How Much Families Spend on Food by Country — and What That Means for Your Budget

Households in Romania spend 23.1% of their total budget on food and non-alcoholic beverages. In Luxembourg, the figure is 9.3%. That is a two-and-a-half-fold difference, and it is the kind of statistic that gets turned into a headline about which countries "spend the most on food."

The headline is misleading. After adjusting for national price levels, real food expenditure is roughly 3,100 purchasing power standards (PPS) per person in Romania and 3,200 in Luxembourg. Those figures indicate a similar volume of price-adjusted expenditure; they do not show that people eat the same weight of food, consume the same calories, or have diets of the same quality. The 2.5× gap in budget share mainly reflects differences in purchasing power and in the rest of household spending.

That distinction matters if you came here to find out whether your own food spending is reasonable. Compare yourself with the wrong number and you will reach the wrong conclusion. This article explains what the published figures measure, shows how to build a useful benchmark for your household, and identifies the situations where a national average will not help.

All figures are Eurostat 2024 data unless stated otherwise, with the year attached to every number.

Three different numbers all called "food spending"

Most confusion about food budgets comes from three distinct statistics being quoted interchangeably. They answer different questions and frequently point in opposite directions.

1. Share of household budget. What proportion of everything a household spends goes to food. This measures pressure — how much of the budget food claims before anything else is paid for.
2. Nominal spend per person. The actual euros leaving the account. This measures cash out, and it rises with local prices whether or not anyone eats more.
3. Real expenditure (in PPS). Spending adjusted for national price levels, expressed in purchasing power standards. This compares the volume of expenditure after price differences are removed; it does not measure kilograms, calories, nutrition, or waste.

An expensive country can have one of Europe's highest nominal food bills and still record below-average price-adjusted expenditure. Switzerland illustrates the point in the table below.

Food spending across Europe, three ways

The source data is for 2024 and covers food and non-alcoholic beverages bought for consumption at home. The nominal and PPS figures below are approximate monthly equivalents of Eurostat's annual per-person values (divided by 12); the share and price index are unchanged. Restaurants and takeaway are excluded — that matters, and we return to it shortly.

Read the rows against each other and the ranking dissolves:

• Switzerland has Europe's largest nominal food bill in this comparison — €4,140 per person per year — but below-average real expenditure (~2,600 PPS per person per year against an EU average of ~2,700). With food prices 58% above the EU average, the euro figure overstates the volume that spending represents.
• Ireland's 9.8% share does not by itself signal abundance. At ~2,000 PPS per person per year, its price-adjusted food expenditure is the lowest in this table. A small budget share can coexist with a low expenditure volume.
• Romania's 23.1% share coexists with high real expenditure in this table. Food prices are low relative to the EU (price level 76), so the nominal amount represents more purchasing power than the euro figure suggests. The large budget share also reflects lower household purchasing power.

The economic principle underneath this is Engel's law: as income rises, the share spent on food generally falls, even when the absolute amount rises. Across countries, however, the share also reflects housing costs, household composition, public services, and statistical conventions. It is evidence of budget pressure, not a score for how well a household eats.

The broader category that reorders the table

The figures above exclude restaurants, cafés, canteens, and takeaway. Eurostat places those services in a broader "restaurants and hotels" category. Adding that entire category changes the ranking, but it does not produce a clean measure of total food spending.

Ireland and Romania sit at effectively the same combined figure — 26.0% and 26.1% — having arrived there from opposite directions. Spain moves above Romania in the combined column. But the shift cannot be attributed to eating out alone: hotel accommodation is included too, and its importance differs by country.

Treat the combined column as a warning about category definitions, not as a restaurant ranking. It shows why the grocery line is incomplete, while also showing why this dataset cannot isolate food away from home.

The practical lesson for your own budget is simpler than the statistics: decide once whether you are measuring groceries or total food, then measure it the same way every month. Households that shift meals between the supermarket and the restaurant without noticing usually conclude their grocery budget is improving when their food spending is flat.

Why the national average is probably not your benchmark

Before you compare yourself to any figure above, four properties of these statistics need stating plainly.

