Is Obamacare the Best Way to Redistribute Income?

by Reihan Salam

One of the chief arguments for the Affordable Care Act is that it would redistribute income from affluent households to less-affluent households. A new Brookings Institution analysis by Henry J. Aaron and Gary Burtless finds that it will do just that:

The Affordable Care Act (ACA) will improve the well-being and incomes of Americans in the bottom fifth of the income distribution. Under our broadest and most comprehensive income measure we project that incomes in the bottom one-fifth of the distribution will increase almost 6%; those in the bottom one-tenth of the distribution will rise more than 7%. These estimated gains represent averages. Most people already have insurance coverage that will be left largely unaffected by reform. Those who gain subsidized insurance will see bigger percentage gains in their income.

Aaron and Burtless emphasize that standard estimates of poor Americans’ incomes don’t capture these benefits, as standard estimates rely on the Census Bureau’s concept of “money income”:

The definition includes pre-tax money wages, net self-employment earnings, pensions, government cash transfers (such as Social Security and unemployment compensation), interest, dividends, rent payments, and other regular and irregular cash income flows. It does not include the health insurance contributions employers make for their workers. Nor does it include government spending on health plans, such as Medicaid and Medicare, which pay for most or all of the health care of the people who are insured. Employers and the government currently spend about $1.5 trillion—13% of total personal income—paying our health care bills. The ACA will increase public spending to help low- and moderate-income people pay their medical bills. If we use a definition that doesn’t count the extra spending as “income,” household incomes obviously will not change very much.

Scott Winship has written extensively on the limits of “money income” as a means of assessing the living standards of poor and middle-income Americans, and Aaron and Burtless address some of the complexities involved. In addition to “money income,” the Census Bureau also uses a concept called “fungible income,” which does include the value of insurance, at least partially:

The fungible value of insurance includes the full premium contribution made by employers in behalf of their workers. It also includes some or all the subsidies provided by the government to participants in public health plans, but only for families who have enough cash income to pay for basic food and housing. If income exceeds this threshold, then part or all of the value of government financed health benefits is included. When families don’t have enough cash income to pay for basic food and shelter needs, none of the value of government insurance is counted as income. The key point is that the “fungible income” approach assigns to low income households little or no income value from publicly financed health benefits.

Only when we count as income the full cost to the government of providing additional health insurance to the poor does the ACA have a meaningful and positive impact on the post-ACA incomes of the very poor. [Emphasis added]

The authors state that had the original version of the ACA, in which states that refused to expand Medicaid would have been denied all federal Medicaid funds, been put into effect, the gains to low-income households would have been larger. They also observe that while Americans in the top three-quarters of the income distribution will experience small proportional drops in income, Americans in the bottom quarter will experience larger proportional gains.

Back in May, I raised questions about the ACA as a strategy for redistribution. If our goal is to increase incomes at the bottom of the income distribution, is providing low-income households with medical insurance the best way to do it? In theory, raising income through other means, like wage subsidies, would make low-income individuals more attractive customers for innovative, low-cost medical providers. Yet this approach is rarely seen as a good substitute for coverage expansion, in part, one has to assume, because advocates of coverage expansion are convinced that low-income individuals aren’t in a position to make good long-term decisions regarding medical care. This could be true. Sendhil Mullainathan and Eldar Shafir have argued that scarcity can undermine our ability to make good decisions, as it leads us to focus on our most pressing, immediate need without reference to downstream consequences, etc. But of course the goal of wage subsidies, and other measures designed to raise low-end incomes, is to alleviate this scarcity problem across the board. While it is unlikely that a modest increase in disposable income will suddenly make navigating the health system a breeze, one has to wonder if we’re tackling the interrelated (but not identical) problems of the stresses associated with poverty and those associating with managing health problems in the right order.