The british government cannot make up its mind about marriage. A white paper on the family was due to be published early this year. But thanks to the divisions it has exposed within the government's ranks it has been shelved until after the election. Ministers such as David Blunkett and Paul Boateng have argued that the government should actively support the conventional form of marriage. But they have clashed with another faction, led by Tessa Jowell, which argues that marriage is only a "lifestyle choice"-no better or worse than non-married cohabitation and other kinds of family structure.
The "lifestyle choice" faction has not won every battle in recent times-witness the government's decision to drop the "no-fault" divorce reform which would have made divorce even easier. And the weight of evidence is beginning to change some minds. When we first wrote on this subject a few years ago, our articles attracted a torrent of abuse; it was considered a betrayal of the progressive cause to support marriage. Now, some of our critics are reconsidering.
But the overwhelming momentum of policy since the 1960s has been-and continues to be-in the direction of removing any privileges attached to conventional marriage. Despite Tony Blair's increasingly warm words about marriage, the actions of his government have continued this momentum. It has, for example, abolished the married couples' tax allowance. The value of this allowance was no longer large, but its abolition was a significant symbolic step. Even before its abolition the tax and benefit system as a whole was biased against marriage. The same allowance was available under another name to unmarried parents, whilst various welfare benefits continue to reward lone parenthood. (Since many lone parents have a partner, the extra benefits they receive is often a form of subsidy to cohabitation.)
The bias against marriage is also evident in Gordon Brown's recent reforms to the welfare system. The new Working Families Tax Credit (WFTC) was inspired by the American Earned Income Tax Credit (EITC). Research by Nada Eissa and Hilary Hoynes at the University of California has shown that the EITC encourages divorce and extra-marital childbearing among the poor in the US. This is because lone parents get a larger tax credit than would be the case if they were living with an employed spouse. The WFTC will have a similar effect in Britain.
Marriage remains a valuable institution, for the individuals concerned, for their children and for society. The evidence is overwhelmingly in support of all three of these propositions. Indeed, there are few hypotheses in the social sciences which receive such confirmation from serious research.
Yet it seems to be only parts of the right-wing press which consistently support these findings. Most of the liberal establishment, which usually takes social science findings seriously, simply ignores them. Instead, liberals proselytise the idea that all family structures are equally valid, to the harm of the poor in particular. Sometimes the argument is pushed even further and marriage is seen as actively disadvantageous to women and in some way incompatible with the modern woman's role in the job market. The increased power and status of women over the past 50 years has certainly affected the respective roles of men and women within marriage, but it has done nothing to undermine the case for the institution.
There are two further points which we must make clear. First, it is not our intention to moralise. What goes on behind the net curtains to preserve marriage is not our concern. Our support for the institution is based on its benefits to both individuals and society. Second, most children, whatever their family background, grow up as well-adjusted members of society. Everyone knows lone parents or cohabiting couples with charming and successful children. But on average the children of married parents do better.
Consider some analogies. Almost all rail journeys end safely, but that does not mean that a rail system in which investment has been neglected for decades is just as safe as one where it has been maintained. Driving with a safety belt does not guarantee protection in the event of an accident and in some cases may even be a disadvantage because it may trap the driver inside the car. But, on average, it is safer to drive with a safety belt than without. The list of examples is endless. The common theme is that most of the time, the outcomes are happy ones. But the probability of an unhappy one rises in certain circumstances. This is exactly the case with family structure.
Despite the rise of cohabitation, the institution of marriage is still popular in Britain. The demographers John Ermisch and Marco Francesconi suggest that almost 90 per cent of all young women in Britain today will ultimately get married. Of those who choose to cohabit, 60 per cent will soon marry their first partner, and a further 25 per cent will eventually marry someone else. The rise of cohabitation does not mean the disappearance of marriage, and it is often used as a "trial run" for marriage. But the spread of cohabitation has led to a growth in family disruption, lone parenthood and the number of step-families. These trends are most common amongst the poor, who most need a stable family.
In liberal circles it has become unfashionable to use the language of marriage when discussing relationships. The ambiguous term "partner" is preferred to the legal terms "spouse," "husband" or "wife." This is partly the influence of radical feminists who believe that marriage is inherently patriarchal. More commonly, using the language of marriage is thought to stigmatise cohabiters or single parents who are often doing a good job in difficult circumstances. Whatever the motivation, this represents the de facto abolition of marriage as an institution.
