Should couples split bills 50/50?

If one partner earns 30,000 and the other earns 90,000, a 50/50 split of a 2,000 monthly cost takes 40 percent of the lower earner's income and 13 percent of the higher earner's. The two people are paying the same number and experiencing completely different lives.

This is the core objection to flat splitting, and it is not an ideological one. It is what the numbers do. The lower earner ends up with no discretionary money, no savings capacity, and no ability to absorb a surprise, while nominally being treated equally. Over years that difference compounds into genuinely unequal outcomes in retirement savings, credit, and the practical freedom to leave.

The usual alternative is proportional splitting: each person contributes the same percentage of income rather than the same amount. In the example above that is roughly 500 and 1,500 rather than 1,000 each. Financial counsellors recommend it widely, though it is worth being clear that this is a reasoned position rather than a research finding.

The case for flat splitting is not nothing, though. It is simple, it needs no disclosure of exact salaries, and it does not require renegotiation every time someone gets a raise. For couples with similar incomes, especially unmarried couples who want to stay financially independent, those are real advantages.

The strongest study in this area did something unusual: it randomised. Newlywed and engaged couples were assigned to keep separate accounts, keep joint accounts, or do whatever they wanted, and were followed for about two years. Couples assigned to separate accounts, and those left to do as they liked, showed the usual decline in relationship quality across the first two years of marriage. Couples assigned to a joint account did not[1].

Random assignment matters enormously here. The obvious objection to every earlier finding, that happier couples pool their money rather than pooling making them happier, does not apply when a coin decides. That makes this one of the few genuinely causal results in relationship finance.

It is still one study, on a specific population, over a limited period. Newlyweds are not representative of couples generally, two years is not a marriage, and the study cannot tell you what happens to couples with large income gaps, prior debts, children from earlier relationships, or a history of financial abuse.

What it does suggest is that the question people argue about, the split, may be the less consequential one. Two couples can both use a 50/50 ratio and be doing something structurally different depending on whether there is a shared pot behind it.

Financial disagreement is not just another topic couples fight about. Research on marital conflict has repeatedly found that disagreements about money predict divorce more strongly than disagreements about children, in-laws, or household labour[2].

One explanation is duration and recurrence: money conflicts tend to be long-running, involve high stakes, and resurface monthly rather than resolving. Another is that they are frequently proxies. An argument about who pays for what is often an argument about respect, autonomy, or an unstated belief about what each person's contribution is worth.

That proxy quality is what makes the arithmetic solution insufficient on its own. Couples who agree a proportional split and still argue every month are usually not disagreeing about percentages. They are disagreeing about whether unpaid work counts, or whether a higher earner has more say, or whether one person's spending is treated as reasonable while the other's is treated as indulgent.

It also means the fix is often procedural. Couples who report doing well with money tend to have a recurring, scheduled, unremarkable conversation about it rather than a very good arrangement they set once and never revisited.

Four things change the answer more than the ratio does.

Unpaid labour is the first. If one partner does substantially more childcare or household work, a cash-only accounting treats that contribution as worth zero, and most people who do it can tell you exactly how that feels. Any split that ignores it is measuring one kind of contribution and calling it the whole picture.

Career cost is the second. A partner who reduced hours, moved cities, or stepped back for the household is carrying a long-term earnings loss that no monthly split reflects. Treating current income as the fair basis quietly charges them twice.

Marital status is the third. In many jurisdictions marriage creates shared property and support rights that unmarried cohabitation does not. Financial independence is a much more reasonable default when the law will not treat you as one unit if things end.

The fourth is the uncomfortable one. Financial control is a recognised form of abuse, and complete pooling can be one of its instruments. Any arrangement where one partner cannot see the accounts, has no independent money, or must ask permission for ordinary spending is a problem regardless of how the split was calculated.

One more, less obvious: whose spending gets described as a need and whose gets described as a want. That labelling is where a lot of these arguments actually live.