The True Cost of CoverageA look at ACA subsidies and their impact on individual rates December 2015/January 2016
Author’s Note: The views expressed herein are those of the author alone and reflect current information as of September 2015. They do not represent the views of the Society of Actuaries, the author’s employer or any other body.
A more detailed examination of the technical components discussed in this article, along with some suggestions on how actuaries can contribute to the public good by correcting simplified explanations and common misconceptions, was published in the May 2014 edition of Health Watch;1 insights, language and numerical examples from the Health Watch article are included in this article.
It was not a slow news day. America awoke on July 22, 2014, to growing foreign and domestic problems on multiple fronts. Evidence was mounting that pro-Russian separatists in eastern Ukraine were responsible for a sophisticated surface-to-air missile attack on a commercial Malaysian flight from Amsterdam. Israeli attacks on Hamas continued to dismantle the tunnel warfare infrastructure in the Gaza Strip. A swiftly emerging and dangerous terror group had just burned a church in Mosul, Iraq, and threatened remaining Christians to leave the region, creating a mass humanitarian crisis in the process.
Closer to home, government leaders were trying to find a solution to an unprecedented influx of unaccompanied migrant children crossing the southern border. As reporters struggled to stay current with the various volatile situations, many commentators were making the unusual case for President Barack Obama to adjust his short-term schedule, which included multiple events deemed to be unofficial business. It was an unusually chaotic day in the midst of an already busy news cycle.
Absent from the fractured dialogue—and seemingly forgotten—was discussion of the 2014 health insurance market changes prescribed by the centerpiece legislation of the Patient Protection and Affordable Care Act (ACA), which had been prominent in the front page and political news cycle since 2010. With some of the more potent controversies quelled, the daily ACA news updates began fading in the spring of 2014. After the final tallies were in for the initial enrollment period, there was little else to report until 2015 premium rates were released.
Despite the widely publicized glitch in the rollout of health exchanges in November 2013, the national criticism had quieted after technical fixes were applied to exchange websites, an extension of the enrollment deadlines was implemented (coupled with successful promotion efforts), and delays or relaxations were granted to some of the more criticized policy elements. With the exception of a Supreme Court decision in June dealing with employee health benefits and religious liberty of closely held corporations, prominent news discussion of the ACA was waning and being crowded out by ostensibly more urgent matters.
The news hit the wire about 10 a.m. on the East Coast, and it seemed to catch everyone by surprise. A three-judge panel of the D.C. Circuit Court, generally regarded as the second-highest court in the land, had ruled that the Internal Revenue Service (IRS) had broadened the ACA language “an Exchange established by the State” to also include fallback exchanges established by the federal government (in states that did not establish an exchange) with regard to the issuance of government subsidies (technically “tax credits”) to assist individuals with health insurance premiums and benefit cost sharing. Less than two hours later, a three-judge panel in the 4th Circuit ruled nearly the opposite.
BEFORE THE ACA
Prior to the Affordable Care Act, benefits and rates for individual coverage varied substantially by state and issuer. In many states, issuers developed unique benefit designs, set their own rating factors, and may have considered underwriting factors for health status and claims experience in determining the premium rates offered to an individual purchaser. Under the ACA, individual rates may vary only for benefit plan, age, geographic area, family members covered and tobacco use. All issuers must use the same age factors as prescribed by the state. The state determines the geographic regions and may define parameters for developing area rating factors.
America was suddenly talking about the ACA again. With split appellate court decisions on a weighty matter, speculation of Supreme Court interest naturally entered the discussion.2 Legal pundits were arguing the importance of literal language versus “overwhelming intent” of enacted legislation. News organizations were sifting through old transcripts and video reels looking for a smoking gun on what exactly congressional intent was, although it seemed evident to some that these details were beyond the scope of understanding of the average member of Congress at the time the legislation was considered. At home, Americans watched as various midday news anchors appeared to be reading transcripts and trying to decipher the content as they read, reminiscent of the 2012 Supreme Court ruling regarding the “individual mandate” and the requirement of states to expand their Medicaid programs.
While the 2012 Supreme Court decision generally upheld the ACA, Medicaid expansion was dealt a minor blow as individual states were given the option to expand Medicaid eligibility or not. A Supreme Court ruling similar to the D.C. Circuit Court, on the other hand, would likely be disastrous for the ACA in states that did not establish their own exchanges. The logic goes something like this: If low- and middle-income enrollees (87 percent of market) are not eligible for subsidies, health insurance becomes “unaffordable,” which negates the individual mandate tax penalty. Without the so-called carrot (the subsidies) and stick (the tax penalty), the incentive to purchase insurance is dramatically reduced for subsidy-eligible individuals in federal exchange states.
