When a price changes, demand doesn't just move — it splits. The substitution effect is the shift between goods caused by changed relative prices: as one good becomes cheaper relative to others, you substitute toward it. The income effect is the change in quantity demanded because your real purchasing power changed: a price cut makes you effectively richer, so you may buy more of that good or reallocate across your basket. The total observed change in quantity is the sum of both. For normal goods the two reinforce each other: a price fall raises quantity via substitution (switch into this good) and via income (you have more real income). For inferior goods the income effect opposes the substitution effect: a price fall still tempts substitution toward the good, but your higher real income may push you toward superior alternatives. When the income effect dominates, you get the backward-bending labour-supply curve or a Giffen good — quantity demanded falls when price falls.
John Hicks and Roy Allen formalised the decomposition in the 1930s using indifference curves and compensated demand. The practical payoff: you can't interpret a demand response without asking how much is "switch because relative prices changed" versus "spend differently because I feel richer or poorer." Tax policy, minimum wages, and pricing strategy all depend on which effect dominates for which segment. A wage increase can reduce hours worked if the income effect is large — workers hit their target income and take more leisure. A subsidy on a staple food can reduce consumption of that staple among the poorest if the income effect pushes them toward slightly better substitutes. The same price move can have opposite effects in different populations.
Labour supply is the classic application. A higher wage makes work more attractive relative to leisure (substitution effect: work more) but also raises real income (income effect: consume more leisure if leisure is normal). For low wages the substitution effect usually dominates — people work more when the wage rises. For high wages the income effect can dominate — the backward-bending supply curve. Tax policy has the same structure: cutting marginal tax rates has a substitution effect (work more; keep more of the next dollar) and an income effect (you're richer; work less). The Laffer curve and the debate over revenue impact turn on the relative size of these effects. In consumer demand, Giffen goods are the edge case: a price rise for an inferior staple (e.g. rice for very poor households) can increase quantity demanded because the income effect — "I'm poorer, I substitute toward the cheapest staple" — outweighs the substitution effect away from the now-costlier good.
Section 2
How to See It
Look for settings where a single price or wage change triggers a response that seems to run against "cheaper means more." When people work less after a raise, buy less of a good after a subsidy, or shift into a product whose price went up, you're seeing income and substitution effects at work. The diagnostic: decompose. Would this person have changed behaviour if we held their real income constant and only changed relative prices? That's the substitution effect. The rest is the income effect.
Business
You're seeing Income & Substitution Effects when a retailer drops the price of its private-label cereal and sees a large shift from national brands. Part of the volume gain is substitution — the relative price of the store brand fell. Part is income — the same basket now costs less, so some customers feel richer and buy more cereal overall or reallocate to other categories. Pricing and shelf placement are bets on which effect dominates. Premium positioning targets customers for whom the income effect pushes toward quality; value positioning targets those for whom substitution dominates.
Technology
You're seeing Income & Substitution Effects when a SaaS company bundles a previously paid feature into the base plan. Existing users don't pay more, but their effective "price" for that feature fell to zero. Substitution: they use the bundled feature instead of alternatives. Income: their real budget for tools is unchanged, but the value they get per dollar rises — they may add more seats or upgrade elsewhere. Churn and expansion metrics mix both. Ignoring the decomposition leads to misattribution: you might credit "better product" for gains that are mostly income-effect reallocation.
Investing
You're seeing Income & Substitution Effects when rising energy prices crush demand for discretionary spending. The income effect dominates for many households: the same nominal income buys less, so they cut back on restaurants, travel, and non-essentials. Substitution shows up in shifts from gas-guzzlers to efficiency and from in-person to remote. Sector allocation depends on whether you're betting on income-effect pain (discretionary) or substitution-effect gains (substitutes for the thing that got expensive).
Markets
You're seeing Income & Substitution Effects when a minimum-wage increase is followed by a mix of more hours (some workers enter or increase effort) and fewer hours (some employers cut jobs; some workers choose more leisure). The net result depends on the relative size of substitution effects (labour vs leisure, or labour vs automation) and income effects (higher wage → target income with fewer hours). Cross-country and cross-segment variation in labour-supply elasticity is largely a story of different income and substitution effect magnitudes.
Section 3
How to Use It
Decision filter
"Before interpreting a demand or labour response to a price or wage change, ask: how much is substitution (relative prices) versus income (purchasing power)? The mix determines who gains, who loses, and how durable the response is."
