The label on a shipping crate carrying the U.S. health insurance market should read “Fragile" or "Handle with care". The fragility of any national health insurance framework comes from the complex interconnectedness of multiple stakeholders and financing elements. Perhaps most entwined into the fabric of U.S. health insurance is the financing support that the federal government provides to all major forms of comprehensive health insurance. Currently, over $1 trillion of federal support comes to this market, with most of it akin to a defined-benefit type of financing system. From subsidies in the individual market (under-65), to federal funding of Medicaid, to tax deductions in employer-sponsored health insurance, current federal support keeps the health insurance market (dysfunctional as it may be) intact.
- The American Health Care Act (ACA replacement proposal), if signed into law, will fundamentally change federal financing of healthcare, especially for the Medicaid and individual insurance segments.
- We expect a decline in enrollment in the individual (2 million–4 million) and Medicaid (4 million–6 million) segments, resulting in a decline in premiums.
- Profitability will likely improve, as the replacement plan can result in an improved risk pool in the individual market.
- Our ratings on U.S. health insurers remain unaffected by this bill.
S&P Global Ratings believes that it is highly likely that "repeal and replace" will alter substantially how U.S. healthcare is financed, consumed, and regulated. On March 6, 2017, the first house bill--The American Health Care Act--was released as a possible replacement for the Affordable Care Act (ACA). This proposed replacement bill still has to traverse the legislative process, and there could be changes along the way. We analyzed the key aspects of this replacement bill to understand its potential impact on the insured rates, insurers' premiums, and insurers' profitability. We expect changes in federal financing to the individual and Medicaid segments to reduce insured rates, and somewhat dampen organic growth potential for insurers. As for profitability, we expect margin improvements for insurers in the individual market and continued low-single-digit margins in the Medicaid segment. The employer/group and Medicare Advantage segments should be mostly unaffected by the replacement bill (see table 1).
Table 1: Net Impact Of ACA Replacement Plan On Number Of Insured, Insurers' Top And Bottom Lines
|Insured rate||Negative: Individual and Medicaid segments will see drop in enrollment|
|Insurers' premiums/revenues||Negative: Somewhat tempered potential for future organic growth|
|Insurers' profitability||Neutral to positive: Individual market margins should improve, while other segments should hold steady|
Table 2: Federal Support Entwined In U.S. Health Insurance Framework
|Insurance segments||Federal funding of health insurance (bil. $)|
|Tax exclusion for employer-based coverage*||266|
|Small-group tax credits*||1|
|Individual (subsidies and related spending)¶||43|
|Children's Health Insurance Program¶||14|
|Small-group tax credits*||1|
*2016 estimated. Source: "Federal Subsidies for Health Insurance Coverage for People Under Age 65: Tables From CBO’s March 2016 Baseline," CBO, March 2016. ¶2016 actual mandatory outlays. Source: "The Budget and Outlook: 2017 To 2027," CBO, January 2017.
Individual (Under-65) Insurance Segment: Highest Impact From Potential Replacement PlanInsurers' current positions in the individual market
In aggregate, Medicaid accounts for about 20% of insurers' total premiums, and like the individual segment it has had strong growth in the past couple of years. More states have moved to a managed Medicaid framework, in which they contract with private insurers to manage their Medicaid programs. Thirty-two states (including DC) expanded Medicaid eligibility for their residents post-ACA. In general, Medicaid has been a profitable (albeit, at lower margins) segment for insurers.
Although insurers saw top-line growth since 2014, profitability has been a struggle for the industry. Other than a handful of insurers, most reported underwriting losses in this segment. In aggregate, we estimate the industry had underwriting losses of over $3 billion in 2014 and $4.5 billion in 2015 for this segment. We expect 2016 and 2017 to be better than 2015, in aggregate, but still below insurers' target profitability levels (see "The ACA Individual Market: 2016 Will Be Better Than 2015, But Achieving Target Profitability Will Take Longer," published Dec. 22, 2016, on RatingsDirect).Current federal funding for the individual market
The federal outlay for the individual market was about $42 billion in 2016, most of which is related to the income-based subsidies provided to the insured in this segment. ACA introduced two forms of income-based subsidies to the individual market:
- Advanced premium tax credit (APTC): Individuals (not eligible for Medicaid) between 100%-400% of the federal poverty line (FPL) receive a tax credit that reduces the premium they need to pay for the health plan. This tax credit is linked to the second-cheapest silver plan available on the exchanges. The amount of tax credit also changes with the price of the insurance plans (or cost of the benefit) in the marketplace.
