KEY POINTS:

Major drivers of 2017 premium changes include: the underlying growth in health care costs, the sunset of reinsurance program funds, how assumptions regarding the composition of the 2017 risk pool differ from those assumed for 2016, and the one-year moratorium on the health insurance provider fee.
 
Other factors potentially contributing to premium changes include the repeal of the expansion of the small group market and the shift to narrower provider networks.
 
Average premium rate changes may not represent the rate change experienced by a particular consumer. A number of factors can result in a consumer’s premium differing from the average rate change, including changes in: age, tobacco status, geography, benefit design, family status, and subsidy eligibility.

 
 

 
1850 M Street NW
Suite 300
Washington, DC 20036
202-223-8196 | www.actuary.org

© 2016 American Academy of Actuaries All rights reserved.

Members of the Individual and Small Group Markets Committee include: Karen Bender, MAAA, ASA, FCA, chairperson; Barbara Klever, MAAA, FSA, vice-chairperson; Eric Best, MAAA, FSA; Philip Bieluch, MAAA, FSA, FCA; Joyce Bohl, MAAA, ASA; Frederick Busch, MAAA, FSA; April Choi, MAAA, FSA; Sarkis Daghlian, MAAA, FSA; Richard Diamond, MAAA, FSA; James Drennan, MAAA, FSA, FCA; Rebecca Gorodetsky, MAAA, ASA; Audrey Halvorson, MAAA, FSA; Juan Herrera, MAAA, FSA; Raymond Len, MAAA, FCA, FSA; Rachel Killian, MAAA, FSA; Kuanhui Lee, MAAA, ASA; Timothy Luedtke, MAAA, FSA; Barbara Niehus, MAAA, FSA; Jason Nowakowski, MAAA, FSA; James O’Connor, MAAA, FSA; Bernard Rabinowitz, MAAA, FSA, FIA, FCIA, CERA; David Shea, MAAA, FSA; Steele Stewart, MAAA, FSA; Martha Stubbs, MAAA, ASA; Karin Swenson-Moore, MAAA, FSA; David Tuomala, MAAA, FSA, FCA; Rod Turner, MAAA, FSA; Cori Uccello, MAAA, FSA, FCA; Dianna Welch, MAAA, FSA, FCA; and Tom Wildsmith, MAAA, FSA.
For a print-ready PDF of this page, click here.

Drivers of 2017 Health Insurance Premium Changes

The 2017 health insurance premium rate filing process is underway. This issue brief outlines factors underlying premium rate setting generally and highlights the major drivers behind why 2017 premiums could differ from those in 2016. It focuses primarily on the individual market, but some factors that are unique to the small group market are highlighted as well.
 

Premiums Reflect Many Factors

Actuaries develop proposed premiums based on projected medical claims and administrative costs for pools of individuals or groups with insurance. Factors that affect proposed premiums include:

WHO IS COVERED—THE COMPOSITION OF THE RISK POOL. Pooling risks allows the costs of the less healthy to be subsidized by the healthy. In general, the larger the risk pool, the more predictable and stable premiums can be. But the composition of the risk pool is also important. Although the Affordable Care Act (ACA) now prohibits insurers from charging different premiums to individuals based on their health status, premium levels reflect the health status of the risk pool as a whole. If a risk pool disproportionately attracts those with higher expected claims, premiums will be higher on average. If a risk pool disproportionately avoids those with higher expected claims or can offset the costs of those with higher claims by enrolling a large share of lower-cost individuals, premiums will be lower.

PROJECTED MEDICAL COSTS. Most premium dollars go to medical claims, which reflect unit costs (e.g., the price for a given health care service), utilization, the mix and intensity of services, and plan design. Unit costs and utilization can vary by geographic area and from one health plan to another depending on the ability and leverage of the insurer to negotiate fees with health care providers.

OTHER PREMIUM COMPONENTS. Premiums must cover administrative costs, including those related to insurance product development, sales and enrollment, claims processing, customer service, and regulatory compliance. They also must cover taxes, assessments, and fees, as well as profit (or, for not-for-profit insurers, a contribution to surplus).

LAWS AND REGULATIONS. Laws and regulations, including the presence of risk-sharing programs, can affect the composition of risk pools, projected medical spending, and the amount of taxes, assessments, and fees that need to be included in premiums.


