Cancer in the family: One spouse's diagnosis can lower household income

By matching tax returns with health data from Canada, co-researcher Vincent Pohl, assistant professor of economics at UGA, was able to determine that a spousal cancer diagnosis results in a decline in household income by 5 percent among men and 9 percent among women, on average.

"The average annual household income for the working-age couples we studied was about $100,000, so the loss of income per family is about $5,000 to $9,000, which is a pretty substantial decline," he said. "In a situation where one household member has a devastating diagnosis, it leads to the whole household suffering economically."

One reason for the steep decline is attributed to what's known as the caregiver effect -- when one family member reduces their own employment in order to support another.

"We thought that the household's lessened income could happen in one of two ways," Pohl said. "One is that the person who is diagnosed might not be able to work because they are getting treatment or they're too sick to work. The second is what happens to their spouse: Does the spouse work more to make up for the lost income or does the spouse also reduce his or her labor supply in order to take care of the spouse that is diagnosed with cancer? We find the latter, that spouses reduce their labor supply and therefore have lowered income levels, which leads to the household having lower income levels as well."

The research finds a spouse's cancer diagnosis leads to up to a 3.5 percent reduction in earnings for men and up to a 6 percent reduction in earnings for women. Because family caregiving isn't covered by insurance, this loss of income directly affects the household's bottom line, Pohl said.

"What we need to think about, in terms of policy implications, is how we can protect not just individuals from the consequences of getting sick, but their entire family. That's not really something that existing policies do," he said. "If you think about disability insurance, it's a function of an individual's inability to work. It doesn't take into account that family members might have to take care of an individual and therefore might also lose their job or reduce their working hours and thus their income."

Published in the Journal of Health Economics, the study also finds little difference in work reduction between men and women whose spouses are diagnosed with cancer -- something that stands in contrast with some earlier studies.

"We didn't find a big difference, although we do see slightly bigger effects in relative terms for wives than for husbands. The decline in employment was about 3 percentage points in both cases, but it's from a lower level among wives because women were less likely to work to begin with," he said. "We see that if women have a cancer diagnosis, there are a substantial number of husbands who quit their jobs to take care of their wives and vice-versa. This applies to all age groups."

While cancer comes in many forms, spousal behavior does not. The aggressiveness of the disease does not initially affect a spouse's decision to reduce their working hours, although it can change how soon a spouse returns to his or her previous employment level, Pohl said.

"If we look at prostate and breast cancer diagnoses, we see that the family income and labor supply pick up again after two or three years, which is typically the time it takes to treat the cancer," he said. "With more severe forms, like lung cancer, we see more persistent effects."

Pohl and co-researcher Sung-Hee Jeon of Statistics Canada were able to measure household income and cancer diagnoses by using administrative data from multiple sources, instead of relying on surveys as previous studies have done. By tracking working-age individuals for five years following a cancer diagnosis, the research shows how reliably household income is affected by health shocks.

"Our data allowed us to examine behavior on a level that's representative for the entire country of Canada. We observe annual earnings and income and potential cancer diagnoses for 1.4 percent of all households in Canada. That's something no one has been able to do," Pohl said.

Despite the robust data, the effects may not be transferable to the U.S., where health care is handled differently than in many developed nations.

"One reason why we don't see that the spouse works more, potentially, is that health insurance is not provided through jobs in Canada," he said. "In the United States, we could expect that if one spouse is diagnosed with a disease, the other spouse has to keep their job in order to keep health insurance for the family."