Low-Income Programs Are Not Driving the Nation’s Long-Term Fiscal Problem
Programs Outside Health Care Projected to Decline As Share of Economy
Revised October 28, 2013
Low-income programs are not driving the nation’s long-term fiscal problems, contrary to the impression that a narrow look at federal spending during the Great Recession might leave. In fact, virtually all of the recent growth in spending for low-income programs is due to two factors: (1) the economic downturn and (2) rising costs throughout the U.S. health care system, which affect costs for private-sector care as much as for Medicaid and other government health care programs. Lawmakers should bear these facts in mind as they struggle over how to address the nation’s long-term fiscal challenges.
The first cause — increased spending on safety net programs because of the recession — is both appropriate and temporary. Congressional Budget Office (CBO) projections show that federal spending on low-income programs other than health care has started to decline and will fall substantially as a percent of gross domestic product (GDP) as the economy recovers. By the end of the decade, it will fall below its average level as a percent of GDP over the prior 40 years, from 1973 to 2012. (See Figure 1.) Since these programs are not rising as a percent of GDP, they do not contribute to our long-term fiscal problems.
Specifically, federal spending for low-income programs outside health care (including refundable tax credits such as the Earned Income Tax Credit) averaged 2.1 percent of GDP over the past 40 years. This spending peaked at 2.9 percent of GDP in fiscal year 2010, a substantial increase. But it has already fallen in 2011 and 2012, and CBO projects that it will return to the prior 40-year average of 2.1 percent by 2017 and then fall further, to 1.7 percent of GDP, by 2023 — a level well below the prior 40-year average.
The above figures include expenditures for both mandatory (entitlement) and discretionary (annually appropriated) programs. Federal spending for low-income discretionary programs is virtually certain to fall as a percent of GDP in the coming decade because, under the Budget Control Act’s funding caps, overall non-defense discretionary spending will fall to its lowest level on record as a percent of GDP, in data that go back to 1962, even if the sequestration budget cuts are cancelled. If low-income discretionary programs remain a constant share of total non-defense discretionary spending, their costs will necessarily fall as a share of GDP as well.
Figure 2 shows the major components in 2013 of federal spending for low-income programs other than health programs.
Background: Fiscal Issues and the Role of Health Care
The nation faces a serious but manageable long-term fiscal problem as a result of a large projected imbalance between revenues and expenditures. Under current policies, total program expenditures (all spending except interest payments on the debt) will decline as a percent of GDP for the next five years, as the economy continues to recover from the recession, but then climb for the following two decades as the population ages and health care costs continue to rise throughout the U.S. health care system. These two factors — the aging of the population and the growth of health care costs — explain all of the long-term growth of program expenditures as a percent of GDP.
Revenues, meanwhile, will also grow slightly faster than GDP under current policy, but not by enough to counteract rising interest rates as the economy recovers. As a result, the total imbalance between expenditures and revenues will worsen after 2018, causing the debt to climb slowly but steadily as a percent of GDP. Debt cannot grow indefinitely as a percent of GDP without eventually causing economic harm and falling living standards. This is why the nation’s long-term budget trends are considered unsustainable.
Here, we examine whether low-income programs will rise in cost as a percent of GDP and thereby contribute to our long-term fiscal problems.
To be sure, Medicaid is projected to rise significantly in cost, relative to GDP, for several reasons. To begin with, costs throughout the U.S. health care system — in both the public and private sectors — have been growing faster than GDP for several decades. Medicaid isn’t the causeof this systemwide cost growth; over the past decade, in fact, per-beneficiary costs have risen more slowly in Medicaid than under private insurance, a trend expected to continue over the next ten years. Moreover, Medicaid costs per beneficiary are substantially lower than those under private insurance (after adjusting for differences in beneficiaries’ health status), because Medicaid pays providers much lower rates and has lower administrative costs. But systemwide health care cost increases, driven in part by medical advances that improve health and lengthen life but add to costs, are expected to push up health care costs faster than GDP across the board in coming years and decades, including in Medicaid.
A second reason that Medicaid costs will rise faster than GDP is the aging of the population. Older people have much higher average health care costs than younger people. Elderly and disabled beneficiaries account for 24 percent of Medicaid beneficiaries but 64 percent of program costs. As the population ages, the number and share of Medicaid beneficiaries who are elderly will increase, raising program costs.
