The Journal
How to split bills when one partner earns more
An equal split of the bills is not neutral: the same pounds are a larger share of the smaller income, and that is arithmetic, not a survey finding. This piece does the sums on the page, sets out the proportional split as a five-step procedure, says what to use as income when one of you is self-employed or on commission, works it through at the statutory maternity pay rates in force on 20 September 2026, and argues that the real decision is what happens to the money left over.
Equal is not the same as fair: what a 50/50 split leaves each of you with
The rent does not know what either of you earns, and nor does the gas bill. Each is a single number, and if you halve it, each of you pays the same pounds out of a different income. That is arithmetic inside one household, not a survey finding about households in general, so this piece does the sums.
Take two people. One brings home £3,200 a month after tax, the other £1,800. Their shared costs — rent, energy, council tax, broadband, food — come to £1,600 a month. For scale: the Office for National Statistics puts average UK household spending at £676.60 a week in the year to March 2025, about £2,930 a month. Housing (net), fuel and power, a category that excludes mortgage interest and council tax, was the largest at £118.40 a week, 18% of the total.
Split equally, each pays £800. For the higher earner that is 25% of take-home pay, and £2,400 is left. For the lower earner it is 44%, and £1,000 is left. The same £800 is a quarter of one income and nearly half of the other. The incomes stood at roughly 1.8 to 1; what is left after the bills stands at 2.4 to 1. An equal split does not preserve the gap between you. It widens it.
Nor is this a niche problem. An ONS analysis of the 2024 Annual Population Survey covers opposite-sex couples in the UK where both partners' gross weekly wages were known. The man earned more in 70.1% of them, the woman in 25.8%, and the two earned the same in 4.2%. In about 96 couples in 100, then, somebody earns more, and that count leaves out the one-earner household, the widest gap there is.
Two named rules, and a third that needs a name
Two ways of dividing a bill have settled names. In an equal split each person pays the same amount. In a proportional split each pays the same percentage of their income, so the amounts differ.
Social psychology has older names for the ideas underneath. Morton Deutsch's 1975 paper in the Journal of Social Issues set out three principles for sharing things out: equity, equality and need. A 2017 Frontiers in Psychology paper by Ali Kazemi, Daniel Eek and Tommy Gärling summarises when a group reaches for each: equity "is associated with effectiveness and productivity"; equality is likely to be applied whenever "enjoyable social relations and harmony are dominant group goals"; and need-based allocations are endorsed where the focus is "individuals' welfare, a sense of concern and responsibility for others". Mind the vocabulary: equity here means getting out in proportion to what you put in, the rule of the workplace. Paying in according to your means is a different idea, and sits nearer to need.
Margaret Clark and Judson Mills drew the companion distinction in the Journal of Personality and Social Psychology in 1979. In an exchange relationship, receiving a benefit creates a debt, to be settled by returning something comparable. In a communal relationship, of which their example is the family, benefits are given in response to need and create no such debt. In their first experiment, being repaid for a favour made people like the other person more when they preferred an exchange relationship and less when they wanted a communal one. A couple who settle every bill to the penny are running exchange rules inside a communal relationship, which is one reason an even ledger can still feel wrong.
Then there is a third rule, and the name for it is ours, not an established term: the equal-leftover split. Each of you ends the month with the same amount of money to call your own. In the example, £3,400 is left after the bills, so each keeps £1,700, which means the higher earner pays £1,500 of the £1,600 and the lower earner pays £100. Pooling every pound produces the same result. If what you mean by fair is that neither of you is the poorer one inside your own home, neither named rule delivers it, and this one does.
How to split bills by income: the proportional split in five steps
Step 1. List the shared costs and turn them into one monthly figure. Rent or mortgage, council tax, energy, water, broadband, insurance, food, and anything else you agree is joint. Divide annual bills by twelve so they are in the total. In the example: £1,600.
