In addition to creating possibilities of trades with different performance times, commitment allows specialization within the marriage. First, it allows one partner to specialize in household production and the other in market production, an arrangement which can result in greater joint wealth than if both worked outside the home and hired someone else for household production. Traditionally, the husband specialized in paid employment and the wife specialized in unpaid homemaking.⁽⁶³⁾ This kind of specialization by its nature creates asymmetries, the most important of which is that the person specialized in marriage-specific work has more to lose from divorce. Learning how to cook Nat's favorite dishes is a lot less valuable when he is gone, but learning how to bring home more wages is just as useful after divorce as it was before.⁽⁶⁴⁾ Because expertise in production of marriage-specific assets is, by definition, less portable than expertise in income production, neither spouse should be eager to specialize in traditional homemaking tasks under a liberal divorce regime.

Even when both spouses have decided to specialize in paid income, the rules of divorce influence how time will be spent on the job and at home. People have choices on what to learn. A person can develop expertise discussing the books his spouse enjoys or can learn how to do his job better. If she divorces him, the time he spent learning how to do his job better will not lose its value, but the time he spent on her favorite books will. Family goodwill is not likely to survive breakup of the family, and financial assets can be divided by a court. By contrast, career advancement will survive divorce and will not be divided. Traditional divorce rules told spouses they might have to share income after divorce. The modern, unilateral, no-fault, minimal-alimony, divorce regime tells them they will get to keep most of the income they produce after divorce. As a result, the pursuit of career advancement has a higher expected benefit today. The modern rules press both spouses to devote their time away from family and cash income. Nat and Dot may, quite rationally, wish to avoid incentives for selfish career building at the expense of family,⁽⁶⁵⁾ without wanting to bind themselves to predetermined roles. They can do this if they are allowed to commit strongly to the marriage.

Second, commitment allows idiosyncratic specialization within household production.⁽⁶⁶⁾ Many household tasks, including those related to automobiles, clothing, home entertainment, food, children's schooling, travel, and medical care, can be done better and more efficiently if a person studies and develops personal contacts, but do not require both partners' detailed attention. The committed couple can divide up these tasks with less worry about whether the particular package of expertise chosen by each will be attractive to a future mate.

Third, commitment provides insurance. Two violinists might each feel that they have an 80% chance of making a decent living at the violin. They, like many others, might be sufficiently risk averse that a 20% chance of professional failure after many years of training would be unacceptable. However, if they band together, committed for life, their chances of having to give up the violin to make a living drop to a more acceptable 4%.⁽⁶⁷⁾ By marrying, and committing to a lifetime of joint income production, they can both pursue their love of the violin.⁽⁶⁸⁾

Nat and Dot could plan to specialize emotionally in the marriage--"forsaking all others,"⁽⁶⁹⁾ as the old marriage service puts it. There is some evidence that a marriage has a slightly better chance of survival under the traditional rules.⁽⁷⁰⁾ A couple making every effort to increase their odds could rationally opt out of the modern regime. Courts have apparently failed to see that their dissolution decisions have important incentive effects on other marriages.⁽⁷¹⁾ In that blindness, they have gang aglee in approaching dissolution as a matter of determining how best to deal with a marriage once it has broken down or how to clean up a messy situation. A rational couple might see the incentive benefits to which courts have been blind.

Religion provides another dimension of reasons for commitment. If both spouses are Roman Catholic, for example, they may wish to bind themselves so that their legal constraints reinforce their religious convictions.⁽⁷²⁾ If just one is Roman Catholic, that one has all the more reason to include a clause restricting divorce since otherwise the non-Catholic spouse could use the threat of divorce as a bargaining chip.⁽⁷³⁾ Both parties might agree that such asymmetry in the marriage would be undesirable. If the law allowed a binding marriage, churches might require their members to select such a marriage. Parties could then signal their intentions and desires by membership in a church that required certain commitments for a religiously valid marriage.

Nat and Dot might also wish to bind themselves out of concern for the children they hope to have. There is enough credible evidence that divorce is harmful to children that a couple might wish to constrain each other for the benefit of their children.⁽⁷⁴⁾

Thus, there are good reasons why some people might want to enter a traditionally binding marriage.⁽⁷⁵⁾ In arguing that some couples might choose restrictive divorce rules, we do not argue that a majority of couples would (or should) choose such a marriage.⁽⁷⁶⁾ A marriage allowing divorce only for fault is not always a better choice than a marriage allowing no-fault divorce. A fault requirement creates a disincentive for fault to someone who wants to stay married, but an incentive for fault to someone who desires divorce. Misbehavior might either increase or decrease. Which has the greater potential and which is the greater concern are both issues that depend on the parties' behavioral inclinations and values. These factors are so idiosyncratic that society should let the parties decide between the two, balancing the possibility of inducing fault against the benefits of increased commitment. One couple might have a less happy, and hence less likely to succeed, marriage if either partner felt trapped by the marriage, that is, with substantial obstacles to exit.⁽⁷⁷⁾ Another couple might want to raise the price of divorce to make it less likely.⁽⁷⁸⁾

B. Societal Interests in Free Divorce

1. Externalities

There are times when society rightly interferes in individual decisionmaking. Substantial negative externalities, spillovers onto third parties, which can lead to both injustice and inefficiency, are reason enough for society to refuse to enforce private agreements. Do such externalities exist in marriages that restrict divorce?

One obvious external cost is the cost of determining whether one party is indeed at fault. The divorcing parties would expend more judicial resources determining fault than are expended by no-fault couples. A court can grant a no-fault divorce with little fact finding. In addition to the costs it imposes on the parties, fault divorce imposes costs on the courts. Why should we pay this price?

It may be, of course, that the benefits to the parties are worth the increased judicial cost. This is true of enforcement of the vast majority of commercial contracts. Why single out marital contracts for special nonenforcement? Society has long enforced other kinds of conditional agreements, many of which impose huge costs on courts, and the trend seems to be in favor of hearing disputes. In light of the trend toward willingness to intercede in other areas of life, is there a good reason to refuse enforcement of marital contracts as a class? The case has not been made that the ratio of costs to benefits for enforcing no-divorce-unless- fault marital contracts is substantially worse than for other contracts.⁽⁷⁹⁾

The judicial costs will never be incurred if the couple never divorces. Requiring fault could reduce the likelihood of divorce,⁽⁸⁰⁾ so it is hard to tell which divorce regime requires more court time. Indeed, the quantity of divorces may be much more important than whether fault needs to be determined, since usually the most contentious and time-consuming issues will be the terms of divorce, not the grounds. This suggests a proxy principle. The parties are in many ways a proxy for societal interests. If the couple decides that those greater costs of divorce are outweighed by the lower likelihood of divorce, then the external costs of divorce might also be outweighed by the lower likelihood of divorce. If they think their marriage has a better chance if divorce is allowed only for fault, society ought to trust their judgment. The question then becomes whether there is any social interest not adequately proxied by private interest, so that following private choice would lead systematically to the wrong societal decision.

Requiring the traditional grounds for divorce would result in more unhappy couples staying together. The unhappiness of the couple, however, is not enough reason to prevent them from binding themselves, unless a case for paternalism can be made. We need societal harms beyond the unhappiness of the choosers to establish the case against private choice. The injustice of spousal abuse could be a cost to the rest of society of keeping couples together, for example, but the traditional grounds for divorce included mistreatment, and so this objection would not arise.

Detrimental effects on children are an obvious justification for not enforcing a couple's agreement. Some experts take the position that children are better off if the couple divorces. Said Gary Sandefur, when asked about the anti-no-fault movement, "The worst thing for kids is to be around a constant state of warfare."⁽⁸¹⁾ Not all experts agree, however. As noted above,⁽⁸²⁾ Judith Younger was so convinced that divorce (even by unhappy couples) is bad for children that she advocated the "marriage for minor children" that could not be dissolved during the minority of children.⁽⁸³⁾ One writer claims that the consensus of recent findings is that divorce helps children if the marriage involves "physical abuse or extreme emotional cruelty," but hurts them otherwise, even adjusting for the income loss which commonly accompanies all divorces.⁽⁸⁴⁾ For the benefit of children, or for other reasons, it may be appropriate for a court in a given case to refuse to enforce an agreement limiting the grounds of divorce. The necessity of that discretion does not, however, justify a categorical rule against enforcement.⁽⁸⁵⁾

2. Fairness

What if one party prefers the default regime? If the law honors agreements, it does two things. It creates the opportunity for gains from trades away from the default marital rights and the possibility that the gains from marriage will be divided unevenly. This is not necessarily a change for the worse; the gains from marriage are not equal under the current rules.⁽⁸⁶⁾ But it is troubling on fairness grounds that even as it increases net welfare the agreement could decrease the welfare of one spouse.⁽⁸⁷⁾ If the law allows bargaining over the marital contract, it gives an advantage to the party who is better at bargaining.⁽⁸⁸⁾ In addition to other reasons that a man can have an advantage in selfish bargaining,⁽⁸⁹⁾ bargaining favors the person who has "a more individuated sense of self,"⁽⁹⁰⁾ the person whose other-regarding preferences are weaker, the person whose utility curve is less intertwined.

