Drivers of 2021 Health Insurance Premium Changes: The Effects of COVID-19
Download a PDF version here.
The 2021 individual and small group health insurance premium rate filing process is underway. Actuaries develop proposed premiums based on their projections of medical claims and administrative costs for pools of individuals or groups with insurance. Projected medical claims reflect unit costs and utilization levels, as well as the mix and intensity of services, all of which can vary by geographic area and from one health plan to another. The composition of risk pools is also important, as medical claims will reflect the health status of individuals in the risk pool. Laws and regulations—such as benefit requirements, issue and rating rules, and risk-sharing programs—can affect the composition of risk pools and projected medical spending, as well as the amount of taxes, assessments, and fees that need to be included in premiums.
The COVID-19 pandemic has introduced new uncertainties into the development of premium rates for 2021. Each year, the American Academy of Actuaries Individual and Small Group Markets Committee publishes a public policy issue brief outlining the major factors driving premium changes for the next plan year. Unlike those previous issue briefs, this year’s issue brief focuses primarily on the impact of COVID-19 on the 2021 premium rate filings.1
The typical rating factors still apply, but issues surrounding the COVID-19 pandemic are a major consideration for rate setting and will impact both the individual and small group markets. To date, the effects of the pandemic have varied significantly by region, both in acuity and duration. While this issue brief broadly addresses COVID-19 considerations, we anticipate that the actual impact of these considerations on rate filings will reflect specific regional and market conditions.
COVID-19 Impact on 2020 Claims Experience
The Affordable Care Act (ACA) rate review process typically requires issuers to develop premium rates based on the plan year of experience, which is two years prior to the pricing plan year, adjusted to reflect expected differences between the experience plan year and the pricing plan year. Thus, 2021 rate development will primarily be based on 2019 (pre-U.S. pandemic) experience with adjustments to project the experience forward to 2021. Issuers consider emerging experience from the current plan year while setting rates, so 2021 rate development would normally be informed by year-to-date 2020 plan experience. COVID-19 has introduced considerable uncertainty into that 2020 experience, and this uncertainty is extremely likely to continue into 2021.
COVID-19 is resulting in high-cost hospitalizations, and these costs have the potential to be material. Direct COVID-19-related health spending is highly dependent on the percentage of the population that is infected and the percentage of those individuals who are hospitalized. For instance, the USC-Brookings Schaeffer Initiative for Health Policy estimates that a COVID-19 infection rate of 5% could increase claims in the commercial insurance markets by about 1%; while a COVID-19 infection rate of 60% could increase commercial claims by 4% to 11%.2 Cost-sharing for COVID-19 testing and related services is being waived pursuant to federal legislation.3 Some carriers are additionally waiving cost-sharing for COVID-19 treatments and/or certain telehealth services as well.
Direct spending for COVID-19 will be offset, at least in part, by reductions in other services. The pandemic has led to significant social distancing requirements, and utilization of many services such as office visits has declined dramatically. In addition, non-emergency hospital services, which include elective surgeries that typically generate significant revenues for providers, have also declined, due to social distancing, state restrictions on elective procedures, and a desire to free up space for COVID-19 patients. Emergency services have experienced a decline, possibly due to patient concerns around contracting the virus. Some practices have expanded availability of telehealth services in order to fill in some of the gaps in office visits. However, many services cannot be provided through telemedicine, particularly those elective surgeries that help support hospitals’ and physicians’ financial stability.
Due to the widespread nature of social distancing requirements, there is evidence that the decrease in costs due to deferred and avoided services has occurred across the nation and is not limited to areas hard hit by COVID-19.4 However there is significant variation at the state and local level as to when and how medical providers may begin offering these services again and many providers may be subject to capacity restrictions. Additionally, it is unclear how many of the missed services will return and how many will be eliminated outright. While there are still many sources of uncertainty as to the long-term impact of COVID-19 on other medical care, it is clear at this point that these deferred and avoided services have reduced health care utilization in the first half of 2020. To date, it appears likely that the impact of deferred and avoided care has outweighed cost increases in the commercial market related to direct COVID-19 diagnosis and treatment costs, including cost-sharing waivers in most areas.
While medical care has been significantly affected in early 2020, prescription drug spending appears less likely to be significantly impacted, at least for now. Pharmacy spending could decrease if people are unable to afford their prescriptions due to loss of income and if patients continue to avoid going in for office visits. On the other hand, prescription drug spending could increase if there are new COVID-19 drug therapies and/or a vaccine becomes available.