As a founder
Price and product changes shift both relative prices and effective income. A discount doesn't only pull share from substitutes; it makes the customer's budget go further, which can raise total spend or shift it to other categories. Design pricing experiments so you can separate the two: e.g. same discount in a context where "real income" is held constant (one-off voucher) vs where it isn't (permanent price cut). Segment by income elasticity: for high-income users the income effect of a small price change is tiny; for budget-conscious users it can dominate and reverse the direction of response. Bundle and tier design is a bet on whether your category is normal or inferior in your segment — and on whether substitution or income drives the margin.
As an investor
Evaluate policy and macro shocks through the lens of income vs substitution. Subsidies and taxes change both relative prices and real incomes. The net effect on a company depends on its product's place in the basket: necessity vs luxury, and complement vs substitute to the thing that was taxed or subsidised. Labour-intensive businesses facing wage mandates are exposed to both substitution (automation, offshoring) and income effects (workers may supply less labour). Model which effect dominates in the relevant segment and horizon.
As a decision-maker
Use the decomposition when setting or reacting to prices, wages, or taxes. A "perverse" response — less quantity when price falls, or fewer hours when wage rises — is usually the income effect outweighing the substitution effect. That's not irrational; it's structural. Design interventions with the split in mind: if you want to encourage a behaviour, strengthen the substitution effect (make alternatives relatively costlier) or offset a negative income effect (e.g. earned income tax credit alongside minimum wage). Avoid one-size-fits-all predictions; the same policy can have opposite signs in different populations.
Common misapplication: Treating every demand response as if it were pure substitution. When real incomes change, people reallocate across the whole basket. A price cut that "should" raise quantity can lower it if the good is inferior and the income effect is large. Similarly, attributing all labour-supply changes to wages ignores that people also respond to wealth and target income — the income effect can make labour supply backward-bending.
Second misapplication: Assuming the decomposition is stable. Income and substitution effects vary by segment, initial income, and time horizon. Short-run substitution is often small (habits, contracts); long-run substitution is larger. The same policy can look benign in one period and perverse in another as effects unfold.
Amazon's pricing and Prime strategy implicitly exploit the income–substitution split. Lower prices and free shipping change both relative prices (substitute toward Amazon vs other retailers) and effective income (savings reallocate to more purchases or other categories). Prime locks in substitution toward Amazon while the membership fee and basket size target customers for whom the income effect of delivery "savings" reinforces loyalty. Bezos's focus on long-term customer value over short-term margin fits a world where income and substitution effects compound over time.
Netflix's pricing and content bets assume that for most subscribers, streaming is a normal good: as real income or perceived value rises, consumption and willingness to pay rise. Price increases test the balance: substitution (to other streamers or activities) vs income (higher price reduces "real" entertainment budget). Hastings's emphasis on engagement and retention is a bet that the substitution effect away from Netflix stays small relative to the income effect and habit — and that relative to alternatives, Netflix keeps winning the substitution game.
Section 6
Visual Explanation
Income and substitution effects: a price fall splits into substitution (move along the indifference curve toward the cheaper good) and income (shift to a higher curve as real income rises). For normal goods both increase quantity; for inferior goods the income effect can oppose the substitution effect.
Section 7
Connected Models
Income and substitution effects sit inside demand theory and link to supply–demand, elasticity, and marginal reasoning. The models below either reinforce the decomposition (supply and demand, elasticity), extend it to related goods (complements and substitutes), or clarify how to use it (marginal cost/benefit, utility). Tension arises when incentives or norms push behaviour away from the clean income–substitution prediction.
Reinforces
Supply and Demand
Supply and demand determine equilibrium price and quantity. Income and substitution effects determine how quantity demanded moves when price or income changes — i.e. the shape and shift of the demand curve. A supply shock that raises price triggers both substitution away from the good and (for normal goods) an income effect that further reduces demand. The two models are inseparable for predicting market outcomes.
Reinforces
Elasticity
Elasticity measures the total responsiveness of quantity to price (or income). That total response is exactly the sum of the income and substitution effects. High substitutability amplifies the substitution effect and often makes demand more elastic. Income effects can make demand less elastic for necessities (small income effect) or create perverse elasticity for inferior goods. Decomposing into income and substitution explains why elasticity varies across goods and segments.
Tension
Revealed Preference
Revealed preference infers preferences from observed choices. Income and substitution effects assume stable, well-defined preferences that generate those choices. When behaviour seems inconsistent with the standard decomposition — e.g. framing or context changes that flip choices — revealed preference can still be salvaged, but the clean split into "income" and "substitution" may blur. The tension is between parsimonious structural interpretation and behavioural complexity.
