Demand System Estimation
Also known as: Consumer Demand System, System of Demand Equations, Complete Demand System, Demand System Analysis
Demand system estimation jointly models how a consumer or population allocates a budget across a complete set of goods, estimating a system of equations — one per good — that relate each good's expenditure share or quantity to all prices and total expenditure. Unlike a single-equation demand curve, a demand system imposes the cross-equation restrictions implied by consumer theory: adding-up (shares sum to the budget), homogeneity (no money illusion), and Slutsky symmetry (consistency of cross-price effects). Classic functional forms include Stone's Linear Expenditure System, the Rotterdam model, and the Almost Ideal Demand System, and the system is estimated with seemingly unrelated regression or full-information methods.
Key highlights
- Imposes and tests the restrictions of consumer theory (adding-up, homogeneity, symmetry), yielding internally coherent elasticities suitable for welfare analysis.
- Captures cross-good substitution and complementarity through a full matrix of price responses, which single-equation demand models cannot.
- Joint estimation exploits the correlation of errors across equations, improving efficiency over equation-by-equation regression.
- Flexible functional forms (Rotterdam, AIDS, QUAIDS) can approximate a wide range of preferences and support exact welfare and cost-of-living measures.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use demand system estimation when you need a complete, theory-consistent picture of how consumers reallocate spending across many goods in response to prices and income — for computing own- and cross-price elasticities, evaluating the welfare and distributional effects of taxes and subsidies, or modeling substitution in policy and indexation analysis. It is the right tool when cross-good substitution and the budget constraint matter, when you have price and budget-share data across households or over time, and when you want to test or impose homogeneity and symmetry. It is less suitable when goods are highly disaggregated and differentiated with many zero purchases (where discrete-choice demand models are preferred), when only a single good is of interest, or when data lack the price variation needed to identify cross-price effects. Functional-form choice (LES, Rotterdam, AIDS, QUAIDS) should match the flexibility and aggregation properties the application requires.
Strengths & limitations
- Imposes and tests the restrictions of consumer theory (adding-up, homogeneity, symmetry), yielding internally coherent elasticities suitable for welfare analysis.
- Captures cross-good substitution and complementarity through a full matrix of price responses, which single-equation demand models cannot.
- Joint estimation exploits the correlation of errors across equations, improving efficiency over equation-by-equation regression.
- Flexible functional forms (Rotterdam, AIDS, QUAIDS) can approximate a wide range of preferences and support exact welfare and cost-of-living measures.
- Requires good price variation and reliable budget-share data; with few goods or collinear prices, cross-price effects are poorly identified.
- Imposing a global functional form can force restrictive substitution patterns and may be rejected by the data (homogeneity and symmetry are often statistically rejected).
- Aggregation across heterogeneous households can bias estimates unless the form admits exact aggregation or demographics are modeled.
- Many disaggregated goods and frequent zero purchases (corner solutions) violate the interior-solution assumption underlying share systems.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
Why estimate a system of equations instead of separate demand curves?
Because a consumer's choices across goods are linked by a single budget and a single set of preferences. Estimating separate demand curves ignores that what is not spent on one good is spent on others (adding-up), that demand should be free of money illusion (homogeneity), and that cross-price effects must be mutually consistent (Slutsky symmetry). A system imposes these links, exploits the correlation of errors across goods for efficiency, and delivers a coherent matrix of elasticities suitable for welfare and policy analysis.
What are adding-up, homogeneity, and symmetry, and are they always imposed?
Adding-up requires the predicted budget shares to sum to one; it is structural and always built in (one equation is dropped). Homogeneity of degree zero means scaling all prices and income leaves quantities unchanged. Slutsky symmetry requires compensated cross-price effects to be symmetric. Homogeneity and symmetry are theoretical restrictions that can be either tested as hypotheses (they are frequently rejected in data) or imposed to obtain theory-consistent, more efficient estimates; practice varies by application.
Which functional form should I use — LES, Rotterdam, or AIDS?
The Linear Expenditure System is simple and globally regular but restrictive (no inferior goods, additive preferences). The Rotterdam model is flexible and well suited to testing homogeneity and symmetry directly in differenced data. The Almost Ideal Demand System and its quadratic extension QUAIDS are the modern default: flexible, consistent with exact aggregation, and yielding exact welfare measures, with QUAIDS adding curvature in Engel curves. The choice depends on whether you need flexibility, exact aggregation, nonlinear Engel curves, or a specific test.
Sources
- 1.Stone, R. (1954). Linear expenditure systems and demand analysis: an application to the pattern of British demand. The Economic Journal, 64(255), 511–527.
- 2.Deaton, A., & Muellbauer, J. (1980). An almost ideal demand system. The American Economic Review, 70(3), 312–326.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 22). Demand System Estimation. ScholarGate. https://scholargate.app/economics/demand-system-estimation