price of a good rises, and the quantity bought also rises.
A student says this disproves the Law of Demand.
Give three different explanations that are each fully consistent with the Law of Demand, and one that genuinely is an exception. For each, say whether it involves a movement along or a shift of the demand curve.
The student has forgotten the phrase "other things remaining equal". The Law of Demand describes a movement along a demand curve. If something else also changed, the curve itself moved, and what we observe is the net effect of two separate changes.
1 · Income rose (normal good). Higher income shifts the demand curve rightward. If that shift outweighs the price rise, the quantity bought rises. → shift (plus a movement along the new curve).
2 · The price of a substitute rose by more. If coffee rose sharply, demand for tea shifts rightward, and more tea is bought despite tea's own price rising. → shift.
3 · Tastes changed in the good's favour. A favourable report, a change of season, a fashion — the curve shifts rightward. → shift.
In all three, the demand curve is still downward sloping. We simply observed two points on two different curves and mistook them for two points on one.
The genuine exception — a Giffen good. For a strongly inferior good, a price rise makes the consumer poorer in real terms, and because the good is inferior, that income effect pushes her to buy more of it. If the income effect outweighs the substitution effect, quantity demanded rises with price. → movement along an upward-sloping demand curve.
This is the crucial distinction. In cases 1–3 the law holds and something else moved. In the Giffen case the law itself fails — the demand curve is genuinely positively sloped.
The methodological lesson. To test the Law of Demand you cannot simply observe price and quantity over time — too many things move at once. You must isolate the effect of price with everything else held constant. This is why "ceteris paribus" is not a formality but the substance of the claim.
Market demand for a good at a particular price is the total demand of all consumers in the market taken together at that price.
Fix a price. Add up the quantities. Repeat for every price.
We add horizontally (quantities, measured along the x-axis) — never vertically. Prices are not added; the price is the same for everyone.
The identical technique is used in Chapter 4 to get the market supply curve from individual firms' supply curves. Learn it once, use it twice.
Two consumers: d₁(p) = 10 − p and d₂(p) = 15 − p.
For p ≤ 10 (both buy): (10 − p) + (15 − p) = 25 − 2p
For 10 < p ≤ 15 (only consumer 2 buys): 15 − p
For p > 15 (nobody buys): 0
Check at p = 10: d₁ = 0, d₂ = 5, market = 5. And 25 − 2(10) = 5. ✓
(a) Two consumers:
d₁(p) = 20 − p for p ≤ 20, and 0 above;
d₂(p) = 30 − 2p for p ≤ 15, and 0 above.
Find the market demand function.
(b) There are 20 identical consumers, each with
d(p) = 10 − 3p for p ≤ 10/3, and 0 above. Find market demand.
(a) Consumer 2 leaves the market above p = 15; consumer 1 above p = 20. So there are three price ranges:
For 0 ≤ p ≤ 15 (both buy):
(20 − p) + (30 − 2p) = 50 − 3p
For 15 < p ≤ 20 (only consumer 1): 20 − p
For p > 20: 0
Check at p = 15: d₁ = 5, d₂ = 0, market = 5. And 50 − 3(15) = 5. ✓
(b) All 20 consumers are identical, so simply multiply one consumer's demand by 20:
D(p) = 20 × (10 − 3p) = 200 − 60p for 0 ≤ p ≤ 10/3
= 0 for p > 10/3
Two consumers. Compute the market demand at each price.
| Price (₹) | d₁ | d₂ | Market demand |
|---|---|---|---|
| 1 | 9 | 24 | ? |
| 2 | 8 | 20 | ? |
| 3 | 7 | 18 | ? |
| 4 | 6 | 16 | ? |
| 5 | 5 | 14 | ? |
| 6 | 4 | 12 | ? |
Add the two quantities at each price (horizontal summation):
| Price (₹) | d₁ | d₂ | Market demand |
|---|---|---|---|
| 1 | 9 | 24 | 33 |
| 2 | 8 | 20 | 28 |
| 3 | 7 | 18 | 25 |
| 4 | 6 | 16 | 22 |
| 5 | 5 | 14 | 19 |
| 6 | 4 | 12 | 16 |
Market demand falls as price rises — the market demand curve, like the individual ones, is downward sloping.
The Law of Demand says quantity moves in the opposite direction to price. It says nothing about by how much.
