The University of Michigan
555 South Forest Street
Third Floor
Ann Arbor, MI 48104-2531

T 734-936-9842
F 734-998-6341



The answers may warrant a fresh look at how policymakers craft solutions to extending health insurance coverage to uninsured Americans.

Interview with Helen Levy, Economist and Researcher at the University of Michigan

  University of Michigan Department of Health Management and Policy Economist and ERIU researcher Helen Levy spends her time exploring the reasons why some households do not have health insurance. Recently, Levy has examined how people spend their money, and the relationship this has to policies for expanding health coverage. Her findings should influence the debate about the merits of tax credits or vouchers for health insurance. Before coming to Michigan, Levy was an assistant professor in the Irving B. Harris Graduate School of Public Policy Studies at the University of Chicago, where she focused on health economics, public finance and labor economics and other social issues related to the uninsured. Levy, who recently co-authored a paper for ERIU, "What Do

People Buy When They Don't Buy Health Insurance: And What Does that Say about Why They are Uninsured?" shares her views on the spending patterns of people without health insurance and what that portends for developing solutions to making coverage more affordable.

By last count, some 45 million Americans are without health care coverage. Are they all poor and unable to afford health insurance?

LEVY: About one quarter of the uninsured are families earning less than the federal poverty level, or about $19,000 for a family of four. Another fifteen percent are between 100 and 200% of poverty. But 40 percent are in families with incomes that are more than twice the federal poverty line, or about $38,000 a year for a family of four. Another 20 percent live in families that make more than three times the poverty line, at least $57,000 a year for a family of four.

What do the uninsured spend their money on? And how does this differ from families who are insured?

LEVY: Overall, uninsured and insured households spend most of their money on housing. That's the big-ticket item for everyone. Transportation - including car payments, car insurance, etc. - follows, and then food and utilities.

When you look at the lowest income uninsured, what are the biggest differences in how they are spending their money compared to households with insurance?

LEVY: If you look at the poorest 25 percent of all households in terms of how they spend their money, you find that the uninsured household is spending a bigger share of its budget on housing, about $90 more per quarter than an insured household, and spends about $45 more on food per quarter. To a lesser extent the poorest uninsured households are spending more on education, about $35 more per quarter, and a little bit more on alcohol and tobacco, or about $20 more each quarter.

Why do low-income uninsured families spend more on housing?

LEVY: Half of the difference is because the uninsured are less likely to own their homes. So, they have higher payments for their housing costs. We know the uninsured don't have more space per person. One may think these households are paying more for housing because they decided that instead of buying health insurance they are going to have a bedroom for each of their kids but that's not the case. The uninsured have on average one-third fewer rooms per person than an otherwise comparable insured household does.

Could this also suggest that quite a few of these individuals are young, single adults? Should we spend resources on trying to get younger people insured, as they are less likely as a group to be sick?

LEVY: Our research controls for age so in theory none of the residual differences in age are due to youth per se. However, a large number of the uninsured are young adults. Being uninsured means something much different and less risky for this group than it does for a 58-year-old with a chronic condition. But this doesn't mean they shouldn't be insured. It just means they should buy a cheap, very high deductible policy that covers them for catastrophic expenses.

Ok, let's move up the income ladder. Do the differences change between how higher-income uninsured spend their money compared to those with insurance?

LEVY: Looking at the top quarter of the income distribution - these are pretty well-off households and about 10 percent of the uninsured in our sample fall in this group. We find real differences in their spending compared to the lower-income households. Well-off uninsured households are spending more than comparable insured households on transportation, more specifically, cars, and on furniture and appliances.

So what does this suggest?

LEVY: Something very straightforward: being uninsured if you are well off is really a different thing than being uninsured and poor. It looks like the well-off uninsured like to spend their money on things that you wouldn't think of as the basic needs of life. They spend a lot more on things that you could argue are discretionary, whereas the poor uninsured are spending more on housing and food.

Can you tie income and personal preferences together, and how that affects policy solutions regarding the uninsured?

LEVY: Yes. We know that income is an important factor in the decision to buy health insurance, but it is not just about income. It's also about price. How does the price of housing affect the probability you will be buying health insurance? That is an area that has not been investigated. Once we are done looking at income and prices, we're left with differences in preferences. The most important of those is health status, which affects the price people pay and their income. So these things interact with each other. There are other components of preferences, such as an individual's risk tolerance and one's attitude about health and medical care. These things are difficult to quantify.