One answer is that tremendous potential gains from trade are worth the distributional costs, which have not in any case been proved. In addition, bargaining occurs during the marriage and at divorce whether the law allows premarital contracts or not.⁽⁹¹⁾ The question is when the bargaining will occur, not whether it will occur. Hence, the issue resolves into whether the weaker bargainer would be better off bargaining before marriage. After marriage and as the marriage deepens, we would expect the more interdependent spouse's sense of self to become even less individuated, leading to even less bargaining power. If that is so, earlier bargaining may be better for the weaker negotiator.⁽⁹²⁾

Put another way, bargaining at divorce occurs in the shadow of background rights to alimony, children, and so forth.⁽⁹³⁾ Equally important, bargaining during marriage occurs in the shadow of the possibilities of divorce.⁽⁹⁴⁾ The key question is who establishes those background rights regarding divorce or alimony. It could be argued that the traditional rules of alimony and fault protected wives by establishing a decent initial position, a veto to divorce, and a possibility of alimony from which to bargain. Allowing them to bargain away that protective set of rights before marriage could have worsened the condition of women. Whatever its validity in the past, that argument has little force in most states today. Background rights accorded women by current law provide women with little protection. Women could hardly do worse bargaining for themselves, fixing their own background rights by premarital contract.⁽⁹⁵⁾

Indeed, following the analysis of Amy Wax, one might expect women to do better for themselves than the law has done for them. How well women will fare in bargaining depends on their options available at the time of bargaining. Wax argues that, for a number of reasons, aging diminishes women's options faster than men's.⁽⁹⁶⁾ If she is right, women can strike the best bargains when they are young. Further, Wax shows that, given unequal initial positions, the gains from marriage can be split more evenly if there are more gains to be had;⁽⁹⁷⁾ "the more love, the more the possibility for equality."⁽⁹⁸⁾ If love is hottest early in the relationship, the gains from marriage are largest then and bargaining then is most likely to yield a fair division.

C. Who Should Initiate the Reform?

Assuming the law should enforce agreements regarding the grounds for divorce, should this reform be accomplished through the common law or by statute? Not all improvements ought to be made by judges. A change of the magnitude urged here should not be made lightly and there are reasons that it should not be made by courts alone.

First, the grounds for divorce have long resided in the legislative bailiwick. It is unlikely that many of the statutes specifying the grounds on which courts can grant divorces could fairly be read as allowing parties the contractual freedom of deleting "incompatibility," "irreconcilable differences," or "irretrievable breakdown" from the enumerated grounds. Had legislators thought private parties could pare back the list, they probably would have specified which grounds were optional. Second, even if courts might claim the authority to modify the legislative list by virtue of their inherent judicial discretion not to grant a divorce, the issue deserves the kind of public debate that does not occur in the courtroom. Third, it will do little good for isolated judges to enforce agreements. Although some decisions encourage private contracting,⁽⁹⁹⁾ a few scattered decisions are not enough to keep contracting from being risky. Some courts may recognize the utility of strict enforcement in long-term contracting,⁽¹⁰⁰⁾ but if the parties cannot predict which kind of judicial temperament they will face twenty years down the road, the wisdom of a few particular courts is little help. For all of these reasons, legislative authorization of private agreements regarding the grounds for divorce is essential.

D. Busting Up the Judicial Monopoly on Divorce

As long as the issue of contracts on the grounds for divorce is being considered by the public and in the legislature, another question ought to be addressed: whether a couple should be able to avoid judicial involvement in their divorce. Legislatures have granted the authority to form a legal bond of marriage not only to judges, but also to clerics, with little investigation into the issue of who can join the clergy. Why should the same group not also have the power to decree a divorce? "I now pronounce you man and woman. You may no longer kiss the former bride."

Why should judges have a monopoly in the supply of this good--divorce?⁽¹⁰¹⁾ Most couples have married on the assumption that they would not become divorced without a judge approving the request. We should not upset their reliance by simply extending the power to grant any divorce to additional groups. But it would not upset reliance to allow parties, by mutual agreement,⁽¹⁰²⁾ to provide that their marriage could be rent asunder by, for example, a Unitarian minister. Indeed, allowing couples to specify who could grant the divorce might serve as a reliable shorthand way of specifying the grounds for divorce at a low cost to the public. If the parties agree that they can be divorced only if a Unitarian minister agrees, it serves their interests and saves judicial time for judges to refuse to second-guess the decision of the Unitarian Church.⁽¹⁰³⁾

There are some consequences of divorce--such as child custody where the child's interests must be protected--that do require judicial supervision. In addition, courts must have the power to grant divorces in some situations, regardless of any private agreement regarding the grounds or tribunal.⁽¹⁰⁴⁾ Within appropriate bounds, however, there is room for individual choice of divorce forum. Just as private parties are allowed by the use of trusts to break up the probate monopoly on disposition of assets on death, they ought to be allowed to break up the judicial monopoly on granting divorce.

E. Bilateral, No-Fault Divorce and Renegotiation: A Dialogue

Much of our focus has been on unilateral, no-fault divorce and the problems it creates for long-term commitments. Traditional marriage law, however, also disallowed bilateral, no-fault divorce, in which both parties agree to terminate the marriage without allegations of fault. Should courts enforce a marital agreement that does not allow bilateral, no-fault divorce? A closely related question is whether the parties should be allowed to renegotiate their marriage contract. If they begin with a traditional marriage allowing divorce only for fault, should they be allowed, by mutual consent, to switch to a modern, unilateral, no-fault-divorce marriage? If so, it is clear that agreements to exclude bilateral, no-fault divorce are useless, since such provisions are easily evaded by renegotiation.

The authors of this Article disagree with each other as to the answer. Some of the arguments are suggested in the following dialog between a Liberal and a Conservative.⁽¹⁰⁵⁾

Conservative: The law should enforce a couple's agreement not to allow divorce by mutual consent, and invalidate renegotiation of a marriage contract that tried to exclude that possibility. Credible precommitment allows both parties to rely completely on the enforceability of their contract. Through the contract, they can set up the incentives as best fit their marriage and, having done so, can invest heavily without worry that the contract will be found unenforceable.

Liberal: Divorce reform has changed the legal treatment of two critically different types of situations. One involves a faultless couple mutually agreeing to terminate a marriage. The other involves a single party wanting out of the marriage in the absence of any fault by the other. In both situations, a divorce would not have been granted under traditional rules but could be granted today. The two situations should not be treated the same. A couple should be able to terminate their marriage by mutual consent without any finding of fault, regardless of the terms of their original marital understanding. The shift to no-fault beneficially reduced transaction costs, facilitating efficient divorces. Unilateral no-fault is another story. Allowing one party to terminate without fault or consent of the other made inefficient divorces possible. Courts should allow unilateral termination without a showing of fault only if that is consistent with the particular marriage contract.

Conservative: In both situations the courts should honor the agreement of the parties. The parties are surely in a better position to determine whether the costs of commitment outweigh the benefits.

Liberal: The parties may be better at weighing costs and benefits ex ante, but the court has the advantage of viewing the matter ex post. Though enforcing the contract might create good incentives for parties not yet in miserable situations, it creates terrible consequences ("status effects") for parties already in those situations.⁽¹⁰⁶⁾ Some marriages are mistakes.⁽¹⁰⁷⁾ Your reasoning would justify enforcing Antonio's pound-of-flesh promise to Shylock. Courts would not enforce that promise, and this is an easier case because neither party wants the contract enforced. At some point the status-effect benefits of ex post decisions to terminate miserable marriages outweigh the incentive costs of doing so. The incentive benefits here are not large enough to justify enforcing a no-renegotiation provision.

Conservative: Why, then, would they ever specify the no-renegotiation clause? You underestimate the benefits of this commitment. No matter what unfortunate events befall them, they know that they must make the best of it together. This should create a large incentive for behaving with consideration, for misery of the other will surely reflect back.

Liberal: Even if the law allows renegotiation, there is an incentive to behave considerately, for a spouse cannot count on getting the other's agreement to a no-fault divorce. The marginal increase in incentives to behave well is not so large. And as for why a couple would choose it, they might make a mistake. Remember, most players in this game are not sophisticated, repeat players.

Conservative: Under your rules, they might be.⁽¹⁰⁸⁾ We have already talked about some benefits in this Article, but there are more, and probably some that professors or courts have not thought about.

Akrasia--self-control--could be a motive. The couple might fear that in the future they would be tempted to divorce to avoid short-run difficulties but would later regret it, or they might believe that divorce is sinful and they might succumb to sin temporarily. It would be rational for them to bind themselves, just as a reformed alcoholic would prefer not to be allowed to drink free martinis.

Or, it could be that the couple would like to show third parties that they are committed to marriage. A bank would prefer to make loans to a committed couple, and relatives would be more willing to make marriage-specific investments in a marriage less easily dissolved. The couple may wish to encourage these third parties.