It’s unknown how trends will continue through the rest of 2020, but high rates of deferred and canceled care could continue, even as the availability of non-emergent care is increasing in many geographic areas. The arrival of another COVID-19 wave in the second half of the year could further increase care deferrals.5 Claims levels are likely to be impacted by the continued duration and severity of the COVID-19 pandemic, the degree of compliance with social distancing guidelines and any resulting deferred and avoided care, utilization levels of COVID-19 testing and related cost-sharing waivers, the duration of the coronavirus public health emergency, and other factors. To date, the prevalence of COVID-19 has varied dramatically by region, indicating that the costs of covering COVID-19 related claims may also vary by region. These same considerations may affect net impacts on 2021 expenditures as well.
Drivers of 2021 Rate Changes
When developing 2021 health insurance rates, insurers are likely to run multiple scenarios involving different assumptions on if any new COVID-19 waves will emerge later in 2020 or in 2021, the degree of deferred and avoided services, the amount of testing (including antibody testing), the cost and availability of vaccines, and other factors relevant to their enrolled population to inform their premium development.6 Greater degrees of uncertainty could lead to more conservative assumptions and risk margins for some insurers. In many states, health insurers are permitted to file updated rates on a quarterly basis in the small group market, which could reduce the need for conservatism. However, individual market rates are filed annually and cannot be updated during the calendar year.
Changes in Risk Pool Composition Due to Economic Impacts of COVID-19
The composition of the 2021 individual market is likely to be volatile and may see significantly different underlying experience than in 2019; there is likely to be some level of influx of individuals who lost employer-sponsored coverage due to the economic downturn resulting from the COVID-19 pandemic. While many individuals who lose income may qualify for Medicaid, some will not, particularly in states that have not expanded Medicaid. An increase in enrollment may be partially offset by individuals who leave the individual market, particularly non-subsidy-eligible individuals who leave the individual market due to unaffordability or subsidy-eligible individuals who become eligible for Medicaid.7
Even if the net enrollment change is small, the underlying morbidity level may change depending on the characteristics of those leaving and those entering the market. Individuals with employer coverage are generally thought to be healthier than people with coverage in the individual market. On the other hand, coverage transitions can result in adverse selection. For instance, when individuals lose coverage, they must decide whether to purchase coverage, and less-healthy people are generally thought to be more likely to purchase coverage than healthy individuals. During the Great Recession of 2008–2009, COBRA8 coverage was subsidized by the federal government. Although similar subsidies are being considered as part of current legislative efforts, as of this publication, such a provision has not been part of any of the coronavirus relief legislation passed into law. In the absence of significant COBRA subsidies that facilitate the ability of workers losing jobs to maintain their prior employer coverage, previous COBRA experience may be an appropriate proxy for the morbidity of members moving into the individual market.
However, the COVID-19 pandemic may have increased the perceived value of insurance, thereby reducing adverse selection among people moving from employer coverage to the individual market. In addition, healthy uninsured individuals could be more likely to obtain coverage.
Small Group Market
Small employers are less likely to offer coverage than large employers, and the economic downturn has the potential to accelerate this trend. During past recessions, some insurers have seen increased morbidity in insureds among employers that retain coverage, suggesting that employer plans that stayed in force had less-healthy members than those that lapsed. The ACA’s single risk pool provisions create additional exposure to insurers beyond what was present during past recessions, particularly if small employers with healthier workers are more likely to drop coverage. Morbidity increases could also occur if less-healthy COBRA-eligible employees9 who suffer job losses are more likely to sign up for COBRA. This effect could be magnified with the extension of the COBRA election period as well as the extension of the window for timely premium payments during the national emergency period generated by COVID-19.10 As with the individual market, adverse selection in the small group market might be reduced due to the health-related nature of this particular crisis, with more value being placed on retaining health coverage, even if the small group market shrinks due to small employers going out of business.
COVID-19 Treatments and Testing Costs
Ongoing experience in 2020 on COVID-19 has provided insurers with some information on the cost of treatment of COVID-19. However, there is still significant uncertainty regarding COVID-19 treatment costs per case for 2021 due to the possibility of new treatment therapies, antibody tests, and/or vaccines, as well as the overall mix of case severity should the virus persist and doctors refine best practices.
