Net HIE Premiums Growing Faster Than Incomes; This is Likely to Accelerate

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum


203.901.1631 /.1632 / .1627 richard@ / hinds@ /


June 24, 2014

Net HIE Premiums Growing Faster Than Incomes; This is Likely to Accelerate

  • The average premium increase thus far proposed by health insurance exchange (HIE) plans for 2015 is 10%; this compares to 4% to 5% expected income growth for subsidy eligible households
  • We encounter a common misperception that the Affordable Care Act (ACA) keeps households’ net premiums constant as a percent of income. In reality, the first of two health insurance subsidy indexing provisions – ‘regular’ indexing – keeps households’ share of premiums constant, until and unless the second indexing provision (‘additional’ indexing) applies. Thus beginning in 2015 households should expect their net premiums to rise (at least) at the rate of overall premium inflation
  • Additional indexing applies in 2019 and after if total federal premium (and cost-sharing) subsidies paid exceed 0.504 percent of GDP, as is likely. Additional indexing serves to keep total federal premium and cost-sharing subsidies at or near this percent of GDP, in which case households purchasing subsidized health insurance on the HIEs will see their net premiums grow even faster than the underlying rate of premium inflation
  • Most current beneficiaries will have enrolled at a point in time at which their exchange could not offer a great deal of information regarding the nature of the coverage they were purchasing – other than the price. As beneficiaries have begun to gain experience with their coverage, we believe many will be disappointed by features such as narrow networks and high levels of cost-sharing. In effect, we believe beneficiaries’ (and potential beneficiaries’) assessment of the value of coverage may be falling, just as the cost of coverage (as a percent of incomes) begins rising. This feeds our conviction that the HIEs are subject to adverse selection pressures that will force policy changes, and leads us to believe that enrollment growth in 2015 may be limited
  • For the HIEs to succeed (both in terms of avoiding adverse selection and enrolling sufficient numbers of persons to meaningfully reduce the ranks of the uninsured), we believe at least one or more of the following policy steps will be taken: (re-)allowing greater premium differences based on age, (re-) allowing higher out-of-pocket maximums, reducing the scope of mandatory benefits, increasing subsidies, increasing penalties, and/or forcing the merger of adverse HIE risk pools into larger, better balanced risk pools


Total premium rate changes requested by insurers on the 2015 individual health insurance exchanges (HIEs) show an average proposed hike of 10.0%.[1] In this note we demonstrate how growth of this magnitude would cause net (of subsidies) premiums to grow at rates in excess of household income growth (projected at 4 to 5pct in coming years). This reduces the affordability of (even) subsidized health insurance, potentially further undermining the HIEs’ grassroots political support and/or catalyzing adverse selection and related stopgap reforms

Indexing of ACA Premium Subsidies, and How This Affects Affordability of Coverage

The Affordable Care Act (ACA) provides for premium subsidies to eligible households (incomes between 100% and 400% of the Federal Poverty Level (FPL)) purchasing coverage on the HIEs. Subsidies are set such that a household’s share of premiums for the second cheapest ‘silver’ plan available in its region is capped at a given percentage of its income. The maximum percentage of income paid by a household varies by income; households between 100 and 133 FPL pay no more than 2.0% of income, while households between 300 and 400 FPL pay no more than 9.5% of income (Exhibit 1)

Under ‘Regular’ Indexing, Households’ Premium Costs Can (and Likely Will) Grow as a Percent of Income

An underappreciated factor determining net premium growth is the ACA’s set of premium indexing provisions, which formulaically increase the percentage-of-income caps over time. There are two types of indexing: ‘regular’ and ‘additional’. At a given level of income, regular indexing serves to keep a household’s net premium payments constant as a percent of the total premium, rather than as a percent of income. This is done by increasing households’ percentage-of-income cap by the excess of national private market average premium growth over average household income growth in the preceding year. For example take a 150 FPL household facing a 4.0% cap on its net premium as a percentage of income. If in 2014 the national private market average premium grew 6% vs. 2013, and average household income grew 4% over the same period, then the 150 FPL household would see its 4.0% percentage-of-income cap rise by 6% – 4% = 2% to 4.08% in 2015. This has the effect of growing both the household’s net premium and subsidy at roughly the same rate as growth in per-capita healthcare costs; by extension, the household’s net premium growth would exceed its growth in income. Regular indexing may (and by all accounts, will) take effect starting in 2015

‘Additional’ Indexing is Likely to Compound the Effect, Further Raising Net Premiums as a Percent of Income

Additional indexing will take effect as early as 2019 if two criteria are met: 1) national private market average premium growth exceeds CPI in the preceding year (i.e. real per-capita healthcare cost growth is > 0); and 2) aggregate total HIE subsidies (sum of premium subsidies and cost-sharing reduction (CSR) subsidies[2]) exceed 0.504% of GDP in the preceding year. If triggered, an HIE household sees its applicable cap on net premiums as a percentage of income grow by a factor of 1 plus the real growth rate in per-capita healthcare costs. Additional indexing is designed to contain aggregate subsidies at approximately 0.504% of GDP, and serves to increase the proportion of a household’s net to total premium while shrinking the share of total premium covered by its subsidy. As a result, when added to the regular indexing effect, additional indexing causes household net premium growth to outstrip income growth by a wider margin. These provisions are summarized in Exhibit 2 alongside their basic effects for a household at a fixed percentage of FPL