Liberal: Ruling out renegotiation is an extreme form of commitment. It may help to think about this another way. Consider the protection offered the party who does not want out of the marriage. Under current law, a married person has a right to stay married, but that right is protected with only a liability rule: the other spouse can get a divorce by paying a judicially agreed price.⁽¹⁰⁹⁾ I would change the law so that parties could choose to protect their right to stay married with a property rule. A divorce would not be granted unless the party wanting out agreed to buy the other's right to stay married for her price, no matter how unreasonable. You would take yet another step, allowing parties to make their right to stay married inalienable.

Your position runs contrary to standard contract doctrine, which allows renegotiation by mutual consent.⁽¹¹⁰⁾ If couples are allowed to prevent renegotiation, that would make marriage contracts much more binding than ordinary contracts and partnership agreements.⁽¹¹¹⁾

Conservative: You are right that I need to distinguish marriage contracts from ordinary contracts. Even in ordinary contracts, though, renegotiated terms are void if executed under duress, and might be void if there is no fresh consideration.⁽¹¹²⁾ Sailors cannot get their employer to agree to higher wages by threatening to breach their contract and let the fish they caught spoil.⁽¹¹³⁾ The law looks to whether the higher wages are justified by extra work and whether the sailors were relying on being judgment-proof or on the court being unable to detect their breach.

Liberal: I do not argue that the court should honor a modification executed involuntarily or lacking consideration,⁽¹¹⁴⁾ but it would be a rare case when two parties both wanting a divorce did not both receive fresh consideration in the divorce.

Conservative: Marriage is a relational contract and breach is often hard to detect and compensate. A husband might threaten to be unpleasant to his wife unless she agrees to a divorce, and the court could not detect his breach of marital expectations. To preclude that possibility, the wife might want to insist on a marriage that disallowed bilateral, no-fault divorce and add a clause that renegotiation of this term would be void. The idea is like specifying liquidated damages; the parties do not trust the court to resolve a breach correctly without help. Moreover, if renegotiation is possible, the bargaining costs never end, and, in particular, issues that were supposed to be settled in the agreement can be reopened when the parties' tempers are hot and they know which parties benefit from the deal as a result of chance circumstances. The veil of ignorance is torn away; the violinists of our earlier example have discovered which one will be the star.

Liberal: Your argument just brings up a bigger problem: the husband who cannot get his no-fault divorce will commit fault to get it. He will beat his wife or commit some other fault that the law should do its best to prevent.

Conservative: This is your strongest argument, because now you have identified a spillover effect beyond just the two parties: the rest of us do not like wife-beating or adultery, even if the two parties have foreseen this and accepted the possibility.

But that is why we have criminal law. Assault is a crime, and many cities have special laws and enforcement mechanisms for domestic violence. Adultery, too, is a crime, and though rarely prosecuted (except in the military) states could step up their enforcement.⁽¹¹⁵⁾ After all, if assault or adultery is ground for a divorce, there must be evidence available for a criminal prosecution.

Liberal: Domestic violence is hard to prosecute, though. We must accept that criminal law is too blunt an instrument for dealing with these offenses. They are crimes today and yet most often go unpunished. Moreover, remember that when the issue is divorce the burden of proof for fault is much less than when the issue is incarceration. And even if we could count on punishment, do we really want him to answer, by his behavior, the question whether it is worth six months in jail to be free of his wife?

Conservative: How about using the terms of divorce, then? If the marital contract specifies that someone divorced for adultery gets none of the marital assets, our reluctant husband will hesitate to use those grounds.

Liberal: If he wants out badly enough, he will not hesitate. But let me shift our attention to a milder externality, the undermining of honesty. In the old days, when two people wanted a divorce, they would manufacture evidence of fault.⁽¹¹⁶⁾ The husband could pretend to be caught in a compromising situation in a motel somewhere, and the judge would agree to a divorce on grounds of adultery. Thus, courts did not prevent a bilateral, no-fault divorce, resources were wasted and ethics were compromised manufacturing evidence, and the integrity of courts was undermined.⁽¹¹⁷⁾

Conservative: If the couple wants a divorce badly enough, that is one method to evade their contract, I agree. But the public declaration of fault is embarrassing, and this method not only requires certain transaction costs,⁽¹¹⁸⁾ but also cooperation between two people who presumably are on bad terms. If the terms of divorce penalize the party at fault, the other party is unlikely to honor a side agreement to refund part of the court's award later. And a vengeful party would be tempted to turn state's evidence later to punish the other party for falsifying evidence.

Liberal: History undermines your argument. The law required fault, criminalized abuse, and considered fault in alimony awards, yet parties still fabricated fault. And judges winked at the fabrication. The fact remains that an exclusion of divorce by mutual consent increases the incentive for fault and pretended fault.⁽¹¹⁹⁾ Because I see no compelling reason to enforce provisions barring renegotiation, I conclude that renegotiation should be allowed in order to reduce judicial acceptance of lying and to reduce incentives for conspiratorial, faulty behavior. The corollary to this conclusion is that provisions stating that the contract cannot be renegotiated should be void ab initio.

The reader may decide who wins the argument. The problem of renegotiation remains an active subject of study even in the context of commercial contracts,⁽¹²⁰⁾ and we have not been able to resolve it here.

IV. Enforcing Contracts Regarding Matrimony

A. A Traditional Division of Labor with a Sharing of Income

We now turn to contractual terms regulating behavior during the course of marriage. Suppose Linda and Paul agree that they want to allow divorce for any reason, but wish to form a binding agreement as to certain terms of marriage. They intend that during wedlock Paul will provide the financial income and Linda will raise the children and take care of the household. They agree that half of Paul's income (enforceable by garnishment) will be deposited in an account belonging to Linda and that amount will be reduced by any income she receives from employment. In addition, they agree that they will remain sexually faithful to each other and that the price of infidelity will be the payment, from the unfaithful to the faithful partner, of half of Paul's previous year's taxable income, as declared to the Internal Revenue Service. Should and would courts enforce this deal?

First, could any sensible couple desire such an arrangement? Yes. In a traditional American marriage, the husband owed a duty of support to the wife and children. This duty arose only after a formal registration of marriage. Sexual fidelity was expected of both spouses, as was avoidance of excess in vices such as drinking and gambling. Divorce was granted for desertion or violation of these duties, and the party at fault was penalized in the terms of divorce. If divorce was avoided, inheritance laws guaranteed the surviving spouse some share of the deceased spouse's assets.⁽¹²¹⁾ A couple might wish to reproduce a package of marital expectations similar to what was for a long time an attractive package to many couples.

There is at least a possibility that a traditional division of labor will produce "a larger marital pie than a comparable dual earner arrangement."⁽¹²²⁾ Moreover, as to dividing those gains, there is some evidence that women with children do not fare well in marriages where both spouses work outside the home. At least as measured by equality of leisure time between husband and wife, wives working at home in traditional roles have a fairer division of the marital workload than their working counterparts.⁽¹²³⁾ A woman wanting to have children might sensibly prefer a single, full-time job to a "double day."

With looser social norms and marriages between people of more heterogeneous beliefs, it may be more important than in the past to have legal enforcement of behavior within the marriage.⁽¹²⁴⁾ Contingent financial payments might be appropriate for couples committed to each other for the long term. Both parties could feel that even after sexual unfaithfulness they wish to stay together. But they want to create a disincentive for unfaithfulness, a disincentive that does not penalize the faithful partner more than the unfaithful one. They also want to give the faithful partner some compensation for the behavior of the other. A contingent payment could serve to make the faithful partner feel that some justice has been done without terminating the commitment. This agreement avoids reliance on the legal process of divorce as the punishment for unfaithfulness.

Is such a contract legally viable? Relatively recent statutes lay the groundwork for increased judicial receptivity. Unlike the grounds for divorce, contracts on the terms of matrimony could be enforced by judges without any new legislative authorization. Section 3 of the UPAA lists the kinds of provisions allowed. These include anything with respect to property, control of property, disposition of property upon separation, death, or any other event; modification or elimination of spousal support; wills, trusts, etc. as part of the agreement; life insurance; choice of law for governing construction of the agreement; and "any other matter, including their personal rights and obligations, not in violation of public policy or a statute imposing a criminal penalty."⁽¹²⁵⁾ Subsection 3(a)(8) is a catch-all, which the official comment says allows choice of residence, career plans, and children's religion.⁽¹²⁶⁾

Notwithstanding this possible authorization under the UPAA, Paul and Linda cannot feel secure that courts would enforce their agreement. Courts might enforce the financial terms, but they do not enforce most terms of behavior during an ongoing marriage.⁽¹²⁷⁾ Given current judicial attitudes, a package of mutual obligations that was once popular, though enforced only by social norms, is of dubious legality.

