How Much Should a Family Spend on Christmas?

Between 2 September and 16 October 2025, three separate US surveys asked American adults roughly the same question: what will you spend on holiday gifts this year. NerdWallet, polling through The Harris Poll, got $1,107 per person. Gallup got $1,007. The National Retail Federation, with the largest sample of the three at 8,247 consumers, got $627.93. Deloitte, asking per household and counting food and gatherings as well as gifts, got $1,595.

None of them is wrong. They are measuring different things, and the gap between them is the first useful fact in this article: there is no national number that tells you what Christmas should cost your household. The second useful fact is what happened to the households that went into the season without a number of their own. A little over a third of them borrowed to get through it.

Why the published averages disagree

Every autumn these figures are reported as though they were the same measurement taken by different people. They are not. Four decisions change the answer before anyone is asked anything: whether you count per person or per household, whether you count gifts only or the whole season, whether you average across all adults or only across the ones who are buying gifts, and when in the autumn you ask.

The NRF and NerdWallet figures are the clearest illustration. NRF divides planned gift spending across every consumer it surveyed, including the ones spending nothing. NerdWallet divides it across gift-buyers only. That single methodological choice accounts for most of a $479 difference, and neither organisation did anything improper to produce it.

The same disagreement exists in the UK. Depending on which survey you read, the 2025 season cost the average UK household £780, or it cost families £1,626. Both numbers circulated widely in December. They describe different households counted in different ways.

The average is the wrong number anyway

Suppose the surveys did agree. A national mean still would not tell you what to spend, because it is an average across households that differ in the two variables that actually decide the answer: how many people you are buying for, and what you can pay for afterwards.

The surveys themselves show the first variable moving. NRF found households with children planning to raise their gift budgets by more than $30 on average against the previous year. Deloitte found the typical household buying eight gifts, down from nine. A family of five and a couple with no children can both spend the national average and have completely different Decembers.

The second variable never appears in the headline at all. An average is a statement about spending; it says nothing about what the spending was paid for with. That is the number that matters, and the outcome data is where it shows up.

What happened after last Christmas

On 5 January 2026 LendingTree published its survey of more than 2,000 US consumers on how the season had actually gone. 37% had taken on holiday debt, averaging $1,223 — up from $1,181 the year before and the highest figure since 2022. Among parents of children under 18 it was 48%, averaging $1,324.

The repayment picture is the part worth sitting with. 63% of those who borrowed expected it to take three months or more to clear, and 40% were carrying it at an interest rate of 20% or above. 59% reported stress about it and 47% said they regretted how much they had spent. Credit cards financed 62% of that debt, buy-now-pay-later 35% and store cards 32%, with many households using more than one.

And then the line that reframes the whole question: 41% of those who took on holiday debt said they were still digging out from the previous year's bills when they did it. For a substantial minority of households, Christmas is not an annual event with a start and an end. It is a rolling balance that each December tops up.

The UK pattern is smaller in scale and identical in shape. YouGov, surveying 2,093 British adults on 6–7 January 2026, found 19% had borrowed to cover Christmas 2025 — 68% of them on credit cards and 29% on buy-now-pay-later, with gifts the leading category at 73% of borrowers. Most kept it modest: 38% borrowed under £250. It is still a December decision being paid for in spring.

The budget is not the problem. The running total is.

Here is the finding that should change how you approach this. In a Harris Poll of 2,084 US adults for the AICPA, fielded 28–30 October 2025 — before the season, not after it — 25% said they usually make a holiday budget and openly admitted they probably would not stick to it. Among 18-to-34-year-olds it was 33%. A further 8% said they were not at all sure how much they planned to spend at all.

These are people who know how to make a budget, have made one, and can already predict its failure in October. The plan is not what is missing. Dan Snyder, who directs the AICPA's personal financial planning division, named the mechanism in the release:

When spending is driven by emotions rather than a plan, it can get out of hand.Dan Snyder, CPA/PFS, Director of AICPA Personal Financial Planning

True, and incomplete. A monthly grocery budget is also driven by appetite and still mostly holds, because you find out where you stand every time you check the account. A Christmas budget has four properties that a monthly one does not, and each of them breaks the feedback rather than the intention.

It is spread across ten weeks, not one shopping trip

Holiday spending starts in October and finishes in late December: a toy here, a delivery there, three separate online orders, the food shop that is 40% larger than usual, the trip to see family. No single purchase feels like the one that broke the budget, because none of them is. The total exists only if something is adding it up as you go, and for most households nothing is.

