Consumer Spending

Household consumption makes up roughly 50-70% of GDP in most economies, with the US near the top of that range and China an outlier near 40%. The composition tells a sharp story: low-income households spend over 50% on food; high-income households spend under 15%. Housing absorbs 20–35% in most advanced economies. Healthcare and education have risen as shares everywhere over the past 50 years.

60%+
Consumer spending share of GDP (advanced economies)
12.9%
US household food share of spending
32.9%
US household housing share
Engel's Law
Food share falls as income rises — 19th century empirical regularity

Key insights

🍽️

Engel's Law holds robustly

Ernst Engel's 1857 finding — that food share of household budgets falls as income rises — is one of the most robustly verified regularities in economics. Nigeria spends about half of its household budget on food; the US ~13% including restaurant meals; Singapore ~7%. Within-country, the same pattern holds across income deciles. The mechanism: food consumption is roughly capped by physiology while income can grow without limit.

🏠

Housing has risen as a share

US housing has climbed to about 33% of household spending. Similar pattern across the OECD. Drivers: rising land prices in big cities; quality improvements (larger homes, better amenities); slow productivity growth in construction. Healthcare and education have risen in parallel for similar Baumol-cost-disease reasons. Manufactured goods and food have fallen as shares.

🌐

Composition convergence has limits

Chinese household consumption is shifting toward the OECD pattern as incomes rise — food share fell from 53% (1980) to 30% (2023). But composition won't fully converge: cultural preferences (dining out vs cooking), housing tenure differences (urban renting vs owning), healthcare-system design (single-payer vs out-of-pocket), and welfare-state coverage all leave durable cross-country differences in observed spending patterns.

US household budget composition (2023)

% of total household spending, BLS Consumer Expenditure Survey

Show data table
US household budget composition (2023)
% of spending
Housing32.9
Transport17
Food12.9
Healthcare8
Insurance & pensions12.4
Entertainment4.7
Education2.2
Apparel2.6
Other7.3

Key Finding: Housing and transport together account for half of household spending. Food share remains under 15%.

Food share of household spending — selected countries

% of consumption spending on food; survey concepts differ between countries

Show data table
Food share of household spending — selected countries
Food share %
Nigeria49
Bangladesh40
India30
Indonesia35
Russia29
China30
Brazil18
Japan16
Germany11
UK8
USA13
Singapore7

Key Finding: Nigeria spends nearly 50% and Bangladesh about 40%; the US, UK and Singapore under 15%. The US figure includes eating out, which lifts it above the UK's.

Methodology & caveats

Spending vs consumption

Household spending (cash outlays) differs from consumption (use of services). Imputed rent for homeowners is consumption but not spending. Employer-paid healthcare is consumption but not household spending. National accounts use 'household final consumption expenditure' which includes imputed rent. Cross-country comparisons should specify which concept is used.

Budget surveys vs national accounts

Two main sources: household budget surveys (BLS CEX in US, FES in UK, OFCS in EU) and national accounts derived from production statistics. The two often disagree — surveys typically capture 70–85% of national-accounts consumption. The gap reflects under-reporting (alcohol, tobacco), elite under-sampling, and methodology choices.

Quality adjustment over time

Comparing 'spending on food' in 1960 vs 2024 is harder than it looks. Food today includes more processing, more variety, restaurant meals. Housing includes air conditioning, internet, more square footage. Hedonic and chain-weighted indices try to adjust for this; long-run shares should be read as the broad pattern rather than precise point comparisons.

Why national figures rarely match

Two countries can report the same food share for different reasons. National-accounts totals include imputed rent on owner-occupied homes, which inflates the denominator and lowers the food share; household surveys often exclude it. Restaurant meals may be in food or in a separate hospitality line. A figure labelled "food" can therefore differ by several points with no difference in diets, which is why the USDA and BLS numbers for the US do not agree.

What would change the picture

Engel's Law predicts a falling food share as incomes rise, but price shocks can reverse it for a time. In a year of high inflation, poorer households see their food share rise because they cannot cut volumes much. Ageing works more slowly: older households spend more on health and less on transport and clothing, shifting the basket in rich countries regardless of income.

Cite this page

GlobeChart. “Consumer Spending.” Last updated 2026-10-31. https://globechart.com/economy/consumer-spending/

Please also credit the original publisher named under each chart: BLS Consumer Expenditure Survey 2023; USDA ERS, household surveys. Spotted an error? Tell us.