How has this situation arisen? How can this indifference to marriage have captured large swathes of the government and the media? The current climate of opinion is partly the result of a cohort effect. The ideas of the 1960s have been carried forward by those who came of age at that time and are now reaching the peak of their influence in politics and society. There are signs of a reaction against their liberalism in younger age groups. Official statistics show that the divorce rate has been falling in recent years. After reaching a peak in 1993 of 14.2, the annual rate of divorce has now fallen to 12.9 per 1,000 marriages. However this downward trend is not yet firmly established and the marriage sceptics still hold much of the intellectual and political high ground.
The secular law in Britain defines marriage as "the union of one man with one woman, voluntarily entered into for life, to the exclusion of all others." At a minimal level, marriage is merely a set of public vows that a couple makes to each other. However, even in the absence of external enforcement, such vows may be taken very seriously, and conscience may cause husband and wife to honour their vows through the bad times-especially in a culture where importance is attached to keeping promises.
But our age has been infected by a philosophy of self-gratification, which erodes the influence of guilt, duty and conscience in interpersonal relationships. This is exemplified in a recent public pronouncement by a well-known writer who had just left his wife and children. Divorce, he declared, is "morally legitimate" because it "accords respect to an individual's needs against the devouring claims of family life." Within such a moral framework there is no such thing as a binding promise.
Trends in family life are not merely personal indulgences. Private decisions have public consequences. And such decisions have led to the number of divorces more than doubling since 1970-almost one in four children born in 1979 is estimated (by Social Trends 1999) to have been affected by divorce.
These decisions impose costs on the rest of us. Family break-up often damages children, even when it makes one or both parents happier. In addition to the direct harm to children, the subsequent deviant behaviour of some of the victims creates social costs. The British Youth Lifestyle Survey shows the percentage of boys living with both natural parents who never offend is 58 per cent. This falls to 51 per cent for boys living with a lone parent and to 43 per cent for boys living with one natural and one step-parent.
In most countries, marriage vows are reinforced by external factors. These may include social ostracism of those who break their vows, financial costs for those who separate-for example, legal penalties for a spouse who is judged to be mainly responsible for the breakdown of the marriage. There may also be positive support for marriage in the form of social and economic privileges.
In recent years, as we have seen, many of the external factors which reinforce marriage in Britain have been weakened or eliminated altogether. For example, the divorce rate rose sharply in the early 1970s as a result of cultural and legal factors including the 1969 Divorce Reform Act which introduced a new ground for divorce, "irretrievable breakdown." It also rose sharply after 1984 when a new law reduced the minimum period before divorce was possible from three years to one. These changes are often seen as an advance in freedom since they allow the individual to escape from an unhappy relationship with minimum cost or delay. However, the fact that individuals can now easily and unilaterally abandon a marriage makes couples more wary of commitment.
Despite the dilution of the legal marriage contract, it is still the case that formal marriage usually involves a higher degree of personal commitment, and provides more security, than is the case in cohabiting relationships. This commitment can be of great benefit to the couple involved, to their children and to society. Over the past decade there has been a flood of academic work on the benefits of marriage, disproving many common beliefs. Much of this material originates in the US, where large, well-funded research projects have investigated family life in great depth. Research in other western countries confirms most of the US findings. The main conclusions of this research can be summarised as follows.
Mental and physical health. On average, married people are healthier and have lower mortality rates than single, cohabiting, divorced or separated people. They live more regular and secure lives, and engage in less substance abuse and other harmful activities. They suffer from less anxiety, depression and other mental ailments. These findings apply to both sexes. In 1972, the American feminist Jesse Barnard claimed that "marriage may be hazardous to women's health." This claim is still widely repeated. For example, in the Observer magazine recently the psychologist Dorothy Rowe claimed that the "one group of women prone to depression is married women; the one group of men not prone to depression is married men." Whilst it is true that on average men gain most from marriage, a survey of the evidence by the American sociologist Linda Waite and co-author Maggie Gallagher shows that women also benefit. Married women engage in fewer risky forms of behaviour than single or divorced women, and they receive more financial and emotional support than other women. They are less likely to become disabled in old age or enter a nursing home, they live longer and are less likely to commit suicide. They also report higher levels of emotional satisfaction with their partner than cohabiting or other women. A major study by the US National Institute for Mental Health found that the annual incidence of depression was 1.5 per 100 for married women. For other types of women the figures were as follows: single 2.4, divorced once 4.1, divorced twice 5.8, and cohabiting 5.1.