Major Drivers of 2017 Premium Changes

UNDERLYING GROWTH IN HEALTH CARE COSTS. The increase in costs of medical services and prescription drugs—referred to as medical trend—is based on not only the increase in per-unit costs of services, but also changes in health care utilization and changes in the mix of services. Medical trend is expected to rise slightly faster than in previous years but remain low relative to historical levels. Some uncertaintyremains regarding the causes of the recent low medical trends and whether they will continue. Structural changes to the health care payment and delivery system might be contributing to slower medical spending growth—such as a greater focus on cost-effective care.
 
Costs for prescription drugs continue to increase and are anticipated to again outpace the costs for other medical services. More high-cost specialty drugs are expected to come to market (e.g., new drugs to treat cancer). Some drugs (e.g., Crestor, Benicar, Symbicort) are coming or have recently come off patent and will over time reduce drug costs; however, price decreases aren’t necessarily immediate because generic competition for drugs coming off patent is often limited or slow to be adopted. The impact could be further mitigated if patients are moved by their physicians to newer, higher-cost alternative drugs.

SUNSET OF REINSURANCE PROGRAM FUNDS. The ACA transitional reinsurance program provides payments to plans in the individual health insurance market, with payments declining over the three years of the program, from 2014 to 2016. The year 2017 will be the first year in which there is no reinsurance in the individual market supported by contributions from health plans under the ACA.1 By offsetting a portion of claims, the reinsurance program lowered premiums, and each year the gradual reduction in reinsurance funding resulted in a corresponding increase in premiums. The final impact of the program on premiums will occur in 2017, when projected claims are expected to increase by 4 to 7 percent due to the reinsurance program ending in 2016.

CHANGES IN THE RISK POOL COMPOSITION AND INSURER ASSUMPTIONS. The ACA requires that insurers use a single risk pool when developing premiums. Therefore, as in previous years since the ACA’s enactment, premiums for 2017 will reflect insurer expectations of medical spending for enrollees both inside and outside of the marketplace (i.e., exchanges). Health insurance premiums are set at the state level (with regional variations allowed within a state) and are based on state- and insurer-specific experience regarding enrollment volume and composition. Changes in premiums between 2016 and 2017 will reflect expected changes in the risk profiles of the enrollee population, as well as any changes in insurer assumptions based on whether experience to date differs from that expected in assumptions underlying prior premiums.
 
Although enrollment in the marketplaces has increased somewhat over time, it is uncertain the extent to which the enrollee risk profile has changed as a result. According to the Department of Health and Human Services, marketplace enrollment at the end of the open enrollment period increased from 8.0 million in 2014 to 11.7 million in 2015 and 12.7 million in 2016.2 Average health costs for a given population in a guaranteed issue environment generally can be viewed as inversely proportional to enrollment as a percentage of the eligible population. Higher individual market participation rates will tend to be associated with lower average costs, and lower participation rates with higher average costs. This is because those previously uninsured individuals with greater health care needs are more likely to enroll and to enroll sooner than those with lesser needs. Higher take-up rates typically reflect a larger share of healthy individuals enrolling. The Kaiser Family Foundation estimates that 2016 enrollment represents 46 percent of the potential enrollment.3
 
Premiums for 2017 will reflect insurer expectations for enrollment changes from 2016 to 2017 as well as any adjustments to assumptions if 2016 enrollment differed from expected. Insurers that expected higher enrollment in 2016 than what actually occurred might need to adjust their assumed average costs upward for 2017; those that expected lower-than-actual enrollment might need to adjust their average costs downward. In addition, there will be downward pressure on premiums if insurers expect significantly increased enrollment in the market as a whole in 2017.
 