Another reason that Medicaid costs will continue to rise significantly is the continued erosion of employer-based health coverage. Over time, the share of low-income people able to get coverage through their (or their families’) employers has fallen, so more of them have turned to Medicaid for coverage.
Finally, the coverage expansions in the Affordable Care Act — both in Medicaid and for subsidies to help near-poor and many middle-income families afford coverage in the new health insurance exchanges — will raise expenditures for means-tested health care programs. (It’s important to note that CBO projects that these increases will not add to deficits because the costs are fully offset under the Affordable Care Act, primarily through savings in Medicare and new revenues.)
For these reasons, total expenditures for low-income programs shrink only a little as the economy recovers, then stabilize over the rest of this decade as a percent of GDP, and climb further as a share of GDP in decades after that. The continued growth of low-income health programs cannot be offset forever by continued shrinkage of other low-income programs.
However, if one examines costs for low-income programs other than health care programs, the picture changes markedly. Low-income programs outside of health care will decline in cost as the economy recovers and are not projected to rise in future years as a percent of GDP.
- In fiscal year 2012, total federal expenditures for low-income entitlement (or mandatory) programs outside health care equaled 1.8 percent of GDP. This was about 40 percent higher than the average for the 40-year period 1973-2012, which was 1.3 percent of GDP. The costs of these programs have indeed risen significantly in the last few years. For low-income discretionary programs outside health care, total costs equaled 0.8 percent of GDP in 2012, about the same as the 40-year average for those programs. Together, mandatory and discretionary low-income expenditures outside health care totaled 2.6 percent of GDP in 2012, about 24 percent above the 40-year average of 2.1 percent.
- But the recent increases are largely driven by the economic downturn and temporary program expansions under the Recovery Act. CBO projections show that total expenditures for low-income entitlements outside health care will decline steadily as a percent of the economy as the economy recovers, falling from 1.8 percent of GDP in 2012 to 1.2 percent by 2023, below their prior 40-year average. CBO’s long-term fiscal projections also assume that mandatory programs other than Social Security and health care will continue to decline as a percent of GDP after 2023.
- Spending for low-income discretionary programs is also expected to decline significantly over the coming decade. Under the Budget Control Act’s funding caps, non-defense discretionary spending — which averaged 4.0 percent of GDP over the past 40 years — will fall substantially over the coming decade as a percent of GDP, from 4.0 percent in 2012 to 2.7 percent by 2023 even in the absence of sequestration. Such a large decline makes it virtually inevitable that spending for low-income discretionary programs will decline as well. We assume that low-income discretionary programs will fare no worse or better than non-defense discretionary spending as a whole, and so will share in this projected decline. On that basis, we project that expenditures for low-income discretionary programs will fall from 0.8 percent of GDP in 2012 to 0.6 percent by 2023.
- In sum, since both mandatory and discretionary low-incomeprograms outside health care are expected to shrink below their historical average, total expenditures on low-income programs outside health care will fall over the coming decade to about 1.7 percent of GDP, well belowtheir 40-year average of 2.1 percent (see Figure 1).
After Growing Due to Recession, SNAP Enrollment and Costs Will Fall in Coming Years
The Supplemental Nutrition Assistance Program (SNAP, formerly known as the Food Stamp Program) is one of the largest federal non-health programs that assists the low-income population, especially during weak economic times, and one that House Republicans have targeted for deep cuts, which they have justified in part by implying that recent SNAP spending growth is a symptom of a problem that will continue without such cuts. These claims are incorrect. SNAP spending exemplifies the pattern we have described above: SNAP costs grew during the economic slump because of increased need, have leveled off in the past two years, and are expected to fall in coming years.
SNAP costs and caseloads grew substantially in recent years. (See Figure 3.) Threemain factors were responsible for the increases: the economic downturn, a temporary benefit increase implemented as part of the Recovery Act (the increase expires on October 31, 2013), andan increase in the percentage of individuals eligible for food stamps who actually receive them.
