UPDATE: I’ve added new CBO coverage estimates and a few points of clarification below.
UPDATE II: And a correction appended below!
Josh Barro critiques AEI health policy scholar Tom Miller’s critique of the Republican approach to health-system reform:
Though Miller spends 6,000 words laying out his health policy vision, he never explicitly notes two facts about his approach: It would cost a lot of taxpayer money to implement, and it wouldn’t achieve universal health coverage. The former makes the approach a nonstarter for so many conservatives that no Republican Congress would ever implement it. The latter makes it unsaleable to the broader electorate.
And Josh makes a reasonable point regarding the political dynamics that have stymied GOP health-system reform efforts:
The failure of conservative politicians and think tanks to advance serious alternatives on health policy reflects their complete lack of interest in fixing health policy: They don’t want to spend money, they don’t want to change Medicare in ways that affect elderly Republican base voters, they don’t want to cut the incomes of Republican-voting doctors, and they don’t want to change the (often overly expensive) health coverage situations of the overwhelmingly insured Republican electorate.
Where Josh and I disagree is in our assessment of the attractiveness and seriousness of some of the ACA alternatives that have emerged out of the think tank world.
But first, it is important to keep in mind that while the goal of the Affordable Care Act is to achieve universal coverage, the Congressional Budget Office projects that it will fall far short of that goal. In March of the last year, the CBO revised its projections:
CBO and JCT’s projections of health insurance coverage have also changed since last March. Fewer people are now expected to obtain health insurance coverage from their employer or in insurance exchanges; more are now expected to obtain coverage from Medicaid or CHIP or from nongroup or other sources. More are expected to be uninsured. The extent of the changes varies from year to year, but in 2016, for example, the ACA is now estimated to reduce the number of people receiving health insurance coverage through an employer by an additional 4 million enrollees relative to the March 2011 projections. In that year, CBO and JCT now estimate that there will be 2 million fewer enrollees in insurance exchanges. In the other direction, CBO and JCT now estimate that, in 2016, the ACA will increase enrollment in Medicaid and CHIP slightly more than previously estimated (but considerably more in 2014 and 2015), and it will reduce the number of people with nongroup or other coverage by 3 million less and the number of uninsured people by 2 million less than previously estimated.
Compared with prior law, the ACA is now estimated by CBO and JCT to reduce the number of nonelderly people without health insurance coverage by 30 million to 33 million in 2016 and subsequent years, leaving 26 million to 27 million nonelderly residents uninsured in those years (see Table 3, at the end of this report). The share of legal nonelderly residents with insurance is projected to rise from 82 percent in 2012 to 93 percent by 2022. According to the current estimates, from 2016 on, between 20 million and 23 million people will receive coverage through the new insurance exchanges, and 16 million to 17 million people will be enrolled in Medicaid and CHIP. Also, 3 million to 5 million fewer people will have coverage through an employer compared with the number under prior law. [Emphasis added]
The CBO released new coverage estimates in May 2013, which estimates that the ACA will reduce the number of nonelderly people without health insurance coverage by 25 million in 2016 (31 million under the ACA, 56 million without it), and the number of uninsured noneldery people will remain in the 30-31 million range until the end of the projection in 2023. That is, the CBO revised the amount of coverage expansion under the ACA downward since March 2012. Also, it should go without saying that much of the residual uninsured population will consist of unauthorized immigrants. One interesting conceptual question, which has gone largely unexplored, is whether the uninsured will be worse off relative to the insured in the ACA world than under the pre-ACA status quo. Given that the uninsured will remain quite large, this seems like an important issue for policymakers.
In April, Douglas Holtz-Eakin and James Capretta analyzed the potential cost and coverage expansion effect of an ACA replacement plan released by Capretta in December of last year using the American Action Forum’s Health Economic Policy Simulation System (HEPSS). Critics of Capretta’s approach will take this analysis with a generous heaping of salt, but it finds a strong coverage expansion effect due to its reliance on default insurance plans:
The assignment of persons eligible for the tax credit who fail to sign up with a plan to a default insurance carrier would substantially reduce the number of uninsured Americans. According to the HEPSS estimates (discussed further below) and as shown in Table 1, the number of uninsured in the U.S. would fall from about 53 million today (under the age of 65) to about 29 million in 2015. That compares very favorably to the CBO estimate of the uninsured under the PPACA, which is 37 million in 2015.
