Potential impact of individual market reforms

Articles

A wide range of changes to stabilize the individual market have been proposed. This special report examines the impact some of the initiatives could have on claims costs and enrollment by the uninsured.

Enrollment in the individual market has now reached close to 18 million people, roughly 10 million of whom are on the exchanges.1 However, affordability remains a concern: between 2014 and 2017, the average silver plan annual gross premium (before tax credits) for someone 40 years old increased by an aggregate ~$1,300, and the average annual deductible for the same plan rose by an aggregate ~$900.2 The average individual market enrollee (a 40-year-old with an income of $60,000) now pays ~$3,400 per year in premiums and faces the possibility of paying up to ~$10,250 if he or she hits the out-of-pocket maximum—in other words, 10% to 30% of that person’s income after housing, food, and transportation expenses.

Cost appears to be a major factor explaining why many healthy individuals have opted to sit on the market’s sidelines,3 which has had a negative impact on many health insurers’ financial performance. In 2015, average per member per month (PMPM) medical costs were $515 in the individual market, compared with $449 in the small group market (once adjusted for actuarial value).4 Across the individual market, health insurers reported losses of about $2.7 billion in 2014 and $7.9 billion in 2015; losses for 2016 could reach $8.9 billion.5 However, a subset of insurers have found ways to be profitable in the individual market,6 and the 2016 financials reported to date suggest that the trajectory of losses may be slowing.7 Nevertheless, broader sustainability questions remain.

Achieving a stabilized individual market has the potential to keep insurers in the market and further boost competition. In the early days of the exchanges, there was considerable interest in and market entry into the exchanges, which suggests that the expectation of a well-functioning market could potentially attract more insurer competition.8

Given this backdrop and the presence of a new administration, discussions on how to redesign the market have come to the forefront. Many interested parties have proposed reforms, including Congress, the new administration, state governments, industry participants, consumer groups, and various think tanks.9 Below, we have attempted to categorize and model the potential impact of many (although not all10) of these reforms. A combination of initiatives could potentially have a compounding effect on improving the functioning of the individual market.

Potential actions

Changes to the Affordable Care Act (ACA) itself—or to how it is implemented—may be pursued by the federal government, individual states, or a combination of the two. A wide range of changes have been proposed, which we have categorized based on the type of intended impact (Exhibit 1). We see four major categories:

  • Promote appropriate enrollment: changes that attempt to prevent “gaming” through loopholes in the rules governing enrollment and payment, with the goal of reducing claims (by limiting inappropriate high-cost enrollment) and improving premium realization per enrollee (by increasing member retention within the year)
  • Stabilize risk pools: changes that restructure the form in which risk is pooled or external funding is applied to make both overall premiums and the medical cost trend more consistent and predictable
  • Maximize market participation: changes that encourage the remaining uninsured (who tend to have lower medical expenses) to participate in the market, thereby reducing average premiums
  • Reduce cost through coverage redesign and payment innovation: changes that better align consumer and provider incentives to reduce both supply- and demand-side inefficiency and encourage innovation through competition

A holistic attempt to reform the individual market may include initiatives from several or all of these buckets, and many of the combinations could have a compounding effect on overall impact (Exhibit 2).11 Nevertheless, we think it is useful to examine the initiatives in isolation, as well as in combination, to understand what the magnitude of their potential impact could be. In the following sections, we examine their potential impact in terms of reductions in average claims costs and increases in enrollment. As noted in the sidebar, our modeling uses a multi-year approach that recognizes the dynamic relationship between enrollment and affordability over time.

Note: This report focuses solely on whether the proposed reforms would increase enrollment and lower costs in the aggregate and for the average enrollee (and, if so, by how much). It does not analyze whether specific subsets of individual market enrollees would be better or worse off as a result of the reforms.