The single biggest reason for SNAP’s growth was the effect of the recession and the slow recovery on the economic circumstances of millions of Americans. SNAP spending rose considerably when the recession hit. That’s precisely what the program is designed to do: quickly help more low-income families during economic downturns as poverty rises, unemployment mounts, and more people need assistance. As a recent CBO report confirmed, “the primary reason for the increase in the number of [SNAP] participants was the deep recession from December 2007 to June 2009 and the subsequent slow recovery.”
Moreover, two recently published economic studies conclude that the rise in SNAP rolls in the Great Recession and its aftermath is consistent with the pattern of what occurred during and after previous economic downturns. The key difference is that this time, the jobs slump is far worse.
With the deep, prolonged recent recession and ensuing weak recovery, SNAP has become increasingly important to the long-term unemployed. It is one of the few resources for people who have exhausted their unemployment benefits and still been unable to find a job. Both the duration of unemployment and the prevalence of long-term unemployment — defined as being unemployed for 27 weeks or longer — have been sharply higher in the current economic slump than in any previous downturn, with data going back to the late 1940s.
The second factor in recent SNAP spending growth is the large but temporary increase in SNAP benefits enacted as part of the 2009 Recovery Act in order to reduce hardship and deliver high “bang-for-the-buck” economic stimulus. The increase, which accounted for 20 percent of the growth in program costs between 2007 and 2011, according to CBO, will end on October 31.
A final significant cause of increasing SNAP enrollment and costs: the share of individuals eligible for food stamps who actually receive them has risen. This percentage fell from 75 percent in 1994 to 54 percent in 2002 and has now returned to 75 percent. Of particular note, the percentage of eligible individuals in low-income working families receiving SNAP rose steadily from 43 percent in 2002 to about 65 percent in 2010, the highest on record.
Despite these clear reasons for SNAP’s growth, some people have looked at the recent trends in SNAP enrollment and costs and assumed SNAP costs will remain at their current high levels or continue growing after the economy recovers. Analysis shows that such assumptions are off the mark.
As in many other areas of budgetary analysis, the year one picks as a starting point is important. Looking back only ten years at SNAP costs, for example, provides a skewed picture of program growth because SNAP participation and costs had plummeted a decade ago, due in part to a large decrease in the proportion of eligible families receiving SNAP in the latter half of the 1990s.
The 1996 welfare law was intended to encourage work, but in the first years after its passage, many families that moved from welfare to low-wage work were cut off SNAP when they left welfare — even though they remained eligible for SNAP benefits — because of problems in state administrative systems. That result was contrary to what Congress intended. Aggravating this problem, some states instituted administrative practices in those years that had the unintended effect of making it harder for many working-poor parents to participate in the program, largely by forcing them to take substantial time off from work for visits to SNAP offices every 90 days to reapply for benefits.
A bipartisan consensus emerged that policies that made it difficult for people to continue receiving food stamps if they left welfare for low-wage work were misguided and contrary to welfare reform goals because they diminished work incentives. Accordingly, Congress enacted significant, although relatively modest, changes on a bipartisan basis in 2002 and 2008 to lessen barriers to SNAP participation among working-poor families, as well as modest improvements in benefits that primarily helped such families.
Some critics have claimed that the fact that SNAP enrollment has not declined in tandem with the unemployment rate indicates most of SNAP’s recent enrollment growth is unrelated to the economy. The reality, however, is that SNAP enrollment and costs remain high because the job market remains weak. The recent reductions in the unemployment rate overstate the improvements in the labor market. Most tellingly, the proportion of the adult population with a job (the employment rate) has barely improved since the depth of the recession. And, even as the overall number of unemployed workers has declined, the number of unemployed workers not receiving Unemployment Insurance, and likely qualifying for SNAP based on their low incomes, has increased. Moreover, the historical record shows that declines in poverty and SNAP enrollment typically lag behind declines in the unemployment rate following recessions. As CBO has observed in other recent recessions, “decreases in [SNAP] typically lagged improvements in the economy by several years.”
In coming years, caseloads and spending are expected to begin declining as households’ economic circumstances improve and the Recovery Act’s benefit increase ends. SNAP caseload and spending growth already have slowed substantially as the economy has begun to recover. SNAP spending in the first half of calendar year 2013 was only 1.5 percent higher than the same period in 2012.