Once again the question must be asked: Have we changed so much that what once was commonplace is now unacceptable? Is the agreement itself contrary to public policy?⁽¹²⁸⁾ Certainly the goals are not illegal under the criminal law. As for other public policy, does society suffer if Paul and Linda realize, in addition to a lasting marriage, their wish of substantial behavior devoted to the family instead of the office? Having one spouse stay at home may generate positive externalities. The presence of one spouse in the home should reduce the costs of police protection paid for by others and may even reduce the need for police protection for all neighbors, since there would be more monitoring during workdays.

A single-earner marriage generates less taxable income than a dual-income marriage, but that does not, of course, mean that it creates less societal wealth. Indeed, the fact that the parties have chosen it suggests just the opposite.⁽¹²⁹⁾ If tax revenues are too low, the tax rules should be revised--perhaps by increasing the rates on single-earner, married households--rather than discouraging the parties from generating the additional wealth just because society has difficulty taxing it. On the other hand, if the problem is that couples would choose single-earner marriages because of lower taxes, perhaps two-earner households should be given a deduction.

Are the spillover effects on children from the division of market and household labor harmful enough to prevent a traditional marriage? Surely not. Traditional roles allow parents to spend more time with their children. Beyond the issue of "quantity time," parents often do a better job educating and nurturing their children than temporary caretakers. In addition, other children benefit when a sick child stays at home with his or her parent rather than being sent to school or day care.

Although the contract between Paul and Linda does not itself contravene public policy, it still might not be worth enforcing. Judges have long been reluctant to enforce such terms because of the cost to the courts, the difficulty of enforcement without invading the sanctity of the marital home, and the possibility that enforcement would increase conflict within marriage. But although such problems might arise in the enforcement of many terms relating to the conduct of a marriage, they do not arise in the enforcement of Paul and Linda's agreement. Their agreement calls for the payment of money on certain contingencies that are no more difficult to prove than many contingencies in commercial contracts. Nor does the agreement call for judicial invasion of the marital home. Whatever courts may do with more problematic provisions, the agreement between Linda and Paul ought to be enforced.

B. More Difficult Cases

1. Judges in Bedrooms

What about those more difficult marriage-contract cases? Consider the invasion contemplated by the court in Favrot v. Barnes. The husband tried to avoid paying alimony by claiming that his wife committed marital fault by seeking sexual intercourse three times daily when a premarital agreement said they would "limit sexual intercourse to about once a week."⁽¹³⁰⁾ Should courts be free to ignore such provisions?

Although merely hearing the evidence might in another era have caused substantial discomfort to finders of fact,⁽¹³¹⁾ the imposition on their sensibilities no longer seems important. As for the married couple, it is they who bear the burden of the judicial intrusion into their affairs. It might not be wise for a couple to argue their bedroom in public, but it is not so clearly unwise that the law should always refuse to hear the matter. The primary costs of intrusion are, after all, borne by the husband and wife, and it is they who may best know whether the possibility of court intervention would increase conflict or reduce it. We cannot be confident they are wrong in concluding that the benefits of an agreement are worth the potential sacrifice in privacy needed to enforce it. The law's refusal to get involved leaves control in the hands of the more powerful spouse, a situation which the couple might sensibly wish to avoid. Given the increased importance of legal enforcement when social constraints are few, it is time for courts to stop refusing enforcement on the paternalistic ground that interference will be harmful to the marriage.

2. Marital Discord

A similarly unacceptable reason for refusing to enforce a marriage contract is that it will lead to marital discord. Current legal default rules are as likely to cause trouble as many provisions denied enforcement by the courts. Consider the oral agreement in Koch v. Koch⁽¹³²⁾ that the husband's mother would live with the couple in their home. The court ruled that the agreement was invalid because it was oral and said that even if it were written it would be unenforceable, as tending to cause contention. But the parties should know better than the court what would lead to discord. Ordinarily, we expect that settling contentious issues in advance will lead to a more harmonious marriage. Moreover, it is not clear why the court should prize marriage over all other relationships. Mr. Koch might have felt that his relationship with his mother was more important than that with Mrs. Koch, and she might have been willing to accept him on those terms. Courts have no business deciding that the nonmarital relationships given primacy by the parties are secondary to the marriage.

3. Verification

This is not to say that all provisions should be enforced. The cost of enforcement to the court is one legitimately considered by the court. Since we do not charge litigants the full costs to the public of operating the courts, the couple may too readily resort to law. Verification of terms such as those in Favrot v. Barnes may be expensive. Courts may rightly refuse to enforce agreements when society will have to pick up a substantial tab for monitoring.

This problem may, however, be overstated. The court may not end up having to pay much of those costs. It is up to the aggrieved party to provide evidence for his position, and if proof is too difficult and the plaintiff has no evidence, the court will dismiss the case. In addition, as can be seen from the Rodney King case,⁽¹³³⁾ videotapes are reducing the cost of monitoring and increasing the reliability of the evidence of misbehavior. Moreover, some kinds of contractual provisions would be simple to verify. Is Christmas spent with the set of parents that the wife chooses? Do the children attend the school the husband desires?⁽¹³⁴⁾ These conditions are simpler than many of those encountered in commercial law regarding the quality of a product or the timeliness of delivery. When judges confront provisions for which compliance is too difficult to monitor they can refuse enforcement on the ground that their involvement would be a waste of judicial resources. They need not, and should not, refuse to enforce all provisions governing the behavior of the parties during marriage.

C. Standard Contract Doctrines: Vagueness and Consideration

A number of standard doctrines of contract law would invalidate some marriage agreements even if they were granted the status of ordinary contracts. Vagueness is one. Are the duties of a husband or wife clear enough to be contractible? Vagueness is not special to marriage, however. Employment contracts are often just as vague, yet are commonly enforced.

One contract doctrine that has been used to limit the enforceability of marital agreements is consideration. Courts have said that a husband's promises to pay a wife for housework and other domestic services are void as against public policy, and lack consideration.⁽¹³⁵⁾ Modern marriage does not itself impose on the husband or wife a duty to clean the house or perform other domestic duties. Therefore, agreements to pay for such services do not lack consideration and should be enforced. The reasons and assumptions of marriage vary widely and the essential core of marital duties has shrunk to almost nothing. Because times have changed, courts should be loath to hold that any promised performance is not consideration.

Some people, those with "big hearts," have the capacity to bring a large emotional contribution of love and affection to a marital alliance. Suppose such a woman is interested in a man with a smaller heart who is willing to offer money to supplement his meager affection. If they marry, the law's refusal to recognize her contribution as consideration results in unfairness to the woman because she cannot enforce his promise after she has performed. If, knowing the promise is unenforceable, the woman refuses his offer, the result is unfairness to the man. Justifiable limitations contained in ordinary contract doctrine may be applied to marriage contracts as well, but these doctrines should be applied fairly.

D. Public Policy, Externalities, and Illegalities

There will, of course, be provisions that should not be enforced for other reasons of public policy.⁽¹³⁶⁾ Suppose a couple agreed that the wife would raise the children, would not get a job or prepare herself for a paying job, would not get any of the husband's income during the marriage, would not be entitled to any other support during the marriage except at the whim of the husband, and would not get any property or alimony upon divorce, which would be allowed without showing of fault.⁽¹³⁷⁾ Although this arrangement is not all that far from the current legal regime,⁽¹³⁸⁾ the contract should not be enforceable. The problem is that this agreement sets up a substantial possibility that the wife would become destitute and a ward of the state.⁽¹³⁹⁾ Such an agreement casts enough negative spillovers on society for the law to refuse enforcement.⁽¹⁴⁰⁾

Courts should remain free to exercise their discretion to refuse enforcement in a number of other situations. Some matrimonial agreements will be too costly for a court to monitor. An agreement requiring the wife to illegally smoke marijuana each day with her husband would no more be enforceable than an employment contract requiring that an employee smoke marijuana with her boss.⁽¹⁴¹⁾ A provision allowing a man to beat his wife, even if they call it "boxing," is an easy example of a private agreement that courts should be free to ignore.⁽¹⁴²⁾

E. Remedies

Supposing that agreements as to the conduct of the marital partners can create enforceable rights, what remedies might courts use to enforce the terms? The biggest problem is that many marital agreements are contracts for personal services, and courts do not generally enforce such contracts with specific performance. Money damages are available, but given the difficulty of measuring nonmonetary benefits from marriage and the possibility of judgment-proofness, money damages would often be insufficient.⁽¹⁴³⁾ In the example of Paul and Linda above, however, the contract can be adequately enforced by monetary damages. Such debts could be enforced without enjoining the performance of personal services. The contingent payments have the important advantage that they can be set large enough to compensate for nonmonetary losses to the parties and save the court the cost of having to estimate those losses.

Consider again the example of the agreement that the husband will provide no support.⁽¹⁴⁴⁾ What should be done when the contract is unenforceable as against public policy?⁽¹⁴⁵⁾ Unfortunately, no one provision can be singled out as the faulty provision and refused enforcement. Any single term might be allowed in the context of other provisions designed to minimize the chances of negative externalities, but taken together they are not acceptable. What should a court do? One approach would be to apply a default marriage rule, but that requires setting up a legal rule that determines when a contract is so bad that the court should shift to the default rule instead of just refusing to enforce the bad provision. It may also take the parties a long way from what they originally agreed.