• Only the euro and PPS columns are per person. You may multiply those amounts by household size for a rough cash comparison, but the percentage column is a national aggregate share and must not be multiplied. Larger households also benefit from economies of scale, so even the per-person multiplication is not a prescription for what a family "should" spend.
• They cover all households, including single people and pensioners. There is no "average family" in this data.
• The denominator includes an estimated rental value for owner-occupied homes. National accounts add imputed rent to total household consumption. Your own budget almost certainly doesn't. If you own your home outright, your self-calculated food share will look higher than the national figure for that reason alone.
• Income changes the meaning of the share. In the United States in 2024, the lowest-income fifth of households spent $5,498 on food — 33.0% of before-tax income — while the highest-income fifth spent $16,989, or 6.4%. These US income shares are not directly comparable with the European consumption shares above, but they clearly show why one national average cannot describe households at very different income levels.

That last point deserves emphasis. A country average is useful context, not a household target. Your own rolling three- or six-month history is the stronger benchmark because its definitions and household circumstances stay more consistent.

A note on comparing across sources

National figures are rarely built the same way, and mixing them produces nonsense. The US Department of Agriculture reports that American consumers spent 9.7% of disposable personal income on food in 2025 — 4.8% at home and 4.9% away from home. Placing that 9.7% beside the EU's 13.2% suggests Americans spend less on food. The comparison is invalid: one is a share of income, the other a share of consumption expenditure, and the American figure includes restaurants while the European one does not. Always check the denominator before comparing two food statistics.

Build a useful benchmark for your own household

You cannot reproduce a national-accounts percentage exactly from an ordinary household budget: Eurostat includes adjustments such as imputed rent for owner-occupied housing. What you can build in about fifteen minutes is a consistent cash-spending share for your own household, then use the country table as rough context.

1. Set your numerator: food and non-alcoholic drinks consumed at home. Include groceries, non-alcoholic beverages, and staples. Exclude alcohol, tobacco, restaurant and takeaway meals, school and workplace canteen payments, cleaning products, toiletries, nappies, and pet food. Supermarket receipts mix all of these together, which is the single most common source of error.
2. Set a stable denominator: household cash spending for the same period. Include housing payments, utilities, transport, clothing, healthcare, leisure, and other day-to-day spending. Exclude savings, investments, debt principal, and income tax. This is a practical household definition, not an exact replica of national accounts.
3. Use a three-month average, not one month. A single bulk purchase or a holiday period will distort any single month.
4. Divide and interpret carefully. Compare the result primarily with your own previous period. Use your country's percentage only as context, especially if you own your home or your household differs from the national mix.
5. Record the number and the date. The comparison that will actually change your behaviour is against your own figure three months from now.

If separating grocery receipts into food and non-food sounds tedious, it is — once. After the first pass you are maintaining categories rather than rebuilding them, and any app that lets you split a single receipt across categories will handle it. Our guide to running an expense audit covers the same separation problem in more depth.

Worked example: a household that was measuring the wrong thing

The Marin household is illustrative, not a real customer, but the three errors it makes are the ones that turn up most often.

Four people in Portugal. Their budgeting app shows €780 in "groceries" for June, against €3,400 in total household cash spending. That is 22.9%, above Portugal's 18.2% national share — a reason to check the categories, not proof that the household spends too much.

Three corrections change the answer:

• €95 was restaurants and takeaway, logged to groceries because it came from the same debit card. Under Eurostat's definition this belongs in a separate category.
• €60 was non-food supermarket spending — cleaning supplies, shampoo, pet food. Bought at the grocery store, but not food.
• €120 was a freezer restock intended to last beyond June. It remains genuine food spending; its timing is a reason to use a three-month average rather than delete or reallocate the purchase.

Removing only the restaurant and non-food items gives €625, or 18.4%. That is close to the national share, but it is still a one-month cash figure built differently from Eurostat's aggregate. The freezer purchase stays in the numerator and should be smoothed by adding the surrounding two months.

The Marins did not change their spending; they corrected its classification. The revised figure does not prove that their budget is right or wrong. It tells them that the original 22.9% was not a sound basis for cutting food while €95 of restaurant spending remained invisible.

An absolute yardstick: what a nutritious diet costs

Budget shares tell you about pressure but never answer "is this enough?" For that you need a cost estimate for an adequate diet. The USDA publishes exactly that, monthly, at four cost levels. These are United States figures and do not transfer to other countries, but the structure is instructive everywhere.

For a reference family of four — two adults aged 19–50 and two children aged 6–8 and 9–11, with all meals prepared at home — the May 2026 costs were:

The Thrifty figure is published directly by the USDA as a reference-family total; the other three are our sum of the published per-person costs for those four age groups, which the report's own guidance permits without adjustment for a four-person household.

