Michael Ross and Fiore Sicoly asked partners to estimate their own share of joint responsibilities. Added together, the two estimates came to more than the whole — each person, independently and honestly, believed they were carrying more than half. The explanation is not vanity, it is availability: you remember your own effort in detail because you were present for all of it, and you see a fraction of your partner's, mostly the finished result. Every ledger in every relationship is kept with one eye closed.

That is worth knowing before you argue about fairness, because it means both of your counts can be sincere and wrong at the same time. It also explains why score-keeping never ends in a settlement. Once a tally exists, attention reorganizes around it — you begin noticing evidence for your own side and filing your partner's contributions as normal, expected, not really countable. The ledger grows because it is being maintained, not because the imbalance is growing.

Margaret Clark and Judson Mills drew a distinction between exchange relationships, where contributions are tracked and expected to be repaid, and communal ones, where you respond to need without keeping accounts. Close relationships run on the exchange model tend to fare worse — not because generosity is morally superior, but because tracking is exhausting and it turns every kind act into a transaction with a receipt. What actually shrinks a tally is not a more accurate count. It is hearing, out loud and specifically, that something you did registered with the other person. None of which means a genuine imbalance should go unmentioned — it means the repair is a conversation about one specific thing, not a running total produced mid-argument.

Tonight, do the opposite of counting. Each name three things the other did in the past week that you noticed and never mentioned — small ones count, and the smallest are usually the best evidence. Then say out loud that you are closing the ledger, which sounds silly and works anyway. If something really is unbalanced, put it on the calendar as its own conversation for another day, with one specific ask instead of a tally.

SourceRoss, M., & Sicoly, F. (1979). Egocentric Biases in Availability and Attribution. Journal of Personality and Social Psychology, 37(3); Clark, M. S., & Mills, J. (1979). Interpersonal Attraction in Exchange and Communal Relationships. Journal of Personality and Social Psychology, 37(1).