- Cost-sharing reduction (CSR): Individuals (not eligible for Medicaid) between 100%-250% of FPL receive a CSR that decreases out-of-pocket costs (e.g., deductible, coinsurance, etc.).
These subsidies perform almost as a stabilizer for the market. They create a floor below which the insured rate will likely not drop because the subsidies are linked to the actual price in the market. A prime example of the subsidies performing as planned was the 2017 open enrollment (OE4). Average premiums increased by 25% in 2017. Such an increase would have resulted in a fairly significant decline in enrollment if not for the subsidies. Close to 80% of the exchange enrollees received an income-based subsidy linked to the actual price or market premium; so as premiums went up, so did their subsidies. This effectively hedged them against the sharp premium increase. We estimate OE4 ended with 12.2 million enrollees, which was about 5% lower than the 2016 open enrollment (OE3). This decline was in line with our previously forecasted range for 2017 of a potential 4% year-over-year enrollment growth on the upper end, and an 8% decline on the lower end (see "Slowing Down: ACA Insurance Marketplace Growth May Halt In 2017," published Oct. 13, 2016, on RatingsDirect).
What may be the impact of a replacement funding plan?
- Net decline of 2 million to 4 million insured in this segment
- Increased affordability for the eligible younger population, and reduced affordability for the older population
- Improved profitability for insurers in this business segment as the risk pool will have a greater proportion of younger enrollees
- Larger differences between states both in terms of insured rates and benefits covered by insurance plans
Given the interconnected nature of the health insurance market, we evaluated the replacement financing proposal, along with other proposed changes (see table 3) such as potential reduced actuarial values, continuous coverage mandate, and wider rating age bands. Other aspects such as guaranteed issue and ability to add children to their parents'' plan until 26 years of age will remain in the replacement plan.
Table 3: Key Potential Changes Between ACA And Proposed Replacement Plan
|Key factors considered in analysis||Currently under the ACA||Proposed replacement plan*||Proposed effective date of replacement*|
|Federal financial support||Income-based federal APTC and CSR; uses the actual price of the second-cheapest silver plan in an individual's area as benchmark for amount of subsidy.||Fixed amount of age-based refundable tax credits (e.g., $2,000 for a 21-year-old increasing to $4,000 for a 64-year-old) that will not differ by area of residence. Limits tax credit based on modified adjusted gross income level.||Jan. 1, 2020|
|Permissible age variation in insurance premiums (rate bands)||3:1||5:1||Jan. 1, 2018|
|Plan benefit design/Levels of coverage||Metallic plans, such as Bronze (60% AV), Silver (70%), Gold (80%), Platinum (90%)||No national minimum; states will decide the requirements for their respective state||Jan. 1, 2020|
|Individual mandate||Individual has to pay a tax penalty if not enrolled in a qualified health plans (with some exceptions)||Insurance premiums can be 30% higher if "continous coverage" is not maintained||ACA mandate is repealed effective December 2015; Continous coverage requirement starts 2019|
*Based on American Health Care Act (draft as of March 6, 2017).
Insurers are no longer allowed to underwrite or price based on medical conditions of the insured, but they are allowed to use a rating band, where they can charge differently depending on the age of the individual. This is based on the premise that younger individuals, in general, use medical services less than the older population. Under the current ACA rules, an insurer is allowed to charge 3x as much for a 64-year-old as for a 21-year-old. The replacement plan widens this rate band. The wider rate bands (5:1) will reduce premiums for the eligible younger population, while likely increasing premiums for the older population.