The near simultaneous and opposite court decisions not only brought the ACA discussion back into the public arena, but heightened awareness of the importance of the subsidies to the overall framework of the law. Unfortunately, since the law’s inception, general discussions of subsidies, even among experts, have sometimes misrepresented the inherent mathematical ramifications for individual consumers with vague explanations.
In fact, supporting comments for market viability often specifically claimed that “young people would enroll in exchanges due to generous subsidies.” Many young people did respond to an aggressive marketing campaign and purchased health insurance for the first time, but the math indicates that the subsidy distribution is dramatically tilted toward older people. The 18–34 demographic represented 28 percent of the market at the end of the 2014 open enrollment, short of the 39 percent targeted expectation; the results are similar for 2015, and anecdotal evidence indicates attrition later in the year may have particularly affected a younger demographic.
Following the early misconceptions regarding subsidy allocations by age, the prevailing tale in 2015 and 2016 may be that “subsidies will partially offset rate increases.” This is true, but only for a minority of subsidy-eligible policyholders who have purchased richer benefit coverage. In reality, many people with exchange coverage will actually see their net premium rates decrease due to rate increases as the federal government’s share of the cost increases. This paradox is illustrated later in this article.
An Understanding of the ACA Impact on Individual Rates
The ACA includes several provisions that impact gross premium rates, rate relativities and net costs paid by individuals. The benefit and rate provisions apply to all individual coverage offered through exchanges or purchased directly from issuers. The ACA generally enhances benefits and attempts to standardize health care coverage by requiring compliant plans to meet an actuarial value (AV) criterion, which is the average value of the plan benefits relative to the total allowed costs based on a federally prescribed model. This allows consumers to compare benefit values across issuers and is intended to increase price transparency.
Plans must meet “metal level” AV requirements. Bronze plans have an AV of 60 percent; silver plans have an AV of 70 percent; gold plans have an AV of 80 percent; and platinum plans have an AV of 90 percent. A +/−2 percent variation in AV is allowed to meet the metal level criterion. Issuers have flexibility in designing benefits packages to meet the AV criterion, but they must meet some specific minimum requirements such as maximum out-of-pocket limits.
Access to insurance is guaranteed, and health status can no longer be used as a rating variable. More comprehensive coverage, combined with guarantee issue and new industry taxes, increases underlying costs and associated premiums. The ACA disallows gender-based rating and prescribes a standard age rating curve with a 3:1 maximum age rating limit, which is intended to lower the premium costs for older people but will increase the premium costs for younger people, particularly young men.
The rate impact obviously varies by age and gender, and may create a market with higher premiums than before for young people, all else being equal. President Obama’s announcements that allow for further extension of pre-ACA benefits preserved the pre-ACA age and gender rating structure for individuals and groups in states, and with issuers that elected this extension option. Hence, it is likely that younger people insured before 2014 and rated on a steeper age curve will have a greater propensity to keep their current plan than older people.
The ACA offsets some of the upward force on premiums by creating penalties for those not retaining adequate insurance, as well as through encouraging healthier people to enroll by providing premium and benefit subsidies to some based on income levels if they purchase coverage through an exchange. As results vary significantly by age, an understanding of the subsidy mechanics is important to understand enrollment ramifications and, ultimately, the long-term implications of the ACA on the individual market.
Premium Subsidy Illustrations
A simplified but lengthy example was constructed in the Health Watch article to illustrate the net premium and cost-sharing impacts segmented by age, benefit plan and income level. Three individuals of different ages were assumed to represent a sampling of the population. Similar examples are used below to illustrate the impact of 2015 rate changes.
|Figure A: Monthly Premium of Second-Lowest Carrier|
Figure A illustrates a sample 2014 gross premium structure for individual health coverage before premium subsidy reductions for three individuals before and after the ACA. The sample premiums reflect the federal age curve and reasonable pricing assumptions. (Note: This example assumes health insurance was purchased prior to ACA, so that the given individuals had a choice of renewing on their pre-ACA plan or purchasing an ACA-compliant plan. The current plan has a 50 percent actuarial value, and while the illustration reflects different rates for the current—also referred to as “transitional” or “grandmothered”—plan due to benefits and age slope, it does not reflect that rates may also be relatively lower due to preferred underwriting status at the time the policy was issued prior to 2014. This example also shows only a single plan for each metal level. Most individuals will have a choice of issuers, plans and rates for each metal level.)