The cost of testing is also uncertain and could be significant if insurers are required to cover the cost of testing for public health and occupational safety reasons, which goes beyond the diagnostic testing of an individual for diagnosis and treatment typically covered under health insurance. Reopening the economy requires more and more frequent COVID-19 testing. Employers may have to frequently test employees until a vaccine is available. At this time, it is unclear who will pay for these tests—health plans, employers as a business expense—or whether there will be federal funding. The coverage of testing for public health reasons rather than for the diagnosis and treatment of an individual may require government funding, otherwise it could add to insurance premiums.
Medical services deferred during 2020 could be delayed into 2021. However, the amount of services provided could be limited by capacity restraints, whether due to social distancing guidelines or facility-specific limitations. Many providers could seek to increase capacity by extending hours in order to accommodate more of the deferred services. This could even potentially result in minimal deferred service load in 2021 if providers can clear any service backlog in the second part of 2020. On the other hand,patients may be reluctant to seek care, especially if COVID-19 infections are ongoing and additional significant waves of infections occur, pushing more deferred care into 2021 or increasing outright care forgone.
There is evidence that some essential services are also being deferred.11 This leads to the potential that individuals with chronic conditions could see a degradation of their health status, resulting in higher future costs on a per-member basis. In addition, many preventive services such as vaccinations and cancer screenings are being avoided, which could lead to increased future illness or condition severity.12
While telehealth services have been used to help fill in some of the service gaps in 2020, there is still uncertainty as to whether that increase will continue to replace certain office visits or whether treatment patterns will return to pre-COVID-19 levels. Prior to COVID-19, telemedicine services were typically reimbursed at a lower unit cost than similar in-office services. In some cases, providers are currently receiving the standard in-office payment rates, and changes to reimbursements may influence provider incentives relative to telemedicine. Additionally, telemedicine is not likely as comprehensive as an in-person visit for certain services, and as such could lead to increased utilization if it takes longer to identify and treat health conditions in this medium. There could also be increased utilization due to the convenience of telehealth compared to an office visit if providers begin to offer the option more broadly.
Mental health services may be more in demand. Stay-at-home orders have separated individuals from their normal support systems and social interactions. Economic factors, including the loss of some or all of household income, as well as increased child care and home schooling responsibilities, have put significant strain on household mental welfare. There are concerns that alcohol use has risen since the start of the pandemic,13 which could increase the need for substance use disorder services. In addition, health- related factors also contribute to patient stress. Beyond the increased general worry about health, patients that spend long periods of time on ventilators are showing signs of post-traumatic stress disorder (PTSD). These and other stressors have resulted in a general increase in anxiety and depression,14 which will likely result in increased short- or long-term mental health service needs, much of which is reasonable to assume will continue into 2021.
Provider Reimbursement Rates
As they face financial difficulties during the COVID-19 outbreak, providers, especially hospitals, may have success negotiating (and renegotiating) higher payments from payers, depending in part on what federal relief they might have received. Some payment increases may be temporary in nature, thereby affecting 2021 costs only minimally. However, to the extent that there are changes in service utilization that impact the mix of services that providers rely on to achieve desired profitability targets, some services could see more permanent payment rate increases as providers seek to offset the lost revenue streams.
Increases in Medicaid enrollment are likely to result in providers who accept Medicaid seeing a greater proportion of Medicaid patients. These providers may attempt to make up for lower Medicaid payment rates by negotiating higher commercial payment rates.
Medical Loss Ratio (MLR) Impacts
To date, claims in 2020 have widely been reported to be significantly below priced levels. If claims over the full year continue to remain below expectations, premium rebates may be required under the MLR. The MLR calculation is done as a three-year average. 2018 and 2019 saw significant MLR rebates in the individual market, which could increase the potential for rebates if 2020 claims levels are lower than expected. Also, the MLR is one- sided: Carriers do not receive payments if claims come in significantly higher than 2020 expectations.
Issuers may consider projected MLR rebates when setting their 2021 rates, especially given the level of MLR rebates expected for 2019.15 This consideration could be given additional weight if the issuers anticipate owing rebates for 2020 given their expectations regarding the net impact of COVID-19. For instance, issuers could reduce the level of conservatism in rate filings to reduce the possibility of owing rebates for 2021. On the other hand, issuers may be less concerned about having rates that end up being too high, relying on the MLR to return any excess premiums to enrollees, particularly in markets where an issuer is able to maintain a competitive position without significant reduction to rates.