The better approach is to cy pres the agreement, reforming it to something acceptable. The objective, of course, is to decide what allowable agreement is closest to the parties' intent. The possible modifications will depend on the posture of the suit. If the wife sues for support during marriage, the court could require the husband to support the wife. If the suit is the husband's request for divorce, the court could award property or alimony notwithstanding the agreement. The judge reviewing the contract and facing a claim of breach or request for divorce and finding an unenforceable term should ask what effect that has on the overall agreement. In some cases, it might be that the closest agreement is no agreement at all, and the court would declare an annulment. Or the judge might find that an unacceptable term is stricken and the rest of the contract stands. Or the judge might find that many or all of the terms are stricken, in the extreme case leaving the parties in the default marriage. It might even be appropriate to find that the parties had created one of a number of standard-form contracts allowed by law.

F. Third-Party Recognition of an Agreement as a "Marriage"

We have focused on traditional marriage, but the questions raised will apply to other individualized marriage contracts. Homosexual partnerships, polygamy, kept mistresses, and other relationships might all be arranged by contract, except that provisions relating to crimes such as sodomy or prostitution would not be enforced.

These last examples suggest another issue not yet discussed: whether third parties should be forced to honor agreements the law holds binding on the couple. We do not suggest that the Catholic Church or the Internal Revenue Service must, or even should, recognize an unorthodox marriage. As a general matter, third parties can decide for themselves how they will treat individualized marital contracts. Some employers now treat a long-term homosexual (or other) relationship as equivalent to marriage for the purposes of spousal benefits.⁽¹⁴⁶⁾ But these matters ought, like the terms of the marital contract, to be left to the parties involved in the employment contract. Such third parties have made use of the legal default rule for marriage on purpose and it would be unfair to add new obligations to those originally agreed by contract.⁽¹⁴⁷⁾

A more difficult question is whether the government itself would be bound to recognize private marital agreements as marriages.⁽¹⁴⁸⁾ The answer depends on the government and the particular legal issue in question.⁽¹⁴⁹⁾ Clearly, the federal government is not bound by a state's decision to set couples free to define their marriages.⁽¹⁵⁰⁾ The federal government could, if it wished, decide that a contractually sexless marriage does not qualify as a marriage for social-security purposes, although it might then ask itself how many other married persons have sex so rarely that they should be denied their spousal benefits.

But state and local governments are not so clearly above state law.⁽¹⁵¹⁾ Would recognition of private marital contracts preclude prosecution for crimes against nature? Not necessarily. The law in some states today prohibits married and unmarried persons alike from certain acts. The issues of whether the acts are and should be illegal has been and will still be separable from the issue of whether the couple is married. What acts, consensual or not, should be crimes is beyond the scope of this Article.

One crime, however, we cannot avoid discussing: the crime of sale of sexual services. Clearly some contracts could be read, as some marriages could be seen today, as mere sales of sexual services. Perhaps there is a way around this problem. One of the key concepts behind marriage, one of the only provisions that has remained stable during the sexual and divorce revolutions, is that marriage, however temporary it actually is, is not intended to be temporary. Hence, it would be consistent with the idea of marriage and all variations of past marital law for courts to refuse to enforce as marriages agreements intended to be temporary. Marriage is a relationship of open-ended commitment, and explicitly temporary arrangements are not marriages. Therefore, supposedly "marital" contracts for temporary arrangements need not be recognized as legal marriages. Similarly, temporary contracts for the sale of sexual services need not be recognized as marriages and, therefore, may be prosecuted under prostitution laws.

The most important third parties are the children. Clearly they have no opportunity to determine whether to recognize their parents' marriage. Therefore, their interests need not be subject to the control of their parents' agreement. A provision that the husband shall have no financial responsibility for the welfare of the children is not enforceable because it is not in the best interests of the children and it might result in children being wards of the state.⁽¹⁵²⁾

Our message is to let the parties decide what "marriage" means to them. If contracts restricting grounds for divorce are upheld, it may be all the more important that parties be able to agree also on what that enduring marriage will be like. That does not, however, mean that the parties can decide what marriage means for tax purposes, for spousal-privilege purposes, for private-pension or employment purposes, or for child support. Ours is a baby step away from the status quo.

V. Dissolution Terms

Premarital agreements often specify the terms of divorce, including provisions dealing with the distribution of property, custody of children,⁽¹⁵³⁾ and perhaps what behavior is permitted the parties after divorce.⁽¹⁵⁴⁾ If the parties have agreed on the division of property at divorce, courts will often uphold their agreement.⁽¹⁵⁵⁾ Hence, this dimension of divorce law does not cry out for reform. But a well-formed agreement on property division, however important and useful,⁽¹⁵⁶⁾ in many cases cannot provide enough security for the simple reason that the homemaker cannot count on there being enough property at the time of divorce to provide comfortable, lifelong income. Nor will courts help much, as substantial permanent alimony is not usually awarded.⁽¹⁵⁷⁾ For these reasons, a person wishing to specialize in risky or nonportable production needs an agreement that provides some portion of the other spouse's income.

Suppose Michael and Cheryl, both embarking on high-risk sports careers, agree in writing that unilateral divorce will be allowed without a showing of fault, and that after unilateral or bilateral divorce they will each give 35% of their income to the other until death or until the historically lower-earning divorcé(e) remarries. They also agree on two possible adjustments. First, a higher-earning spouse unilaterally dissolving the marriage without showing a listed type of fault would have to pay 15% more than if fault were shown, for a total of 50-50 income sharing. A lower-earning spouse unilaterally dissolving the marriage without showing fault would receive 10% less than if fault were shown. Similarly, the postdivorce contribution from the higher-earning spouse to the lower will be increased by 15% if the higher-earning spouse committed infidelity during the marriage and will be decreased by 10% if the lower-earning spouse committed infidelity during the marriage. One party could legally divorce the other, but if this agreement were enforced the higher-earning former spouse would still have continuing obligations.⁽¹⁵⁸⁾

Could this be an attractive contract? If enforceable, it would allow the parties to do for themselves what the public has not seen fit to do for all couples: create a disincentive for fault. Such a contract might be especially attractive in a jurisdiction not allowing constriction of the grounds for divorce. Imposing a cost on a party who dissolves the marriage without proving fault could be used as an imperfect substitute for restricting the grounds for divorce to fault-based grounds.⁽¹⁵⁹⁾

Michael and Cheryl's agreement might be enforceable. Courts do allow antenuptial agreements some leeway.⁽¹⁶⁰⁾ The biggest obstacle to individualized tailoring of divorce consequences is uncertainty over whether courts will enforce agreements that base property division on fault.⁽¹⁶¹⁾ A court might refuse enforcement for various reasons. A court in a jurisdiction disallowing private limitation of the grounds for divorce might say that it should not allow private agreements attempting to circumvent the law to achieve the same illegal goals by legal means. Parties should not, the court might say, make something that is legally a ground for divorce into a nonground by penalizing someone who sues for divorce on the forbidden ground.⁽¹⁶²⁾ Another basis for refusing to enforce this agreement is that alimony or other payments might encourage divorce. The terms of the agreement, by providing security after divorce, reduce the price of divorce to one party, and thus could be found to contravene public policy.⁽¹⁶³⁾

Need for agreement relating to consequences of divorce sometimes arises from religious conviction.⁽¹⁶⁴⁾ For example, an Orthodox Jewish woman who thinks she might wish to remarry following divorce must find a way to assure that she can get a religious divorce if she becomes civilly divorced. Without the religious divorce, she cannot remarry except in a civil ceremony. Once she is married, it is difficult for a former wife who wants a religious divorce from her husband to induce him to appear before the Beth Din, the rabbinical court, if he can get a civil divorce unilaterally. Antenuptial contracts, "ketubahs," are used to compel such appearances.⁽¹⁶⁵⁾

As mentioned above, courts sometimes refuse to enforce marital promises on the ground that they lack fresh consideration. In the context of promises by married persons relating to divorce, the judicial search for fresh consideration can be especially misguided. Suppose a husband is eager to maximize the opportunity for his wife to invest in the marriage. Knowing that she fears the loss of income that might accompany divorce, he executes a written promise to share his income with her even after divorce for as long as she is not married to anyone else. Under standard doctrine, she has provided no new consideration and his promise is unenforceable. But his promise was made to allow his wife to devote her time to the marriage, which certainly redounded to his benefit. To refuse to enforce the promise is both unfair to the wife who has made career choices in reliance upon it and is harmful to husbands wanting to provide their wives security for their own benefit. A single promise by one spouse may create consideration for both the promisor and promisee.