For example, we assumed that average premiums for a 21-year-old would decline by 20% (given the wider age bands and some reduction in actuarial value of plans) as a result of the replacement plan. Using the average national premium price from the 2016 marketplace, a 20% decrease would mean annual premiums of $2,625 compared to average annual premiums of $3,281 for the 21-year-old. However, the wider rate band would mean premiums for a 64-year-old could increase almost 30%, resulting in annual premiums of $13,125 (5x higher than the 21-year-old) compared to $9,844 (currently, 3x higher than the 21-year-old). The proposed tax-credits of $2,000 in the replacement plan, although not covering the entire premium cost for a 21-year-old, would reduce it by almost 75%. The proposed $4,000 tax credit for a 64-year-old falls well short of the potential premium cost, reducing premiums by only 30%.
Therefore, we expect the replacement plan to result in an increase in the younger (21-35 years) insured, and a higher decline in the older age (45-64 years) insured. We estimate the net impact would be a 2 million–4 million reduction in enrollees in 2019.
Another impact of the replacement scenario is the possible amplified variance among states. First, depending on how much flexibility states use to create minimum actuarial values, in some instances it may feel like a walk down memory lane. "Mini-med" or limited-benefit plans that were prevalent before the ACA in the individual market segment may make a comeback in some states, whereas other states may choose to maintain close to ACA-level essential benefits. Second, because the proposed replacement tax credits aren't linked to the underlying costs, individuals in states with higher premiums will see a less-effective benefit than states with lower premium rates (see chart 2). Health insurance premium rates can differ significantly among states. For example, the average premium for a silver plan in Alaska was almost three times as expensive as one in Hawaii. The current ACA system of APTC links the financial support to the underlying cost (price of the second-cheapest silver plan), and can be effective regardless of the insurance costs of each state. Thus, the potential shift to age-based prefixed tax credit will hurt states with higher-than-average premium rates, resulting in greater differences in insured rates among states.
The proposed replacement plan, similar to ACA, includes a disincentive for not buying insurance. The proposed plan changes the penalty of not having health insurance from a tax penalty to a "continuous coverage" premium penalty. As per the "continuous coverage" rule, an insurer can charge 30% higher premiums if an individual has lapsed coverage for more than 63 days. Continuous coverage may be a more effective disincentive in stopping individuals from jumping in and out of coverage based on need. But it will not solve the affordability issues and may prove to be a disincentive even for a lower-morbidity individual to sign up after missing coverage. However, the bigger impact may be felt more immediately if this proposal is signed into law soon. The ACA mandate penalty is repealed effective 2015, and the continuous coverage mandate isn't in place until 2019. So although the ACA mandate penalty wasn't very strong, it is still better to have some disincentive than none. This gap may lead to higher-than-expected lapses in 2017, which could be somewhat concerning for insurers already struggling with profitability issues in this segment.
As for health insurers' credit profiles, premiums will decline as enrollment declines, but profit margins will likely improve. This is because, starting in 2019, there will likely be a higher proportion of the younger population in the insured ranks, which is different from the current individual risk pool.
We didn't include high-risk pools in our analysis. The proposed bill appropriates $15 billion in 2018 and 2019, followed by $10 billion annually, to be distributed across states. States can use this "patient and state stability fund" for a few different things, including setting up a high-risk pool. High-risk pools are expensive endeavors. They have limited track records for success and may end up being cost-prohibitive to maintain in the longer term. We don't expect $15 billion across 50 states and DC to be sufficient for states to run high-risk pools. Instead, a reinsurance program, which was the most effective of the three ACA premium stabilization programs (referred to as 3Rs) may work better. A possible reinsurance program is mentioned as the "default federal safeguard" in the proposed rule.
Medicaid Insurance Segment: High Impact For Medicaid Expansion From Potential Replacement PlanInsurers' current positions in the Medicaid market
In aggregate, Medicaid accounts for about 20% of insurers' total premiums, and like the individual segment it has had strong growth in the past couple of years. More states have moved to a managed Medicaid framework, in which they contract with private insurers to manage their Medicaid programs. Thirty-two states (including DC) expanded Medicaid eligibility for their residents post-ACA. In general, Medicaid has been a profitable (albeit, at lower margins) segment for insurers.Current federal funding for the Medicaid market
Medicaid is funded jointly by the federal and state/local governments. The federal outlay for Medicaid was about $368 billion in 2016, which is approximately 60% of the total spending on this program. This federal share of Medicaid expenditure is greater for the beneficiaries who are newly eligible as a result of ACA's Medicaid expansion. Federal funds match 100% of the expenditure for the expansion in 2016, reducing to 90% for 2020 and beyond. (Under ACA, states could expand Medicaid eligibility to 138% of the federal poverty line. So far, 32 states, including DC, have undertaken this expansion.)
