Health Affairs Blog
First, the Centers for Disease Control and Prevention (CDC) released a report on Problems Paying Medical Bills Among Persons Age 65: Early Release of Estimates From the National Health Interview Survey, 2011-June 2015. There has been much discussion lately about the high and rising cost-sharing obligations imposed on consumers both by ACA marketplace plans and by employee health benefit plans. What the CDC report documents, however, is that the number and percentage of adults under 65 having problems paying medical bills has declined steadily since 2011 and has dropped further in the first six months of 2016.
In 2011, the National Health Interview Survey (NHIS) estimated that 56.5 million persons under the age of 65 in the United States—21.3 percent of the total—were in families (which may have included one individual) that had problems paying medical bills. By the first six months of 2015, that number had fallen to 44.5 million — 16.5 percent.
The percentage of children who lived in families that had problems paying their medical bills dropped from 23.2 percent in 2011 to 18.1 percent in 2015. The percent of adults under age 65 having problems paying medical bills dropped from 20.6 percent in 2011 to 15.9 percent in 2015. The percentage of persons under age 65 having problems paying medical bills dropped among the uninsured, among those with public and private insurance, and among all racial and ethnic groups tracked by the NHIS.
The decline in problems paying medical bills between 2011 and the first six months of 2015 was steeper among persons under age 65 with incomes below 100 percent of the poverty level (from 32.1 percent to 24.5 percent) and those with incomes between 100 percent and 200 percent of the poverty level (from 34.6 to 27.1 percent) than it was for persons with incomes above 200 percent of the poverty level (from 15.2 percent to 12.2 percent).
For the poor and near-poor the drop off was sharpest between 2013 and 2015 when the premium tax credits and cost sharing reduction payments—and in many states, the Medicaid expansion—became available. The percent of persons under age 65 having problems paying medical bills fell from 29.3 to 24.5 percent for the poor (under 100 percent of poverty) and from 32.9 to 27.1 percent for the near-poor (100 to 200 percent of poverty) during those years. It would seem that the coverage expansions are having their desired effect of making health care more affordable, particularly for poor Americans.
CBO Projects ACA Will Cause Labor Supply Decline
The second document is a working paper released by the Congressional Budget Office (CBO) describing how it estimates the effects of the ACA on the labor markets. The CBO estimates that by 2025 the ACA will reduce the labor supply, measured as total compensation paid to workers, by 0.86 percent. Because the ACA affects people who earn lower wages more than those who earn higher wages, this represents a 1.7 percent decline in total hours worked. This is equal to 2 million fewer full-time-equivalent workers by 2020.
The estimate is based on the CBO’s projection of the combined effects that federal programs created under and taxes imposed by the ACA will have on labor markets. The largest projected effect will be caused by the ACA’s coverage expansions, which will account for a 0.65 percent reduction in total compensation. The amount of premium tax credits, for example, fall off steadily between 133 and 400 percent of poverty. A four-person family whose income rose from 250 to 300 percent of poverty would experience a decline in premium tax credit assistance equal to a marginal tax rate of 16.8 percent.
While premium tax credits decrease gradually, cost-sharing reduction payments phase out in steps. A family of four with an income of 150 percent of poverty would receive cost-sharing reductions worth on average $3,038, but if their income would increase $1 the value of the subsidy would drop by $882. Combining the effect of the premium tax credits and the cost-sharing subsidies, a family moving from 175 to 225 percent of poverty would experience an effective marginal tax rate of 30.1 percent. Because premiums are projected to increase more rapidly than inflation generally between now and 2025, the marginal tax rate will be half again larger in 2025.
The CBO estimates that the disincentives that this high marginal tax rate poses will mean that hours worked among the subsidized population will be 12 percent less in 2025 than it would otherwise have been, but since this population only accounts for 3 percent of total earnings, this translates into a 0.35 percent reduction in aggregate labor supply, measured by total compensation. Reduction in months of employment by individuals who work only part of the year will further reduce economy-wide earnings by .08 percent, thus the subsidies will in total account for .43 percent of the reduction in labor supply.
Two other provisions, according to the CBO, will also reduce the labor supply. First, the insurance reforms prohibiting health status underwriting will allow individuals between 55 and 64 to purchase affordable insurance and thus to retire early. The CBO estimates that by 2025, 225,000 workers will have retired early, reducing total workforce compensation by .17 percent. The Medicaid expansions will also reduce workforce participation as some individuals may reduce their hours to become eligible for Medicaid or cease working because they are eligible for Medicaid, but others who currently eligible for Medicaid will work more because they can now be covered by Medicaid even though they earn more. The net effect of this will be to reduce labor supply measured by compensation by .03 percent.
The taxes imposed by the ACA will also, the CBO believes, reduce labor supply. The Medicare surtax imposed on high-income workers will increase the marginal tax rate of higher-income workers from 44 to 45 percent, but since the 8 percent of workers subject to the tax earn 38 percent of the economy’s total earnings, this will reduce labor supply by 0.12 percent. The CBO projects that the employer mandate penalty will reduce labor supply by .06 percent, but not because it will cause employers to move from full-time to part-time employees, but rather because employers will reduce compensation to pay for the tax. The high premium (Cadillac plan) excise tax will raise the cost of labor slightly, reducing the labor supply by 0.03 percent. Finally, the individual mandate tax penalty will reduce the labor supply, the CBO estimates by 0.01 percent. The CBO acknowledges that its estimates are very uncertain, and that the reduction of labor supply could be as little as 0.4 percent or as much as 1.3 percent.
