In my very first Inequalities blog post I considered the argument that the United States has a more generous safety net for the poor than conventional comparisons with European states would suggest because we spend so much on public health care programs for the poor, elderly, and disabled. Subsequently, I reviewed work by Richard Burkhauser and others that suggests that accounting for health care benefits to low-income households substantially reduces the estimated rise in income inequality over time.
The New York Times has an interesting piece weighing the arguments for and against counting health care benefits in our poverty statistics. Right now the monetary value of government health programs are not counted as income in the official poverty statistic (which also does not count food stamps or any other non-cash benefits) and they are not included in the Census Bureau’s supplemental poverty measure. Now, however, the Congressional Budget Office is including health care benefits at their full dollar amount in their estimates of taxes and spending.
This may seem like so much bean counting, but it’s not. There are tricky conceptual questions here and lots of moving pieces – over time several things have been happening simultaneously. The price of health care for everyone has been rising much faster than inflation, the quality of health care has also been increasing (because we have more surgeries, drugs, and therapies then in the 1960s when Medicaid was enacted), and the social safety net has mainly been declining in other areas benefiting those at the very bottom of the income distribution. The net effect is that government health care looks much more generous for the poor in absolute and in relative terms.
All that said, health care may not, on average, be valued by poor households nearly as much as it costs for the government to provide it to them. As Tim Smeeding says in the article, “you can’t eat health care.” For families that need money for housing, food, and utilities, cash or food stamps may be much more valuable than health insurance (which they may not choose to use at a particular point in time). On the other hand, as some health economists like Jon Gruber point out, the security provided by health insurance benefits may be much more valuable to the poor than what the government pays for them. Having the security of health insurance protects people from very catastrophic and totally unaffordable expenditures that would otherwise lead to death and irreversible disability. Even when these benefits are not used, people gain peace of mind from knowing they are protected.
How can we get to the bottom of this? I have two suggestions. First, we need to get better at measuring the value of health insurance from the perspective of low-income families. The first promising route is to take advantage of real world scenarios in which households are offered the choice between taking health care benefits or taking cash instead. To some extent, we can observe this in the labor market when individuals are faced with the choice between taking job A, which provides higher wages but no health insurance benefits, and job B, which provides lower wages but includes health insurance benefits. This “wage-fringe tradeoff” provides insights into how much money people are willing to give up in order to get benefits. The wrinkle in the United States is that employers have a greater incentive to offer health insurance benefits, because those benefits are not taxed (unlike wages). This matters less for low-income families, because they generally pay little if any income taxes. Still, we see that some low-income workers gravitate to jobs where they can get health insurance benefits, and these workers tend to be older and less healthy. To deal with this selection issue, we could focus on mandated coverage benefits as a natural experiment (for example, national health reform includes such a mandate on employers), and observe how much wages shift in response to mandated coverage.
Second, once we have a credible estimate of how much low-income workers, on average, are willing to trade in income for health insurance benefits, we need a framework for incorporating this valuation into policymaking. This is trickier than it seems. There is likely to be a very wide range in how much these benefits are valued among low-income families, and setting the valuation at the average could have the effect of majorly discounting the value of benefits to very sick low-income individuals. There are also issues related to myopia: some people may not value benefits enough, because they are misinformed about their objective risk of getting sick and requiring medical attention. Still, gaining a better understanding of what health care is worth to the poor relative to other things that the welfare state could be doing to improve their lives could help to shape public priorities: as a society we might decide to trade in a little bit of health for a lot more wealth.