What Is Marriage For?

No one calls a good thing a defect — in Russian the word for marriage does double duty.

— Folk irony

When marriage joined fields, debts and kin groups, the question of compatibility was simpler. The young couple was only required not to disturb the bookkeeping. Modern societies usually describe marriage as recognition of love. Family law, in various jurisdictions, sees a different set of objects: property, inheritance, medical decisions, parenthood, taxes, support after a split and the procedure for leaving. Both descriptions are true. The symbolic meaning hides the machinery of the institution whenever it is taken for the whole description. So the question is not whether a stamp in a document can create love, but which risks are left resting on a private promise without it, and which new risks the protection itself creates.

A contract for an unknown future

An ordinary contract lists what each side must do. In a long shared life it is impossible to describe everything in advance: nobody knows who will fall ill, lose a job, care for a child, move, or inherit money. Economists analyse such agreements as incomplete contracts: instead of listing every future action, they allocate rights and procedures for the unforeseen. The theory was built first for firms, ownership and control; applied to marriage it works as an analytical frame, not as a literal identity. In jurisdictions where marriage governs property, inheritance, parenthood and exit, the institution really does part of this work. It does not say who washes the dishes on Tuesday, but it sets the rules by which the law will read property, parental duties and liability when the arrangement falls apart.

This lowers transaction costs. A couple does not need to sign a separate contract before every joint purchase and every year of childcare. The package applies by default.

A long-term investment is vulnerable. One partner may cut back on work for the children, counting on future shared income. If the other can leave right after the investment is made, the promise “we will split everything” is not convincing enough. A legal institution makes some promises more credible, because breaking them carries external consequences. Community property, support after a split and inheritance rights — where they exist — do not guarantee fairness, but they change the price of an opportunistic exit. Marriage can be useful even to a couple who trust each other completely. It insures against bad faith, and against changes in circumstances, memory and power.

Insurance is not bought because a fire is expected. It is bought because a fire is expensive.

Residual rights over the unforeseen

An incomplete contract does not merely omit details. It leaves someone the right to decide what to do when an event occurs that the text never mentions. In the theory of the firm such residual rights are tied to ownership and control: the owner of an asset has the last word in a situation nobody wrote down. In the family analogy the objects of control are not people but decisions and resources: the flat and the account, where to live, decisions about a child’s residence and contact within the limits the law sets, a career track, a circle of social ties, and the time invested in care.

Take a move for one partner’s career. Before the decision both can promise that the gains will be shared. After the move one gets a promotion and a new professional circle, the other gets a lost job and dependence on a future promise. The joint project may have raised total income and at the same time shifted the distribution of control. If the rules of property and exit do not compensate the second partner’s investment, a rational person will be less willing to invest in advance — or will invest and end up exposed.

This is where underinvestment comes from. The couple could gain more from specialisation, a move or childcare, but a participant fears that once the irreversible step is taken the terms will be renegotiated. Love reduces that fear when the trust is warranted, but it does not remove the possibility that circumstances change. Legal protection makes part of the future gain portable through conflict, and can therefore support an investment that would otherwise be too risky.

None of this requires treating spouses as potential fraudsters. An incomplete contract exists even between people acting in good faith: they can fall ill, change their minds, acquire unexpected obligations, or remember an old promise differently. A default rule is needed against betrayal and against diverging versions of the past.

Marriage allocates residual rights, not quantities of romance. Before any large investment it is worth settling who owns the accumulated capital, how household work that never appears on a payslip is compensated, and how decisions about children and housing are made in a conflict. The institution answers these questions, sometimes crudely, but the absence of an explicit answer leaves the decision to whoever holds more actual power.

The same lock holds in the bad

The mechanism that protects investments also creates a lock-in. The more expensive the exit, the safer it is to make long investments — and the harder it is to leave a destructive union. The history of marriage shows both sides. A ban on divorce, or its practical unavailability, plus economic dependence and the risk of losing contact with the children, held families together very effectively. A low divorce rate said nothing about the quality of the relationships; the door was locked from the outside. When divorce became easier, some unions fell apart. That can be read as the destruction of an institution or as the discovery of a hidden demand for exit. The data allow different moral stories. The economic mechanism is simpler: the price of dissolution went down.