Overall approach to impact sizing

  • For each initiative, we attempt to understand how it could change both consumer participation and underlying claims in the individual market.
  • We estimate changes based on consumer research, historic market performance, detailed cost analysis, other segment analogies, and McKinsey’s overall experience.
  • We model these changes in enrollment and claims PMPM on a multi-year basis to understand the likely impact over time. For example:
    • For initiatives aimed at increasing coverage uptake among the uninsured, we model the changes in claims PMPM that would result from a larger risk pool.
    • For initiatives aimed at reducing claims directly, we model the impact on increased market participation if claims reductions are translated to lower premiums.
  • For each category of potential initiatives, we discuss the impact in terms of changes in enrollment and claims compared with today’s individual market.

Promote appropriate enrollment

Some consumers choose to enroll in a plan only when they need healthcare services (e.g., by claiming eligibility for a special enrollment period (SEP) because of a putative job loss or move) and then later drop coverage. Others purchase plans with limited coverage and then try to switch to a better plan when they need healthcare services. Conversely, some consumers purchase individual market plans even though they are eligible for Medicare or Medicaid (often, to obtain better coverage for high-cost conditions or because they receive third-party payments to enroll in individual market coverage).12 These types of behavior increase overall costs for all consumers in the individual market.

A variety of approaches have been proposed to address these issues, including:

  • Improve the special enrollment period (SEP) verification processes13
  • Prohibit individual market enrollment for those eligible for other types of health insurance
  • Require premium payment at the beginning of each month14
  • Limit the grace period for enrollees
  • Introduce late-payment penalties for those who stop paying their premiums mid-year

Major drivers of change

  • Removing the ~4% of enrollees who enter during the SEP inappropriately and have ~3X higher-than-average claims costs from the market1
  • Removing the <1% of enrollees who should be covered by some other type of insurance and have expenses ~10X higher than the average from the market
  • Encouraging the ~5% of enrollees who stop paying prematurely (but can afford to keep paying) and have ~2X higher than average costs to pay for the entire year; this would amortize their claims over 12 months

1. Potential impact of discouraging eligible individuals from enrolling during SEP would need to be assessed.

The goal of all of these approaches is to limit inappropriate high-cost enrollment while continuing to encourage eligible people to enroll (admittedly, in any of these approaches an inherent tradeoff is being made).15 Our analyses suggest that initiatives such as these could, in the aggregate, achieve a claims reduction of up to ~10% of average PMPM claims costs (some of the reduction would result from costs being amortized over a broader base as average member months are increased). However, the initiatives would likely have minimal enrollment impact—there would be an initial enrollment decline of ~1% as inappropriate enrollment is reduced, followed by an enrollment increase of ~2% if claims reductions are translated to lower premiums.

Stabilize risk pools

Instability in the market’s risk pools—whether from changes in the mix of consumers purchasing plans, uncertainty about where plans stand in the zero-sum risk-adjustment mechanism, changes in market rules to allow extensions of pre-ACA individual market plans, or shortfalls in risk corridor payments—has likely caused participating health insurers to be conservative in their pricing assumptions. The resulting uncertainty has created a vicious circle: it led to higher premiums for consumers, which then reduced enrollment rates, which then resulted in premium hikes, and so on. This uncertainty is likely to persist in the future, given questions about the payment of cost-sharing reductions and reinsurance collections.16 Insurance markets without large, stable risk pools and appropriate funding sources can become volatile. An easy solution to this problem would be to simply inject new funding at the level required to quickly right-set losses, but this is likely not feasible or sustainable over the longer term. 

