Indifference curve analysis derives consumer behaviour from axioms about preferences rather than assuming the law of demand from the outset. By splitting a price change into an income effect and a substitution effect, the model shows exactly WHEN demand curves slope downward and when (in extreme cases) they do not. The claim is that inferior and Giffen goods — the exceptions — showcase the model's analytical power because they are precisely the cases that cannot be handled by simple marginal utility reasoning. This discussion assesses that claim.
Theoretical derivation. A fall in Px rotates the budget line outward from BL1 to BL2. Equilibrium shifts from A on IC1 to C on IC2. Construct the auxiliary line BL′ parallel to BL2 and tangent to IC1 at B. The move A → B is the substitution effect — ALWAYS POSITIVE for the good whose price has fallen, because the consumer switches toward the relatively cheaper good along a given indifference curve. The move B → C is the income effect — positive for a normal good, negative for an inferior good.
Total effect = SE + IE. Three cases:
(i) Normal good: SE > 0, IE > 0, TE > 0 (strongly positive). Demand curve downward-sloping.
(ii) Ordinary inferior good: SE > 0, IE < 0, |SE| > |IE|, so TE > 0 (modestly). Demand curve still downward-sloping but less elastic than for a normal good.
(iii) Giffen good (extreme inferior): SE > 0, IE < 0, |IE| > |SE|, so TE < 0. Demand curve upward-sloping.
For the claim. The IC model is the ONLY standard framework that generates these three cases from the same primitives. Simple MU analysis cannot deliver case (iii) — diminishing MU of a single good alone cannot produce an upward-sloping demand curve. By identifying the Giffen case as the point where the (negative) income effect outweighs the (positive) substitution effect, the IC model gives an exact theoretical criterion that empirical work can test. Jensen and Miller's 2008 field study found Giffen behaviour among very poor Chinese households for subsidised staple foods (rice in Hunan, wheat in Gansu) — exactly where the model predicts it should emerge, because the staple took a large share of the household budget (large negative IE) and real substitutes were scarce (small SE). The theory ALSO explains why Giffen responses are rare in developed economies: there, inferior goods take only a small share of the budget, so |IE| is small; and substitutes exist, so |SE| is large. In this sense the exceptions do indeed 'prove' the rule — they confirm the predictions of the IC model in the exact circumstances it specifies.
Against the claim. The claim overstates the case in several ways. First, the Giffen result is NOT a counter-example to the model, so calling it an 'exception' is misleading — it is a direct implication of the IC framework. Second, the IC model's success here is PURELY QUALITATIVE: it tells us the sign of the effect but not its magnitude, so empirical elasticities remain necessary for policy work. Third, there are OTHER cases where the IC model is less successful: it cannot explain Veblen goods (where high price itself drives demand through conspicuous consumption), speculative demand, or quality signalling under asymmetric information. Fourth, behavioural economics shows that observed demand often departs from IC predictions for reasons unrelated to inferiority — framing, loss aversion, mental accounting — and these cases do NOT prove the usefulness of the IC model; they expose its limits.
Further considerations. The policy relevance of the Giffen result is substantial. A subsidy on an inferior staple (bread, cassava, rice) may REDUCE consumption of the staple because the subsidy raises real income enough for consumers to switch to preferred alternatives. Food-security interventions that assume 'cheaper → more consumed' can therefore mis-target the poorest households. This real-world consequence of the IC model's logic is more consequential than the theoretical neatness of the three-case taxonomy.
Judgement. The claim is PARTLY correct. Inferior and Giffen goods DO showcase the IC model's power to generate counter-intuitive but theoretically grounded predictions, and the Jensen-Miller evidence confirms that the model identifies the right conditions for Giffen behaviour. But the exceptions do not stand alone — they live inside the model's framework rather than disproving it, and the model has important limitations elsewhere (Veblen, behavioural, static assumptions). The IC model is best regarded as a useful qualitative framework for understanding when and why consumer behaviour departs from the standard downward-sloping demand curve, complemented by empirical elasticities and behavioural evidence for quantitative policy work. On that reading, the inferior/Giffen cases are a strength of the model, but neither its only strength nor a sufficient defence of its wider use.