New Study Confirms How the Welfare State Perpetuates Poverty

Ronald Reagan had an old adage about the nine most terrifying words in the English language: “I’m from the government and I’m here to help.” Recently, a new paper reinforced that truth and adds to the existing literature showing how America’s welfare state often traps generations in a cycle of poverty.

At its core, a complicated set of welfare programs and tax breaks generate sizable incentives for many low-income Americans not to increase their incomes and improve their station in life. This “poverty trap” results in well-intentioned government programs hurting those they were designed to help.

Marginal Tax Rates

The study, published by the National Bureau of Economic Research (NBER), examined marginal net tax rates on American households. Their analysis included the phase-out effects of various government programs and the extent to which those phase-outs discourage work.

For instance, consider the $1,200 payments in March’s coronavirus “stimulus” legislation. Individuals with incomes under $75,000 qualified for the full $1,200 payment, while the payment was reduced by 5 percent for each dollar of income over $75,000. As a result of this phase-out rate, individuals with incomes over $99,000 receive no payment.

For every additional dollar of income a person making $75,000 received, he will lose five cents of his “stimulus” payment, plus have to pay regular federal income taxes (likely at a 22 percent rate), payroll taxes (7.65 percent), and state and local income taxes where applicable.

Put another way, the coronavirus checks gave people making between $75,000-$99,000 added incentive to reduce their income. By working fewer hours, sheltering income from taxes, or both, people could “maximize” their “free” payment from the federal government.

The researchers studied data from 2016, well before this year’s “stimulus” (or the coronavirus). But as the paper’s introduction notes, many other federal programs and laws have similar distortionary effects:

Earn $1 too much two years back and your Medicare Part B premiums will rise by close to $800. Earn $1 too much and, depending on the state, lose thousands of dollars in your own or your family’s Medicaid benefits. Hold $1 too much in assets and forfeit thousands in Supplemental Security Income. Earn an extra dollar and receive thousands of dollars in Obamacare subsidies. Earn $1 beyond Social Security’s earnings ceiling and watch your Social Security payroll tax drop to zero. Earn $1 too much and flip onto the Alternative Minimum Tax (AMT), reducing your marginal income-tax bracket from a rate as high as 37 percent to 28 percent. Earn $1 too much and lose 22 cents, in the Earned Income Tax Credit and the list goes on.

If this appears to look almost like a game, you’re not far off. It’s hard not to view it as the government picking winners and losers through its various program parameters. Made an extra $1 of income? Too bad — now we’re taking your subsidy away. Do not pass go, do not collect $200.

Worst Effects on the Poor

Unfortunately, these distortionary effects hit poor and near-poor households the hardest. As Gene Steuerle of the Urban Institute has documented, phase-outs of programs like cash welfare, food stamps, the Earned Income Tax Credit, and Obamacare subsidies mean households making roughly $10,000-$40,000 can lose almost as much in government benefits as they gain in added income by working additional hours.

The new NBER study further quantifies this phenomenon. It finds that, both over the current year and over their lifetimes, individuals in the lowest income quintile (the bottom 20 percent) face higher marginal tax rates than those in the next three income quintiles (from the 20th percentile of income through the 80th percentile). Essentially, when taking the phase-out of government benefits into account, the poor face more disincentives to work than the middle class.

It gets worse. One in four low-income households (those with income in the bottom 20 percent) face lifetime marginal net tax rates of more than 70 percent. As the authors put it: “One in four of our poorest households, regardless of age, make between two and three times as much for the government than they make for themselves in earning an extra $1,000.” Given the construct of the modern welfare state, it seems less logical to ask why poor people wouldn’t work and instead to ask why they would.

There is, however, one silver lining in the paper: States can help undo the damage caused by poorly crafted federal policy. The NBER researchers found that a “typical household can raise its total remaining lifetime spending by 8.1 percent by moving from a high-tax to a low-tax state.”

Break the Cycle of Poverty

Obamacare didn’t create the phenomenon of the welfare state discouraging work, but it did make this worse. The Congressional Budget Office noted repeatedly that the phase-outs in the law’s insurance subsidies penalize individuals who earn more income. Its most recent in-depth analysis, conducted in February 2014, concluded the law would reduce the labor supply by the equivalent of about 2.5 million full-time jobs.