Two adults are buying separately

This is the failure mode specific to couples and the one households most reliably underestimate. Two people shopping from the same budget without a shared total do not make one budget — they make two private ones that are each individually reasonable. Each partner knows their own spending and assumes the other is tracking theirs. Both are right about themselves and wrong about the household, and neither discovers it until the statements arrive. It is the same structural problem we wrote about in creating a shared family budget, concentrated into the most expensive eight weeks of the year.

Buy-now-pay-later moves the arithmetic into next year

36% of the AICPA's respondents planned to use Klarna, Affirm or Afterpay, and 35% of LendingTree's debt-takers did. Split into four payments, a $200 purchase presents itself as $50 — which is not a lie, but it is not a total either. The December you are budgeting for and the January that pays for it are no longer the same ledger, and the running total you most need is the one no single app is holding.

Christmas is filed as an exception

Richard Thaler's work on mental accounting describes how people sort money into separate notional pots and apply different rules to each. A season explicitly framed as special gets its own pot, and the rule attached to that pot is that the usual rules do not apply. The household that will not spend $60 on a weeknight takeaway will spend $60 on a stocking filler without the same hesitation, because the money has been mentally reclassified. That is normal, it is not a character flaw, and it is also why a December cap needs to be written down somewhere outside your head.

Setting a number you can actually hold

The usual advice is to pick a budget and stick to it, which is exactly the instruction that a quarter of households already expect to fail. A better sequence runs backwards, starting from the constraint rather than from the wish list.

1. 1.Start in January, not in December. Ask what you could clear in January without borrowing and without touching savings you have earmarked for something else. That figure, not the national average, is your ceiling. Everything below is how you spend it.
2. 2.Write the number down before November. A ceiling that only exists as a shared understanding between two adults is not a ceiling. Both of you should be able to say it from memory by the first week of November.
3. 3.Split it into categories, not one pot. NRF's split of the 2025 season is a reasonable starting ratio: roughly 70% gifts, 30% everything else — food, decorations, cards. Most households discover their real failure is the 30%, because nobody budgets for it.
4. 4.Give each child a number, not each gift. Per-child totals are easier to keep even than per-gift limits and remove the end-of-season audit of whether things came out fair.
5. 5.Decide who logs what, out loud. The rule only has to be that every purchase gets recorded by whoever made it, the same day. The household that does this has a running total; the one that does not has two private estimates.
6. 6.Check it weekly, not on 26 December. Four checks in a season is enough to catch the overrun while there is still something you can do about it. One check at the end is a post-mortem.

If you want the category list in more detail, the same logic applied to the ordinary year is in our pieces on family budget categories and the monthly household expenses list.

If you are already past the number

Some readers will find this article in December, after the fact. The research still has something useful to say, and it is not that you should have started earlier.

The first thing is to keep recording. An overrun you are tracking is a known quantity you can plan repayment against; an overrun you stopped looking at in the second week of December becomes the January surprise, and the surprise is what produces the 59% reporting stress. The second is to establish the real total across every method you used — card, store card and each buy-now-pay-later plan separately — because the 35% who used instalments have part of their total sitting outside the account they are checking. The third is that the repayment rate matters more than the amount: with 40% of last season's borrowers carrying it at 20% or above, the difference between clearing it in two months and letting it ride into spring is a real sum of money, not a moral question.

How this works in CapKin

The limits first, because they matter for this particular use. CapKin's caps are monthly, and Christmas is not — a season that runs from November into December is two budget periods in CapKin, so a single season-wide total is not something it will compute for you. You can set a gift category cap in each month, or read both months' category totals and add them yourself. Category targets also have to sum to no more than the household's overall monthly cap, so making room for a December gift category means either raising that cap or lowering something else for the month. That is a deliberate constraint, not an oversight: the whole product is built on the overall cap meaning something.

The rest of the limits are the ordinary ones. CapKin has no bank connection, no card and no way to move money, so nothing appears in it unless a household member enters it. It gives no financial advice and will not tell you whether your Christmas number is sensible — that judgement is yours. It is in beta, and free while it is.

