One of the hidden costs of welfare reform, both at the federal and state levels, is addressed by a new study at Cornell University. It finds that the marginally employed – people whose low pay makes them eligible for welfare payments – are almost four times more likely to be violent with their families than workers not on welfare or even than other unemployed workers.

The study also found that unemployed individuals who use alcohol to excess are at greater risk of committing family violence than the employed who abuse alcohol or nondrinkers, with the greatest risk among employed welfare recipients.

Race and ethnicity were not found to be significant in predicting family violence, when other factors were controlled in the analysis. Partnership stability, having a physical or mental limitation which could restrict the ability to hold a job or a previous history of alcohol abuse were also not significant.

In a paper published in the March issue of the Journal of Epidemiology and Community Health (Vol. 55, No.3), Eunice Rodriguez, an assistant professor in Cornell's Department of Policy Analysis and Management, says her analysis indicates that marginal employment coupled with welfare are a very significant factor in family violence. However, she says, unemployment on its own, appears to have no effect on the likelihood of family violence, when alcohol abuse, income, education, age and other factors are taken into account.

The research article was selected as one of the best papers from the BMJ Publishing Group journals and will be reprinted in the Western Journal of Medicine in May. Rodriguez, an epidemiologist who studies the role of social supports on unemployment and marginal employment, previously presented her latest research at the International Conference on Research for Social Work Practice in Miami in 1998.

Welfare reform, passed by Congress in 1996, instituted policies that emphasize job placement and requirements that welfare recipients work in many cases. Rodriguez says that evidence indicates that welfare reform accounts for 44 percent of the employment rate gain from 1992 to 1996.Rodriguez collaborated with Kathryn Lash and Jennifer Lee of the New England Medical Center and Pinky Chandra, now a consultant with the Cornell Institute for Social and Economic Research. They analyzed data on 4,780 married or cohabiting individuals who were in both the 1987 and the 1992 National Survey of Families and Households. In 1987 the subjects were older than 16, and still working. By 1992 they were still in a relationship.

Rodriguez says the study sought to disentangle the effects of employment, unemployment, partnership instability and alcohol use/abuse in the risk of domestic violence. Previous research has been inconsistent in finding a relationship between unemployment and alcohol use and between unemployment and violence.

Family violence is a public health problem that takes a $857 million toll annually on the healthcare system. Experts estimate that between 2 million and 4 million women annually are physically battered by their partners. The researchers defined domestic violence as physical violence by both partners during an argument. These incidents represent 60 percent of family violence cases in the sample. The researchers did not study family violence in which only one partner was the victim.

One explanation for the increased risk of family violence among working welfare recipients is the additional stress of working in low-skill jobs while coping with poverty and childcare, Rodriguez speculates. Job conditions, low job security and the stigma of receiving welfare could also contribute to this relationship, she says. The number of children in a family also increases the likelihood of violent family arguments, the study found.

She notes that an increase in female employment tends to foster different forms of relationships for women, which may generate tensions that could increase the likelihood of marital violence. "This is particularly relevant given our fast-changing economy and increasing employment demands on young parents, including those receiving welfare benefits," she says.

"In any event, we have identified a group that is particularly at risk for family violence. As a result, we think it is critical to monitor the impact of welfare reform on family violence," Rodriguez says.

The study was supported, in part, by the Cornell Agricultural Experiment Station via a Hatch Grant from the U.S. Department of Agriculture.