More recently, of course, lawmakers in the CARES Act provided a $600 per week federal supplement to unemployment insurance, further discouraging work. Because more than two-thirds of unemployed individuals can now make more money on unemployment than they did while working, businesses face difficulty recruiting furloughed employees back to their jobs.

At a time our country faces a massive recession brought on by the coronavirus lockdowns, America’s welfare state exacerbates that stagnation. Reforming the system to eliminate work disincentives could save taxpayer funds. More importantly, it would encourage all Americans to embrace the dignity of work.

This post was originally published at The Federalist.

Medicaid vs. Cash for the Poor

In a Think Tank post last week, I wrote that in trying to sell Obamacare’s Medicaid expansion to states, the administration “has made no attempt to argue that expansion represents the most economically efficient use of those dollars—that the funds could not be better used building bridges, returned to citizens,” or other things. A study released this month raises questions about the very utility and efficiency of Medicaid coverage.

The study, from MIT’s Amy Finkelstein, Nathaniel Hendren, and Erzo F.P. Luttmer, used data from an Oregon health insurance experiment—in which some low-income citizens received access to Medicaid and some did not, based on the results of a random lottery—to estimate the utility of Medicaid coverage. They found that beneficiaries valued Medicaid at 20 cents to 40 cents on the dollar; in other words, for every $1,000 the states and federal government spent on health coverage, the average beneficiaries felt like they were receiving goods or services valued at $200 to $400.

In response, Bloomberg View’s Megan McArdle asked whether the poor would prefer cash benefits to Medicaid. It seems like a fanciful question; for one thing, programs providing cash benefits—such as the Earned Income Tax Credit—historically suffer from a large incidence of fraud. But say a state wanted to offer residents such a choice. Would the Obama administration allow it?

The state waiver program included in Obamacare provides flexibility only to the extent that states provide coverage at least as generous to as many people as the health-care law itself. A state cannot target resources only to certain groups—individuals with disabilities, for example—or provide slimmed-down coverage to some beneficiaries. Under that logic, it seems that if a state wanted to provide residents a choice between a smaller cash benefit and Medicaid insurance coverage, the administration would not permit such a measure—even though, according to the MIT study, the average beneficiary would prefer this outcome, and taxpayers would benefit as well.

While the Obama administration talks about its flexibility, that appears to apply only when such flexibility promotes the administration’s objectives. The president said early in his tenure that for too many years “rigid ideology has overruled sound science.” Now, an MIT study shows “sound science” questioning the efficiency and utility of Medicaid coverage for beneficiaries. Will the administration react to this scientific evidence, or will its own ideology prevent the consideration of more innovative, and potentially more effective, policies?

This post was originally published at the Wall Street Journal Think Tank blog.

JCT: At Least $27 Billion in Improper Insurance Subsidy Payments

CongressDaily has an interesting article this morning on a significant source of over-spending in the health care law – excessive health insurance subsidies provided to individuals and families.  The issue arises from the fact that the new insurance subsidies are based on an individual’s (or family’s) most recent tax return – so that subsidy levels beginning in January 2014 will be based on reported income for 2012.  As might be expected, a family’s circumstances can change significantly during this time lag for a variety of reasons – a change in job, significant raise, divorce, birth, or death, to name just a few.

The health care law established a reconciliation process intended to recapture any subsidy over-payments – but the law capped the amount of such repayments at $250 for individuals and $400 for families.  As the article reports, raising these limits to $1,000 for an individual and $2,000 for a family would result in an additional $27 billion in repaid subsidies.  This $27 billion in subsidy overpayments represents a significant portion of the $464 billion total devoted to insurance subsidies in the new health law.  But the figure does NOT represent the full level of overpayments – it excludes the revenue repaid under the law’s existing provisions (capped at $250 per individual and $400 per family), and it does not include any additional revenue that might be repaid if the repayment cap were lifted entirely, rather than merely raised to $1,000 per individual and $2,000 per family.

