Gareth Olds grew up toeing the poverty line.
For years his parents struggled to support three children in Anchorage, Alaska, where food costs run high. His stepfather held down a steady but low-paying job as a dental assistant, and his mom performed a series of odd jobs. The kids pitched in. "Delivering phone books is the worst way to spend a Sunday when you're a kid," says Olds, now an assistant professor in the Entrepreneurship Management unit at Harvard Business School.
The family lived paycheck to paycheck. And yet, when Olds was a teenager, his parents garnered the nerve and bootstrapped the funds to launch a vocational training program for dental assistants. Revenue from the new business pushed them out of poverty and into the middle class. Olds, a budding economist, was intrigued.
“The rate of new business births rose by 13 percent among households that qualified for SCHIP. The survival rate of new businesses rose by 8 percent.”
"They didn't have any experience starting a business, and they couldn't get a loan, because they didn't have the credit," says Olds, who joined the HBS faculty in July. "I wanted to know what it was that allowed them to save startup funds—and what allowed them to take the risk of launching a company."
Curiosity eventually led to his doctoral dissertation: a series of studies showing a definitive link between the social safety net and entrepreneurship. In short, Olds finds that an increase in access to public welfare services leads to an increase in the formation of new companies. (The paper cites a related example: author J. K. Rowling, who wrote the first Harry Potter novel while receiving welfare benefits in the United Kingdom.)
"Social insurance programs can promote entrepreneurship by reducing the risks of business ownership and relaxing credit constraints," Olds writes in the 2014 paper "Entrepreneurship and Public Health Insurance."
Studying The Social Safety Net
Olds grew up on Medicaid, a Federal health insurance program for low-income individuals and families in the United States. At the same time, his family received food purchasing assistance from the Food Stamp program. He's pretty sure that these safety nets helped lead to the formation of his family's company. But while his personal experience motivated the idea for the study, he insists that it didn't shape the research.
"One family's story isn't an economics paper," he says. "People do get nervous if you have a personal story associated with your research. They worry you're bringing an emotional perspective to what should be a scientific endeavor. So a big part of my paper was convincing my advisers that what I was seeing was reproducible and scientifically valid. It also just happened to be personally valid for me."
“Having the net made these people more willing to walk on the high wire.”
Olds focused his initial study on the State Children's Health Insurance Program (SCHIP). Established in 1997, the program provides health insurance to uninsured children in moderate-income families.
To research the link between SCHIP and entrepreneurship, Olds studied 1992-2011 data from the United States Census Bureau's Current Population Survey and Survey of Income and Program Participation. He compared data for households that fell just above the SCHIP income eligibility threshold with those that fell just below it, before and after the program took effect. This mimicked the effect of experimental treatment and control groups, a common research technique in cases when an actual experiment isn't practical or ethical. (In this case households that qualified for SCHIP were the treatment group.)
The data showed that SCHIP had a significant positive effect on entrepreneurship. The program increased the self-employment rate by 23 percent among eligible households compared with non-eligible households. The rate of new business births rose by 13 percent among households that qualified for SCHIP. The survival rate of new businesses rose by 8 percent.
Olds took care to find out whether the businesses were serious, sustainable endeavors. "Economists tend to think in terms of employment growth potential," Olds says. "What I found was that the largest area of growth was newly incorporated firms. Eligible households were 31 percent more likely to have an incorporated business than ineligible ones, which is larger than the effect when you consider all firms. This means the distribution shifted: There are more firms overall because of the policy, but there are proportionally more incorporated firms. These were new ventures that people were serious enough about that they were willing to take the significant step of incorporation."
He also discovered that the share of total household income from self-employment versus outside wages increased 16 percent. That is, these new businesses were successful enough to contribute significantly to household income. "The newly self-employed are working longer hours per week, and more weeks per year, and are making more money from these firms," Olds says. "These aren't dogs that shouldn't have been firms in the first place. These are people's dreams that they've been held back from before."
Food Stamp Entrepreneurs
Similar results bore out in a study of the link between entrepreneurship and eligibility for the Supplemental Nutrition Assistance Program (SNAP), which Olds details in the 2014 paper "Food Stamp Entrepreneurs."
His research examined what happened after a significant expansion of SNAP threshold levels in the mid-2000s, which loosened eligibility requirements such that more people qualified for the food stamp program. He found that newly eligible households were 20 percent more likely to include an entrepreneur as a result of the policy. Incorporated businesses ownership increased by 16 percent among the newly eligible.
The paper also addresses the common perception that public assistance leads to strategic sloth. "There's this fear that people go on the program and fudge the system so that they can stay on it," Olds says. "The concern is that people will deliberately keep their incomes low so that they can 'camp out' and keep the benefits."
As with his health insurance study, Olds acknowledges personal motivations for looking into the effects of SNAP. "I've never seen my mother working harder than when we were on Food Stamps, so I was skeptical that people would reduce their labor supply after receiving benefits," he says.
Studying Census Bureau data from 1996-2011, he found that the expansion of SNAP correlated with a 5 percent increase in the average length of a workweek. "I found no evidence of camping out," he says. In fact, many of the people in the study never actually took advantage of the public benefits programs for which they were newly eligible, Olds says. Yet they took the risk of starting new companies even though their budget constraints had not changed, and even though they never signed up for SCHIP or SNAP. It was enough just to know that the social safety net was in place.
"Having the net made these people more willing to walk on the high wire," Olds says. "It didn't make them want to walk on the net."
In the next stage of his research, Olds is researching how the social safety net influences the potential labor pool for start-up firms. Initial results indicate that public benefits provide risk assurance among start-up employees, too.
"Let's say I have a comfortable management position at a large firm, but I'm thinking of taking a chance on joining a new small business," Olds explains. "I want to get in on the ground floor of what could become a wildly successful company, but I also want to make sure my kids have health insurance. If I have this outside option, if I have access to public health insurance, then I'm more likely to join. The founders of the next Google or Facebook are happy because they can get good labor. And the new employee is happy because the safety net is there."