Similarly, courts should hesitate to invalidate agreements providing for security in the event of divorce on the ground that they tend to encourage divorce.⁽¹⁶⁶⁾ A provision that a spouse obtains substantial property on divorce has the possible effect of encouraging divorce, but it also has the effect of encouraging investment in the marriage. It ought to be up to the parties to weigh the costs and benefits of such a provision.⁽¹⁶⁷⁾

On the other hand, a common precondition to the enforcement of ordinary contracts is that the parties have adequate time before execution for consultation and careful deliberation regarding the terms and legal effect of the contract. Such judicial oversight is especially appropriate for premarital agreements. If a man springs a premarital agreement on his fiancée while the extended families are assembling in the chapel, she does not have enough time to think through her options and the agreement should not be upheld against her wishes.⁽¹⁶⁸⁾ Legislatures could minimize the unpredictability of an "adequate time" doctrine by providing a statutory safe harbor, that agreements executed more than one month before the wedding, for example, would not be unenforceable on this ground of procedural unconscionability.

VI. Implementation

Now that we have discussed all three parts of the marriage contract--terms of wedlock, grounds for divorce, and terms of dissolution--we will discuss some considerations that apply to all three.

A. Choice Hurts

Making choices is painful.⁽¹⁶⁹⁾ Some restaurants have no menu, offering only one item for dinner and saving their customers the time and aggravation of deciding what to order. Increasing the marital options open to couples will increase the costs of determining what marriage is appropriate. Not only will individual decisionmaking be more costly, but negotiations with the marital prospect will also be more costly since there is more to negotiate. In addition to rising costs, and as a result of that price increase, some marriages that would have been happy will not occur. On the other hand, some marriages that were not allowed or facilitated under current rules will occur, so the direction of the net change in the number of happy marriages is difficult to predict.

One way to diminish the costs of negotiation is to offer a menu of legislatively approved, alternative, standard-form contracts. Not only would this save drafting,⁽¹⁷⁰⁾ but the alternatives could be the focus of educational efforts that would help people learn about their options. Proposals in various states including Indiana,⁽¹⁷¹⁾ and a bill enacted in Louisiana in 1997,⁽¹⁷²⁾ add a new type of marriage license, a "covenant" license, dissolution of which is more difficult than a standard "contract" license. The law should offer more than those two choices.

If a state, in the course of offering a menu of options, changes the default marriage from the existing law, it should be made clear that the new default does not apply to existing marriages. One of the great injustices of previous reforms was that couples entering marriages under a regime promising some security were deprived of that security when the law changed. That mistake ought not to be repeated.⁽¹⁷³⁾

B. Forum Shopping and the Need for Federal Law

Suppose Indiana decides to allow agreements restricting divorce and an Indiana couple, Henry and Marian, state that their fault-only divorce agreement shall be governed by the law of Indiana regardless of their future residence.⁽¹⁷⁴⁾ What happens if the couple relocate to Nevada? Will Nevada refuse to grant a unilateral, no-fault divorce to one of them?⁽¹⁷⁵⁾ If the agreement and others like it are to be useful, Nevada should refuse. And under ordinary conflicts rules, a Nevada court would honor the contract's specification of Indiana law, unless it were contrary to Nevada public policy.⁽¹⁷⁶⁾ But divorce law does not follow the rules for contracts. Courts have considered marriage to be status rather than contract. Historically, states have felt a powerful interest in marriage.⁽¹⁷⁷⁾ Because of that concern, states have applied local law rather than the law of the state in which the couple were married.⁽¹⁷⁸⁾ Although divorce reform shows that states no longer have such a strong interest in whether couples are married or not,⁽¹⁷⁹⁾ the courts have not ceased viewing the issue as one of status to be decided by local law. Nevada would feel as free to ignore Indiana divorce law today as it has in the past. If Nevada refuses to honor the Indiana agreement,⁽¹⁸⁰⁾ such agreements become less useful. Spouses can protect themselves by remaining in Indiana, but that protection comes at some cost.

A mutually agreed relocation is not the only contingency threatening the agreement. Suppose Henry takes a trip to Nevada. If he establishes domicile, his wife, Marian, has no power to stop him from getting a Nevada divorce that will be recognized in Indiana. In 1942, the Supreme Court held in Williams v. North Carolina⁽¹⁸¹⁾ that other states must recognize a unilateral, fault-based divorce validly granted by a Nevada court, despite Nevada's lack of personal jurisdiction over the unwilling spouse. The Nevada court may assert jurisdiction over Henry and his status even though it has no jurisdiction over Marian back in Indiana. Not only does Nevada have the power to invalidate the Indiana restriction on divorce, it has an incentive to do so. By offering divorces not available elsewhere, Nevada promotes its tourism industry. Seeing a market opportunity and eager to supply to the demand, it might well follow its own law.⁽¹⁸²⁾

Contractual restrictions on the grounds for divorce can be circumvented by one spouse as long as some state will ignore them.⁽¹⁸³⁾ To protect their marriage-law requirements, states have enacted anti-evasion statutes.⁽¹⁸⁴⁾ It is not equally easy, however, for states to protect their rules, or their citizens' private agreements, limiting divorce. The problem lies partly in the federal constitution's requirement that "Full Faith and Credit . . . be given in each State to the public Acts, Records, and judicial Proceedings of every other State."⁽¹⁸⁵⁾ Since a marriage is less clearly a "judicial proceeding" than a divorce decree,⁽¹⁸⁶⁾ states may be free to ignore marriage agreements binding in other states.⁽¹⁸⁷⁾

To give Nevada sufficient interest in Henry's status for its decision to be binding on Indiana, Henry must be domiciled in Nevada. This requires that Nevada be his permanent place of residence and that he intend to reside there for the indefinite future. This domicile requirement provides Indiana with an opportunity to protect its divorce limitations. As part of its divorce-reform law authorizing private agreements on the grounds for divorce, Indiana could make it a crime of desertion to unilaterally move out of the marital home and take up residence in a new home with the intent to remain there in the indefinite future.⁽¹⁸⁸⁾

However, even a strong law against desertion might fail to give Indiana couples sufficient confidence in their divorce-limiting agreements to allow comfortable reliance. There always exists a possibility that the couple will later move to a state lacking a strong desertion statute. After such a move, it would be possible for one spouse to establish domicile in Nevada without fear of criminal prosecution.

If one state does not honor marital agreements and other states cannot protect their agreements, a national law forcing states to honor foreign agreements may be needed to make premarital contracts adequately reliable.⁽¹⁸⁹⁾ A national law legalizing agreements is not necessary. Nor do we advocate a national law requiring Illinois to honor a marital agreement by an Illinois couple that Indiana law would govern. What would be needed, however, is a national law that says agreements regarding the grounds for divorce that are effective in the couple's domicile at the time of execution must be honored by other states.⁽¹⁹⁰⁾ On the other hand, even without a national law, states might feel some pressure to recognize marital contracts because some spouses would refuse to move to states not willing to honor the agreements they are relying upon. At a minimum, however, it may be necessary to prevent states from offering divorces to one spouse without the consent of the other spouse.

This issue shows the importance of the distinction between contract and status. If marriage is status, and status is determined by the various states, then moving from one state to another changes the terms of marriage and divorce, changing with them the incentives for behavior, the fairness of the arrangement, and the happiness of the couple. Indeed, even if the couple does not move, the mere possibility of moving undermines the incentive structure and their happiness. By contrast, if marriage is contract, then the relationship remains the same, for the most part, regardless of the couple's migration.

The flip side of this forum-shopping point is that if states do honor marital agreements from other states, then it only takes one state allowing contracts to put pressure on others. If one state passes the law, and couples see the options as attractive, that state will become a magnet for marriage ceremonies. The tourism business alone might cause others to follow suit in a race to the top. Virginia says it is for lovers,⁽¹⁹¹⁾ but other states with freedom of marital contract may claim greater hospitality.⁽¹⁹²⁾

VII. Conclusion

In the past, it was clear why few people in England and America executed a marital contract. Religious and social norms defined a "marriage" and the gender roles within a marriage. The law did not allow contractual variations out of keeping with the religious and social norms. As a matter of public policy, it was thought that the interests of society required the fostering of a certain type of marriage, in the interests of child rearing and stability, and the desires of the individual needed to be subordinated to social order.⁽¹⁹³⁾ Thus, social norms, religious rules, and legal doctrine prevented people from entering into a marriage agreement that might allow easy dissolution or unusual roles.

Some of the reasons why it was once difficult to contract out of a traditional marriage are clear. But many of those restraints on individual liberty have weakened or disappeared. Prevailing opinion has changed. Although still limited, a marriage today is considered much more a matter for the two parties concerned, not for society, to structure. Yet few people entering their first marriage memorialize their shared understandings, obligations, and aspirations by contract. The rarity of individualized contracts is partially explained by inertia and partly by lack of awareness that being bound by marital contracts can be a good thing for both parties. But another reason is that no one can be sure that courts will enforce the contracts. The law has not kept pace with societal sensibilities. The freedom valued by society and offered by expanded social norms is not enabled by the law.

