Income Elasticity

How income elasticity works

Imagine your salary jumps 10%. What do you buy more of? Probably a few extra restaurant meals, maybe a slightly bigger apartment, certainly more streaming subscriptions. What do you buy less of? Maybe instant ramen — once you can afford fresh pasta, the cheap stuff loses its appeal. What barely moves at all? Salt. You eat the same amount whether you make $30,000 or $300,000.

Income elasticity captures all three patterns in one number:

Y_E = (percent change in quantity demanded) ÷ (percent change in income)
    = (ΔQ / Q) ÷ (ΔY / Y)

The sign and magnitude classify the good:

     Y_E              Type                  Examples
   ─────────  ─────────────────────  ──────────────────────────────────
   Y_E > 1     Luxury / superior      Foreign vacations, fine dining,
                                       luxury cars, art, jewellery
   Y_E ≈ 1     "Unitary"               Many discretionary services
   0 < Y_E < 1  Normal necessity       Food (in aggregate), clothing,
                                       housing, electricity
   Y_E = 0     Perfectly inelastic    Salt, basic medicine, postage
   Y_E < 0     Inferior               Instant noodles, used clothing,
                                       public bus, generic store brands

Two goods can have the same own-price elasticity but completely different income elasticities — and the difference dictates which one grows when wages rise. Beef has Y_E ≈ +0.6 (normal necessity); fine wine has Y_E ≈ +1.7 (luxury). A 10% rise in real GDP grows beef demand 6% but wine demand 17%.

Worked example: a luxury watch maker in a recession

Suppose a luxury watch brand sells 40,000 units a year at $5,000 average price ($200 million revenue). Industry studies put the income elasticity of luxury watches around Y_E ≈ +2.5 — well into "luxury" territory. The economy enters recession; real disposable income falls 4%.

Predicted change in unit sales:

%ΔQ = Y_E × %ΔY
     = (+2.5) × (−4%)
     = −10%

New quantity = 40,000 × 0.90 = 36,000 watches
New revenue (same price) = $180 million

The brand loses $20 million in a 4% downturn — much steeper than the macroeconomy's slip. Operations planners building inventory for the next two years need exactly this calculation.

Now run the same recession through a grocery chain selling staple food (Y_E ≈ +0.3):

%ΔQ = (+0.3) × (−4%) = −1.2%

The grocery loses about 1.2% of unit sales — barely visible. That asymmetry is why luxury equities (LVMH, Tiffany, Ferrari) plunge in recessions while consumer staples (P&G, Coca-Cola, Walmart) hold up. Equity analysts call them "cyclicals" vs "defensives" and the Y_E number is the distinction.

Real-world income elasticities

Estimates below blend the Bureau of Labor Statistics' Consumer Expenditure Survey (cross-section, U.S.), the World Bank's International Comparison Program (cross-country), and academic surveys. Magnitudes shift across studies; the qualitative ranking is robust.

Good	Income elasticity Y_E	Classification	Notes
Salt	0.05	Necessity (near-zero)	Quantity barely moves with income
Cigarettes	0.10	Necessity / addictive	Slightly lower in higher-income strata
Food at home (aggregate)	0.20–0.40	Normal necessity	Engel's law in action
Electricity (residential)	0.30–0.50	Normal necessity	Larger homes, more appliances
Public transit	−0.36	Inferior	Quantity falls as income rises
Instant noodles	−0.20	Inferior	Replaced by fresh pasta, restaurants
Beef	0.60	Normal necessity	Higher Y_E in lower-income countries
Restaurant meals	1.40	Luxury	Cooking at home is the substitute
Foreign vacations	1.80	Luxury	Highly cyclical
Luxury watches	2.50	Strong luxury	Falls hardest in recessions
Private aviation	3.50+	Extreme luxury	Collapses in downturns

The same good can shift category across countries. Beef in the U.S. is a normal necessity (Y_E ≈ +0.6); in Vietnam, where most households still rarely eat it, beef behaves like a luxury (Y_E ≈ +1.5+). World Bank cross-country regressions show this pattern repeating for cars, air travel, and beef across nearly every developing economy.

Engel curves: the geometry of income response

An Engel curve plots quantity demanded against income, holding prices constant. The shape captures the entire income response, not just one local elasticity number.

