I n his 1985 book Family and Nation, the late Senator Daniel Patrick Moynihan wrote, "No government, however firm might be its wish, can avoid having policies that profoundly influence family relationships. This is not to be avoided. The only option is whether these will be purposeful, intended polices or whether they will be residual, derivative, in a sense, concealed." As Moynihan knew, government policies have real effects on the lives of families, producing often unintended consequences. Sometimes this is the result, as Moynihan implied, of policymakers not fully understanding the scope of their actions. But, just as often, it can be the result of changing patterns within the American family itself. Indeed, over the past generation the American family has changed dramatically, but the policies designed to mitigate the risks it faces have remained frozen in time, many of them operating on rules developed in the midst of the Great Depression. As a result, the most vulnerable families in the new economy all too often wind up with limited protection in times of need.

Recent efforts to fashion policies to respond to the strains on modern families have focused, for the most part, on providing job-protected leave to help balance work and family responsibilities. After years of partisan wrangling, President Bill Clinton signed the Family and Medical Leave Act (FMLA) in 1993; since then, progressives have focused on expanding the FMLA to more workers, allowing workers to take time off to attend school activities and doctors’ appointments, and providing workers with sick leave for their own or a relative’s illness.

The FMLA, though overlooked by many opinion-makers, was a breakthrough in updating the American social contract for today’s families. It amended the Fair Labor Standards Act of 1938 to allow workers to take time off to care for a new child or a sick relative without losing their jobs. Clinton realized its impact; it was the first law he signed, and he mentioned it in every State of the Union. But the ability to take time away from one’s job, while a critical component of modern family policy, is not comprehensive. To give families the security they need to raise healthy, productive members of society, we need also to address the financial risks parents incur just for being good parents–when they take time out of the workplace, require a flexible schedule to raise children, or get hit with high health care or child care expenses.

For this, progressives should turn to one of the most important innovations of the last century: social insurance. In the 1930s, progressives established a suite of social insurance schemes to help families share the risks of the industrial economy: the risks associated with the inability of the breadwinner to earn the family income because of old age, death, a temporary layoff, or disability. These social insurance programs continue to provide families essential support. But today we need to create new elements in the social insurance system–as well as reform the protections now in place–to confront the new risks families face. Our current social insurance system–a patchwork of programs put in place over the course of decades–was designed to help nuclear families in which a breadwinner worked one job his entire career while the homemaker cared for the children and any ill relatives. Today’s American family and today’s workforce are markedly different. Both two-earner and single-parent families operate in a volatile, winner-takes-all economy; families often are expected to raise a younger generation and care for an older one, while saving to prepare for the current one’s future; and workers at all skill levels face a career of increased mobility and volatility.

Contrary to what conservatives have argued, Americans do not need to replace social insurance with some version of the private accounts of President George W. Bush’s Social Security plan. Rather, they need a way to continue sharing the all-too-real risks families bear as parents across the country attempt to raise kids in a new economy that provides them no margin for error. This should be the central goal of our social insurance system in the twenty-first century.

Families in the early twentieth century lived in a transformative period and confronted new challenges analogous to those we face today. In a quickly urbanizing and industrializing society, old patterns of life were breaking down. Many families had left agricultural lifestyles for wage labor in cities, making them dependent on a single breadwinner’s earnings. Extended family networks, capable of providing additional economic support for individuals in times of need, were replaced by the rise of the nuclear family. Better public health and health care had increased the lifespan of Americans by ten years from 1900 to 1930.

President Franklin D. Roosevelt responded to these changes, as well as the exigencies of the Great Depression, by creating a series of policies to soften the rough edges of the industrial economy for workers and their families. In addition to employment innovations like the creation of the minimum wage, the 40-hour workweek, and the ban on child labor, he also created a system of social insurance to guard against "the hazards and vicissitudes" of life in the new economy.

Social insurance programs had precedents both at home, where there had been a national system of Civil War pensions, and in Europe, dating back to Bismarck’s Germany. While governor of New York, FDR had implemented the first comprehensive system of unemployment relief and an extensive program for industrial welfare. Roosevelt sought a similar program for the entire nation. The result was the Social Security Act of 1935, centered around old age and unemployment insurance (UI) programs to prevent catastrophic drops in families’ income resulting from old age, widowhood, or cyclical downturns (Disability insurance was added in 1956).

From its beginning, the system combined two competing strands: earnings-based benefits and benefits designed purely to keep families afloat. Social Security’s old-age program keyed individual benefits to lifetime earnings, but the benefit structure was progressive: Lower-income workers received relatively more per dollar earned, while higher-income workers received relatively less. In 1939, before it began paying out benefits, Social Security was changed to provide a spousal benefit–married couples would receive 150 percent of the benefits of a single worker with the same earnings–and a survivor’s benefit for dependent children and spouses. In addition, the Social Security Act contained Aid to Dependent Children, designed to provide aid to needy single-parent households. The family benefits became a central feature of the Social Security program. Even today, the National Women’s Law Center finds 50 percent of people receiving Social Security benefits actually collect these benefits as either a widow or widower, spouse or child of a worker, or a disabled worker–rather than as an individual receiving Social Security benefits based on his or her employment history.

The system was, of course, built around the American family and social assumptions of that particular time. The Social Security family benefits went to spouses, whether they had children or not, because in the 1930s most mothers were in fact wives and only 15 percent of married women were in the paid workforce (and even then they were under pressure to leave to make way for male breadwinners). Wives were ineligible for the benefit unless they remained married for a fixed amount of time, in an attempt to prevent women from marrying just for benefits. Even if a former spouse was entitled to a benefit after divorce, it was less than what the worker received, because it was thought that a woman living alone could survive on less than a man (one participant in the debates at that time argued that a woman could do her own housekeeping, while a man would have to eat in restaurants). Single women were assumed to be widows and expected not to work; Aid to Dependent Children provided help to these women as long as they were not married and did not work. To ensure that funds benefited only those with a strong labor-force attachment (not women working for "pin money"), Social Security benefits were keyed to lifetime earnings, and UI eligibility was tied to past work history and whether the worker was seeking full-time employment. Moreover, Social Security’s old-age program, Unemployment Insurance, and later Social Security Disability Insurance, were designed for manufacturing workers who would spend 40 hours a week for 40 years working for a single employer; as a result, the programs covered only income interruptions associated with retirement, temporary layoffs, and, later, disability.

The system worked well for the families for whom it was originally designed, keeping the wolf from the door even when an assembly line closed, an injury occurred, or the breadwinner grew too old to work in the mill. As a result of the old–age insurance program, the official poverty rate for people age 65 and over dropped to 10 percent by 2003, from 35 percent in 1959 (the first year the federal government kept records using a standardized measure of "poverty"). And Social Security is still enormously important to America’s elderly; it accounts for about 90 percent of total income for both men and women ages 65 and older. It allowed people to start new companies or switch jobs knowing that if the business failed, their families would not be destitute. The system of social insurance has, in short, been one of the most successful government initiatives ever undertaken, demonstrating how Americans can act together to improve the lives of all citizens.