Insurers have more information now than they did last year regarding the risk profile of the enrollee population and are using that information to adjust their 2017 assumptions accordingly. Because the ACA risk adjustment program shifts funds among insurers depending on the relative health status of an insurer’s population to that of the entire market, premiums need to reflect not only an insurer’s expected claims, but also any expected risk adjustment receipts or transfers. In other words, premiums should reflect the risk profile of the entire state risk pool, not just the insurer’s expected enrollment. When filing premiums for 2016, insurers had information on their enrollee demographics and health spending in 2014 and 2015, but lacked information regarding the risk profile of the market as a whole. Since that time, the Centers for Medicare and Medicaid Services (CMS) released information regarding payments and receipts under the risk adjustment, reinsurance, and risk corridor programs for the 2014 plan year, as well as interim risk adjustment information for 2015. These data provide insurers more information regarding how their enrollee risk profiles compared to those of the whole market. Analysis of the risk adjustment data suggests that some insurers may have set premiums low relative to the market-wide risk profile.4 In addition, the risk corridor results reveal that for many insurers, 2014 premiums were too low relative to actual claims. Some of this understatement was likely due to the implementation of the transitional policy that allowed individuals to keep their prior non-ACA-compliant coverage. Risk profile assumptions for 2017 premiums will reflect these results to the extent that they have not already been factored into prior premium increases.
 
As mentioned above, subsequent to 2014 premiums being finalized, states were allowed to adopt a transitional policy that allowed non-ACA-compliant plans to be renewed in 2014. The policy was subsequently extended until the end of 2017. A majority of states allowed insurers to renew non-ACA-compliant policies and most, but not all, have allowed the extension through 2017. In states with the transition policy, ACA-compliant plans exhibited less favorable experience because lower-cost individuals were more likely to retain their prior policies. Insurers already knew about the transition policy when developing their 2015 premiums, so any related premium increases likely have already been incorporated. To the extent they have not, or if the impact of the transitional policy is expected to change over time, assumptions for 2017 premiums will be revised accordingly.
 
Although more information is now available to insurers regarding their own enrollment and claims experience as well as the market-level experience, some uncertainty remains. The individual market undergoes considerable enrollment turnover as individuals move among different plans within the individual market or among individual market coverage and other coverage (e.g., employer coverage, Medicaid).5 This turnover limits the ability to use 2014 and 2015 experience data to project risk profiles in 2017. Furthermore, CMS emphasized that the 2015 interim risk adjustment results are preliminary, incomplete, and could change materially when the final risk adjustment process is performed later in the year. In addition, experience for insurers with limited market share in 2015 might not be indicative of future experience as they may be more subject to random fluctuations or selection bias.

HEALTH INSURER FEE. The health insurance provider (HIP) fee was enacted through the ACA. The HIP fee is scheduled to collect $11.3 billion in 2016, and insurers built this cost into their premiums. The Consolidated Appropriations Act of 2016 included a moratorium on the collection of the health insurer provider fee in 2017. Insurers will remove the cost of this fee in their 2017 premiums, resulting in a reduction in expected premiums by about 1 to 3 percent, depending on the size of the insurer and their profit/not-for-profit status.
 
REPEAL OF EXPANSION OF THE SMALL GROUP MARKET. In the current small group health insurance market, small employers are those employing up to 50 employees. The ACA as originally enacted called for the expansion of the definition of small employers to include those with up to 100 employees for plan years beginning in 2016. Insurers set their 2016 premiums based on an expectation that this expansion would occur. However, in October 2015, the Protecting Affordable Coverage for Employees Act was signed into law, resulting in the definition of small employers remaining at 50 or fewer employees except in states that elect to extend the definition to include employers with up to 100 employees. To date, California, Colorado, New York, and Vermont have chosen the higher number. All other states remain at 50.
 
For states remaining at 50, the impact on 2017 premium changes depends on whether and how the 2016 premiums were adjusted based on the expectation of groups sized 51-100 entering the market. Some insurers did not expect the experience of groups sized 51-100 to differ significantly from smaller groups and therefore did not materially adjust their 2016 premiums for this expansion.6 In that case, there would be little or no impact on 2017 premium increases. Others expected higher levels of claims for groups sized 51-100 choosing to purchase ACA-compliant plans, in part because the healthier groups would have an incentive to forgo ACA-compliant plans and self-insure instead. Groups sized 51-100 are rated in part based on their own experience. Those with poor experience would have higher-than-average premiums while healthier groups would have lower premiums. Because small group, ACA-compliant plans are not rated based on experience, the healthier groups would have faced premium increases had they purchased ACA-compliant plans. These potential premium increases might have caused many of the healthier groups to self-insure, while groups with poor experience would be more attracted to ACA plans. The removal of this adjustment in 2017 will exert a slight downward pressure on premiums for insurers that adjusted 2016 small group premiums upward based on this expectation.7
 
For insurers in the four states that elected to include employers with up to 100 employees in the small group market, there would be no impact on 2017 premiums unless the insurer’s assumptions as to the impact of the expansion on experience have changed. Such a change in assumptions may be unlikely because they will have had very little 2016 experience at the time 2017 premiums are filed.