The useful part is the spread. Same country, same month, same family, every plan nutritionally adequate — and a 1.65× gap between the cheapest and the most generous. There is no single correct grocery budget. There is a range, and where you sit inside it is a choice about convenience, variety, and how much cooking you want to do, not a matter of getting it right or wrong.

Two limits worth stating: these plans assume every meal is cooked at home, which most households do not manage, and they are national averages with no adjustment for regional price differences.

What to do with the number once you have it

A benchmark is only useful if it leads somewhere. Three cases:

Your share is close to the benchmark

Do not cut the food budget automatically. If money is tight, inspect the largest categories first. At EU level in 2024, housing accounted for 23.6% of household consumption expenditure and transport 12.7%, although your own mix may be very different. Food attracts attention because you buy it frequently, not necessarily because it offers the largest saving.

Your share is well above the benchmark

Work through the causes in order of size before cutting anything:

1. Check the measurement first. Run the Marin corrections above. A seemingly high food share may include restaurant or non-food purchases.
2. Check the denominator. A high food share can mean low total spending rather than high food spending. If your income dropped, food takes a bigger slice without your grocery bill changing at all — and cutting food will not fix an income problem.
3. Separate waste from consumption. Food you throw away is the only category where spending less has no cost at all. Track it for two weeks before assuming the answer is cheaper brands.
4. Then look at prices and habits — where you shop, how often you shop, how much is convenience food.

Your share is well below the benchmark

This is not automatically good news. Compare against the adequate-diet range rather than congratulating yourself on the share. Persistently spending below what a nutritious diet costs in your area is a signal to examine, not an achievement — and if that is where you are, the USDA's Thrifty plan exists precisely because eating adequately on a constrained budget is a real problem with real research behind it.

When to ignore the national average entirely

Some households cannot be usefully compared to any national figure. If two or more of these apply to you, stop benchmarking against your country and benchmark against your own history instead.

• Medical or allergy-driven diets. Coeliac, severe allergies, and prescribed diets carry a structural cost premium that no average reflects.
• Household size at the extremes. One person or six-plus people; per-capita averages fit neither.
• Large price differences inside your country. Capital cities and remote rural areas can differ enough to make the national figure meaningless.
• Infants or teenagers. Formula costs and teenage appetites both break the average in opposite directions.
• Multi-generational or fluid households. If the number of people eating changes month to month, per-capita figures do not apply.
• Substantial home production. A productive garden or smallholding shifts food from the budget to the labour column.

Common mistakes

• Comparing a household total to a per-capita figure. The single most frequent error, and it makes every family look extravagant.
• Comparing across incompatible sources. Share of income and share of consumption are different denominators; one including restaurants and the other not makes it worse.
• Treating nominal spending as consumption. Higher grocery bills in an expensive country do not mean more food.
• Judging a single month. Bulk buying, holidays, and school terms all distort. Use three months.
• Letting the supermarket define "food." Shampoo and pet food bought at the grocery store are not food, and leaving them in inflates your share by a few points.
• Cutting food to solve an income problem. If the share rose because income fell, the grocery bill is a symptom.
• Chasing the lowest possible number. Nutritional adequacy is the floor, not the share.

How Lumy fits in

Lumy is a family budgeting app that can produce the numbers this article asks for, provided your categories are set up to separate food from the rest of the supermarket receipt. Expenses can be recorded through chat, a receipt photo, or voice, and category totals over a chosen period give you the numerator and denominator for the calculation above.

Lumy is offline-first: financial data is stored locally by default with optional cloud sync, and the app does not ask for banking credentials. If you would rather not connect an account at all, our guide to tracking spending without a bank connection covers that workflow.

To be clear about what the app does not do: Lumy does not benchmark your household against Eurostat or USDA figures automatically. The comparison in this article is a manual calculation, and it will stay manual. Availability and limits depend on your plan and app-store offering — see the pricing page and the privacy page for current details. A spreadsheet does this job perfectly well too, and a household that already has one does not need an app to run this analysis.

What to do today

1. Pull your last three months of grocery spending from wherever you record it.
2. Split out anything that is not food eaten at home — restaurants, alcohol, cleaning products, toiletries, pet food.
3. Divide the corrected food total by your household cash spending for the same three months.
4. Compare the result first with your previous period; use your country's row only as rough context.
5. Write the figure and today's date somewhere you will find it again, and recalculate in three months.