What it does is the thing this article argues decides the outcome: it gives two adults one running total instead of two private ones. A gift category carries its own cap, the dashboard shows what is spent against it, and both partners log into the same budget — so the question “where are we against the number” has an answer at any point in November, not a reconstruction in January. Entries go in by voice or by typing, in the words you would use to a person, which is what makes recording a purchase in a shop take seconds rather than being deferred to an evening that never comes. The mechanics of that are in the piece on voice expense tracking, and the reasoning behind having no bank connection at all is in tracking expenses without linking a bank account.

For households where children are buying presents too, a child logs their own spending as pending until the household owner approves it, which turns a gift budget into something a child participates in rather than something that happens above their head. That is the same mechanism behind pocket money and the reason it doubles as teaching: a season with a visible limit is the most concrete lesson in saving and trade-offs a child gets all year. Every product claim here is stated in full on the product facts page, which is written to be quoted rather than interpreted, and the couples case is on the budgeting app for couples page.

Sources

• —National Retail Federation and Prosper Insights & Analytics. Consumers to Spend Second-Highest Amount on Record, October 2025. Survey of 8,247 adult consumers, fielded 1–7 October 2025, margin of error ±1.1 percentage points. $890.49 planned per person, of which $627.93 on gifts and $262.56 on food, candy, decorations and greeting cards; 1.3% below the prior year's record $901.99; second-highest in the survey's 23-year history.
• —Gallup. U.S. Holiday Shoppers Plan to Spend Briskly, published 28 October 2025. Telephone interviews by ReconMR, 1–16 October 2025, random sample of 1,000 US adults, margin of error ±4 percentage points (±$78 on spending estimates). Average expected gift spending $1,007, against $1,014 the prior year and $923 in 2023; 37% predicted spending $1,000 or more.
• —NerdWallet. 2025 Holiday Spending Report. Online survey by The Harris Poll, 2–4 September 2025, 2,083 US adults, of whom 1,710 planned to buy gifts; accurate to ±2.5 percentage points at 95% confidence. $1,107 average per gift-buyer, a collective $242 billion, an increase of $182 on the previous year.
• —Deloitte. Holiday Spending Declines Amid Economic Uncertainty, 15 October 2025. Online survey of 4,270 consumers fielded 27 August – 5 September 2025. Average planned household spend $1,595, down 10% year on year; eight gifts on average against nine the prior year; 57% expected the economy to weaken in 2026.
• —American Institute of CPAs. AICPA Holiday Spending Survey Reveals Budgeting, Debt and a Bit of Regret, 13 November 2025. Online survey by The Harris Poll of 2,084 US adults aged 18+, fielded 28–30 October 2025, margin of error ±2.5 percentage points. 25% usually make a holiday budget but admit they probably will not stick to it (33% of 18–34s, 16% of over-55s); 8% not at all sure what they planned to spend; 47% anticipated going into debt; 79% planned to use credit cards; 52% did not expect to pay the balance in full; 17% expected it to take more than six months; 36% planned to use Klarna, Affirm or Afterpay; 39% had felt regret about past holiday spending.
• —LendingTree. Holiday debt survey, published 5 January 2026, more than 2,000 US consumers. 37% took on holiday debt averaging $1,223, up from $1,181 and the highest since 2022; 48% of parents with children under 18, averaging $1,324; 63% expected repayment to take three months or longer; 40% carrying it at 20% interest or above; credit cards 62%, buy-now-pay-later 35%, store cards 32%; 59% stressed, 47% regretful; 41% were still paying off the previous year's bills.
• —YouGov. One in five UK adults borrowed to cover Christmas holiday spending in 2025, published 12 January 2026. 2,093 adults in Great Britain, fielded 6–7 January 2026, weighted to ONS population estimates. 19% borrowed; of those, 38% under £250 and 28% between £250 and £499; credit cards 68%, buy-now-pay-later 29%, overdraft 14%; gifts named by 73% of borrowers, food and drink 39%, bills 21%.
• —Deloitte. 2026 holiday retail forecast, 10 September 2026. US holiday retail sales projected at $1.70–1.71 trillion for November 2026 to January 2027, up 4.0% to 4.8% on the prior season; e-commerce $316.1–318.9 billion.
• —Thaler, R. H. (1985). Mental Accounting and Consumer Choice. Marketing Science, 4(3), 199–214; and Thaler, R. H. (1999). Mental accounting matters. Journal of Behavioral Decision Making, 12(3), 183–206.
• —CapKin product facts, last verified 4 August 2026 — the product claims in the final section are stated there in full.