CHANGES IN PROVIDER NETWORKS. Since the ACA marketplaces became operational in 2014, many insurers have been shifting to narrower provider networks to help keep premiums affordable. Narrower networks can give insurers more leverage to negotiate lower provider payment rates, and they also can be used to direct enrollees to more cost-effective and high-quality providers. Not only do broader network plans tend to have higher provider reimbursement rates, but health plan experience for 2014 and 2015 suggests that preferred provider organization (PPO) plans and those with broader network choice have had worse experience than narrower network plans. That has led to some health plans eliminating PPO plans on the marketplaces rather than raise premiums further.
 
These developments have created a mix of complex forces related to changes in provider networks and their effect on 2017 premiums. The elimination by some companies of plans with broad provider networks may put additional upward pressure on premium increases for remaining broad network plans in the market if the insured moving from plans that were terminated are less healthy than the current membership of the persisting broad network plans and if risk adjustment transfers do not fully offset the increase in morbidity or utilization. It is possible premiums could be reduced if the transferred membership is relatively healthier than the current membership and the risk adjustment transfer does not offset this improved health status change.
 
The elimination of broad network plans might put some upward pressure on the premiums of narrow network plans because those enrollees who contributed adverse experience to the richer network plans will likely need to move to a narrower network plan for their coverage if alternative broad network plans are not available or affordable.
 
Additionally, CMS has increased its focus on the makeup of provider networks to ensure that they provide adequate access and do not penalize consumers for using non-network providers of which they had no choice or knowledge (e.g., hospital-based specialists, radiologist, pathology labs, still-listed physicians who actually left the network). Any changes to the composition of existing narrow networks or revisions to the adjudication process of claims related to use of non-network providers in order to meet these expectations could result in upward pressure on premiums.
 

Other Drivers

BENEFIT PACKAGE CHANGES. Changes to benefit packages (e.g., through changes in cost-sharing requirements or benefits covered) can affect claim costs and therefore premiums. This can occur even if a plan’s metal level8 remains unchanged. In particular, changes in benefits or cost-sharing requirements may have been needed to comply with the metal-level determinations using the actuarial value (AV) calculator, which was recalibrated for 2017. Other changes in benefit packages could be made based on market or other considerations. Such changes could put upward or downward pressure on premiums, depending on the particular change. Other plan design features, such as drug formularies and care management protocols, also could affect premium changes.
 
RISK MARGIN CHANGES. Insurers build risk margins into their premiums to reflect the level of uncertainty regarding the costs of providing coverage. These margins provide a cushion should costs be greater than projected. Greater levels of uncertainty typically result in higher risk margins and higher premiums. Changes to the level of uncertainty regarding claim costs or other aspects of ACA provisions can cause changes to the risk margins.
 
The ACA risk corridor program sunsets at the end of 2016. The program was intended to minimize the risk margins built into premiums by having the federal government share in the plan losses and gains beyond a 3 percent corridor. Because of legislative changes enacted after the initial passage of ACA that required the program to be budget neutral, it is unclear how effective the program was in reducing 2016 premium risk margins. However, any reductions that were reflected in 2016 premiums due to this program will be eliminated in 2017 pricing, likely resulting in slightly higher premiums.
 
MARKET COMPETITION. Market forces and product positioning also can affect premium levels and premium increases. Insurers might withstand short-term losses in order to achieve long-term goals. Due to the ACA’s uniform rating rules and transparency requirements, premiums are much easier to compare than before the ACA, and in previous years some insurers lowered their premiums after they were able to see competitors’ premiums. However, underpricing in any one year could drive premium increases higher in future years because, in the long run, premiums need to adequately cover claims and expenses. Health insurers are increasingly focused on local competition, offering coverage only in geographic regions in which they are believe they have a competitive advantage. As such, there may be more price competition in those regions where many health plans are offered, and less price competition where fewer health plans participate.