It is however fair to say that at least $27 billion – and quite possibly a good bit more than $27 billion – will be paid in insurance subsidies to individuals who do not deserve them based upon their income levels at the time they actually receive the payment.  Some of these payments could be a result of innocent mistakes that a family might not have noticed.  But it’s also worth asking whether the law itself encourages individuals not to report changes that would reduce their subsidy eligibility: After all, would you be quick to disclose a change in income that will result in your insurance subsidy being cut, if you knew that the most you would have to pay back for receiving thousands of dollars in taxpayer-funded subsidies you didn’t deserve would be $400?

The Obama Administration recently announced its intention to cut Medicare fraud in half by 2012.  That’s an admirable goal, considering both the skyrocketing federal budget deficit and Medicare’s shaky long-term financing.  But it’s worth examining whether the overpayments associated with what an expert quoted in the article called “an entirely new welfare system” will erase any gains from anti-fraud efforts in Medicare – and whether, at a time of skyrocketing deficits and high unemployment, it is appropriate for the federal government to raise taxes by more than half a trillion dollars to create a system where individuals will receive more than $27 billion in insurance subsidies to which they are not entitled.


Tinkering With Health Credits Eyed As Way To Cut Costs

Wednesday, June 30, 2010
by Peter Cohn

An area of health reform that has received little attention is getting a new look as deficits mount: what happens when someone receives larger health insurance subsidies than they are eligible for because they made too much money.

Under the law, low-income individuals and families can get tax credits to help pay insurance premiums beginning in 2014, when new exchanges come online. The credits are based on income, marital status and number of children — up to 400 percent of the poverty line, when the out-of-pocket expense is capped at 9.5 percent of adjusted gross income and the government picks up the rest.

According to CBO, the credits will cost $464 billion through 2019, the largest expense in the health law. And that’s just in the first six years, meaning the cost should grow considerably beyond 2020.

The kicker is eligibility is based on a household’s most recent tax return. With enrollment beginning in October 2013, eligibility would be determined by 2012 reported income. In January 2014 when the credits take effect, an individual or family could have had much-changed economic circumstances in the intervening years.

It can work both ways: A household could earn significantly less and be eligible for bigger credits due to job loss, divorce, retirement, or a newborn child, for instance. In such cases they can reapply for larger credits, which are awarded monthly. If a household is earning more money, they still get to keep much of their credits received during the year.

There is an IRS “true up” process where the credits are reconciled with actual income, and taxpayers are required to pay back up to $400 in excess credits, or $250 for individuals, which can be a fraction of the overpayments. If income turns out to be above 400 percent of poverty — $88,200 for a family of four based on 2009 guidelines — it can get very expensive. They would have to pay back the entire amount received during the year.

One proposal floating on Capitol Hill would increase the payback requirement from $400 to $2,000 for households and $250 to $1,000 for individuals. The Joint Committee on Taxation scored that idea as raising $27 billion over 10 years, which would barely make a dent in the overall cost but could provide deficit-conscious critics of the law an opening.

Take a family of four earning $44,100, which is 200 percent of the poverty line based on 2009 guidelines. They would have to pay 6.3 percent of income toward health coverage, which under a hypothetical $13,000 insurance plan would total $2,778. The government pays the remaining $10,222 in tax credits, freeing up that money for the family.

Then comes tax season the following April, and due to new income, it turns out the family actually made $55,125. That is up sharply from two years prior, putting that household now at 250 percent of poverty — meaning they should have contributed 8.05 percent, or $4,438, and received tax credits worth $8,562.

That’s a $1,660 difference between credits received and what they were eligible for, of which the family would have to pay back $400 — essentially a $1,260 freebie. Under the alternative scored by JCT, whose authors requested anonymity because the proposal was unofficial, they would have to pay back the full $1,660.

Urban Institute fellow Eugene Steuerle argued the program will likely be error-prone and difficult to administer. “It’s extraordinarily hard to collect from people at the end of the year. That’s why we try to have a tax system that withholds accurately, or over-withholds,” said Steuerle, a Treasury tax official in the George H.W. Bush administration. “They’re really not just creating a new health system; they’re creating an entirely new tax system and an entirely new welfare system.”