COVID-19 could additionally impact state ACA Section 1332 waivers. For instance, some states fund their reinsurance programs through insurer assessments, and these assessments fall primarily on group insurance. The economic downturn could result in declines in group market enrollment and an increase in individual market and Medicaid enrollment. This could lead to a decrease in reinsurance funding with the potential of an increase in reinsurance claims. For states that are allowed to retrospectively adjust the coinsurance or other reinsurance program parameters to either achieve a specified cost to the state (e.g., $50 million) or else align the state’s cost with the assessments collected, reinsurance payments would go down on a per-reinsured claims basis and carriers would therefore bear more risk than anticipated in rate filings. In states with reinsurance programs, issuers will need to consider whether and how program funding will affect 2021 reinsurance reimbursements and net 2021 claims.
States have issued varied guidance to meet the challenges of health care coverage brought on by COVID-19 in several areas, including restrictions on prior authorization, extended grace periods or non-cancellation periods, and implementing increased access to telehealth. Many of these public policies are temporary, but if any are anticipated to continue into 2021, issuers may likely consider the impact on rates.
Other Legislative Actions
In response to the COVID-19 pandemic, several federal relief measures were enacted to provide economic assistance to individuals, businesses, health care providers, and state and local governments. The measures contained some requirements for health insurers, such as waivers of cost-sharing for COVID-19 testing. At the time of this publication, additional legislation is being considered. Depending on what, if any, health insurance- related provisions are included, 2021 premiums may be affected.16
Other Premium Drivers
Changes in Federal Taxes and Fees
The Further Consolidated Appropriations Act, 2020,17 enacted in December 2019, included changes related to three federal tax provisions that may impact 2021 rates. Most prominently, the act repealed the Health Insurance Provider Fee (HIPF) for plan years beginning in 2021. Eliminating this fee could reduce premiums by 1% to 3%. Additionally, the act restored the Patient Centered Outcomes Research Trust Fund fee, which could increase premiums by a little less than $3 per member per year, and removed the medical device excise tax, which should produce minimal downward pressure on claims.
Risk Adjustment Data Validation (RADV)
In the final 2020 Health and Human Services Department (HHS) Notice of Benefit and Payment Parameters for 2020, the Centers for Medicare & Medicaid Services (CMS) significantly extended the timeline for RADV payment transfers, with the initial round of payments based on the 2017 benefit year RADV to be made in 2021.18 In the final 2020 Unified Rate Review Template Instructions, CMS indicated that states may elect to allow issuers to reflect RADV transfers in their 2021 premium rates because amounts will be reflected in 2021 calendar year experience for MLR filings. This could result in additional rate volatility for states and issuers with material RADV transfer amounts as a result of 2017 benefit year RADV transfers.
Rate setting in the ACA-compliant individual and small group markets is complex, and pricing actuaries are considering a wide range of factors when determining rate levels. This year, the COVID-19 pandemic has introduced significant additional uncertainty related to 2020 and 2021 claims levels and needed 2021 premium rates.
For the first half of 2020, increased health spending due to the direct costs of diagnosing and treating COVID-19 appears to have been more than offset by a reduction in non- COVID-19 health services. It’s unknown how trends will continue through the rest of 2020. When developing 2021 health insurance rates, insurers are likely to project claims under multiple scenarios involving different assumptions on if any new COVID-19 waves will emerge later in 2020 or in 2021. The economic impacts of COVID-19 could cause shifts in insurance enrollment along with changes in the risk pool composition related to these shifts. COVID-19 testing and treatment costs, the availability of new treatments and vaccines, increases in mental health and substance abuse treatment needs, changes to telehealth utilization and costs, as well as any changes to provider reimbursement rates also will be considered. In addition, the timing of any subsequent COVID-19 waves will affect whether non-COVID-19 utilization continues to be deferred or forgone in 2021 or whether treatment deferred in 2020 is provided in 2021. While new information continues to emerge regarding the epidemiological, economic, and health care impacts of this pandemic, there is still a wide range of potential effects.
 We refer readers to our 2020 premium drivers issue brief for a fuller discussion of the premium drivers that are typically considered each year.
 Matthew Fiedler and Zirui Song; "Estimating Potential Spending on COVID-19 Care"; USC-Brookings Schaeffer Initiative for Health Policy; May 7, 2020. The percentage increase in claims due to COVID-19 is relative to 2020 projected spending prior to the emergence of COVID-19.
 Centers for Medicare and Medicaid Services; FAQs About Families First Coronavirus Response Act and Coronavirus Aid, Relief, and Economic Security Act Implementation Part 42; April 11, 2020.