No-fault divorce might seem to foster individual choice, in keeping with the spirit of the age, but it does not really favor individual choice. Some people would like to be able to choose to bind themselves in a permanent marriage, yet the law makes it difficult to personalize the contract. One legal size is presumed to fit all. The chief problem comes in being unable to specify the grounds for divorce, either directly or through using the terms of divorce to penalize a spouse who is at fault. The long-established rule against judicial interference in ongoing marriages further hinders the establishment of individually tailored marriages.

Moreover, social norms and families have become weaker. With nonlegal constraints weakening, people need legal institutions to pick up the slack, allowing them to make credible commitments to each other. It is time for legislators and judges to clarify to what extent courts will enforce marital agreements.

We have argued that there is a rational basis for traditional marriage bonds constraining husband and wife. Some people rationally wish to bind themselves to a relationship with each other. They do so not just because of akrasia, the weakness of will, anticipation of which motivated Ulysses to bind himself to the mast. They do so also for the same reason as business partners bind themselves: so that each party can make relationship-specific investments without fear of having them rendered useless by the other's perfidy.

We do not argue that traditional marriage should be the only form allowed. Such an argument would run counter to our claim that couples differ in their need for legally binding commitment. Our point is that no one has a choice and that some couples suffer for it. Marriage is still a matter of status,⁽¹⁹⁴⁾ and its failure to move to contract is causing much harm.

The center of the divorce-law discussion relates to what should be the best rules for custody of children and allocation of marital assets or future income. Many scholars raise the controversy up a level, arguing that one theory or another should be used in deciding how to reform the law of marriage.⁽¹⁹⁵⁾ At either level, the argument assumes that we as a society need to decide what is best for people. Our claim is that the debate regarding whether society or couples should decide what is best for couples has not been resolved satisfactorily.⁽¹⁹⁶⁾ The dispute often looks like an argument over whether states should build sedans or minivans for everyone to drive, or what criteria or theory we should use in deciding which car states should build.⁽¹⁹⁷⁾ The first point of attention should be whether the government ought to let people choose.⁽¹⁹⁸⁾

The legal system should increase private choice in marriage and divorce law. We need not give as much freedom of contract as exists in the commercial context. Indeed, given the lack of sophistication of most people, that would help little. What is essential is that legislatures recognize that people need to be able to commit themselves to each other by judicially enforceable agreements. Statutes could usefully provide forms with several enforceable and reliable options. In addition to modern, no-fault exit without alimony, premised on both spouses developing careers, the traditional, fault-limited divorce with alimony, premised on efficient division of labor, ought to be an option. The law does not need to provide enforcement for every possible kind of marriage, but it should provide clear and dependable enforcement for a few kinds, without any more uncertainty from fairness, unconscionability, and public policy limitations than is present in ordinary contract law. In the interests of individual autonomy, self-control, and self-realization, in the interests of freedom of permanent association, in the interests of privacy and efficiency, couples ought to be allowed to structure their lives as they wish. Much more than it does today, the law should lift the veil of ignorance shrouding marriage.

1. † © 1998 by Eric Rasmusen & Jeffrey Evans Stake.

2. Associate Professor, Department of Business Economics and Public Policy, Indiana University, Kelley School of Business. B.A., M.A., Yale University, 1980; Ph.D., M.I.T., 1984. E-mail: erasmuse@indiana.edu. Homepage: http://php.indiana.edu/~erasmuse. Phone: (812) 855-9219. Fax: (812) 855-3354.

3. Professor, Indiana University School of Law-Bloomington. B.A., University of Illinois, 1975; J.D., Georgetown University Law Center, 1981. E-mail: stake@indiana.edu. Phone: (812) 855-4444. Fax: (812) 855-0555. Our thanks go to Michael Alexeev, Gregory Alexander, Margaret Brinig, F.H. Buckley, Daniel Conkle, Robert Heidt, Kenneth Dau-Schmidt, Allan Farnsworth, Joseph Hoffmann, J. Mark Ramseyer, Larry Ribstein, Lauren Robel, Gene Shreve, Emily Van Tassel, Nathaniel S. Weeks, and the Econlaw discussion list for helpful comments. We also thank T. Lynn Fisher and Gretchen Lyn Grosick for research assistance. Although some readers might try to deduce the relative contributions from the order of author names, the authors wish to discourage it. Readers who know Stake will recognize that three-quarters of the ideas are his, and readers who know Rasmusen will recognize that the other three-quarters of the ideas are his.

4. See Henry Sumner Maine, Ancient Law 141 (Dorset Press 1986) (1861); cf. 2 Max Weber, Economy and Society 671 (Guenther Roth & Claus Wittich eds., University of Cal. Press 1978) (1956) ("[T]he farther we go back in legal history, the less significant becomes contract as a device of economic acquisition in fields other than the law of the family and inheritance."). But see Grant Gilmore, The Death of Contract 3 (1974) ("Contract, like God, is dead."); Roscoe Pound, The End of Law as Developed in Juristic Thought, 30 Harv. L. Rev. 201, 219 (1917) ("[T]he whole course of English and American law today is belying it unless, indeed, we are progressing backward.").

5. Comparing the degrees of freedom in medieval tenancies and the oaths of loyalty once binding tenants to lords to the freedom allowed today (as evidenced by the sheer length of leases), today's law would appear to be less a matter of status. The fact of tenancy told more about the relationship of yore. But see Carl E. Schneider & Margaret F. Brinig, An Invitation to Family Law 348-49 (1995) (arguing current tenancies retain many elements of status).

6. Numerous authors have, however, advocated marital contracts. See, e.g., Richard W. Bartke, Marital Sharing--Why Not Do It by Contract?, 67 Geo. L.J. 1131, 1134 (1979) (proposing model statutory framework that would permit couples to elect a community-property arrangement for their marriage); Theodore F. Haas, The Rationality and Enforceability of Contractual Restrictions on Divorce, 66 N.C. L. Rev. 879, 894 (1988) (proposing model antenuptial agreements that impose financial burdens for initiating divorce); Joan M. Krauskopf & Rhonda C. Thomas, Partnership Marriage: The Solution to an Ineffective and Inequitable Law of Support, 35 Ohio St. L.J. 558 (1974) (discussing partnership agreements between husbands and wives); Marjorie Maguire Schultz, Contractual Ordering of Marriage: A New Model for State Policy, 70 Cal. L. Rev. 207 (1982) (arguing the utility of marriage contracts to govern human interactions); Elizabeth S. Scott, Rational Decisionmaking About Marriage and Divorce, 76 Va. L. Rev. 9, 42-44, 79-81 (1990) (discussing the use of precommitment strategies, including antenuptial contracts, to lower divorce rates); Amy L. Wax, Bargaining in the Shadow of the Market: Is There a Future for Egalitarian Marriage?, 84 Va. L. Rev. (forthcoming May 1998) (Aug. 1997 manuscript at 91-92, on file with authors) ("the ability to negotiate a binding antenuptial agreement would still have salutary effects, because it would arrest the bargaining squeeze and eliminate the potential for opportunism that it presents").

7. Maynard v. Hill, 125 U.S. 190, 210-11 (1888). It is rare for the Court to engage in sarcasm, but the facts of this case make one wonder. The holding in this case appears to be that the law of marriage is up to the legislature, not the parties to the marriage. But the facts are that a man went to Oregon, broke his promise to his Ohio wife to return, and successfully lobbied the legislature for a customized, unilateral, no-fault divorce without notifying his wife.

8. But cf. Jana B. Singer, The Privatization of Family Law, 1992 Wis. L. Rev. 1443 (describing insightfully the trend toward privatization).

9. Domestic-relations cases made up a third of civil filings in state courts of general jurisdiction in 1992. See Brian J. Ostrum et al., State Court Caseload Statistics Annual Report 1992, at 23 (1994).

10. See Michael J. Trebilcock & Rosemin Keshvani, The Role of Private Ordering in Family Law: A Law and Economics Perspective, 41 U. Toronto L.J. 533 (1991).

11. See Lacks v. Lacks, 189 N.E.2d 487, 488 (N.Y. 1963) (holding that a contract to pay a spouse during marriage is void); see also McGuire v. McGuire, 59 N.W.2d 336, 342 (Neb. 1953) (requiring a separation before a party can bring an action for maintenance). The requirement of minimal support is discussed infra text accompanying notes 136-38. See also notes 153, 193.

12. Even where the well-being of children is involved, courts resist intervening in disputes during marriage.

It would be anomalous to hold that a court of equity may sit in constant supervision over a household and see that either parent's will and determination in the upbringing of a child is obeyed, even though the parents' dispute might involve what is best for the child. Every difference of opinion between parents concerning their child's upbringing necessarily involves the question of the child's best interest.

Kilgrow v. Kilgrow, 107 So. 2d 885, 889 (Ala. 1958) (denying injunction restraining wife from taking child to certain school in violation of premarital agreement).

