Love it or hate it, the Affordable Care Act is unprecedented in size, scope, complexity, and uncertainty.  To project its impact, numerous policy, economic, competitive, and behavioral factors must be considered and assumptions made.  From an analytical perspective, the ACA is a simultaneous, non-linear equation from hell.

A number of government organizations, advocacy groups, and researchers have published estimates of how the Affordable Care Act (ACA) will affects health insurance premiums, enrollment, and health care spending. Reports agree that the health reform law’s impact will be broad and significant, but the details are less clear.  Also, the estimates range from rosey scenarios published by liberal thinks tanks funded by foundations to gloomy forecasts from conservative research shops.  Because ACA is so unprecedented and wrapped in uncertainty, researchers can do reasonably rigorous work and still reach wildly different conclusions by using different models, techniques, and assumptions.

Health reform will reduce, by perhaps 40-50 percent, the number of uninsured Americans by expanding Medicaid, providing subsidized coverage through Health Insurance Exchanges (HIX), and requiring almost all Americans to purchase federally approved health insurance. The rates at which people eligible for Medicaid or federal insurance subsidies participate in those programs, called take-up rates, are largely unknown and, of course, greatly influence enrollment estimates. Medicaid take-up estimates, for example, have ranged from 45 percent to 95 percent. The Supreme Court’s decision allowing states to choose whether to expand Medicaid, in NFIB v. Sebelius, adds another layer of complexity to Medicaid enrollment estimates.  Meanwhile, ACA policies are expected to cause at least several million Americans to lose employer-sponsored or individual coverage.

Estimates of rates at which people will switch from private non-group insurance or employer-sponsored insurance (ESI) – called crowd-out – are similarly widespread. Another unknown is how many people who are currently eligible for Medicaid but not enrolled will enter the program, because of better outreach and more awareness about their health insurance options. The woodwork effect, as it’s called, is of significant concern to states because the federal government will reimburse state Medicaid programs much less for new enrollees who were eligible before the ACA Medicaid expansion. Federal Medical Assistance Percentages (FMAP) for spending on the new Medicaid enrollees under the ACA expansion will be 100 percent initially, and will be no less than 90 percent in subsequent years.  However, for individual who are already eligible but not yet enrolled, the federal match rate is as low as 50 percent.

Characteristics of New Enrollees in Medicaid and HIX:

Research on the demographics of new enrollees has been less common and less certain than research on their numbers. In general, the newly insured population is expected to have less income, be less healthy, be more diverse, and have less education than the currently insured population.

The health and demographic profiles of new Medicaid and exchange enrollees are at least as important as how many people will participate. If only the least healthy people enroll in Qualified Health Plans (QHP) on the exchanges and non-exchange health plans, the cost of those plans will rise and further discourage healthier people from joining – a serious, costly problem called adverse selection.

QHPs applying to participate in the Health Insurance Exchanges must submit rates and benefit designs for approval several months before enrollment in the individual market exchanges begins on October 1, 2013. Without reliable information about who will enroll and what they will cost to insure, health plans will have a difficult time knowing how much to charge in premiums.  This is made more difficult by the radically new federal regulatory framework for health coverage.

New Predictions on Post-ACA Premiums and Enrollment:

A report from the Society of Actuaries is one of two new estimates with ominous predictions about the cost of health insurance under the ACA.  See the health reform section of this blog to see my post on the other recent report from Milliman.  Actuarial techniques are the same methods health plans will use to set premiums and therefore are likely to generate reasonably sound estimates.

The Society of Actuaries drew upon the expertise of study authors Randy Haught and John Ahrens of Optum, the parent company of The Lewin Group. The study also used the Lewin Group’s Health Benefits Simulation Model (HBSM) for its projections.

Here’s a quick summary of the report’s findings, before going into more detail:

1. After three years of exchanges and insurer restrictions, the percentage of uninsured nationally will decrease from 16.6 percent to between 6.8 and 6.6 percent, compared to pre-ACA projections.

This estimate assumes all states will opt in to the ACA Medicaid expansion. The report also includes another estimate that assumes no states do so, which is one way analysts have dealt with the great deal of uncertainty about which states will choose to expand Medicaid to people with incomes less than 138 percent of the federal poverty level (FPL).

Enrollment projections under the full-expansion scenario are similar to the Congressional Budget Office (CBO) analysis in March 2012, before NFIB v. Sebelius. Both the CBO and the actuaries expect 31 million fewer uninsured, if all states were to expand Medicaid. The CBO estimated 5 million people with employer-sponsored insurance (ESI) to lose that coverage, whereas the actuaries estimated 2 million people would lose ESI, likely because of crowd out.