Spouse abuse. Marriage is sometimes described as a hitting licence which allows men to batter women. In fact, serious violence amongst married couples is uncommon, and violence of all kinds is much less frequent than amongst cohabiting couples. American researchers, Jan Stets and Murray Straus, estimate that severe violence is three and a half times more common amongst cohabiting couples than married ones. They say that one explanation for the difference may be that some couples "enter cohabitation rather than marriage to keep more of their own independence, only to find that there are frequent arguments over rights, duties and obligations, which may lead to violence." These researchers also argue that "married couples have a greater material, social and psychological investment in the relationship... and may give in to their partners' wishes, believing that they need to make sacrifices to keep the relationship intact."
Equally striking is the evidence on sexual coercion. In their massive study of sexual practices in the US, Edward Laumann and his colleagues at the University of Chicago find that 0.6 per cent of women report having been forced to perform some sexual act by a spouse. Five times this number report sexual coercion by someone that they were in love with but who was not their spouse.
Child abuse. Both physical and sexual abuse are less frequent for children who live with their married, biological parents than in other kinds of family set-up. The riskiest situation for a child is to live in a step-family. Evidence collated by the sociobiologists Martin Daly and Margot Wilson indicates that children living with a step-parent are between 15 and 150 times as likely to be neglected, sexually abused, or murdered as children living with both their biological parents. In Finland, a 1996 study of 9,000 15-year-olds indicated that 3.7 per cent of girls living with a stepfather claimed that he had sexually abused them compared to 0.2 per cent living with their biological father. There is also evidence that abuse is less frequent if the step-parent and biological parent are married. This is presumably because marriage involves a stronger commitment to the new family than does cohabitation.
Stability. Marriages are on average more stable than cohabiting unions are. This applies whether or not children are present. Kathleen Kiernan estimates that 8 per cent of British couples who get married before their first child is born split up within five years of its birth. The figure is 25 per cent for cohabiting couples who marry after their baby is born, and 52 per cent for cohabiting couples who never marry. Using a different data source, John Ermisch estimates that for those cohabiting couples who never marry, around five out of six unions will break up within ten years of the birth of their first child.
A similar pattern can be observed elsewhere in western Europe, although the contrasts are not as extreme as in this country. The most interesting case is that of Sweden. Those who believe that marriage is an outdated institution frequently claim that long-term cohabitation has replaced marriage in Sweden and that Swedish divorce rates are very high among the few people who do marry. These claims are misleading. It is true that cohabitation is widespread in Sweden and is legally recognised by the state. More than half of all mothers have their first child in a cohabiting union. However, a majority of those mothers eventually marry the father and they generally stay married for a long time. Many of those who remain cohabiting break up within a few years. Kiernan estimates that the risk of dissolution for cohabiting parents in Sweden is four times as high as for married parents. Her figures also show that when they involve children, Swedish marriages are the most durable in Europe.
The instability of cohabiting unions is to be expected since many of them involve no lifelong commitment and the option of breaking up is deliberately preserved. People who view themselves as two separate individuals are more likely to leave a relationship than those who view themselves as two halves of a permanent couple. In this country at least, cohabiting couples who stay together usually end up getting married. Thus, even amongst couples who initially cohabit, marriage is still widely seen as the final step in binding the relationship. The role of marriage as a "cement" is one of the reasons put forward by campaigners for the legal recognition of same-sex marriage. Why should gays and lesbians be excluded from an institution which encourages the commitment and stability which many of them desire?
Civilising influence. The decline of marriage is creating an army of males upon whom women cannot, and do not expect, to rely. Numerous studies have found that married men work harder, earn more and are more likely to have a job than other men. The causation is two-way, since having a steady job also increases the likelihood that a man will get married and stay married. Thus, any programme for encouraging marriage must also address the issue of creating jobs for men in depressed areas.
Care. Stable families create a network of reciprocal obligations between generations, siblings and spouses. Despite the spread of nursing homes and social services, care for old or sick adults in our society is mainly the responsibility of close relatives. According to the General Household Survey, 93 per cent of informal care for the old is provided by family members. Friends and neighbours hardly figure. Modern family trends mean that millions of extra men and women in the future will have no close relative prepared to care for them in sickness or old age.