Price doubles. You still buy roughly the same amount. Demand barely responds.
Price rises 10%. Buyers switch to a rival brand. Demand collapses.
Both obey the Law of Demand. But they behave completely differently — and that difference decides whether a shopkeeper who raises prices earns more or less.
Price elasticity of demand is a measure of the responsiveness of the quantity demanded of a good to a change in its price.
Writing the percentages out:
Because price and quantity move in opposite directions, eD is always negative. For convenience we conventionally report its absolute value and drop the minus sign.
It is a ratio of two percentages, so the units cancel out. It does not matter whether quantity is in kilograms or dozens, or price in rupees or dollars. This is what makes elasticities comparable across goods.
An individual buys 15 bananas when the price is ₹5. When the price rises to ₹7, she reduces her demand to 12 bananas.
| Price (₹) | Quantity |
|---|---|
| Old: P₁ = 5 | Old: Q₁ = 15 |
| New: P₂ = 7 | New: Q₂ = 12 |
(Q₂ − Q₁)/Q₁ × 100
= (12 − 15)/15 × 100
= −3/15 × 100 = −20%
(P₂ − P₁)/P₁ × 100
= (7 − 5)/5 × 100
= 2/5 × 100 = +40%
|eD| = 0.5 < 1, so demand for bananas is inelastic — a 40% price rise caused only a 20% fall in demand. Quantity responded proportionately less than price.
| Value of |eD| | Name | Meaning | Typical goods |
|---|---|---|---|
| eD = 0 | Perfectly inelastic | Quantity does not change at all when price changes | Life-saving medicine (idealised) |
| 0 < eD < 1 | Inelastic | %ΔQ less than %ΔP | Salt, foodgrains, necessities |
| eD = 1 | Unitary elastic | %ΔQ equals %ΔP | — |
| eD > 1 | Elastic | %ΔQ greater than %ΔP | Luxuries, branded goods |
| eD = ∞ | Perfectly elastic | Any rise in price drops demand to zero | Idealised competitive market |
Compare the two percentages. Quantity responds more than price → elastic. Quantity responds less than price → inelastic. Nothing more.
Whatever the price, demand stays at q̄. Price never changes quantity, so |e| = 0.
At p̄ any quantity is demanded; at any other price, demand is zero. |e| = ∞.
p × q = constant. Any % change in price causes an equal % change
in quantity. |e| = 1 everywhere.
A straight-line demand curve has a constant slope. Students therefore assume it has constant elasticity. It does not.
Slope is Δq/Δp — an absolute change. Elasticity is a ratio of
percentage changes, so it depends on the p and q you start from. Same slope,
different starting point, different elasticity.
For the linear demand curve q = a − bp, we have Δq/Δp = −b.
Substituting into (2.16b):
The −b is fixed. The ratio p/q is not —
and it is what drives elasticity.
| Position on the curve | p / q | Elasticity |
|---|---|---|
| At the price axis (q = 0) | infinite | e = ∞ |
| Above the midpoint | high | e > 1 (elastic) |
| At the midpoint (p = a/2b) | — | e = 1 (unitary) |
| Below the midpoint | low | e < 1 (inelastic) |
| At the quantity axis (p = 0) | 0 | e = 0 |
The elasticity of demand at any point on a straight-line demand curve equals
Chapter 4 uses the very same geometric trick for the price elasticity of
supply, measuring Mq₀/Oq₀ along a straight-line supply curve.
Necessities (salt, foodgrains, basic medicine) → demand is inelastic. You must buy them whatever the price.
Luxuries (jewellery, foreign holidays, branded goods) → demand is elastic. You can simply do without.
Close substitutes available → elastic. If one brand of pulses gets dearer, switch to another.
No close substitutes → inelastic. There is no substitute for salt.
Demand for food in general is inelastic — you must eat.
But demand for one particular variety of pulses is elastic — if its price
rises you switch to another variety.
The more narrowly a good is defined, the more elastic its demand, because narrower definitions mean more substitutes.
Share of income spent on the good (a larger share → more elastic); whether consumption can be postponed; number of uses the good has; and the time period allowed for adjustment (demand is more elastic in the long run).
This is where elasticity earns its keep. A shopkeeper raises his price — does his revenue rise or fall? Elasticity alone decides.