2. Under the ACA, the individual non-group market will grow 115 percent, from 11.9 million to 25.6 million lives; 80 percent of that enrollment will be in the Exchanges.

Just as take-up rates for newly eligible Medicaid beneficiaries is unknown, take-up rates for subsidized coverage on the exchanges is hard to estimate. The Society of Actuaries report assumes that all people who are eligible for subsidies will enroll in non-group coverage through the exchanges, giving a total exchange enrollment of 21 million people. That figure is similar to estimates from the CBO (23 million) and the Centers for Medicare and Medicaid Services (CMS) Actuary (23.4 million).

Again, that estimate assumes all states expand Medicaid. In states that choose not to expand, some people will instead be covered in the health insurance exchanges. In the CBO’s July 2012 estimate, which accounted for states choosing not to expand Medicaid, expected exchange enrollment increased to 25 million.

Further complicating the issue is the fact that some states have proposed using Medicaid expansion money to allow people to buy private coverage through the exchanges. Department of Health and Human Services (HHS) Secretary Kathleen Sebelius recently supported Arkansas’ plan to do that (although details remain sketchy), and Governors in other states have voiced interest in a similar model.

3. The non-group cost per member per month will increase 32 percent under ACA, compared to pre-ACA projections.

Three main reasons contribute to higher health costs and spending after 2014, when the ACA coverage expansion provisions take effect. First, high-risk pool members will join the rest of the non-group market. Second, employers who choose not to offer employees health insurance will push people who previously had employer-sponsored insurance into the non-group health insurance market. And third, previously uninsured people who become newly insured under the ACA will have higher morbidity rates, according to the report’s estimates.

One important cost-related assumption the study authors made was that people who become newly insured will use services at the same rate as people who are currently insured. In other words, the study assumes there is no pent-up demand for services among the uninsured. Other studies of this issue have assumed there would be some increase in the rate of health care use for people who go from being uninsured to insured. But the actuaries say research on the subject has “mixed results.”

The actuaries expect 45 states and the District of Columbia will see increases in the average per-member, per-month (PMPM) cost for insurance in the individual market. Average PMPM cost does not convert directly into premiums members will be charged, but it is an estimate of how much insurers expect to pay for the average members’ care. If average cost increases, it is a safe bet that premiums will go up, too.

Average cost is expected to drop for five states in the Northeast: Massachusetts, New Jersey, New York, Rhode Island, and Vermont. Most of those states already use community rating in their individual health insurance markets. With community rating, risk is spread evenly across all plan members, regardless of health status or other demographic characteristics. Health plans that do not use community rating would charge a person with poor health, and therefore with higher expected cost, more than a person with good health.

Under the ACA, all health plans will be required to adopt adjusted community rating. For example, they will not be allowed to charge women more in premiums than men, they cannot charge older people more than three times what they charge young people, and they must charge the same premiums for the healthy as the sick.  This means, starting in 2014, men will subsidize women, the young will subsidize the old, and the healthy will subsidize the sick.  The result will be what many are calling premium shock, as the young, men, and healthy see huge increases in health premiums.

Premium Increases by Demographic Group, Wisconsin Case Study:

To allow for a deeper analysis, the Society of Actuaries report looks at how the ACA will affect insurance markets in a particular state. Authors chose Wisconsin because they have background knowledge of the state.

Some interesting findings, drawn directly from the report:

  • The highest percentage of people remaining uninsured under the ACA will be for those under age 19 (60 percent) since the Medicaid expansion does not affect children, those with incomes at or above 400 percent of FPL (71 percent), and those with excellent self-reported health status (43 percent). That suggests there will be some adverse selection.
  • Among those who transition from uninsured to newly insured, the age group of 45-54 years will see the largest increase in average cost (255 percent), followed by people ages 19 to 24 years (98 percent).
  • In the individual (non-group) market as a whole, including the previously insured, the greatest increase in average monthly costs for individuals ages 55 and over (a 68 percent increase), those with incomes at or above 400 percent of FPL (an 83 percent increase), and those with a self-reported health status of “fair” or “poor.” The average increase in PMPM cost is 32 percent.
  • The increase in utilization of services by newly insured people will result in a 2.0 percent total increase in statewide health care spending. Spending is expected to increase most for physician services (2.1 percent), hospital inpatient services (2.8 percent), and medical devices and equipment (3 percent).
  • Those who remain uninsured after the ACA is implemented, there will be an average decrease in PMPM cost of 30 percent. People who choose to go without insurance under the ACA will be healthier, younger, and have higher incomes than the currently uninsured population.

Kudos to the Society of Actuaries for the report – certainly one of the most comprehensive assessments yet of ACA’s likely impact on health insurance premiums.