Figure B: Maximum Premium Contribution
The premium subsidy calculation varies by individual. Figure B displays the first input to the premium subsidy calculation as prescribed by the ACA. Depending on income relative to the Federal Poverty Level (FPL), an individual’s contribution (that is, net premium) is capped based on the benchmark plan (the second-lowest cost silver plan available to that individual on the exchange) available to an individual of his or her age and in his or her geographic region. As mentioned earlier, premium subsidies are not available to individuals with incomes below 100 percent of FPL or above 400 percent of FPL; applicable percentages are linearly interpolated in between the data points in Figure B.3
|Figure C: Premium Subsidy Calculation (Benchmark Plan)|
|Age||FPL Amount*||FPL Level||Maximum Percent of Income||Benchmark Plan (B)||Maximum Contribution (M)||Calculated Subsidy (greater of (B-M) and zero)|
Figure C illustrates the ACA premium subsidy calculation for each individual age. For an income level of 275 percent FPL, the monthly contribution per individual regardless of age is capped at $231.06 ($11,490 * 275 percent * 8.78 percent / 12). As the benchmark plan (assumed to be the silver plan illustrated) rate is lower than the maximum contribution, the 24-year-old is not eligible for a premium subsidy. The older individuals can purchase the second-lowest silver plan for the maximum contribution, or apply the calculated subsidy to purchase another lower or higher cost plan in the exchange offered by any issuer. While this is only one example, and not an exhaustive study, this example demonstrates that the calculation results in higher subsidy dollars for older people with the same income, simply because their rates are higher than younger people.
Figure D: Net Monthly Premium of Second-Lowest Carrier (275% FPL)
Figure D illustrates the net premiums available to an individual at the 275 percent FPL level after subtracting the premium subsidy from the monthly premium. A few things should be noted from the resulting net premiums. First, the rates for the current plans have not changed from Figure A to Figure D, as these plans are not ACA-compliant and therefore not eligible for federal subsidies. The same would be true for individuals with incomes above 400 percent of FPL who purchase ACA-compliant policies. Second, the rates for the 24-year-old also did not change, as no subsidy was calculated in Figure D because the gross premium is below the maximum contribution. Third, the net premium for the second-lowest silver plan (the benchmark plan) is the same for the older individuals because the affordability threshold depends only on income and not on age.
Varying Premium Rates
Notably, the net premium rates vary by income level as illustrated in Figures D’ and D” as lower income leads to a higher subsidy level and more leverage to the net premium rates.
Figure D’: 2nd Lowest Plan Net Premium Calculations (200% FPL)
Figure D”: 2nd Lowest Plan Net Premium Calculations (350% FPL)
Finally, perhaps most enlightening and not at all intuitive, is the finding that at the illustrated income level, the net premiums for the bronze plan by age are inverted due to the leveraging of the premium subsidies (that is, the age 64 individual will pay less than the age 44 individual who will pay less than the age 24 individual). A direct comparison of the current and bronze plans illustrates why young, previously-insured people would more likely remain on current plans while older people would more quickly move to the subsidized exchange plans. Young individuals at this income level may be disillusioned to learn that the mandated coverage that they are strongly being encouraged to purchase is not only more expensive than their previous plan due to age rating compression and other rating changes, but that the premium subsidies are allocated in such a way that the net premium costs for older people are actually lower than the net premiums for younger people for the bronze plan option.
Rate Increase Impact on Net Premiums
The impact of rate increases on net premium changes is thought to be an indicator of market disruption. As most consumers in the exchanges to date have subsidized insurance, it is instructive to understand the net premium changes to subsidy-eligible individuals. To isolate the impact of the rate changes, it is assumed that individual incomes and the contribution percentages (in reality, slightly higher in 2015 versus 2014) in Figure B do not change. Figure A1 illustrates the gross premium rates after a 10 percent increase from Figure A,4 again assuming rates for the second-lowest cost plan available. (Note: Under ACA, the benchmark plan resets each year based on updated carrier rates and plans. Individuals looking for the lowest net premium will need to review their exchange options every year and may need to change carriers to keep the desired net premium.)
|Figure A1: Monthly Premium of Second-Lowest Carrier (with 10% Increase)|
Figure C1 illustrates the premium subsidy calculation for each individual age with gross premium rates at a 10 percent higher level.
|Figure C1: Premium Subsidy Calculation (with 10% Increase)|
|Age||FPL Level||Maximum Percent of Income||Benchmark Plan||Maximum Contribution||Calculated Subsidy|
Figure D1 illustrates the resulting net premiums with gross premium rates at a 10 percent higher level.
|Figure D1: Net Monthly Premium of Second-Lowest Carrier (with 10% Increase)|
Figure E illustrates the net premium change for each plan and individual age. Notably, the net bronze premium rates decrease while there is no change to the silver rates. As most consumers have purchased plans less expensive than the benchmark plans, the impact of rate increases will generally reduce net rates for subsidy-eligible consumers.
