The Congressional Budget Office (CBO) launched its latest mistaken Obamacare-related estimate this week, predicting that a House-passed bill to repeal and replace the embattled law would lead to 23 million more uninsured people by 2026.
Although CBO itself acknowledges that this latest prediction is “especially uncertain,” one thing is especially certain: CBO is wrong.
We explain why at greater length in a new Galen Institute paper that can be found here.
Estimating the coverage effects of big and complex legislation is treacherous work and CBO can hardly be faulted for getting it wrong every time. But the agency’s errors are not only massive – one of their predictions of 2016 exchange-based enrollment missed by 140% – they also are consistent: the agency has repeatedly over-estimated the number of people who would get insurance through the Obamacare exchanges.
Undaunted by failure and unschooled by experience, CBO soldiers on, fearlessly predicting that millions will flock to the exchanges any day now. Agency estimates of 2016 exchange enrollment ranged from 12 to 24 million. The reality: 10 million had exchanged-based coverage as of December 2016.
No matter, CBO forecasts that this year, 15 million people will sign up for Obamacare policies. The actual number as of March 31: 10.4 million, little changed from December. But there’s always next year when, CBO assures us, enrollment will abruptly skyrocket to 18 million.
CBO measures the House-passed bill against this imaginary baseline and finds it wanting. The agency believes that 8 million fewer people will have individual health insurance coverage in 2018 if that bill becomes law.
But here’s our prediction: even if the bill doesn’t become law, 2018 enrollment may fall 8 million short of the CBO prediction. The presumed 8 million reduction in the number of people with individual insurance looks more like an artifact of inflated coverage estimates than the consequence of Obamacare repeal.
One reason CBO gets it so wrong so consistently is its fervent belief that the individual mandate has motivated millions to enroll in coverage. Because the House bill repeals the mandate, CBO believes that the number of uninsured will increase by 14 million next year. We have already looked into the implausibility of CBO’s assumption that 8 million fewer people will have individual insurance in 2018. CBO thinks they will be joined by 2 million people who will drop out of their employer-based plans, and by 4 million people who will torch their Medicaid cards.
CBO isn’t predicting that 14 million will lose coverage next year if the House bill passes, despite how their estimate is habitually mischaracterized; CBO predicts people will cancel their insurance or not sign up at all. They will be covered only if the government continues to threaten them with tax penalties for being uninsured. With the penalties gone, CBO believes, they will choose to be uninsured.
It is plausible that some will drop their job-based insurance in the absence of a mandate. People who get insurance in the workplace earn enough to incur tax liability and the government requires employers to report their uninsured workers to the IRS. Freed from IRS scrutiny, some workers may drop coverage.
But most Medicaid recipients earn little or nothing and few owe the government anything at all in taxes. Most are among the 43.5% of Americans who pay no federal income taxes, and are likely among the 16% of Americans who don’t even bother filing. It seems a bit far-fetched that millions struggling to get by will refuse free health insurance if Congress scuttled the tax on the uninsured.
CBO’s belief in the power of the individual mandate is misplaced. The evidence suggests that it has not induced many people to enroll in health coverage, despite the hopes of its architects. The IRS reports that in the 2015 tax year, 6.5 million uninsured filers paid the tax penalty, 12.7 million got an exemption and additional 4.2 million people simply ignored the penalty. They left line 61 on their form 1040 blank, refusing to tell the government whether or not they had insurance. The IRS processed their forms and sent them their refund checks. In all, that is a total of 23.4 million uninsured people – out of an estimated 28.8 million uninsured – who either paid, avoided or ignored the penalty. That hardly suggests that the mandate has worked.
CBO also forecasts big long-term reductions in the number of Medicaid recipients in the next decade. If the House bill were passed, CBO warns, 14 million fewer people will be enrolled in the program in 2026 than current projections.
The media typically report that this is because the bill would repeal Obamacare’s Medicaid expansion, which allows states to open their programs to nondisabled, non-elderly adults. That is false. The House bill retains the expansion for the 31 states and the District of Columbia that have enlarged their programs and retains the option for the rest of the states.
It does, however, end the Obamacare requirement that the federal government pay 90-100% of the medical costs for this new population, much more than the 50-75% of the costs it pays for the frail elderly, impoverished children, people with disabilities and pregnant women. The House bill ends the preferential financial treatment of the expansion population; it does not end the expansion.
The bill also caps federal Medicaid payments at current rates, adjusted for medical inflation. Instead of growing at an average annual per capita rate of 4.4% as under current law, we estimate that per capita federal Medicaid payments would grow by a little more than 4.2% annually. That incremental difference, combined with repeal of the individual mandate and equalization of federal reimbursement rates, CBO believes, will result in 14 million fewer Medicaid enrollees by the middle of the next decade. Hardly a plausible estimate.
CBO was wrong about coverage gains attributable to Obamacare and it is likely wrong about coverage losses attributable to Obamacare repeal. Lawmakers should recognize the agency’s limitations and debate competing policy proposals on their merits. Those who believe that taxing the uninsured is a good idea should explain why. Those who think the federal government should provide preferential Medicaid reimbursement rates for medical services provided to non-disabled, childless adults should defend the preferences. But these issues should not be decided by an appeal to CBO’s highly unreliable coverage estimates.
Bludgeoning the opposition with CBO numbers does not advance debate. It silences it.