The method of Medicaid funding is linked to the actual costs of the state markets. States pay providers or managed care plans for Medicaid costs and then report those payments to the Centers for Medicare and Medicaid Services. The federal government pays the states a percentage of the costs of those reported medical services. This percentage, known as Federal Medical Assistance Percentage (FMAP), is calculated annually for each state. The actual FMAP differs by state, ranging from about 50%-70% in 2016.What may be the impact of a replacement funding plan?
- Insured population in this segment will likely decline by 4 million–6 million after 2020-2024, as a portion of the "expansion" falls off the rolls
- Increased opportunity for private insurers as more states move to managed Medicaid model for running the program
- Insurers should continue to generate low single-digit profits in this segment, but can be squeezed in the longer term if states reduce reimbursement rates to offset their greater burden of the cost
The overall funding for Medicaid is shifted to a per-capita funding under the proposed replacement plan. This is different from the current form of funding that is close to a defined benefit financing of the program.
The proposal also changes the financing of the expansion population. It allows new enrollees to join the expansion ranks until Dec. 31, 2019. But, after 2019 it doesn't provide federal funding for any new eligible enrollees that aren't already on the roles or for any current enrollee that has more than a month's break in eligibility. This effectively puts Medicaid expansion in "run-off" after 2019. The Medicaid program enrollees generally have a lot of churn due to change in income levels. Starting 2020, even a low level of churn among the "run-off" expansion enrollees will result in a gradual decline in enrollment.
We expect expansion states (and DC) to be most affected in terms of potential reduction in Medicaid enrollees. As ACA's generous federal matching for the continued expansion is changed after 2019, states--depending on how stressed their budgets are--may be unable to continue funding new expansion enrollees. However, we don't expect all of the 32 states to go into "run-off." Assuming some states--especially those that had higher eligibility levels even before ACA--maintain their increased expansion expenditure (perhaps with some changes to benefit design) while others aren't able to, we estimate a drop of about 4 million–6 million Medicaid enrollees from 2020 to 2024.
Table 4: Enrollment Of Medicaid Expansion States (And DC)% change, July-September 2013 to December 2016
|States expanding Medicaid||Total Medicaid and CHIP enrollment, December 2016 (preliminary)||Average monthly Medicaid and CHIP enrollment, July- September 2013||Net change, July-September 2013 to December 2016||% change, July-September 2013 to December 2016|
|District of Columbia||264,849||235,786||29,063||12.3|
|Subtotal for all states expanding Medicaid||51,628,843||37,249,111||13,618,422||36.6|
Source: CMS, Medicaid.gov. N.A.-Not available.
As their share of the expenditure increases, states will likely increase their use of private insurers to manage the program. As Medicaid managed care increases, insurers will likely see increased revenues from this program. But if the expansion states reverse their gains, it will take away some of the recent gains seen by the insurers. We expect insurers to remain profitable in this business segment, but margins may be squeezed in states where budgets force a reduction in managed Medicaid reimbursements.
Employer-Sponsored Or Group Insurance Segment: Mostly Unaffected By Proposed Replacement PlanInsurers' current positions in the group market
In aggregate, the group segment accounts for about 60% of total premium revenues for U.S. insurers. This is by far the largest segment for the industry, and has also proven to be the most stable in terms of earnings. Insurers participate in two forms in this segment:
- Self-insured plans: The employer/group retains the insurance risk and pays a fee to the insurer to administer the program for its employees.
- Fully-insured plans: Insurer retains the insurance risk and gets paid a premium to take the insurance risk for the group's employees.
Current federal funding for the group market
Contrary to the individual and Medicaid markets, the employer/group market doesn't receive direct federal dollars in funding. But premiums paid toward group insurance are tax deductible. The Congressional Budget Office estimates that there was $266 billion of tax exclusion related to this segment in 2016.