If the recalculated figure has moved, you will know whether it moved because of prices, habits, or income — and that is a question worth answering, which "the average family spends X" never was.

Frequently asked questions

How much should a family of four spend on food per month?

There is no single correct figure, and the honest answer is a range. The USDA's four plans for a US family of four in May 2026 ran from $1,018.20 to $1,675.30 a month with all meals cooked at home — all four nutritionally adequate. In the EU, food and non-alcoholic beverages took 13.2% of household consumption expenditure on average in 2024, varying from 9.3% to 23.1% by country. Calculate your own share and compare it to your country's figure rather than looking for one universal number.

Which country spends the most on food?

It depends entirely on which measure you mean, which is why the question is often answered badly. By share of household consumption expenditure, Romania is highest in the EU at 23.1% (2024). By nominal spend per person per year, Luxembourg leads the EU at €4,000, with Switzerland higher still at €4,140. By real expenditure adjusted for price levels, Portugal and Luxembourg are near the top while Switzerland falls below the EU average. A high share signals budget pressure; it does not prove that households buy more food.

Is 20% of income on food too much?

Not necessarily, and the answer depends on your income level rather than the percentage. In 2024 the lowest-income fifth of US households spent 33.0% of before-tax income on food while the highest-income fifth spent 6.4% — the same groceries are a much larger share of a smaller income. Note also that most published European figures are shares of total spending, not income, so check which one you are calculating.

Why is my grocery bill higher than the national average?

Check the definition before assuming the gap is meaningful. The euro and PPS figures are per person rather than per household; the food category excludes restaurants, takeaway, and non-food supermarket items; and national accounts include imputed rent for owner-occupiers in the denominator. These differences mean a household cash budget can provide only a rough comparison.

Does eating out count as food spending?

It depends on the statistic. Eurostat's food category excludes restaurants and records them separately under "restaurants and hotels"; USDA figures for the US include food away from home. Decide which definition you are using and apply it consistently — mixing them is the most common reason two food statistics appear to contradict each other.

How often should I recalculate?

Every three months is enough. Food prices and household habits do not change fast enough to justify monthly recalculation, and a three-month average smooths out bulk purchases and holidays that would otherwise distort the figure.

Sources and methodology

This article was written and reviewed by the Lumy editorial team. European figures were retrieved directly from Eurostat's dissemination API on 31 July 2026 (datasets nama_10_cp18 for expenditure shares and per-capita spending, and prc_ppp_ind for price level indices and real expenditure in PPS), and cross-checked against Eurostat's published Statistics Explained article. United States food plan costs were read from the USDA's primary PDF releases rather than secondary reporting — several third-party sites currently quote an outdated Thrifty Food Plan figure of $229 per week, which the May 2026 release supersedes at $235.00.

Eurostat rounds real expenditure per capita in PPS to the nearest 100, so those values are shown as approximate; several 2024 Eurostat observations are also marked provisional in the dissemination API. PPS expenditure compares economic volume after price adjustment, not the physical quantity or nutritional quality of food. USDA family-of-four totals for the Low-Cost, Moderate-Cost and Liberal plans are our sum of the published per-person costs for the four relevant age groups; the Thrifty figure is published as a family total. Reference years differ between sources and are stated with each figure. National accounts data covers all households, including single-person households, and is not specific to families. The Marin household is an illustrative example created to demonstrate the correction workflow, not a real customer.

This article is provided for general educational purposes about comparing household food spending. It is not financial, nutritional, medical, or tax advice, and it does not replace a qualified professional.

• Eurostat — Household consumption by purpose (2024 reference year; EU-27 shares for food, housing and transport).
• Eurostat — Final consumption expenditure of households by consumption purpose (nama_10_cp18) (country shares and nominal spending per capita, 2024).
• Eurostat — Purchasing power parities and price level indices (prc_ppp_ind) (food price level indices and real expenditure per capita in PPS, 2024).
• USDA Economic Research Service — Food Prices and Spending (share of disposable personal income, 2025; spending by income quintile, 2024).
• USDA FNS — Official USDA Thrifty Food Plan, U.S. Average, May 2026 (issued June 2026).
• USDA FNS — Cost of Food at Home at Three Levels, U.S. Average, May 2026 (issued June 2026).
• Office for National Statistics — Family spending in the UK: April 2024 to March 2025 (published 11 June 2026).
• Lumy features, plan comparison, and privacy policy (product capabilities reviewed for this article).

Published: 31 July 2026. Reviewed: 31 July 2026.