How each income fifth splits $100 of spending
Of every $100 a US household spends, the lowest-income fifth puts $41 into housing and $11 into groceries. The top fifth puts $30 into housing, $6 into groceries — and $17 into pensions, Social Security and insurance.
Tap a category to follow it across all five income groups. Numbers inside the bars are dollars out of every $100.
Show the numbers
The bottom spends more than it earns. The top saves.
Spending is a better guide to living standards than reported income at the bottom. The lowest fifth reports $16,337 of after-tax income but spends $32,612 — twice as much. The gap is filled by drawing down savings, family help, benefits that don't count as income, and under-reported earnings; many in this group are retirees (the second-lowest tenth averages 61 years old). The top fifth spends $140,654 of its $196,794 and saves the rest.
Bottom 40% of households: 22% of all consumer spending, 13% of after-tax income. Top 20%: 39% of spending, 47% of income.
Engel's law: more money on food, a smaller share of the budget
The richest tenth spends 2.4 times as much on groceries as the poorest tenth, yet groceries fall from 12.5% to 5.5% of its budget. Eating out behaves differently: it rises almost sixfold in dollars and holds its share. Once calories are secured, food money goes to convenience and experience.
Income tenths, poorest (1) to richest (10). US, 2022.
The same law, worldwide
The ladder: for every $1 the poorest fifth spends, what does the top fifth spend?
Overall, the top fifth spends $4.31 for every $1 the bottom fifth spends. Goods below that line take a shrinking share of the budget as income rises — necessities. Goods above it take a growing share — luxuries. Two goods sit below $1: the top fifth spends fewer dollars on them than the bottom fifth does. People leave them behind.
Log scale. Top fifth average ÷ bottom fifth average, US 2022. Top-fifth households are larger and more often working-age, which lifts every ratio somewhat; the ranking is what matters.
Show the numbers
What economists measure: income elasticity
Income elasticity is the percentage change in demand for a 1% rise in income. Below 0, a good is inferior — bought less as people get richer. Between 0 and 1 it is a necessity. Above 1 it is a luxury. These are classic estimates from different studies and decades, not one dataset.
The same good changes class as countries get richer: the elasticity of demand for cereals is 0.62 in Tanzania, 0.47 in Georgia, 0.28 in Slovenia and 0.05 in the United States. Broadband has moved the same way — European demand shifted from luxury-like to necessity-like between 2010 and 2024.
Physical goods: from renting to owning, and into a car
Rent takes 16.7% of the poorest tenth's spending and 2% of the richest tenth's. Owner costs (mortgage interest, property tax, upkeep) rise the other way; the lines cross in the middle of the distribution. Car ownership climbs from 64% of the poorest households to 97%.
Income tenths, poorest (1) to richest (10). US, 2022.
Information goods: cheap to buy, unevenly used
Spending on reading — books, newspapers, magazines, e-books — is tiny at every income level, 0.1–0.2% of the budget, and the top fifth spends only 3× the bottom fifth's $71. That is below the 4.3× average, so in today's data reading behaves like a necessity, not the luxury that older estimates (elasticity 1.44 for books) suggested. Entertainment, by contrast, is a clear luxury at 6×.
The constraint on information at the bottom isn't the price of content, which is often free. It's the device, the connection and the skills to use them — the subject of the next chapter.
Digital goods: the phone is universal, the quality of access is not
Nearly everyone in the US has a phone. What income buys is a home connection, always-on use and a choice of platforms. Among adults earning under $30,000, 28% reached the internet only through a smartphone in 2024 — about a third in Pew's 2025 survey — against 4% of those earning $100,000 or more.
Pew Research Center: broadband and smartphones 2024 survey; platforms 2025; cash 2026 (middle point is $50–99k).
TikTok is the only major platform used more by the lowest-income group. Reddit, Instagram and WhatsApp rise steeply with income; Facebook is flat. Pew, 2025.
Globally, the device and the data plan are the gate
The future as a purchase: pensions, education, giving
The steepest line in the whole budget is the future. Pensions, Social Security and life insurance take 1.9% of the poorest fifth's spending and 17.4% of the richest fifth's — $636 against $24,543. Much of it is mandatory payroll contribution that scales with wages, so it reflects earnings as much as choice. Giving also climbs, from 2.7% to 4.6% of spending. Education is U-shaped: high at the bottom (likely students) and at the top.
What it adds up to
Interpretation — drawn from the data above, not measured directly
- What people value stays roughly the same across incomes; its form and price change. Security at the bottom means rent paid and food in the cupboard; at the top it means a pension and a portfolio. Everyone spends on status and belonging — the form changes.
- Necessities press hardest at the bottom. Housing (with utilities), groceries and healthcare take $62 of every $100 for the poorest fifth and $43 for the richest.
- For information and digital goods, price isn't the barrier — access is. Content is cheap or free; the handset, the connection and the skills decide who benefits.
- Higher incomes buy convenience, experience and the future. Eating out, entertainment, travel and owned homes grow faster than spending overall; retirement saving grows fastest of all.
- Spending is not the same as value. Low spending on healthcare at the bottom doesn't mean it is valued less — it means it can't be afforded. These charts show what people buy under a budget.
Sources and caveats
- US Bureau of Labor Statistics, Consumer Expenditure Surveys 2022: means and shares by income quintile, by decile, aggregate shares. Groups are "consumer units" ranked by pre-tax income; dollars are 2022 nominal.
- Income elasticities: Frank (2008) and Samuelson & Nordhaus (2001), as collected in Wikipedia, "Income elasticity of demand"; environmental WTP: Drupp et al., arXiv 2308.04400 (735 estimates, 396 studies); broadband: Orujov et al., arXiv 2512.21356.
- Food: Our World in Data, "Engel's law" (2023); India Household Consumption Expenditure Survey 2022–23 (NSO), summary.
- Digital, US: Pew Research Center — mobile and home broadband (2024), social media fact sheet (2025), cash use (2026). Pew's 2025 survey puts broadband at 54% (<$30k) vs 94% ($100k+).
- Digital, global: ITU Facts and Figures 2025 internet use, 2024 affordability; GSMA, affordability of handsets and data (2025).
US figures are one country in one year; the bottom fifth includes retirees and students whose reported income understates their resources. Global figures come from different years and definitions. Built from a research corpus of 1,440 indexed sources (Wikipedia, arXiv, books, web, X).