In reading the analysis, questions that occurred to me, admittedly a non-economist, included why there is no accounting for the increased employment of health care workers which surely must accompany the coverage expansions; will the individual mandate not cause some otherwise unemployed or underemployed people to seek full-time employment to get employer-based coverage in order to avoid the mandate’s penalty; and how we should regard people who because of the ACA are able to leave dead-end jobs that they tolerated merely for health benefits? Might some of these people, for example, launch their own businesses, which could pay less but be more productive? What about those who take advantage of the ACA to leave low-paying jobs to take care of children or aged parents without compensation?
The CBO report does not say that the ACA will kill jobs. What it does say is that the ACA gives some people, for whatever reason, an opportunity to work less. And it does illustrate the problems that result when an income tax system is used to expand access for health care.
New Enrollment Snapshot
On December 9, 2015, CMS released its fifth weekly snapshot of enrollments through the federally facilitated marketplace (FFM), covering the week of November 29 to December 5, 2015. Plan selection during the fifth week was strong, with a total of 804,338 individuals selecting plans through the FFM and 1,065,877 submitting applications for coverage. Nearly 2.8 million consumers visited healthcare.gov and over one million called the call center. The percentage of new enrollees as compared to renewing enrollees increased to 37 percent, with over one million new enrollees now having selected plans for 2016. CMS also reports that one quarter of 2015 enrollees have now actively reenrolled for 2016.
December 15, 2015 is the last day on which most applicants can enroll for coverage effective as of January 1, 2015. If the past two years are any indication, enrollment will ramp up substantially between the closing date of today’s report (December 5) and December 15. Between December 15 and January 1, most enrollees who had 2015 coverage and have not returned to the marketplace will be auto-reenrolled. This will significantly increase enrollment, although the January 1 enrollment total will probably not be released until early January. Open enrollment for 2016 will continue through January 31, 2016, and total enrollment will continue to grow until that time.
Upgrading The ‘Back-End’ Of The Marketplaces: Policy-Based Payment
For the past two years, the “back-end” of the marketplaces have been functioning on a jerry-built manual payment process: marketplace insurers have submitted enrollment reports and payment requests every month, with continuous adjustments to previous month requests, and CMS has paid premium tax credits and cost-sharing reduction payments (and collected marketplace user fees) based on these manual reports. As of January 216, CMS will finally be implementing an automated payment approach, called policy-based payments. This system will be based on the exchange between federally facilitated marketplaces (FFMs) and insurers of individual-enrollee-specific IC 834 enrollment and payment transaction data. These data will be updated daily.
On December 4, 2015, CMS posted at its REGTAP.info website a bulletin explaining how the transition to policy-based payments will occur. The bulletin sets out the criteria that CMS will use to determine whether insurers are ready to receive policy-based payments. It describes how the transition to policy-based payments will occur in early 2016. It also establishes a policy under which CMS will partially withhold payments from insurers whose data is not deemed ready for a timely transition to policy-based payments.
The bulletin describes the criteria that CMS intends to use to evaluate whether FFM participating insurers are ready for the transition. First, an insurer must have met CMS standards for the activation of IC 834 individual-specific transactions by December 1, 2015. Second, an insurer must have successfully submitted reconciliation files that passed basic validation checks for the most recent month, or if the most recent month did not pass those checks, for the two prior months. Third, an insurer’s most recently submitted manual payment submission cannot vary by more than 10 percent from the policy-based payments to which it would be entitled. Currently 93 percent of enrollees are enrolled with an insurer that is within the 10 percent variance threshold.
The readiness of new insurers entering the FFM will be evaluated based on the readiness of a larger insurer group to which they are related. If an insurer is entirely new to the FFM, it will be subject to partial withholding until its readiness can be evaluated.
Insurers that have not demonstrated readiness for policy-based payments will be subjected to program integrity withholding until they demonstrate readiness. Insurers will continue to submit manual payment requests. Insurers that have not demonstrated readiness will be paid based on those manual requests.
Beginning in April, 2016, however, CMS will calculate also an upper bound payment amount for each insurer based on all non-cancelled FFM enrollments. Payments to insurers that are not ready to participate in the policy-based payment system will, starting in April, amount to 75 percent of the lesser of their manually submitted payment amount or the calculated upper bound amount. As of July, the payments to non-ready insurers will drop to 50 percent of this amount. The difference between the amount claimed and these amounts will be withheld. When an insurer is ready to begin submitting and accepting policy-based payments, the withheld amounts will be paid in full to the extent payments are in fact due.
For the first three months of 2016, policy-based payments will be adjusted for all insurers, ready or non-ready to match payments requested through the manual process. As of April, CMS will begin actual policy-based payments to insurers that are ready to participate in the process. Payments will from that date be policy-based unless there is an extreme discrepancy (above 25 percent) between the policy-based amounts and manual payment requests. CMS will continue to reconcile 2015 data using its manual monthly restatement process for six months into 2016 to allow resolution of outstanding claims and appeals and will then audit insurer data to ensure the accuracy of 2015 payments and close out 2015.