For comparison, the CBO projects that the number of uninsured under the pre-ACA status quo would be 57 million in 2015, and so the ACA would reduce the number of uninsured by 20 million while the ACA replacement plan would reduce the number of uninsured by 28 million, though the actuarial value of the insurance under the ACA replacement plan would generally be less generous.
While the ACA relies on Medicaid expansion and the individual mandate to reduce the number of uninsured Americans, Capretta’s plan offers all individuals without access to large employer health plans refundable tax credits and it makes use of default insurance plans to “nudge” the hard-to-reach into coverage:
[E]ven with a widely available tax credit for insurance, some portion of the population would likely still go uninsured, for a variety of reasons (this is often the case with other programs like Medicaid, in which eligible individuals fail to sign up in large numbers). There are a couple of reasons for this phenomenon: some people are hard to reach in terms of public information campaigns, and others tend not to make significant changes to their life arrangements without a clear need and direct, personal intervention.
It is possible, however, to boost insurance coverage among the hard-to-reach population without resorting to coercive mandates or requirements. Under the framework proposed here, this could be done through what might be called “default insurance.” States would be responsible for designating several insurance plans as default options to which persons who are eligible for a refundable tax credit would be assigned (on a random basis) if they failed to sign up for coverage on their own.
The key to making this concept work is that the premiums for default insurance would need to be set to the value of the tax credit so that persons who were assigned to such plans would not be charged any additional premium. And to keep the premiums equal to the credits, the insurance plans must be given the authority to set their up-front deductibles accordingly so that the cost of the coverage does not exceed the federal credit.
These default plans would, however, be catastrophic plans with larger-than-normal deductibles. But low-income individuals would be shielded from the cost of routine medical expenditures by Capretta’s reformed Medicaid program.
Josh suggests that Tom Miller intends to strip healthy low-income adults of insurance coverage. First, Josh quotes a passage from Miller’s article:
Income-based financial support from taxpayers to address health-care vulnerabilities should distinguish better between short-term emergency assistance and longer-term arrangements for those suffering from chronic or permanent conditions. Reformers need to have in mind a socially conscious “floor” for the former — both in dollar amounts and eligibility duration for catastrophic coverage — while avoiding the danger of setting a ceiling on personal responsibility, economic initiative, and social mobility. Short-term assistance can provide stability and opportunity to rebound from health-related misfortune, but it should not dull the incentives to regain more self-sufficiency after the crisis is over.
And he translates it as follows:
In other words: Reduce Medicaid for healthy adults to “catastrophic” coverage covering major health events instead of the broad coverage it is today, and impose a time limit on its use, so that healthy people who end up long-term unemployed with almost no income also end up uninsured.
One problem is that Josh leaves out the preceding paragraph, which reads as follows:
Finally, reforming the health-care safety net for low-income and medically vulnerable Americans involves more than just reshaping the future financial structure of Medicaid and preserving the sustainable core of Medicare. One of the overriding principles of reform should be to distinguish between societal commitments to basic income support (more transparently called “welfare” in previous decades) and promises to ensure access to essential health-care services for people facing particularly challenging medical conditions. We should be more generous in targeting persistent, condition-based problems and vulnerabilities. Subsidizing access to health-care choices in a competitive marketplace (through high-risk pools, for instance) is preferable to directly providing services through government channels, although markets alone do not and cannot meet every medical need. [Emphasis added]
That is, the passage Josh is criticizing is not actually about Medicaid, but rather about the disabled* — both those who are temporarily disabled (and in need of short-term emergency assistance) and the permanently disabled (those suffering from chronic or permanent conditions). In fairness, this passage could have been written more clearly.
