Hearing on Poverty and Health

Committee on Health, Education, Labor, and Pensions
Subcommittee on Primary Health and Aging
United States Senate

Thank you, Chairman Sanders and Ranking Member Paul for the opportunity to speak with you today about the relationship between poverty and health, and how government should address these goals.

Any sincere effort to grapple with the problems of poverty must begin with the understanding that poverty has been the natural state of affairs throughout human history. Only in the past few hundred years have humans struck upon the antidote to poverty. Rather than begin our inquiry with the question, “What are the causes of poverty and how can we eradicate them?”, we must instead begin by asking, “What are the causes of prosperity and how may we promote them?”

This was the very aim of Adam Smith’s volume An Inquiry into the Nature and Causes of the Wealth of Nations — known to most as The Wealth of Nations — published in 1776. Smith demonstrated that trading with others leads to enormous gains in innovation and productivity, and thereby greater wealth. Figure 1 illustrates how rapidly the United States’ market economy has made new and often life-saving products available to people who previously could not afford them.

U.S. households officially classified as “poor” today have access to amenities that not even the wealthiest people in the world could access just 100, 50, or even 20 years ago. Nearly all of the U.S. poor (99.6 percent) have refrigerators, 78 percent have air conditioning, 65 percent have one or more DVD players, 62 percent have clothes washers, 55 percent have cellular phones, 53 percent have clothes dryers, and 17.9 percent have big-screen televisions.1 To highlight these numbers is not to deny that poverty is a problem. It is to highlight that a market economy is the remedy.

Figure 1


The benefits of this market process can be seen in U.S. health statistics. Figure 2 shows the actual and projected survival rates of men after age 60 in both the top and bottom halves of the earnings distribution from two birth cohorts: men born in 1912 and men born in 1941.2 One interesting feature of Figure 2 is that the “gap” in survival rates between the top and bottom halves of the earnings distribution is larger for men born in 1941 than for men born in 1912. Put differently, the gap in survival rates between higher- and lower-income males is growing. But that is not even the most interesting aspect of Figure 2.

Much more interesting is that men born in 1941 who were in the lower half of the earnings distribution (the dashed green line) are living longer than did men in the top half of the earnings distribution among those born in 1912 (the solid red line). In other words, the lower-income males born in 1941 are living longer than the higher-income males born 29 years earlier. We should all be able to celebrate this progress: both upper- and lower-income workers are living longer; and today’s lower-income workers are living longer than yesterday’s upper-income workers.

As a threshold matter, then, governments should not pursue policies (and should eliminate existing policies) that inhibit economic exchange and wealth creation.3 Unfortunately, governments the world over maintain policies that reduce economic activity and thereby perpetuate poverty, often for the benefit of a privileged few. Such policies include government-imposed barriers to trade, which leave all nations poorer and trap Third World residents in lives of privation far worse than that known to the U.S. poor. These policies also include high marginal tax rates. In the United States, excessive marginal tax rates destroy anywhere from 25 cents to $1.65 of economic activity for every dollar of tax revenue the U.S. government collects.4 Excessive tax rates mean fewer jobs, less opportunity, and fewer goods and services for Americans to consume.

Figure 2


Poverty and Health

If we seek to improve lives by improving population health, it is not sufficient to identify a social factor that is associated with health outcomes and throw taxpayer dollars at it. We must first identify the causal relationships between various factors and health outcomes. Second, we must identify policies that yield improvements in those factors and whose benefits exceed the costs.

Figure 3, created by economist David Meltzer, demonstrates the difficulties inherent in the first task. The economic literature shows a correlation between poverty and health, but this relationship is complex. The existence of a correlation between A and B does not tell us whether A causes B, whether B causes A, or whether some third factor causes both. Poverty may cause some people to suffer poor health, while poor health may drive some people into poverty. And indeed many other factors are also correlated with health, including education, social status, health behaviors (e.g., smoking, exercise), genetics, access to medical care, and more. The arrows in Figure 3 show the causal connections between the many factors associated with health. Factors such as income, insurance status, education, and health behaviors not only influence health status but are influenced by health status. These factors may also exert an influence on each other.

Figure 3


With so many complex interactions between the factors associated with health, establishing the relative influence of any one factor requires controlling for all the others. In complex phenomena like human health, that means conducting a randomized trial. Such trials are expensive and often impractical. Yet without them, policymakers who attempt to maximize health by focusing on factors with which it is most correlated may neglect other factors that have a greater causal influence on health.

Even if policymakers can overcome this hurdle, it is not sufficient to create new government programs that would deliver improvements in a known determinant of health. Policymakers must also ensure that the benefits of such programs exceed their costs, and that they deliver the greatest improvement in health per dollar spent. Most important, in judging the efficacy of anti-poverty programs, policymakers must look at all of the program’s effects, both seen and unseen.5 Unfortunately, such accounting is usually lacking.

On the benefits side, this means not looking solely at the consumption that the program enabled. We must also subtract the private charity and self-help for which the program substituted. Crowd-out is a persistent phenomenon with government anti-poverty programs. Economist Jonathan Gruber has estimated that, in effect, six out of every 10 new enrollees in Medicaid and the State Children’s Health Insurance Program would have had health coverage anyway.6 If the aim of these programs is to expand health insurance coverage, only four of those 10 new enrollees count toward that goal. Elsewhere, Gruber has estimated that “church spending fell by 30 percent in response to the New Deal, and that government relief spending can explain virtually all of the decline in charitable church activity observed between 1933 and 1939.”7

Likewise, the costs of government programs go far beyond the tax dollars required to fund them. The costs also include the economic activity destroyed by those taxes, other behavioral changes the programs produce, and any additional economic distortions.