Child development. On every measure of achievement and emotional condition, children living with their married parents do better on average than other children. This applies to both adopted and biological children. The children of lone parents on average perform less well on all measures than those with married parents. Part of this gap is explained by the fact that lone parents as a group are relatively poor, but much of it is due to other disadvantages. One US study, by Sara McLanahan and Gary Sandefur, found that children who have lost a parent through death perform quite well according to most measures of wellbeing. (This may be because the missing parent is still in a sense present in the family.)
Divorce and children. Although divorce can benefit children in the case of severe parental conflict, it frequently damages them emotionally and harms their life chances. The conventional view at one time was that parents should stick together for the sake of the children even if one or both of them is unhappy. For decades this view was criticised by therapeutic professionals, who claimed that children are better off if unhappy parents separate. But the latest research suggests that the traditional view was often right. In a study analysing the life history of 471 families over a 12-year period, two of America's leading experts, Paul Amato and Alan Booth, find that children in "low-conflict" families are usually better off if their parents stay together. Such families account for 70 per cent of all divorces in their sample. Only in high-conflict families do they find that divorce is on average better for the children. They conclude: "Our study suggests that the worst situation for children to be in is either a high-conflict marriage that does not end in divorce or a low-conflict marriage that does end in divorce. Consequently for the children's sake, some marriages should not be salvaged. But in marriages without severe conflict and abuse, future generations would be well served if parents remained together until the children are grown."
It must be repeated that all these findings refer to averages and may not apply in individual cases. Many cohabiting unions are very successful, some divorces are good for the children, and many lone parents and stepfamilies bring up their children well. However, despite these caveats, the evidence in favour of marriage as an institution is now overwhelming. There is thus a clear public interest in strengthening the institution of marriage. If the present decline in marriage could be reversed, not only would the quality of life of many adults and children be improved, but claims on the public purse would be significantly reduced. This is a powerful argument for using the tax and benefit system to support marriage. Like any insurance company, the state should offer better terms to citizens whose behaviour will reduce future claims on the exchequer. Subsidising marriage is, in principle, no different from offering reduced insurance premiums to non-smokers or to people who fit locks on their windows.
Some people would regard government support for marriage as an unwarranted intrusion into private life. Even if one accepts this view, it does not follow that the existing policy towards marriage is optimal. At the least, the government should eliminate the bias against marriage in the tax and benefit system. The government should also ensure that the public is aware of the benefits of marriage. Information about these benefits should be made a central part of sex education in schools, with a stress on not entering marriage and parenthood lightly. This in itself might begin to revive the fortunes of an institution whose decay has been a big factor behind so many personal and social problems.
In the face of all the evidence, the marriage sceptics fall back on two points. First, anyone who disagrees with their view of marriage is said to be hostile to women's empowerment or blind to the effect this has had on modern life. It is certainly the case that the financial independence of women has made divorce easier to contemplate than it was 50 years ago. But it does not follow that easy divorce is a good thing for women or anyone else-in fact, this more hostile context for marriage provides a good reason for bringing more countervailing pressure to bear in its support. As we have argued, women's greater independence and self-confidence has implications for relationships within marriage but should not affect the institution itself.
Moreover, women's participation in the job market has increased all over western Europe, yet divorce rates elsewhere are still lower than in Britain. And within living memory, in parts of Britain such as east Lancashire, traditionally high female labour force participation coincided with high rates of stable, conventional marriage. We are not proposing a return to patriarchy. Conventional marriage is compatible with a more equal relationship between men and women.
This brings us to the second point, the counsel of despair that things have gone so far that nothing can be done to change them. Simply accept what has happened and celebrate the diversity of family structures. This argument reveals a fundamental lack of understanding of the dynamics of social systems.
Policy changes need not be large themselves in order to have a big impact. We need a series of small measures-some of which have been mentioned above-to nudge the system of family structure back towards a "tipping point," where the popularity of marriage once again becomes self-reinforcing. Family structures are a complex system, arising from the actions of individuals whose behaviour is influenced directly by tradition, the tax and benefit system, and the behaviour of others. The more people who are married, the more likely it is that an individual will stay in or convert to the married state in any given period.
Such systems are hard to control with precision. But marriage is amenable to direction through the cumulative impact of many small changes. The Long March begins with the first small step.