[UPDATE: I had a brief conversation with Miller and he explained that the passage in question was gently suggesting more thinking about new parameters. That is, we need to reformulate the triage for who most needs to be covered — including who needs to be covered more generously, like the disabled and those with chronic or permanent conditions — in light of resource constraints. His concern is that in focusing on Medicaid as a vehicle for coverage expansion, we wind up offering more people coverage that only appears to be comprehensive while actually delivering a shallow menu of services that doesn’t adequately serve the needs of those who really need the most help. His intention was not to offer specific, explicit “time limits” at this point, but he wanted to introduce the concept of some outer bounds for the low-income healthy.]
Miller’s article doesn’t pretend to offer a comprehensive ACA alternative, but elsewhere he has endorsed refundable, fixed-dollar credits for people who don’t have access to job-based coverage, and these refundable, fixed-dollar credits will not be subject to time limits. These refundable, fixed-dollar credits are the foundation of Capretta’s default insurance approach. My impression is that conservative reformers like Miller and Capretta aim to transform Medicare into a premium support program designed to supplement the catastrophic coverage purchased through the refundable, fixed-dollar credit, as suggested in the following passage from “The Defined Contribution Route to Health Care Choice and Competition“:
A move to replace both traditional Medicaid assistance and the tax preference for ESI with defined contribution payments would open up new possibilities for explicit and beneficial coordination between the Medicaid program and the coverage normally offered to working-age Americans. For example, the financial contributions from taxpayers for health insurance coverage could be restructured so that all working-age Americans and their families receive a baseline amount of assistance equal to their proportionate share of the value of eliminating the current exclusion of employer-paid premiums from federal income and payroll taxes. In most past formulations of this proposal, a fixed, refundable tax credit would be paid to American households. That could be explicitly amended to include those who otherwise would also be eligible for Medicaid. Medicaid could then supplement this financing mechanism for those with especially low incomes who need additional support beyond the base amount (from the reallocated funds of the tax exclusion) to pay for more of their remaining premiums and cost sharing. These add-on Medicaid payments could then be phased down gradually in steps to avoid large disincentives for the beneficiaries to climb the wage and income ladder on their own.
The goal is not to strip healthy low-income adults of Medicaid and to leave them without coverage, as Josh suggests.
[UPDATE: I think it’s reasonable to say that Miller wants to transition healthy low-income adults off of Medicaid as they work longer hours and earn higher wages. Miller believes that welfare reform offers a worthwhile model, but he acknowledges that efforts to apply its lessons to Medicaid face a steep learning curve and that any such effort needs to be sensitive to those with persistent, chronic medical conditions.]
The real issue is this: if Caprettacare achieves as much or more coverage expansion as Obamacare yet it costs substantially less, what are we getting from the additional spending on the ACA? Comprehensiveness is one reasonable answer. Coverage expansion under Caprettacare largely takes the form of low-cost catastrophic insurance designed to protect households against crippling medical expenditures, not to greatly reduce routine medical expenditures. But then again, if the ACA causes more of an economic drag, due to the high taxes necessitated by its high costs (and its regulatory burdens), then perhaps the benefits aren’t worth the costs.
Fortunately for the Obama administration and its Democratic allies, we’re not actually having a debate between Caprettacare and Obamacare. Rather, we are having a debate between Obamacare and a scattered series of proposals, few of which will deliver much in the way of coverage expansion. The reason is, as Josh argues, that Caprettacare will cost something, albeit substantially less than Obamacare, and all too many GOP lawmakers favor nothing over something, even though nothing is not a realistic (or attractive) option.
UPDATE: The basic issue seems to be whether or not one supports refundable, fixed-dollar tax credits for all who don’t have access to job-based coverage. This policy idea entails many complexities, e.g., we have to think through income-based adjustment to the base tax subsidy, etc. But refundable, fixed-dollar credits provide a foundation for coverage atop which the Medicaid program can serve as a supplement. How this supplement will work — will it provide a more or less permanent cushion for less-skilled workers earning persistently low incomes or will it be designed as transitional assistance for all but the disabled and those with chronic or permanent conditions — is going to subject to debate. My view is that refundable, fixed-dollar tax credits for all who don’t have access to job-based coverage is very important, and this position is embraced by very few Republican lawmakers. This represents a serious failure, substantively and politically.