Programs that offer subsidies to those with low incomes or assets also withhold those subsidies when incomes or assets exceed certain thresholds, for example. The potential loss of subsidies can discourage individuals from climbing the economic ladder. Gruber has estimated that the Medicaid program encourages low-income households to reduce their asset holdings by $1,600 to become eligible for the program.8 The “Patient Protection and Affordable Care Act” of 2010 (PPACA) offers large subsidies to help low-income households purchase health insurance. But because those subsidies shrink or disappear when household income exceeds certain thresholds, the law creates effective marginal tax rates in excess of 100 percent on low-income households.9 Those implicit marginal rates are far higher than the marginal tax rates faced by the wealthiest Americans.

The behavioral changes that such programs encourage can have the perverse effect of expanding poverty if they induce Americans not to climb the economic ladder. The fact that the 1996 welfare reforms led to a vast reduction in the number of Americans receiving cash assistance yet was not accompanied by an increase in poverty (which actually fell) suggests that government anti-poverty programs can have very high off-budget costs.

Unfortunately, the political system as an institution does not take the care to identify which social factors promote health, much less target those factors for improvement in a cost-effective way.

The highest-profile example of this is PPACA. President Obama claimed this law will “save lives.” Yet the most reliable research to date suggests that the federal government’s last great expansion of health insurance coverage — Medicare — did not save a single life in at least its first 10 years of operation.10 Congress rushed PPACA into law without bothering to wait for the results of the one study — the randomized, controlled Oregon Health Insurance Experiment11 — that might inform policymakers about PPACA’s benefits and enable them to ascertain whether they could deliver even greater gains in health and financial security for the same or less money.


As stewards of the public fisc, your first task is not to create or expand government anti-poverty programs in response to every perceived need, but to ascertain whether existing programs are wise investments of taxpayer dollars at all. Ideally, that research would capture all of these programs’ costs, which go far beyond outlays to include the economic activity destroyed by the taxes that finance them and the by the incentives such programs create not to climb the economic ladder.

A good place to start would be to build upon the Oregon Health Insurance Experiment by allowing other states to conduct similar experiments. Rather than expand Medicaid eligibility to all Americans under 138 percent of the federal poverty level as PPACA requires, states could use a lottery to extend Medicaid coverage to a predetermined number of residents with incomes below that threshold, and measure the results.

Armed with those results, policymakers could determine whether they would save more lives by expanding Medicaid or by funding smaller programs targeting vulnerable populations with highly effective treatments (e.g., programs offering hypertension screening and treatment for low-income adults). Such experiments would cost the federal treasury less than the Medicaid expansion mandated by PPACA, would reduce future deficits, and could yield further savings while helping to save lives.


1 Robert Rector and Rachel Sheffield, “Air Conditioning, Cable TV, and an Xbox: What is Poverty in the United States Today?” Heritage Foundation Backgrounder no. 2575, July 19, 2011, http://www.heritage.org/Research/Reports/2011/07/What-is-Poverty.
2 Hilary Waldron, “Trends in Mortality Differentials and Life Expectancy for Male Social Security-Covered Workers, by Socioeconomic Status,” Social Security Bulletin, Vol. 67, No. 3, 2007, http://www.ssa.gov/policy/docs/ssb/v67n3/v67n3p1.html.
3 See generally, Johan Norberg, In Defense of Global Capitalism (Washington: Cato Institute, 2003).
4 Christopher J. Conover, “Congress Should Account for the Excess Burden of Taxation,” Cato Institute Policy Analysis no. 669, October 13, 2010, http://www.cato.org/pubs/pas/PA669.pdf.
5 “There is only one difference between a bad economist and a good one: the bad economist confines himself to the visible effect; the good economist takes into account both the effect that can be seen and those effects that must be foreseen.” Frédéric Bastiat, That Which Is Seen, and That Which Is Not Seen, 1850.
6 Jonathan Gruber and Kosali Simon, “Crowd-out 10 years later: Have recent public insurance expansions crowded out private health insurance?” Journal of Health Economics 27 ( 2008): 201-217; http://econ-www.mit.edu/files/6422.
7 Jonathan Gruber and Daniel M. Hungerman, “Faith-Based Charity and Crowd-Out During the Great Depression,” Journal of Public Economics 91(2007): 1043-1069; http://www.religionomics.com/old/erel/S5-ASREC/REC05/Gruber%20-%20Hungerman%20-%20Faith-based%20Charity.pdf.
8 Jonathan Gruber and Aaron Yelowitz, “Public Health Insurance and Private Savings,” Journal of Political Economy 107, no. 6, part 1 (December 1999): 1259.
9 Michael F. Cannon, “Obama’s Prescription for Low-Wage Workers: High Implicit Taxes, Higher Premiums,” Cato Institute Policy Analysis no. 656, January 13, 2010, http://www.cato.org/pubs/pas/pa656.pdf.
10 David Jackson, “Obama: ‘On the precipice’ of health care change, though ‘differences’ remain,” USA TODAY, Dec 15, 2009, http://content.usatoday.com/communities/theoval/post/2009/12/obama-on-the-precipice-of-health-care-reform-though-differences-remain/1. Amy Finkelstein and Robin McKnight, “What Did Medicare Do? The Initial Impact of Medicare on Mortality and Out of Pocket Medical Spending,” Journal of Public Economics 92, July 2008, 1644-1668.
11 Michael F. Cannon, “Oregon’s Verdict on Medicaid,” National Review (Online), July 7, 2011, http://www.nationalreview.com/articles/271252/oregon-s-verdict-medicaid-michael-f-cannon.