He said extra tax liability of up to $2,000 wouldn’t be easy to pay for a family of four earning shy of $90,000.

A Senate Democratic aide said repayments were limited to $400 to ease the burden on poor families, as those earning more than 400 percent of poverty can more easily pay it back. It was also designed that way to avoid similar flaws with Earned Income Tax Credits taken in advance during the year, which many low-income families have turned down for fear their income will end up too large and they would have to pay it back.

Judith Solomon, co-director of health policy at the Center on Budget and Policy Priorities, said increasing the repayment caps could prove a deterrent to getting health coverage through the exchanges.

“For some people, knowing they’re going to have this tax liability, it’s likely they’re going to come in in the middle of the year and say ‘I don’t want a credit anymore,’ ” Solomon said. She added the $400 penalties were probably enough to make credit recipients report changes in income during the year, thus avoiding large overpayments and holding down costs.

Ed Haislmaier, senior research fellow in health policy at the Heritage Foundation, said upping the repayment caps might make the deficit numbers look a little better but wouldn’t resolve basic design flaws in the system. “This is not health policy. This is income redistribution,” Haislmaier said.

Casey/Brown (OH) Amendment (#4371) on COBRA Subsidies

Senators Casey and Brown (OH) have offered an amendment (#4371) regarding COBRA health insurance subsidies.  A vote is possible later today.  This amendment may be subject to a Budget Act point of order for exceeding aggregate spending caps.
  • The amendment would extend eligibility for federal subsidies of COBRA continuation coverage first included in the “stimulus” (P.L. 111-5).  The amendment would allow individuals currently eligible for subsidies – those suffering a loss of employment prior to May 31, 2010, when the subsidies lapsed – to continue receiving a maximum of 15 months of subsidized coverage.  Individuals laid off after June 1, 2010 would receive only six months of subsidies.
  • The amendment includes elimination of the advanced refundability of the Earned Income Tax Credit (EITC) as a pay-for.
  • The Democrat argument that this amendment is paid for is factually dubious on several levels.
  • Although a CBO formal score for the Casey/Brown amendment is not yet available, a six month extension of COBRA subsidies initially included in Section 511 of the House version of H.R. 4213 was scored as costing $6.8 billion.  When included in the President’s Fiscal Year 2011 budget, the Joint Committee on Taxation scored elimination of the advance EITC as saving only $1.2 billion over ten years.
  • While scaling back the subsidy length for newly eligible individuals (i.e. those laid off after June 1, 2010) may reduce the apparent cost of the amendment, many may view this as a budgetary gimmick.  Section 1010 of the Department of Defense Appropriations Act (P.L. 111-118) extended the subsidies from the nine months originally included in the “stimulus” to 15 months.  Given that Democrats have already extended the length of COBRA subsidies once this year, some may question whether the majority will try to lengthen the subsidy duration again at the first possible opportunity.
  • The amendment sponsors are citing an interim report on the COBRA subsidy recently released by the Treasury Department as evidence that the subsidy extension will cost less than the CBO projects.  The Treasury report stated that the program has cost $2.1 billion to date, and served just under 2.2 million households.  However, at the time of the “stimulus” JCT estimated that the subsidy would benefit 7 million individuals.  Even after accounting for an average American household size of 2.59 persons, the 2.2 million households cited in the Treasury report mean that about 5.6 million individuals received some benefit from the subsidies—or nearly 20% fewer than the 7 million projected.  Some may view the smaller cost—because fewer individuals than expected participated in the program in the first place—as evidence that the program is ineffective and should be discontinued, NOT extended.
  • Democrats themselves have raised concerns about the repeated extensions of COBRA subsidies, with little oversight or evaluation by Congress as to the program’s merits—one reason why a six-month COBRA extension was not included in the House-passed extender package.  Rep. Stephanie Herseth Sandlin (D-SD) told CongressDaily last month: “COBRA, that’s never had a hearing. That’s never had a congressional hearing, yet some in our Caucus think that automatically should be extended ad infinitum…There are constituents in our districts who are working and don’t get any help paying for their health insurance and yet we’re going to subsidize COBRA for people who may lose their jobs next month. … So we think some of these provisions warrant far more scrutiny than they’ve received to date.”