 Ateev Mehrotra, Michael Chernew, David Linetsky, Hilary Hatch, and David Cutler; "The Impact of the COVID-19 Pandemic on Outpatient Visits: A Rebound Emerges"; The Commonwealth Fund; May 19, 2020.
 Hayley M. Rogers, Charlie Mills, and Matthew J. Kramer; "Estimating the Impact of COVID-19 on Healthcare Costs in 2020: Key Factors of the Cost Trajectory"; Milliman; April 2020.
 The Health Actuarial Task Force of the National Association of Insurance Commissioners released a template to help rate reviewers evaluate COVID-19 pricing adjustments (retrieved June 4, 2020).
 Health Management Associates; COVID-19 Impact on Medicaid, Marketplace, and the Uninsured, by State; April 3, 2020.
 Consolidated Omnibus Budget Reconciliation Act.
 COBRA requirements generally apply to employers with at least 20 employees.
 On May 4, 2020, the Department of Labor put forth policies to extend COBRA timelines for the duration of the National Emergency Declaration declared by the president. See "Extension of Certain Timeframes for Employee Benefit Plans, Participants, and Beneficiaries Affected by the COVID-19 Outbreak"; Federal Register; May 4, 2020.
 Liz Hamel, Audrey Kearney, Ashley Kirzinger, et al.; KFF Health Tracking Poll-May 2020: Impact of Coronavirus on Personal Health, Economic and Food Security, and Medicaid; Henry J. Kaiser Family Foundation; May 27, 2020.
 See for instance Jan Hoffman; "Vaccine Rates Drop Precipitously as Patients Avoid Doctor’s Visits"; New York Times; April 23, 2020.
 Caren Chesler; "As Pandemic and Stay-at-Home Orders Spread, So Does Alcohol Consumption"; Washington Post; April 2, 2020.
 Hamel, op cit.
 Rachel Fehr and Cynthia Cox; Data Note: 2020 Medical Loss Ratio Rebates; Henry J Kaiser Family Foundation; April 17, 2020.
 See the Academy issue brief, Health Insurance Risk Mitigation Mechanisms and COVID-19, for information on the implications of implementing mechanisms such as risk corridors and reinsurance to address COVID-19 related insurer risks.
 See the Academy’s comments on changes to RADV timing in the final 2020 NBPP and related documents for more information about the potential impacts of this change.
Craig Hanna, Director of Public Policy
Members of the Individual and Small Group Markets Committee, which authored this issue brief, include Barbara Klever, MAAA, FSA—Chairperson; Joyce Bohl, MAAA, ASA—Vice Chairperson; Dylan Ascolese, MAAA, FSA; Eric Best, MAAA, FSA; Alfred Bingham, MAAA, FSA; Brent Bish, MAAA, FSA; Frederick Busch, MAAA, FSA; April Choi, MAAA, FSA; Andrea Christopherson, MAAA, FCA, FSA; Richard Diamond, MAAA, FSA; David Dillon, MAAA, FSA; Beth Fritchen, MAAA, FSA; Rebecca Gorodetsky, MAAA, ASA; Audrey Halvorson, MAAA, FSA; Casey Hammer, MAAA, FSA; David Hayes, MAAA, FSA; Kevin Hurley, MAAA, FSA; Shiraz Jetha, MAAA, FCIA, FSA; Jason Karcher, MAAA, FSA; Neil Kelsey, MAAA, FSA; Rachel Killian, MAAA, FSA; Kuanhui Lee, MAAA, ASA; Raymond Len, MAAA, FCA, FSA; Julia Lerche, MAAA, FSA; Timothy Luedtke, MAAA, FSA; Scott Mack, MAAA, ASA; Ryan Mueller, MAAA, FSA; Valerie Nelson, MAAA, FSA; Donna Novak, MAAA, ASA, FCA; Jason Nowakowski, MAAA, FSA; Bernard Rabinowitz, MAAA, FCIA, FIA, FSA; Paul Schultz, MAAA, FSA; David Shea, MAAA, FSA; Martha Stubbs, MAAA, ASA; Ari Szafranski, MAAA, FSA; Tammy Tomczyk, MAAA, FCA, FSA; David Tuomala, MAAA, FCA, FSA; Roderick Turner, MAAA, FSA; Dianna Welch, MAAA, FCA, FSA; and Thomas Wildsmith, MAAA, FSA.