Step 2. Agree which income figure you are using, and use the same one for both of you. Monthly take-home pay is the honest choice, because that is what the bills are paid from. In the example: £3,200 and £1,800.
Step 3. Add the two incomes together and work out each person's share of the total. £3,200 plus £1,800 is £5,000. £3,200 is 64% of that; £1,800 is 36%.
Step 4. Multiply the shared costs by each share. 64% of £1,600 is £1,024. 36% of £1,600 is £576. Check that they add back to £1,600.
Step 5. Write down what would make you redo the sum. MoneyHelper, the Money and Pensions Service's guidance site, tells couples to review any agreement when something changes, such as a new job or children; a rent increase counts too.
The result: each of you now pays 32% of your income towards the household. The higher earner is left with £2,176 and the lower earner with £1,224. That is still roughly 1.8 to 1, exactly the ratio of the incomes. A proportional split neither widens the gap nor closes it. An equal split widens it; an equal-leftover split closes it. The choice between the three is a choice about the gap, not about the gas bill.
The same five steps work for rent on its own, if rent is the only thing you share.
What counts as income: self-employment, commission and maternity pay
Use take-home pay rather than gross. Income tax is charged in bands above a tax-free personal allowance, so a larger salary loses a larger proportion of itself before it reaches an account; gross figures therefore overstate the higher earner's real capacity to pay.
If one of you is self-employed, a single month proves nothing. Take the last twelve months of profit, subtract what must be set aside for tax and National Insurance, divide by twelve, and treat that as the monthly income. For commission or bonuses, either fold last year's total into the annual figure or run the formula on basic pay, applying the same percentages to each bonus as it arrives.
Maternity leave tests the rule hardest, because income falls on a known date by a known amount. The figures on gov.uk on 20 September 2026 are these. Statutory Maternity Pay is paid for up to 39 weeks: 90% of average weekly earnings before tax for the first 6 weeks, then £194.32 a week or 90% of average weekly earnings, whichever is lower, for the next 33. Tax and National Insurance are deducted. Statutory Maternity Leave is 52 weeks, so a full year of leave ends with 13 weeks that carry no statutory pay at all. A self-employed mother does not receive maternity pay from an employer; Maternity Allowance for the self-employed is between £27 and £194.32 a week for up to 39 weeks. GOV.UK's guide for employers adds that an employer "can offer more than the statutory amounts" if it has a company maternity scheme, so read the contract.
£194.32 a week is about £842 a month before deductions. Suppose the lower earner's take-home falls to £840. Run the five steps again: the shares become 79% and 21%, the higher earner pays £1,264 and the partner on leave pays £336, leaving £1,936 and £504. The formula has done what it promises, and the person at home with a new baby has about £500 a month. In the unpaid weeks it asks nothing of them and leaves them with nothing, which is where proportional stops being an answer. Do the sums before the leave starts.
The argument is never about the gas bill: dividing what is left
Nobody resents the gas bill. What is resented is the position after it: one person booking a weekend away without a thought while the other checks their balance before buying shoes. That is how a couple can agree the split and keep having the argument.
The best evidence that this matters is a study of 10,236 people living with a different-gender partner, published in PLOS ONE in 2019 by Brian Joseph Gillespie, Gretchen Peterson and Janet Lever. Asked whether the way they handled shared expenses was fair, 79% said it was about right and 15.9% said they paid more than they should. The authors found that "fairness evaluations over shared expenses are a stronger predictor of relationship quality than perceived equity in housework", and that feeling unfairly treated yourself predicted more than feeling your partner was. The limits are real: the data came from a 2008 survey on a US news website, which the authors call a "non-representative convenience sample"; it is a snapshot, so it cannot show which way the cause runs; and it measured whether the arrangement felt fair, not how anyone split anything.
British research points at the leftover in particular. Carolyn Vogler and Jan Pahl, writing in The Sociological Review in 1994 and using the Social Change and Economic Life Initiative data, found that even when couples nominally pooled their money, one partner usually controlled the pool. Only one fifth of couples controlled it jointly, and those households had the highest levels of equality between husband and wife in decision making, experience of deprivation and access to personal spending money. The last of those is the thing to watch.