Figure 4 shows SNAP costs as a percent of GDP from 1995 to the present and CBO’s projections through 2023. As it shows, by 2019, costs are expected to decline back to their 1995 level as a percent of GDP and then to edge lower. Because SNAP is projected to shrink as a share of the economy, it is not contributing to the nation’s long-term budget problems. No significant demographic or programmatic pressures will cause SNAP costs to grow faster than the economy.
The story for SNAP is thus the clearest illustration of the story for overall low-income program spending outside of health care: costs have grown substantially in recent years as a percent of GDP, an appropriate response to the worst economic slump since the Great Depression. But as the economy recovers, costs will return to, or even edge below, prior levels as a percent of GDP. These programs therefore do not contribute to the nation’s long-term fiscal problems.
Policymakers and others may hold different views about whether programs for the poor should be maintained at current levels, strengthened, or weakened. But the mistaken assumption that the universe of safety-net programs is experiencing ever-increasing costs is not a sound reason to impose deep cuts in this part of the budget.
 For an example of a paper that creates such misconceptions, see Congressional Research Service, Spending for Federal Benefits and Services for People with Low Income, FY2008-FY2011: An Update of Table B-1 from CRS Report R41625, Modified to Remove Programs for Veterans, October 15, 2012, available at http://budget.senate.gov/republican/public/index.cfm/files/serve/?File_id=0f87b42d-f182-4b3d-8ae2-fa8ac8a8edad as of October 2, 2013.
 Debt can and generally should rise as a percent of GDP during recessions, in order to cushion the loss of household income caused by the recession and keep the recession from being unnecessarily deep and protracted. Recessions and their accompanying debt increases are temporary. During normal economic times, policymakers should try to ensure that deficits are modest enough for debt to remain stable or decline as a percent of GDP. Our projections show that the debt ratio will resume rising in five years and continue to do so through 2040. Although our long-term projections (and those made by CBO and the Committee for a Responsible Federal Budget) are far less explosive than they were a few years ago, they nevertheless show a rising debt ratio under current policies. See our paper on CBPP’s long-term projections here: http://www.cbpp.org/cms/index.cfm?fa=view&id=3983.
 John Holahan et al., “Medicaid Spending Growth over the Last Decade and the Great Recession, 2000-2009,” Kaiser Commission on Medicaid and the Uninsured, February 2011 and John Holahan and Stacey McMorrow, “Medicare and Medicaid Spending Trends and the Deficit Debate,” New England Journal of Medicine, 367:393-395, August 2, 2012.
 As with mandatory programs, our figures for discretionary programs do not assume that the automatic cuts, or “sequestration,” that the Budget Control Act calls for will occur after 2013; if they did, total non-defense discretionary funding would fall to 2.5 percent of GDP rather than 2.7 percent.
 Congressional Budget Office, “The Supplemental Nutrition Assistance Program,” April 2012.
 Marianne Bitler and Hilary Hoynes, “The More Things Change, the More They Stay the Same:
The Safety Net, Living Arrangements, and Poverty in the Great Recession,” draft, May 2013. http://www.econ.ucdavis.edu/faculty/hoynes/working_papers/Bitler-Hoynes-GR-fin.pdf. This study examines SNAP participation through 2011 and finds “[The] safety net programs receiving the most attention through the Great Recession (Food Stamps and UI) exhibit adjustments very consistent with their behavior during previous historical cycles.”
Peter Ganong and Jeffrey B. Liebman, “Explaining Trends in SNAP Enrollment,” August 2013. http://scholar.harvard.edu/ganong/publications/explaining-trends-snap-enrollment/. Ganong and Liebman conclude that changes in employment conditions account for a minimum of 69 percent and as much as 96 percent of the increase in SNAP participation over this period.
 Chad Stone, Jared Bernstein, Arloc Sherman, and Dorothy Rosenbaum, “SNAP Enrollment Remains High Because the Job Market Remains Weak,” Center on Budget and Policy Priorities, Revised July 20, 2013, http://www.cbpp.org/cms/index.cfm?fa=view&id=3996.
 The most recent year for which USDA publishes these estimates is 2010.
 These three factors dwarf the impact of SNAP eligibility changes on program costs. For example, CBO estimates that an eligibility change sometimes referred to as “broad categorical eligibility,” an eligibility change that has recently drawn attention, accounts for only 2 percent of program costs.
 Stone, et. al, op. cit.
 CBO, op. cit. p. 3.