CHANGES IN PROVIDER COMPETITION AND REIMBURSEMENT STRUCTURES. Consolidation of health care providers has been ongoing in many local markets, largely for the purpose of increasing providers’ negotiating power. This trend is likely to continue. Any increased negotiating power among providers could put upward pressure on premiums. On the other hand, mergers of health care plans can have the opposite effect if they increase health plans’ negotiating leverage with providers. It is also notable that insurers are pursuing changes in provider reimbursement structures that move from paying providers based on volume to paying based on value. For example, accountable care organization structures offer incentives to provide cost-effective and high-quality care. Such efforts could put downward pressure on premiums, at least in the short term.

CHANGES IN ADMINISTRATIVE COSTS. Changes in administrative costs will also affect premiums. For instance, changes can result from increased costs associated with ACA implementation or from spreading fixed costs over a different enrollment base than projected. Moreover, as the ACA reforms have gone into effect, the important role that brokers can play has been acknowledged, and reductions in commissions that may have been expected generally have not been realized. However, some plans have decided to eliminate or dramatically reduce commissions for marketplace sales, at least outside of the open enrollment period. This may help reduce premiums, not only because of the lower administrative costs, but also due to the expectation that there may be less adverse selection during the year. On the other hand, some health plans are finding that increased regulatory requirements associated with the administration of provisions in the ACA are increasing their administrative costs. These costs all need to be reflected in premiums. Depending on the circumstances in any particular state, these changes in marketing and administrative costs can put upward or downward pressure on premiums. However, the ACA’s medical loss ratio requirements limit the share of premiums attributable to administrative costs and margins.

CHANGES IN GEOGRAPHIC FACTORS. Within a state, federal rules allow health insurance premiums to vary across geographic regions established by the state. Insurers can use different geographic factors to reflect provider cost and medical management differences among regions, but are not allowed to vary premiums based on differences in health status (which should be accounted for by the single state risk pool construct and risk adjustment process). An insurer might change its geographic factors due to changes in negotiated provider charges and/or in medical management of some regions compared to others. A decision to increase or decrease the number of regions in which the health plan intends to offer coverage in 2017 within a state could also result in a change in its geographic factors. Another key reason for changes in geographic factors could be new provider contracts that reflect different relative costs. A realignment of these differences could result in changes across the rating regions within a state.
 

Summary

The 2017 health insurance premium rate filing process is underway and how 2017 premiums differ from those in 2016 will depend on many factors. Key drivers include the underlying growth in health costs and the sunset of the reinsurance program, each of which will increase premiums relative to 2017. Another key driver is changes to the risk pool composition and insurer assumptions from 2016. Insurers have more information than they did previously regarding the risk profile of the enrollee population and will revise their assumptions for 2017 accordingly. However, some uncertainty remains, as a market equilibrium in terms of enrollment levels and risk profiles likely has not yet been reached. The one-year moratorium on the health insurer fee will reduce premiums relative to 2016.
 
Other factors potentially contributing to premium changes include the repeal of the expansion of the small group definition and modifications to provider networks. In addition, changes to provider reimbursement structures, benefit packages, risk margins, administrative costs, and geographic region factors can affect premium changes. Insurers also incorporate market competition considerations when determining 2017 premiums.
 
Premium changes faced by individual consumers will also reflect increases in age, and any changes in geographic location, family status, or benefit design. In addition, if a consumer’s particular plan was discontinued, the premium change will reflect the increase or decrease resulting from being moved into a different plan, which could be at a different metal level. Average premium change information released by insurers or states could reflect the movement of consumers to different plans due to their prior plan being discontinued.
 

PREMIUM CHANGES FROM A CONSUMER PERSPECTIVE

Premium changes are often the most visible and discussed aspect with respect to the ACA impact on health insurance. However, premium changes can be measured using different approaches, making it difficult to compare premium changes among health insurers, among plans offered by an insurer, or among consumers.
 
In addition, the average premium change within a specific insurer may not represent the premium change experienced by a particular consumer. The ACA requires that premiums vary only by age, tobacco use, geographic location, family status, and benefit design. Premium changes from a consumer perspective can then result from underlying medical trends and other aggregate premium factors, as well as changes in these consumer-specific factors. The following situations could result in a premium change a consumer experiences differing from the average premium change reflected in a premium rate filing.