The ACA introduced a tax (often referred to as the "Cadillac tax") to reduce this exclusion. The "Cadillac tax" has a 40% tax on high-cost employer health plans. The tax threshold (or amount above which the tax would become effective) was set at $10,200 for single coverage and $27,500 for family coverage. In December 2015, Congress delayed the implementation of the Cadillac tax until 2020.What may be the impact of a replacement funding plan?
- No impact in the near term as "Cadillac tax" is further delayed to 2025
- If implemented in 2025, no significant change in insured rate in this segment
The Cadillac tax, which is the same in the replacement proposal, shows an interest in taxing premiums on employer plans. If implemented in 2025, we expect minimal change in the insured rate for this segment due to the tax. It is very likely that "buy-downs" (purchase of reduced-benefit plans) will increase in an attempt to limit the tax impact. This will accelerate the trend of increased cost sharing with the consumer. But we don't expect a meaningful amount of employees to drop coverage because of this tax.
As for insurers' revenues, the biggest impact will be on the "fully-insured" segments. Insurers' premiums will likely decline in this segment as the groups increase "buy-downs." We expect a minimal impact on the "self-insured" segment because insurers only receive an administration fee. But we remain cautious regarding any unintended impact on this segment, given this is the largest and most mature segment for the industry. Any unexpected, meaningful reduction in premiums can have a negative effect on margins, not due to medical claims cost issues, but because of the increased administration cost strain from a lower top line.
Medicare Advantage (MA) Insurance Segment: Limited To No Impact From Potential Replacement Plan
In aggregate, the MA segment accounts for about 12% of total premium revenues for U.S. insurers. The proposed replacement plan doesn't affect MA. We believe MA may be the "safest" program in terms of being protected from potential changes. This program is highly popular with seniors and generally has bipartisan support in Congress. We expect year-to-year MA reimbursement rates to be an ongoing issue (nothing new), but one that should be manageable for the industry. A potential shift of the MA program to a premium support program could also provide potential upside as it could open up the market to further managed-care penetration depending on the details of the program.
No Rush Delivery On This Package
The key to successful execution of a replacement plan is not just a plan that takes into account the interconnectedness of the marketplace, but one that also provides adequate time to adapt to the proposed changes. Sudden changes or lack of clarity in rule adds to uncertainty in this already complex framework. A two-to-three-year pathway (as provided for several of the key factors in the replacement plan) to implementation would provide insurers adequate time to modify products and strategies to the new rules.
However, this replacement bill isn't a done deal. The legislative process may take time, and tweaks to the bill are also likely. At this time, it is clear that the most affected segments will be the individual and Medicaid markets. Overall, we expect a decline in insured rates and insurance industry revenues due to loss of enrollees in those two segments. As for future growth, insurers will have to turn to MA, which remains the only clear path for organic growth.
Additionally, we cannot overemphasize the benefit from diversification in a rapidly changing industry. As is the case with any major industry reform, diversified companies will see a less-adverse impact than those concentrated in a single business segment. In this instance, the diversified health insures with less (as a percentage of their total revenues and earnings) exposure to the individual and Medicaid lines will see minimal adverse impact to their credit quality from a replacement plan. However, such a plan may affect insurers more who are concentrated in the individual market, which has the likelihood of becoming a smaller piece of the health insurance pie.
Data Behind Our Individual Market Forecast
For the individual market (under-65, non-Medicaid), we are forecasting a 2 million–4 million drop in enrollment from implementation of the replacement bill. We looked at various data points for our forecast. Three cohorts that weighed heavily on our forecast were:
- As per census data, about 10 million are uninsured in the 18-35 year age group. Based on our analysis we estimate a portion (about 35%-40%) of them will enroll under the proposed replacement plan. (100% enrollment of this population is unlikely. Even today, all subsidy qualified individuals haven't enrolled, which is why outreach remains important to this age group).
- Based on data from the U.S. Department of Health and Human Services, about half of the exchange population is between 45 and 64 years of age (we estimate that to be about 5 million). We assumed the majority of these enrollees will drop coverage in the proposed plan, since most currently get an income-based subsidy.
- A portion of current off-exchange nonsubsidy enrollees, most of whom we assumed to be in the 40-64 age-band, will drop coverage because of the impact of the wider rate band.