So settle the leftover on purpose. You can each keep what is left, which accepts the income gap inside the household. You can fund joint savings by the same percentages as the bills and keep the rest. Or you can each take an equal personal allowance and treat the rest as the household's, which is the equal-leftover rule by another route. Whichever you choose, copy one piece of MoneyHelper's advice: agree a spending limit above which a purchase needs a joint decision.
The transfer that never happens, and how to fix it
A fair split can fail for reasons that have nothing to do with the formula. One of you pays the bills from your own account, and the other is meant to send their share. The transfer comes late, or short, or after a reminder. The payer has become a creditor, choosing between chasing and absorbing the loss: an exchange relationship's bookkeeping between two people who are meant to be on the same side.
The fix is to take the remembering out of it. Open an account that exists only for bills, move every shared direct debit to it, and each set up a standing order for your share, dated the day after your own payday. MoneyHelper describes this as dividing money into mine, yours and ours: a joint account for the bills, separate accounts for what each of you wants, and a monthly contribution that can be 50/50 or related to the size of your incomes. It names the proportional option without giving a method. If you are wary, it suggests starting small: a joint account with no overdraft facility, one or two bills, and a review after a few months.
Two cautions from the same guidance. Joint account holders are both responsible for any debt or overdraft on it, and opening one can affect your credit rating, so check both credit records first. If you would rather not share an account, a standing order to whichever of you holds the direct debits works just as well.
When proportional is the wrong answer
A split by income assumes that income measures what each of you can afford and what each of you contributes. Sometimes it measures neither. Large savings on one side are the simplest case; these are the harder ones.
Unpaid work. If one of you earns less because you do the childcare or care for a parent, your income understates your contribution and the formula charges you for the shortfall. MoneyHelper is direct about this where it discusses the main earner paying the other an allowance: the payment "should not be seen as a 'favour'", because looking after the children or working as a carer is a job too. In these households, and in the unpaid weeks of parental leave, equal leftover is the rule that fits. Who does the work rather than who pays for it is the subject of this site's piece on how to split chores with your partner.
Debt brought into the relationship. Loan repayments reduce what a person can really pay, so decide whether they come off before the shares are worked out. Be careful about turning one person's debt into a joint one: MoneyHelper's warning is that each of you is liable for the whole of a joint debt, and remains liable if the other does not pay their share.
A home only one of you owns. Paying a share of a mortgage in your partner's name raises a question rent does not: what you are acquiring. MoneyHelper notes that couples who have lived together and then separate have fewer rights than couples who divorce or dissolve a civil partnership. Take advice before settling into that arrangement.
Housemates. Proportional splitting is a rule for people who share a future. Housemates are in an exchange relationship, and an even split suits it. Liability is a different question from the arrangement: under a single joint tenancy the landlord may ask any named tenant for all of the rent, which "Housemate not paying rent: who the landlord can ask to pay" sets out for England.
One boundary holds throughout. An agreement about shares is made by two people who can each see the numbers and say no. MoneyHelper's position is that everyone has the right to financial independence, and that a partner controlling your money or running up debts in your name is financial abuse. If that describes your household, the question is not which formula to use; MoneyHelper's guide to financial abuse is the place to start.