Changes in Age
All insurers are required to use a prescribed age rating curve (either the federal default curve or a state-established curve) when determining how to vary premiums by age. In other words, premium variations by age are the same regardless of insurer. Most individual consumers will experience a premium increase each year, due to aging one year. Such a change (on the order of 2 to 3 percent per year for individuals older than 24) is rarely included in insurer-level premium change calculations because it does not represent a change in the underlying factors. But it is a change a consumer would experience.

Tobacco Status
In most states, insurers are allowed to charge smokers more than similar nonsmokers, and this surcharge can vary by state and by age. In other words, older smokers can face higher surcharges than younger smokers (or vice versa). In plans that vary the surcharge by age, consumers who smoke will see a premium change due to the change in the tobacco use surcharge. In addition, consumers who have either started or stopped using tobacco products could see a premium change.

Changes in Geographic Location
All states require the use of rating areas prescribed by the CMS. Insurers are not allowed to change the rating areas but are allowed to change how premiums vary across areas due to differences in relative provider charge levels and differing levels of medical management. Such a change may or may not be included in the average aggregate premium change from the insurer’s perspective, but it is a change a consumer would experience depending on where they live. If a consumer moves from one rating area to another, that also may result in a premium change.

Changes in Benefit Design
A plan’s benefit design encompasses both the benefits covered as well as the associated cost-sharing requirements (e.g., deductibles, coinsurance, copayments). Consumers who switch to a new benefit design or are re-enrolled in a different plan due to discontinuance of a plan could experience a premium change due to the benefit design change. If an insurer discontinues offering plans at a metal level, such as platinum or bronze, consumers in those plans may be re-enrolled in the next higher or lower metal level plan, which could significantly impact the premium. Insurers also might change covered benefits or cost sharing (subject to uniform modification provisions of guaranteed renewability) in order to offset medical trend or maintain the metal level.

Family Status
The ACA allows premiums to vary by family size. Family premiums reflect the premiums for each covered adult plus the premiums for each of the three oldest covered children younger than 21. Therefore, consumers with family coverage who experience a change in family composition could face a premium change.
 
Subsidy Eligibility
The ACA provides premium subsidies in the individual market based upon household income. Changes in income alone can result in upward or downward changes in the net premiums that any specific consumer may have to pay, even if there is no change in the underlying premiums. A change in available plans offered in the market also could affect the subsidy an individual receives.

For a print-ready PDF of this page, click here.
 

1 Funding for the reinsurance program came from contributions required by the ACA from all health plans, including not only plans in the individual market, but also those in the small and large group markets, as well as self-insured plans. These contributions were used to make payments to ACA-compliant plans in the individual market.
2 See enrollment reports for 2014, 2015, and 2016; Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation. Enrollment figures are understated because they do not include off-marketplace enrollment in ACA-compliant plans, and overstated because they reflect plan selection only, with or without payment of premium.
3 Larry Levitt, Gary Claxton, Anthony Damico, and Cynthia Cox, “Assessing ACA Marketplace Enrollment,” Kaiser Family Foundation, March 4, 2016.
4 For more details, see the Academy’s analysis, Insights on the ACA Risk Adjustment Program.
5 According to Avalere, only one-third of individual market exchange enrollees in 2016 were in the same plan from 2015.
In states where it was available, the transition policy allowing small groups to retain non-ACA-compliant coverage was expected to mitigate the impact of the expansion of the small group definition on 2016 premium changes.
Some insurers that initially adjusted 2016 premiums based on the expectation of covering groups sized 51-100 removed that adjustment for small group policies issued or renewed in the second and later quarters of 2016. In that situation, only groups renewing coverage in the first quarter of 2017 will see a difference in premiums as a result of the adjustment being removed.
ACA plans are categorized into four metal tiers (bronze, silver, gold, and platinum), based on the relative level of plan generosity. Actuarial value is used to measure plan generosity, and is based on the average share of medical expenses that a plan will cover, as opposed to being paid out of pocket by the consumer. In turn, actuarial value is measured using the AV calculator released by the CMS.