13. See Saul Levmore, Love It or Leave It: Property Rules, Liability Rules, and Exclusivity of Remedies in Partnership and Marriage, Law & Contemp. Probs., Spring 1995, at 221, 225-26 (comparing the lack of remedies in partnership law to the lack of remedies in domestic-relations law); Katherine Silbaugh, Turning Labor into Love: Housework and the Law, 91 Nw. U. L. Rev. 1, 29 (1996) (discussing "the traditional argument that marital obligations cannot be consideration for a contract").

14. "Grounds for dissolution" and "grounds for divorce" are used in this Article interchangeably.

15. Before 1967, several states did allow divorce without a showing of fault if the couple had been living "separate and apart" for a period of time. Divorce on this ground without mutual consent was risky because separation without agreement could have been found to be desertion. No state allowed one spouse to divorce the other against his or her wishes without proving fault on the part of the spouse wishing to continue the marriage.

Separation for cause, without divorce, has been available for a long time. According to Blackstone:

Divorce a mensa et thoro is when the marriage is just and lawful ab initio, and therefore the law is tender of dissolving it; but, for some supervenient cause, it becomes improper or impossible for the parties to live together: as in the case of intolerable ill temper, or adultery, in either of the parties. For the canon law, which the common law follows in this case, deems so highly and with such mysterious reverence of the nuptial tie, that it will not allow it to be unloosed for any cause whatsoever, that arises after the union is made.

1 William Blackstone, Commentaries on the Law of England 428 (photo. reprint 1979) (1765).

16. According to Elizabeth Schoenfeld, "[o]n September 5, 1969, with a stroke of his pen, California governor Ronald Reagan wiped out the moral basis for marriage in America." Elizabeth Schoenfeld, Drumbeats for Divorce Reform, Pol'y Rev., May-June 1996, at 8, 8. Within five years, 44 other states had followed with laws allowing courts to grant divorces sought unilaterally on the ground of "irreconcilable differences" or "irretrievable breakdown." See id. For a brief history of the development of no-fault, see Schneider & Brinig, supra note 2, at 71-94.

17. "Unilateral" divorce includes four kinds of situations: (1) divorce against the wishes of one spouse based on the fault of that spouse, (2) divorce against the wishes of one spouse not based on any fault by that spouse, (3) (rarely) divorce requested by one spouse where the other spouse is unavailable and his or her preference is unknown, and (4) divorce sought by one party where the other party does not contest the divorce. Cases in the last category are similar to bilateral (or mutual) divorces, in that they are not contested. A huge number of cases are uncontested.

18. "No-fault" divorce refers in this Article to divorce that parties may obtain without a showing of fault. The reader should keep in mind, however, that fault still plays a role in the consequences of divorce in some states. A pure "no-fault divorce" state would neither require nor allow the parties to show fault in any part of a divorce proceeding.

19. See Mary Ann Glendon, Abortion and Divorce in Western Law 108 (1987); Harry D. Krause, Family Law in a Nutshell 339 (2d ed. 1986); Leora Friedberg, Did Unilateral Divorce Raise Divorce Rates? Evidence from Panel Data, 88 Am. Econ. Rev. (forthcoming 1998) (Jan. 1997 manuscript at tbl.1, on file with authors) (listing five states which do not allow unilateral, no-fault divorce: Arkansas, Delaware, Mississippi, New York, and Tennessee); Marsha Garrison, How Do Judges Decide Divorce Cases? An Empirical Analysis of Discretionary Decisionmaking, 74 N.C. L. Rev. 401, 420 n.64 (1996) ("Only three states . . . restrict no-fault divorce to cases involving spousal agreement.") (citing Miss. Code Ann. § 93-5-2(1) (1994); N.Y. Dom. Rel. Law § 170(6) (McKinney 1988); Tenn. Code Ann. § 36-4-101(12) (1991) (requiring separation agreement for couples with minor children)); Scott, supra note 3, at 17 n.23 (stating that only two states require consent of both spouses to a no-fault divorce). Unfortunately, tables categorizing divorce grounds sometimes fail to identify states such as Virginia that do not allow unilateral, no-fault divorce. See Linda D. Elrod & Robert G. Spector, A Review of the Year in Family Law: Of Welfare Reform, Child Support, and Relocation, 30 Fam. L.Q. 765, 807 tbl.4 (1997). It is stunning that the Family Law Quarterly summary table ignores the important difference between unilateral and mutual no-fault divorce.

20. In writing Mandatory Planning for Divorce, Stake presumed that the reform allowing unilateral divorce would not soon be reversed. See Jeffrey Evans Stake, Mandatory Planning for Divorce, 45 Vand. L. Rev. 397, 409-10 (1992). In writing their casebook, Schneider and Brinig said "Is no-fault divorce desirable? No one advocates returning to the old regime . . . ." Schneider & Brinig, supra note 2, at 94.

21. Some of the recent reforms in Britain have followed this pattern. In 1996, the first major change in marriage law since 1969 in England and Wales required mandatory "cooling off" periods, delaying most divorces twelve to eighteen months instead of the earlier seven-month average. But in other ways, Britain is out of phase. As in earlier American reforms, the role of fault has been sharply diminished. See Robin Knight, In Great Britain, an Easier Out, U.S. News & World Rep., Sept. 30, 1996, at 60.

22. See Hanna Rosin, Separation Anxiety: The Movement to Save Marriage, New Republic, May 6, 1996, at 14, 14. In one state, Louisiana, the bill has been signed into law. See infra note 169.

23. "Terms of dissolution" and "terms of divorce" are used in this Article interchangeably.

24. See Smith v. Smith, 547 N.E.2d 297, 300 (Ind. Ct. App. 1989) (stating in dictum that the "trial court may order spousal maintenance only after a showing of incapacitation"). However, divorcing spouses "crafting their own agreements may provide for maintenance without such a showing." Id.

"[W]hile a divorce court is prohibited from fashioning an award of spousal maintenance containing a provision that the award is not subject to modification, divorcing couples are perfectly free to craft their own agreements--as did the parties in the present case--for an award of maintenance that is not subject to modification."

Roberts v. Roberts, 644 N.E.2d 173, 175 (Ind. Ct. App. 1994) (alteration added) (quoting Bowman v. Bowman, 567 N.E.2d 828, 830 (Ind. Ct. App. 1991)). Thus, divorcing spouses have more flexibility in crafting their property settlement by mutual agreement than do divorce courts by commands.

25. This assumption is suspect. If men's filings for divorce increased more than women's, that would suggest that more men desired to leave a non-faulty wife than vice versa, and that the new law benefited more men than women.

26. This assumption is also suspect in that courts (dividing assets "equitably") often look primarily to the financial contributions of each spouse, undervaluing nonmonetary contributions by women.

27. See Paula England & George Farkas, Household Employment and Gender 163-64 (1986) (describing the sex gap in wages); Victor R. Fuchs, Women's Quest for Economic Equality 44-45 (1988) (stating that women work fewer paid hours than do men); Milton Regan, Family Law and the Pursuit of Intimacy 155 (1993) (stating that "women are more likely than men to make career sacrifices in order to meet family responsibilities, and therefore often are economically dependent on men") (citing Fuchs, supra, at 140).

28. This proposition might be tested empirically if it could be determined whether men became more willing to marry after divorce reform. If they did become more willing to marry, that would support the claim that reform reduced the ex ante costs of marriage for them. Overall marriage rates have diminished, however. See Margaret F. Brinig & Steven M. Crafton, Marriage and Opportunism, 23 J. Legal Stud. 869, 884-85 (1994) (discussing the evidence of fewer women getting married--91 per 1000 in 1988 compared with 147.2 per 1000 in 1967--and the woman's lack of bargaining power in marriage).

29. This legislative wealth transfer was challenged in litigation, without success. See In re Marriage of Franks, 542 P.2d 845, 850 (Colo. 1975) (en banc) (rejecting argument that the no-fault divorce grounds in the Uniform Dissolution of Marriage Act violated the contract clause of the state constitution); see also In re Marriage of Walton, 104 Cal. Rptr. 472, 475-76 (Ct. App. 1972) (holding that a no-fault divorce was not an impairment of wife's contract). Given the current scope of the Takings Clause, perhaps that should have instead been the basis of the challenge. See Dolan v. Tigard, 512 U.S. 374, 391 (1994) (stating that the burden is on the government to make findings of fact showing rough proportionality in exaction cases); First English Evangelical Lutheran Church v. County of Los Angeles, 482 U.S. 304, 318 (1987) (holding that a deprivation of land-use rights is no less a taking simply because it is temporary); Hodel v. Irving, 481 U.S. 704, 717-18 (1987) (determining that it is a taking to stop a person from passing assets by will and by intestate succession even if person still has a right to transfer assets at death by settling a trust); Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419, 436-37 (1982) (stating that no physical invasion is too small to be a taking).

30. Loving v. Virginia, 388 U.S. 1, 17 (1967); see also Zablocki v. Redhail, 434 U.S. 374 (1978) (holding that the state cannot prevent marriage by someone unable to meet his obligation to support his existing children).

31. See Boddie v. Connecticut, 401 U.S. 371 (1971).

32. The new legal regime also created an incentive