Sources
- Average weekly UK household expenditure was £676.60 in the financial year ending March 2025 (April 2024 to March 2025); housing (net), fuel and power was the largest category at £118.40 a week, 18% of the total; mortgage interest payments, Council Tax and Northern Ireland rates are categorised as other expenditure rather than in that category. Used only to show that the worked example's £1,600 of shared monthly costs is realistic (the monthly conversions, about £2,930, are ours: weekly figure × 52 ÷ 12). Main points read on the bulletin page on 20 September 2026; released 11 June 2026. — Office for National Statistics, "Family spending in the UK: April 2024 to March 2025", 2026
- In 2024 the man earned more in 70.1% of couples, the woman in 25.8%, and the two earned the same in 4.2%. UK, Annual Population Survey, January to December 2024; "proportions are calculated from the gross weekly wage of the Household Reference Person and their spouse or cohabiting partner in opposite sex relationships, where the wages for both are known". The 'about 96 in 100' is our sum of the first two figures. Figures read from the spreadsheet attached to the release (created 20 May 2025) on 20 September 2026. The release gives no figure for the size of the gap, so the piece gives none. — Office for National Statistics, ad hoc release 2828, "Sex of highest earner in relationship, UK, selected years 2004 to 2016 and every year 2019 to 2024", 2025
- 10,236 respondents in different-gender cohabiting couples, from a survey posted on a US national news website in 2008; respondents were asked whether the way they handled shared expenses was fair, with the options of paying more than they should, about right, or the partner paying more than they should; 79% reported the division of shared expenses as fair and 15.9% reported unfairness to themselves; "fairness evaluations over shared expenses are a stronger predictor of relationship quality than perceived equity in housework"; unfairness to oneself a stronger predictor than unfairness to one's partner; the authors' own limitation that it was a "non-representative convenience sample" with higher incomes and education than the national population; cross-sectional; perceived fairness measured, not actual splits. Abstract, methods, results and limitations read on 20 September 2026. — Brian Joseph Gillespie, Gretchen Peterson and Janet Lever, "Gendered perceptions of fairness in housework and shared expenses: Implications for relationship satisfaction and sex frequency", PLOS ONE, 2019
- Carries the account of Deutsch's three principles of distributive justice (Morton Deutsch, "Equity, Equality, and Need: What Determines Which Value Will Be Used as the Basis of Distributive Justice?", Journal of Social Issues 31(3), 137-150, 1975) and the three quoted phrases on which group goal goes with which principle: equity "is associated with effectiveness and productivity"; equality "whenever enjoyable social relations and harmony are dominant group goals"; need where the focus is "individuals' welfare, a sense of concern and responsibility for others". Deutsch's own paper is paywalled and was not opened; its bibliographic details and abstract were checked against the ERIC record EJ133723. Read on 20 September 2026. — Ali Kazemi, Daniel Eek and Tommy Gärling, "Equity, Equal Shares or Equal Final Outcomes? Group Goal Guides Allocations of Public Goods", Frontiers in Psychology, 2017
- The distinction between exchange relationships, where receiving a benefit incurs a debt or obligation to return a comparable one, and communal relationships, where benefits are given in response to need and create no specific debt; the authors' statement that "the typical relationship between family members" exemplifies the communal type; and Experiment 1's finding, the one the piece uses. The abstract states the Experiment 1 hypothesis as "the receipt of a benefit after the person has been benefited leads to greater attraction when an exchange relationship is preferred and decreases attraction when a communal relationship is desired", and says those hypotheses "were supported in Experiment 1, which used male subjects". Its manipulation was desire, not expectation: "the desire for a communal relationship was manipulated by using unmarried males as the subjects and having the part of the other played by an attractive woman, who was described as either married or unmarried", on the assumption that the subjects "would desire a communal relationship with the attractive, unmarried woman but would prefer an exchange relationship with the attractive, married woman" (96 subjects). Experiment 2 is the one framed in terms of what was "expected", and it used female subjects, a different manipulation and a request for a benefit rather than a benefit returned; the piece does not draw on it. Text extracted from the PDF hosted by the first author's laboratory at Yale and read on 20 September 2026. The application to couples splitting bills is this piece's, not the paper's. — Margaret S. Clark and Judson Mills, "Interpersonal Attraction in Exchange and Communal Relationships", Journal of Personality and Social Psychology 37(1), 12-24, 1979