The model gives no universally optimal price of exit. Lowering it can weaken the protection of relationship-specific investments while making it easier to leave a harmful union; raising it does the reverse. The right balance depends on which investments are being protected, who bears the risk, and whether more precise ways of compensating them exist.

Law affects a family long before any divorce. It changes the background of everyday bargaining. Rules on dividing property, the recognition of childcare as a contribution, and the regime governing spouses’ earnings all shift each participant’s outside option in advance. Most couples do not quote statutes over breakfast, but the possibility of a split is already shaping their positions.

Alessandra Voena used differences between American regimes of unilateral divorce and property division. In states with equal division of property, the introduction of unilateral divorce was associated with higher household savings and lower employment among married women; a structural model read this as a change in intertemporal incentives and in the allocation of resources between spouses. The result applies to a specific combination of rules, not to “divorce in general”.

Such research does not make law the sole cause of family behaviour. States differ, people choose where to live, and legal reforms arrive together with cultural change. But it shows the direction of the mechanism: rules of exit can change behaviour inside an ongoing marriage, long before any divorce.

One inconvenient fact gets in the way of the symbolic conversation about marriage. The same ceremony under different property regimes creates different incentives. Conversely, in some jurisdictions an unregistered couple can reproduce part of the protections through contracts and joint ownership. The name of the institution matters less than the package of rights the participants actually receive.

The functions of marriage have changed along with technology and law. In agrarian societies marriage organised land, heirs and alliances between families. In many industrial societies it locked in the specialisation of a breadwinner and someone who ran the house and did the unpaid work. Contraception separated sex from childbearing. In countries with developed social protection the state took over part of the insurance against old age, illness and widowhood. DNA tests changed the problem of establishing paternity. Women’s own earnings reduced dependence on a husband. The institution remained, but the set of functions it performs changed. Arguments about “traditional marriage” usually suffer from a missing date. A lifelong union without divorce, a marriage for love, an economic partnership of equals and a contract between families are different traditions, sometimes incompatible ones.

Geography adds further differences: marital property regimes, recognition of cohabitation, maintenance and parental rights vary considerably. The word “marriage” creates different risks in different jurisdictions.

The symbol and the package of rules

A couple may value the ceremony, the public recognition and the sense of commitment. These effects are real, though harder to measure. A ritual tells relatives, and the participants themselves, that the decision has a long horizon. Publicity raises the reputational price of backing out. But the symbol does not replace the content. Two people can believe in love equally and face different consequences depending on the property regime, their incomes and their children. Instead of a general argument about whether marriage is necessary, it is more useful to work out which risks and rights the chosen form of union creates in this particular situation. For a couple with no property and no children the package is one thing. For migration, leaving a job or substantial property it is another.

Cohabitation also creates a legal and factual regime, and its content differs by jurisdiction. Some of the rules arise from statute and settled practice without any explicit choice by the participants. The absence of a signature does not mean the absence of consequences. Marriage does not guarantee love and does not necessarily destroy it. It makes some promises more credible and the exit more expensive. It is neither a sacred essence nor an empty stamp. It is a package of rules for a future the parties cannot yet describe.

Main sources

Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405; Becker, G. S. (1981/1991). A Treatise on the Family. Harvard University Press.

Henrich, J., Boyd, R., & Richerson, P. J. (2012). The puzzle of monogamous marriage. Philosophical Transactions of the Royal Society B, 367(1589), 657–669.

Grossman, S. J., & Hart, O. D. (1986). The costs and benefits of ownership: A theory of vertical and lateral integration. Journal of Political Economy, 94(4), 691–719.

Lundberg, S., & Pollak, R. A. (1996). Bargaining and distribution in marriage. Journal of Economic Perspectives, 10(4), 139–158.

Voena, A. (2015). Yours, mine, and ours: Do divorce laws affect the intertemporal behavior of married couples? American Economic Review, 105(8), 2295–2332.

Hart, O., & Moore, J. (1990). Property rights and the nature of the firm. Journal of Political Economy, 98(6), 1119–1158.