When price rises, P goes up but Q goes down. Expenditure is their product, so the outcome depends on which change is proportionately larger — and that is exactly what elasticity measures.
| # | Price | Quantity | %ΔP | %ΔQ | Expenditure | Elasticity |
|---|---|---|---|---|---|---|
| 1 | ↑ | ↓ | +10 | −8 | ↑ rises | Inelastic |
| 2 | ↑ | ↓ | +10 | −12 | ↓ falls | Elastic |
| 3 | ↑ | ↓ | +10 | −10 | no change | Unit elastic |
| 4 | ↓ | ↑ | −10 | +15 | ↑ rises | Elastic |
| 5 | ↓ | ↑ | −10 | +7 | ↓ falls | Inelastic |
| 6 | ↓ | ↑ | −10 | +10 | no change | Unit elastic |
Expenditure moves in the OPPOSITE direction to price.
Price ↑ → expenditure ↓
Price ↓ → expenditure ↑
Quantity wins the tug of war.
Expenditure moves in the SAME direction as price.
Price ↑ → expenditure ↑
Price ↓ → expenditure ↓
Price wins the tug of war.
Expenditure is UNCHANGED.
The two changes cancel exactly.
A draw.
Example: price rises 10%, e = −0.2 ⇒ %ΔQ = −0.2 × 10 = −2%.
So %ΔE ≈ 10 + (−2) = +8%. Expenditure rises — as expected for an inelastic good.
To raise revenue: raise the price of an inelastic good, lower the price of an elastic good. Getting this backwards costs money.
(a) At price ₹4 the demand is 25 units. Price rises to ₹5 and demand falls to 20 units. Calculate the price elasticity of demand.
(b) For the demand curve D(p) = 10 − 3p, find the elasticity at
p = 5/3.
(a) %ΔQ = (20 − 25)/25 × 100 = −20%
%ΔP = (5 − 4)/4 × 100 = +25%
Since 0.8 < 1, demand is inelastic at this price.
(b) Here a = 10, b = 3. First find the quantity:
q = 10 − 3(5/3) = 10 − 5 = 5
|eD| = 1 — this is the midpoint of the demand curve.
Check: midpoint price = a/2b = 10/6 = 5/3 ✓
(a) Price elasticity of demand is −0.2. If price rises by 5%, by what percentage does demand fall?
(b) Price elasticity is −0.2. How is expenditure affected by a 10% price rise?
(c) A 4% decrease in price caused expenditure to increase by 2%. What can you say about elasticity?
(a) From e = %ΔQ / %ΔP:
%ΔQ = e × %ΔP = (−0.2) × 5 = −1%
Demand falls by 1%.
(b) %ΔQ = (−0.2) × 10 = −2%.
%ΔE ≈ %ΔP + %ΔQ = 10 + (−2) = +8%
Expenditure rises by about 8%. This is the inelastic case — price rose, and since
quantity barely responded, expenditure moved in the same direction as price.
(c) Price fell and expenditure rose — they moved in opposite directions, so demand must be elastic (|e| > 1).
Quantitatively: %ΔQ ≈ %ΔE − %ΔP = 2 − (−4) = +6%
e = 6 / (−4) = −1.5 ⇒ |e| = 1.5, confirming elastic demand.
A government wants to reduce cigarette smoking and also to raise revenue from the tobacco tax. Demand for cigarettes is known to be price inelastic in the short run and more elastic in the long run.
(a) If it raises the price by 20%, what happens to consumption and to total expenditure on
cigarettes in the short run? Use the elasticity–expenditure rule.
(b) Can it achieve both goals at once in the short run? In the long run?
(c) A colleague proposes taxing a good with elastic demand instead, "because people will
cut back more". Evaluate this as revenue policy.
(a) Demand is inelastic, so |e| < 1 and %ΔQ is smaller than %ΔP. Say
e = −0.4: a 20% price rise cuts consumption by only 0.4 × 20 = 8%.
By the elasticity–expenditure rule, for an inelastic good expenditure moves in the same direction as price:
%ΔE ≈ %ΔP + %ΔQ = 20 + (−8) = +12%
Consumption falls 8%; total spending on cigarettes rises 12%.
(b) Yes in the short run — and this is exactly why tobacco is taxed so heavily. Consumption falls (the health goal) while revenue rises (the fiscal goal). Inelastic demand is what makes both possible simultaneously.