Tension
Section 8
One Key Quote
"When the price of a good changes, the consequent change in demand can be split into two parts: one that would occur if the consumer were compensated so as to keep his level of satisfaction unchanged (the substitution effect), and one that reflects the change in his real income (the income effect)."
— John Hicks, Value and Capital (1939)
The quote is the decomposition. Hicks's "compensated" move isolates substitution; the residual is income. That split is what allows economists and operators to predict when a price cut will raise demand, when it might lower it, and when a wage increase will reduce hours worked. The practitioner's job: identify the relevant segment, judge whether the good (or labour) is normal or inferior, and size the two effects.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Most demand and labour responses are a mix of two forces. The substitution effect (relative prices) and the income effect (purchasing power) together determine the sign and size of the response. Ignoring the split leads to wrong predictions: e.g. assuming every price cut increases quantity, or every wage increase increases labour supply. The discipline is to decompose.
Segment matters. The same price or wage change can have opposite net effects in different populations. For some, the good is normal and income and substitution reinforce each other. For others, the good is inferior or the income effect is large enough to dominate — backward-bending labour supply, or less consumption after a subsidy. One-size-fits-all policy or pricing fails when the mix of effects differs by segment.
Time horizon matters. Short-run substitution is often limited by habits, contracts, and information; income effects can show up quickly. Long-run substitution is larger as people adjust. The same intervention can look benign in the short run and perverse in the long run (or the reverse) as effects unfold. Match your decomposition to the horizon of the decision.
Use it in pricing and policy. Price and wage changes are bets on the relative size of income and substitution effects. Design experiments and policies so you can tell which effect is driving results. That improves pricing, labour strategy, and regulatory prediction — and avoids blaming "irrationality" when the outcome is structural.
Inferior goods and labour supply are the high-leverage applications. The counterintuitive cases — less consumption after a subsidy, fewer hours after a wage rise — are where the model pays off. They're not bugs; they're the income effect dominating. When you see a "perverse" response, decompose. The same logic that explains Giffen goods and backward-bending labour supply explains segment-level and policy-level surprises in real data.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A city raises the minimum wage. Some workers increase hours; others reduce hours or drop out of the labour force. Employers cut some jobs and automate others.
Scenario 2
A grocery chain cuts the price of its store-brand bread. Sales of store-brand bread rise; sales of national-brand bread fall; total bread category sales rise slightly.
Scenario 3
After a fuel subsidy is introduced, poor households buy less of the subsidised staple and more of a slightly better substitute. Total calorie intake stays similar.
Scenario 4
A SaaS company raises prices 20%. Churn rises in the low-tier plan but stays flat in the enterprise plan. Revenue per customer rises in both segments.
Section 11
Top Resources
The decomposition comes from microeconomic demand theory. Hicks and Slutsky are the canonical sources; modern textbooks make the mechanics accessible. Applied work on labour supply and consumer demand illustrates how to estimate and use the split. For operators, the main takeaway is to stop interpreting demand and labour responses as single numbers — and to ask, for each segment and horizon, whether substitution or income is driving the result. Policy and pricing that assume "price down, quantity up" or "wage up, hours up" will misfire when the income effect dominates or when the good is inferior.
Hicks's treatment of the compensated demand curve and the income–substitution decomposition. The reference for holding utility constant and defining the substitution effect.
Applications to labour supply: backward-bending supply, wage and tax policy. Clear treatment of income and substitution effects in hours and participation.
Accessible chapter on demand decomposition: substitution and income effects, Slutsky equation, Giffen goods. Good first pass before graduate texts.
Mental Accounting
Mental accounting partitions money into non-fungible buckets. That can make "real income" and "relative price" subjective: a bonus might be spent differently from a wage raise; a discount might be treated as a gain to save rather than to spend. The income–substitution decomposition assumes fungible income. When mental accounting is strong, predicted income effects can be wrong — creating tension between the model and observed behaviour.
Leads-to
Complements & Substitutes
Complements and substitutes define the sign and size of cross-price effects. The substitution effect is about substituting between goods when relative prices change — so the pattern of complements and substitutes determines how large and in which direction the substitution effect goes. Income effects then shift the whole bundle. Understanding complements and substitutes is a step toward quantifying income and substitution effects in multi-good settings.
Leads-to
Marginal [Cost](/mental-models/cost)/Benefit
Optimal consumption and labour supply set marginal benefit equal to marginal cost. A price or wage change shifts both: the substitution effect is the direct change in relative marginal costs and benefits; the income effect is the indirect change through revised total resources and thus revised marginal valuations. Thinking at the margin makes the income–substitution split operational for decisions.