- Using the Social Change and Economic Life Initiative data set: even where couples nominally pooled their money, in practice one partner was likely to control the pool; the pool was jointly controlled in only one fifth of couples, and those households had the highest levels of equality between husband and wife in decision making, experience of deprivation and access to personal spending money (paraphrased closely from the abstract). Verified on 20 September 2026 from the publisher-deposited abstract in the Crossref record for this DOI; the publisher's page sits behind a bot check and the full text was not read, so the piece claims nothing beyond the abstract. In particular it does not claim that personal spending tracks share of income. — Carolyn Vogler and Jan Pahl, "Money, Power and Inequality within Marriage", The Sociological Review 42(2), 263-288, 1994
- Carries the one sentence in section 4 that Income Tax is charged in bands above a tax-free personal allowance. The page covers the tax year 6 April 2026 to 5 April 2027 and states: "The standard Personal Allowance is £12,570, which is the amount of income you do not have to pay tax on", and that "How much Income Tax you pay in each tax year depends on how much of your income is above your Personal Allowance", above a table of rising bands (basic 20%, higher 40%, additional 45%). Those bands are for England, Wales and Northern Ireland; the page says "Income tax bands are different if you live in Scotland", which is why the piece prints no threshold or rate and claims only the banded shape, which holds in both. Read on 20 September 2026. — GOV.UK, "Income Tax rates and Personal Allowances", accessed 20 September 2026
- Statutory Maternity Pay is paid for up to 39 weeks: "90% of your average weekly earnings (before tax) for the first 6 weeks", then "£194.32 or 90% of your average weekly earnings (whichever is lower) for the next 33 weeks"; "Tax and National Insurance will be deducted." Read on 20 September 2026; the £194.32 figure was cross-checked the same day against GOV.UK's "Rates and thresholds for employers 2026 to 2027". The monthly equivalent of about £842 is ours (× 52 ÷ 12). — GOV.UK, "Maternity pay and leave: Pay", accessed 20 September 2026
- "Statutory Maternity Leave is 52 weeks", made up of 26 weeks of Ordinary and 26 weeks of Additional Maternity Leave. Carries the statement that a full year of leave ends with 13 weeks without statutory pay (52 weeks of leave less 39 weeks of pay; the subtraction is ours). Read on 20 September 2026. — GOV.UK, "Maternity pay and leave: Leave", accessed 20 September 2026
- Maternity Allowance for self-employed people is "between £27 to £194.32 a week for up to 39 weeks", depending on Class 2 National Insurance contributions. Read on 20 September 2026. — GOV.UK, "Maternity Allowance: What you'll get", accessed 20 September 2026
- Carries the one quoted clause at the end of the maternity paragraph in section 4: "You can offer more than the statutory amounts if you have a company maternity scheme." This is employer-facing guidance — GOV.UK's "Statutory Maternity Pay and Leave: employer guide" — and the piece attributes it as such rather than presenting it as advice to the employee. Read on 20 September 2026. — GOV.UK, "Statutory Maternity Pay and Leave: employer guide", accessed 20 September 2026
- Every statement the piece attributes to MoneyHelper, paraphrased except for one short quotation: reviewing agreements if something changes (a change of job, having children); a spending limit above which a purchase needs a joint decision; the mine, yours and ours arrangement with a monthly contribution that is 50/50 or related to the size of your income; the trial joint account with no overdraft facility; joint-account holders both being responsible for any debt or overdraft, and being co-scored when applying for credit; the allowance that "should not be seen as a 'favour'" because caring is a job too; both partners being liable in full for joint debts; separating cohabitants having fewer rights than couples who divorce or dissolve a civil partnership; and the right to financial independence with its description of financial abuse. The page names proportional contribution as an option but gives no method, which is the gap the five steps fill. Read in full in a browser on 20 September 2026 (the site refuses automated fetches). — MoneyHelper (Money and Pensions Service), "Should you manage money jointly or separately", accessed 20 September 2026
Barriers is a household planner for tasks, money and receipts. A job carries someone's name, and whoever it lands on can accept it or say no with a reason. The task list is free.
Get Barriers