So if the poor uninsured are buying housing and food because those things matter to them more than health insurance, and the rich uninsured are buying cars and furniture because they care more about those things than health insurance, what is the relevance for policymakers?

LEVY: We need to think differently about the needs of the poor uninsured than we do about the rich uninsured. And, when we think about the poor uninsured, we need to think about the other things they need and where health insurance fits in that priority list.

One thing our data suggest is that at least some people who are uninsured are making a choice about how they want to spend their money. One important lesson is that when we think about policies that extend insurance coverage to low-income uninsured, we have tended to think about it from the context of "Oh, we need to help these people access health insurance at a reasonable price because then they'll buy it." Actually, our results and other research, including some funded by ERIU, show that price elasticity of demand for health insurance is quite low. If someone doesn't have insurance, you have to give them a whole lot of money before they will take it.

What is the biggest myth or misconception about why some people decide to be insured whereas others decide to be uninsured?

LEVY: The biggest misconception is to set everything up as if it were all about constraints. You hear people say, "the uninsured can't get insurance" or "the uninsured can't afford insurance." This is about households making trade-offs between different needs. These are rational economic decisions. People face a lot of demands on their household budget, and the uninsured are making choices about how to spend their money. We'll all be better off if our policies explicitly recognize that.

How does your work bust this myth?

LEVY: By highlighting the fact that things the poor uninsured are spending their money on are basic needs of life. We want to dispel the notion that the poor uninsured lack protection because they are out blowing their extra money on vacations. We need to better understand why we see big increases in discretionary spending for the well-off uninsured. What are logical policies for a group of people who have high incomes, and who aren't buying health insurance? Do we mandate that they buy health insurance coverage? Do we extend further subsidies to them? It raises the question: Who is paying for the medical care that the well-off uninsured receives? That hasn't gotten much research attention.

Why is knowing how the uninsured spend their income so important for policy?

LEVY: The policy debate about the uninsured has been very myopically focused on health insurance markets. The discussion has been really about: what is it about health insurance that makes you have it or not. It's more constructive to think of it as - these households have a whole range of choices. Health insurance is nested within that framework. It is not necessarily the top priority people have. Too often the health insurance policy debate is predicated on the notion that nothing could be more important than health insurance. Based on what people spend their money on, that isn't true.

For policymakers, we ought to be thinking whether health insurance is really the right transfer to give the uninsured. Should we think more generally about income transfers that are not targeted to health insurance? If you are putting tax credits out there for people to take up health insurance, and you are baffled about why they aren't taking them up, understanding how health insurance fits into the larger framework of a household budget will help explain that.

But can tax credits help?

LEVY: A health insurance tax credit was included in the trade adjustment assistance tax of 2002. It provided a 65 percent subsidy to buy health insurance and the take-up rate for that program was less than 10 percent. It was a giant dud. We now are having a discussion around tax credits for health spending accounts and higher deductible health plans. Again, it is a tax credit attempt to get people to buy a particular kind of insurance and it is unclear yet how successful that is going to be. To better understand whether it will be a success or not, you need to have some sense of where health insurance fits into the spending priorities of the households these programs are targeted to.

Is there a better way to target subsidies?

LEVY: I think the tax code is the right place. You can target people based on a fairly reliable measure of income. Certainly it's preferable on that basis to targeting subsidies to providers or employers

Could you put into context how the impact of Hurricane Katrina plays into this in terms of affecting people's priorities for health insurance?

LEVY: The silver lining of Katrina has been to spotlight how multi-dimensional the needs of the poor are. In the world of health policy, we tend to think about "the uninsured" and the problems of access to medical care. But the stories we're hearing are of people who face multiple obstacles: not just a disability, not just poverty, but also lack of transportation and in some cases the lack of a social network to provide support. These are similar to the factors that can impede access to medical care - even without considering health insurance coverage. Our discussions of the uninsured should keep in mind that lacking coverage may be just one of many problems poor people face.

Back to top

Funded by The Robert Wood Johnson Foundation, ERIU is a five-year program shedding new light on the causes and consequences of lack of coverage, and the crucial role that health insurance plays in shaping the U.S. labor market.