Engel curves

   Q ↑                                       Q ↑
     |     /                                    |  ─────
     |    /  ← luxury (Y_E > 1)                  |       \
     |   /                                       |        \
     |  /                                        |         \  ← inferior
     | /     ← normal necessity (0 < Y_E < 1)    |          \      (Y_E < 0)
     |/        flatter than 45°                  |
     +────────→ Y                                +────────→ Y

Three regions show up vividly when you plot share of income (rather than absolute quantity) against income:

• Engel's law on food. The share of income spent on food falls steeply as income rises. Poor U.S. households (bottom decile) spend ~35% of income on food; the top decile spends ~8%. Across countries the same pattern holds — Burundi households spend ~62%; Singaporean households ~7%.
• Housing share is roughly flat. Y_E for housing is near 1 in most studies; the income share stays roughly constant across the distribution.
• Luxury share rises with income. Foreign travel, fine dining, and discretionary services consume a larger share of income at the top.

Engel curves can also bend: a good can be a luxury at low incomes and a necessity at high ones (basic kitchen appliances) or a normal good at low incomes and inferior at high (white bread, ground chuck). A linear elasticity number captures only the local slope; the full curve shows the bend.

Variants and decomposition

	Quantity-based Y_E	Expenditure-based Y_E
What it measures	How many units demanded	How much money spent
Formula	%ΔQ ÷ %ΔY	%Δ(P·Q) ÷ %ΔY
Captures quality upgrades?	No	Yes
Best for	Volume forecasting	Revenue forecasting

The distinction matters because rising income often shifts consumers to higher-priced versions of the same good rather than higher quantities. As wages rise, families don't necessarily eat more rice — they eat more expensive rice. Quantity-based Y_E for rice may be near zero or negative; expenditure-based Y_E is positive. Restaurant analytics, fashion retail, and grocery category management all rely on the expenditure version.

A second decomposition is the Slutsky split between substitution and income effects. Income elasticity sits on the income-effect side. The full demand response to a price change combines both: when bread prices rise, the substitution effect pushes consumers to rice; the income effect (poorer in real terms) reduces spending on every normal good. For inferior goods the income effect partly offsets the substitution effect, and for the rare Giffen good it reverses sign entirely.

What makes Y_E large or negative

• Necessity vs discretionary. The single biggest factor. Necessities (food, utilities, basic medicine) cluster around Y_E ≈ 0–0.5. Discretionary purchases (vacations, watches, art) cluster around Y_E ≈ 1.5–3.0.
• Availability of upscale substitutes. A good becomes inferior when an obviously better-and-affordable substitute exists at higher income. Public buses lose riders to Ubers; instant noodles lose buyers to fresh pasta. Without that upgrade path, low-income goods stay normal.
• Income level of the country. Cars are luxuries in Bangladesh, normal goods in the U.S. Air travel is a luxury everywhere except for the small sliver of the global income distribution where it has saturated.
• Saturation. Y_E drifts toward zero once a good has saturated. Refrigerator demand at very high incomes is inelastic — every household already has one and few buy more.
• Time horizon. Long-run income elasticities can differ from short-run ones because durable-goods purchases lag wage gains by years.

Common pitfalls

• Confusing income elasticity with own-price elasticity. Both can be reported as a single number but they measure responses to different drivers. A good can be inelastic to price (PED ≈ −0.3) while being elastic to income (Y_E ≈ +2.0).
• Using a single Y_E across the income distribution. Income elasticity is local. Cars at the median U.S. income are normal goods; at $30,000 of household income they behave like luxuries; at $1 million they're saturated. The Engel curve bends.
• Forgetting to deflate for prices. "Income" in the formula means real income — nominal-income changes that come from inflation should not move quantity demanded the way real-income changes do. Empirical work that fails to deflate confounds price elasticity with income elasticity.
• Aggregation bias. Aggregate Y_E is a population-weighted average; it can hide opposing responses from different income groups. Cigarettes are mildly inferior in U.S. data — but only because the high-income group cuts back sharply while the low-income group barely moves.
• Mistaking quality upgrades for quantity changes. When wages rise, consumers often buy fancier versions of the same good rather than more units. Quantity-based Y_E and expenditure-based Y_E will diverge sharply. Choose the one that matches the question (volume planning vs revenue forecasting).
• Sign confusion for Giffen and inferior goods. An inferior good (Y_E < 0) is not the same as a Giffen good (price elasticity > 0). Giffen goods are inferior, but most inferior goods are not Giffen — the income effect has to be large enough to flip the price response, which is rare.