In the long run the tension appears. Demand becomes more elastic — people have time to quit, switch, or never start. As |e| approaches and passes 1, the quantity response grows and the revenue gain shrinks; beyond |e| = 1, further tax rises reduce revenue.
So the two goals are compatible in the short run but increasingly in conflict over time. A tax that succeeds completely at its health objective destroys its own tax base.
(c) The colleague has it exactly backwards for revenue purposes.
With elastic demand, expenditure moves opposite to price. Raising the price of an elastic good makes consumers cut back more than proportionately, so total spending — and hence the tax base — falls. It is a poor revenue source.
The general rule for revenue: tax goods with inelastic demand — necessities, goods with few substitutes, addictive goods. This is precisely why salt, fuel, alcohol and tobacco have been taxed by governments for centuries.
But note the normative sting. "Inelastic demand" often means "necessity", and taxes on necessities fall hardest on the poor, who spend a larger share of income on them. The positive analysis says such a tax raises revenue efficiently; whether it is fair is a separate, normative question — the Chapter 1 distinction again.
Starting from nothing but "a person can rank bundles" and "she has limited money", you have derived the market demand curve — a real, usable object that will determine the price of a commodity two chapters from now.
1. What do you mean by the budget set of a consumer?
2. What is a budget line?
3. Explain why the budget line is downward sloping.
1. The budget set is the collection of all bundles that a consumer
can buy with her income at the prevailing market prices — every bundle
(x₁, x₂) satisfying p₁x₁ + p₂x₂ ≤ M. It includes bundles costing
less than her income as well as those costing exactly her income.
2. The budget line consists of all bundles that cost the consumer
exactly her entire income: p₁x₁ + p₂x₂ = M. It is the boundary of the
budget set, with vertical intercept M/p₂, horizontal intercept
M/p₁ and slope −p₁/p₂.
3. Every bundle on the budget line costs exactly M. If the consumer wants more of one good, she must spend more on it — and since her income is fixed and fully spent, she must spend less on the other good, and so buy less of it. An increase in one quantity therefore necessarily accompanies a decrease in the other, which is precisely what a downward slope means.
Formally: the slope is −p₁/p₂. Since both prices are
positive, the slope is negative.
10. What do you mean by 'monotonic preferences'?
21. Explain price elasticity of demand.
10. A consumer's preferences are monotonic if and only if, between any two bundles, the consumer prefers the bundle which has more of at least one of the goods and no less of the other good as compared to the other bundle. In short: more is better.
Monotonicity is what makes an indifference curve slope downward, makes higher indifference curves represent greater satisfaction, and forces the consumer's optimum to lie on (not below) the budget line.
21. Price elasticity of demand measures the responsiveness of the quantity demanded of a good to a change in its price. It is defined as the percentage change in demand divided by the percentage change in price:
It is negative (price and quantity move oppositely) but conventionally reported as an absolute value, and it is a pure number, independent of units. Demand is elastic if |e| > 1, inelastic if |e| < 1 and unitary elastic if |e| = 1.
17. What do you mean by a normal good?
18. What do you mean by an inferior good? Give some examples.
19. What do you mean by substitutes? Give an example.
20. What do you mean by complements? Give an example.
17. A normal good is one for which demand increases as the consumer's income increases, and decreases as income decreases — demand moves in the same direction as income. Examples: milk, branded clothing, cars.
18. An inferior good is one for which demand decreases as the consumer's income increases, and increases as income decreases — demand moves in the opposite direction to income. Examples: coarse cereals, low-quality food items, second-hand clothing.
A good may be normal at low incomes and inferior at higher incomes.
19. Substitutes are goods that can be used in place of each other to satisfy the same want. The demand for a good moves in the same direction as the price of its substitute. Example: tea and coffee — if the price of coffee rises, consumers shift to tea, so the demand for tea rises.
20. Complements are goods that are consumed together to satisfy a want. The demand for a good moves in the opposite direction to the price of its complement. Example: tea and sugar — if the price of sugar rises, the demand for tea falls.
We have one blade of the scissors — the demand curve. Chapter 3 crosses to the other side of the market to ask how a firm turns inputs into output and what that costs. Chapter 4 turns those costs into the supply curve. Chapter 5 brings the two blades together.
Watch for old friends in new clothes: marginal product behaves exactly like marginal utility, the isoquant is an indifference curve, and market supply is built by the same horizontal summation you learnt here.