The Congressional Budget Office (CBO) projects that the criteria for additional indexing will be met in 2018, thereby triggering its effects as early as possible, in 2019. By our estimates this is fair; moreover there exists a considerable margin of safety within which this would hold true. Exhibit 3 summarizes a projection of total HIE subsidies relative to the 0.504% GDP failsafe in 2018. Using figures recently published by the Department of Health and Human Services’s Office of the Assistant Secretary for Planning and Evaluation (HHS ASPE) we calculate aggregate premium subsidies of $21.0B in 2014 (see Appendix I for detail). Aggregate CSR subsidies are estimated by CBO to be a quarter of premium subsidies in 2014 (falling to about 20% by 2018), resulting in aggregate total HIE subsidies of $26.2B this year. If we assume per-capita HIE premium subsidies inflate at the preliminary observed rate of 10.0% in 2015 and thereafter conservatively assume reversion to an estimated future average of 5.7%, and forecast GDP using IMF and CBO growth estimates, then aggregate HIE subsidies would total 0.504% of GDP in 2018 if HIE enrollment growth is only 77% of what CBO expects. Even if 2015 per-capita HIE premium inflation were contained to 5.7%, HIE enrollment growth would need only be 81% of CBO’s projection to trigger additional indexing in 2019

The significance of indexing in coming years can be seen by examining net premium growth and net premiums as a percentage of income at the household level. We illustrate the effects for three prototypical households at 150 FPL, 250 FPL, and 350 FPL in Exhibits 4 and 5. We assume 3.9% real household healthcare cost growth per annum, that household total premium growth matches nominal growth in household healthcare costs, constant medical loss ratio (MLR) and AV, and that FPL inflates at CPI while household incomes grow at 3.5 to 5.0pct. The last assumption is an important break from the ‘all else equal’ effects described in Exhibit 2: because income growth tends to outstrip FPL growth (which historically has fluctuated around CPI), a given ‘150 FPL’ household in 2014 will see a natural indexing year over year as their income as a percentage of FPL creeps up – pegging them to a higher cap on their net premiums as a percentage of income – before the ACA’s indexing provisions are added.[3] The result is net premium growth outpacing growth in underlying household healthcare costs (and total premiums) for all but the wealthiest subsidy-eligible households before additional indexing in 2019, at which point net premium growth for all subsidy-eligible households is expected to routinely exceed 10pct (note the incremental impact of additional indexing can be seen by the difference between the solid and dashed lines). This would result in net premiums eating up an additional few percentage points of income in less than ten years for subsidized HIE households, a presumably unacceptable course from their vantage

Higher (as a Percent of Income) Premiums Means Fewer Enrollees: The Employer-Sponsored Example

The transformation of the employer-sponsored insurance (ESI) market since the ‘90s is a fruitful case study for understanding the expected response of HIE households to excessive net premium growth. As ESI beneficiaries faced continued real growth in premiums of +/- 4pct, they naturally responded by either enrolling in less generous (i.e. lower AV) plans or dropping coverage altogether (Exhibit 6). Households seeking coverage on the HIEs face yet more formidable cost headwinds, as such we fully anticipate declining HIE participation, underwhelming new enrollment, and migration towards less generous plan offerings

The HIEs face major political risks if they proceed to unravel in this fashion, and their structural weaknesses would require hasty action to prevent the program from succumbing to adverse selection[4]. We believe the primary options available to policy-makers include (re-)allowing greater premium differences based on age, (re-)allowing higher out-of-pocket maximums, reducing the scope of mandatory benefits, increasing subsidies, increasing penalties, and/or forcing the merger of adverse HIE risk pools into larger, better balanced risk pools

  1. Please see: “Early Look at 2015 HIE Rates: Avg. Increase = 10%”, SSR Health LLC, June 17, 2014
  2. Cost-sharing reductions are federal subsidies designed to lower cost-sharing and effectively increase plan actuarial value (AV) for eligible enrollees (between 100% and 250% of FPL). This is accomplished by reducing maximum out-of-pocket (OOP) limits as well as potentially reducing OOP expenses for deductibles, coinsurance, and copayments
  3. N.B. modeling FPL growth at CPI and household income growth at >3pct is consistent with CBO’s methodology
  4. Please see: “Buckle Up! A Summary of Adverse Selection Pressures on Health Insurance Exchanges”, SSR Health LLC, May 16, 2013; and “ACA Enrollment and Adverse Selection Pressures – An Update”, SSR Health LLC, February 20, 2014
Print Friendly, PDF & Email