Murray Amendment (#3356) and Medicare Beneficiary Assignment

Before the vote on the Murray amendment regarding TANF and summer employment, I wanted to forward around this summary.  You will note that the amendment uses a Medicare pay-for – “intelligent assignment” of low-income beneficiaries in Part D plans – that MedPAC commissioners have previously criticized as potentially unworkable, and may result in greater “churning” of beneficiaries if subsidy-eligible seniors get re-assigned to a new plan every year.

Despite the pay-for referred to above, CBO has scored this amendment as increasing the deficit by $1.6 billion over six years (2010-2015).  A PAYGO point of order therefore lies against the bill, and is expected to be voted on.


Senator Murray has revised her amendment to the Baucus Amendment (#3336), to extend the TANF Emergency Fund through 2011 and provide funding for summer employment for youth.  Please note that the only change to the summer youth employment provisions is a change to the cost, from $1.5 billion to $1.3 billion.

The Senate will vote on this amendment as part of a series of four votes at 11:30am tomorrow.

Background on new provisions

TANF Emergency Fund

The amendment provides an additional $1.3 billion for the Temporary Assistance for Needy Families (TANF) Emergency Contingency Fund and extends it until March 2011, with payments allowable through 2012.  It also includes “employment services” as a use of funds.  These services are broadly defined as “services designed to help an individual begin, remain, or advance in employment.”

The American Recovery and Reinvestment Act of 2009 (ARRA, P.L. 111-5) created a $5 billion Emergency Contingency Fund within the TANF block grant. The fund helps states pay for additional costs of providing economic aid to families during the current economic downturn for FY2009 and FY2010.

Republicans opposed including the Emergency Contingency Fund in ARRA because it undermined key principles of welfare reform by rewarding states that increased their caseload and exempting new beneficiaries from having to engage in meaningful work related activities.

Intelligent Assignment

The amendment includes language requiring the Centers for Medicare and Medicaid Services (CMS) to provide “intelligent assignment” of low-income subsidy (LIS) beneficiaries’ Part D prescription drug plans.  Under current law, CMS is required to randomly auto-assign LIS beneficiaries to plans, and those beneficiaries can switch from their auto-assigned plan if they prefer another option.  The amendment would require CMS to assign LIS beneficiaries to plans that are in the lowest quartile of costs and “reasonably meets the needs of such Part D eligible individuals as a group.”  This amendment would save approximately $1.6 billion as a result of the requirement that CMS choose low-cost plans for beneficiaries.

Two potential concerns have been raised by this amendment.  First, it would almost by definition appear difficult to provide “intelligent assignment” to Part D beneficiaries “as a group.”  At a 2008 Medicare Payment Advisory Committee (MedPAC) meeting, commissioners noted that “the idea that you could do this in some kind of systematic way for beneficiaries in an open season kind of framework is kind of mind-boggling,” given each beneficiary’s unique medical conditions.

Second, while the amendment is designed to place LIS beneficiaries in low-cost plans, low-income subsidy beneficiaries tend to have higher health and prescription drug costs than the Medicare population as a whole.  As a result, any “low-cost plan” with high LIS enrollment could experience a quick spike in costs in future years, thus losing its “low-cost” designation.  This scenario could create a large amount of “churn,” whereby beneficiaries are switched from plan to plan every year in order to meet the “low-cost” threshold, creating beneficiary confusion and potentially harming access in the process.


We are still waiting for a score on this amendment and will update you as soon as we have it.   The amendment is paid for by eliminating the of the advance refundability of the Earned Income Tax Credit (EITC), found in section 3507 of the Internal Revenue Code.  This proposal was put forward by the Administration in their FY11 budget.  Under current law, recipients of the earned income credit have the option of receiving “advanced payments” of the EITC, delivered through lower employer withholdings in their paychecks, instead of receiving a single payment from the government once a year.  The Administration believes this has been an ineffective program, with at most 3 percent of eligible individuals participating.  In addition, the Administration cites recent research that shows “evidence of significant non-compliance by employers and workers,” meaning fraud.  The Administration’s proposal was scored by Treasury as raising $760 billion, all on the outlay side.