Figure E: Change in Net Premium Rates
Expected Total Cost (ETC) and Plan Selection
No discussion of consumer health insurance purchase decisions is complete without consideration of out-of-pocket cost sharing in addition to premium payments used to obtain coverage. The premiums represented in the figures presented do not present the total consumer cost, as individuals will still have a cost-sharing responsibility in the form of deductibles, copays and coinsurance for the health care they use. In addition, individuals with income below 200 percent FPL who select a silver plan are eligible for cost-sharing subsidies as well as premium subsidies. An individual’s ETC for health care can be thought of as the net premium, calculated in the figures presented, plus the expected net cost sharing, versus the applicable tax penalty if qualifying minimum coverage is not obtained.
|Figure F: Optimal Plan Selection for 2016|
|FPL||Expected Lowest Cost Option|
While not presenting detailed calculations, Figure F illustrates the 2016 plan selection decision based on ETC for different age and income levels corresponding to the gross premiums presented, using standard utilization and cost models to estimate benefit cost sharing.
While issuers’ experience and models will differ, several general comments can be made about likely decisions based on age and income:
- Individuals with low incomes (below 200 percent FPL) will overwhelmingly select silver plans to take advantage of the cost-sharing subsidy.
- To avoid the rate change due to age rating compression, many high-income young people with coverage likely would stay on their current plan for as long as possible.5 They are the most likely to drop coverage when the transitional plans are no longer allowed unless they have a high need for services.
- Middle- to high-income young people are the most likely to go without coverage, assuming they don’t expect to need it. As the penalty is a percentage of income, at high-income levels, the penalty will exceed the gross premiums (which do not vary based on income), and high-income individuals will likely purchase at least the minimum required coverage to avoid paying the penalty.
IN HEALTH WATCH
Read the article, “Implications of Individual Subsidies in the Affordable Care Act—What Stakeholders Need to Understand” in the May 2014 issue of Health Watch. Author Greg Fann, FSA, MAAA, discusses the details of the ACA provisions of federal subsidies that affect consumers’ cost of coverage in the individual market, and breaks down how the net effect of these provisions will shape consumers’ decisions to buy a new level of coverage, retain current coverage or elect to be uninsured (or underinsured according to the ACA definition) despite new tax penalties.
The premium subsidy calculations in the individual exchanges represent perhaps the least transparent aspect of the ACA. The resulting federal outlay is dependent on how many subsidy-eligible individuals enroll; the subsidies vary considerably by age and income level, the gross premiums offered in the marketplace, ACA awareness, unemployment rates and overall economic conditions. In addition, subsidy calculations change each year as plans and rates are updated. A proper understanding of the subsidy mechanics is required to formulate detailed enrollment projections.
The exchange subsidy calculations are quite complicated and unfortunately not well understood. As we heard in the arguments leading up to the Supreme Court decision, subsidy eligibility plays a crucial role in the affordability consideration for many potential customers and, ultimately, impacts the overall viability of the market. Actuaries should play a leading role in educating the public about relevant technical matters such as how the ACA subsidies actually function. Otherwise, the only usefulness of the news we hear about subsidies might be to give Don Henley another good song idea.
- 1. SOA.org/Professional-Interests/Health/hlth-detail.aspx. See Publications: Health Watch–May 2014. ↩
- 2. In June 2015, a 6-3 Supreme Court ruling upheld the eligibility of subsidies in federal exchange states. ↩
- 3. Maximum Percent of Income is 2014 values. These amounts are indexed and are slightly higher in 2015. ↩
- 4. In this example, it is assumed that the benchmark plan does not change. In reality, many 2014 benchmark plans are higher than the 2015 benchmark plan. ↩
- 5. It is assumed that the “current plans” are no longer available in 2016. If available, the model indicates that they would have the lowest ETC for 24-year-old individuals with incomes greater than 200 percent FPL. ↩