Among the methods proposed to stabilize the risk pool:

  • Extend reinsurance mechanisms
  • Introduce high-risk pools (HRPs)
  • Merge non-high-risk Medicaid expansion enrollees with the individual market population

There is an inherent value in creating risk pool stability to alleviate insurers’ concerns about high-dollar claimants. Actuaries can more accurately price insurance plans if large and volatile claims are capped or removed. An overall reduction in volatility is possible through reinsurance (capping the high-dollar cases) or HRPs (removing the high-dollar cases). Assuming that either mechanism is self-funded within the market-level individual risk pool, reinsurance provides more predictability in a more efficient way (Exhibit 3).17 Moreover, a reinsurance mechanism is similar to the medical stop-loss policies used in the employer-sponsored insurance market, and health insurance actuaries have many years of experience with them. By contrast, it is more challenging to predict the cost impact of HRPs, since it varies considerably based on prospectively applied eligibility criteria that can result in significant high-dollar claims that remain in the non-HRP pool. Hence, removing volatility through reinsurance should improve pricing accuracy and insurer participation in the market. In addition, reinsurance enables high-risk people to be given the same plan choices and prices as low-risk people, avoiding some of the historical challenges with HRPs.

Our analyses also suggest that initiatives that do not require new funding, such as merging a subset of the non-high-risk Medicaid expansion enrollees with the individual market population, could potentially help to stabilize the risk pool. By taking Medicaid beneficiaries who have incomes between 100% and 138% of the federal poverty level (FPL) and are not currently considered to be high risk and merging them with the current individual market population, average claims costs PMPM could be reduced by around ~15%. If these savings were translated into lower premiums, the result could be a secondary enrollment increase of ~5%. However, real-world experience to date is insufficient to understand whether these potential savings could be realized. For example, Arkansas, through its private option, merged non-high-risk Medicaid expansion enrollees of all FPLs into its individual market population, but it realized only a 2% premium savings net of trend in the first year.18 It is yet to be seen how other states’ attempts—e.g., Michigan’s 1115 waiver, which calls for giving 100% to 138% FPL Medicaid expansion enrollees access to qualified health plans (QHPs), beginning in April 2018—will play out.

Major drivers of change

  • Reducing volatility by capping or removing high-dollar cases through a transfer of funds, with reinsurance proving a more efficient mechanism than high-risk pools
  • Combining a subset of the non-high-risk Medicaid expansion population1 with the individual market, yielding a healthier risk pool, with claims savings (and potentially lower premiums) dependent on the risk difference between the markets

1. Assumes only beneficiaries with incomes between 100% and 138% FPL are merged, and the highest-risk enrollees are carved out to remain in Medicaid. Impact of potential changes to provider reimbursement could result in additional savings. Impact of introducing cost-sharing to Medicaid beneficiaries would need to be assessed.

Maximize market participation

Although close to 18 million people now purchase coverage through the individual market, another 10 million remain uninsured.19 Increasing coverage uptake among the persistently uninsured would improve the risk pool and set in place a virtuous circle of lower premiums leading to higher enrollment, leading to even lower premiums, and so on.

As our consumer research shows, there are a variety of reasons the persistently uninsured decide to remain without coverage, which suggests the need for a diverse set of initiatives to effectively encourage enrollment. For example, for those uninsured making an economic tradeoff, raising the cost of staying uninsured or lowering the cost of enrolling could potentially change their behavior. Even so, it is not likely that everyone would be convinced to enroll—our consumer research shows that ~30% of the uninsured opt to remain uninsured for factors other than economic reasons.20 However, the research also indicates that the remaining 70% do make an economic choice to remain without coverage and may be likely to enroll if insurance becomes more affordable or if the cost of being uninsured grows. Yet, the bar for making coverage more affordable is quite steep for the 75% of the uninsured who are subsidy-eligible. Given the current subsidy structure, a decrease in gross premiums for these consumers would, in most cases, not translate into a lower net premium.21 This limits the potential for uninsured uptake significantly.

Some initiatives to increase enrollment include:

  • Introduce continuous coverage protection with a transitional high-risk pool or other late enrollment fee
  • Introduce auto-enrollment into the lowest-price option22
  • Widen the age rating curve
  • Allow lower actuarial value plans for all individuals (e.g., copper plans)23

Major drivers of change

  • Increasing the average perceived cost of choosing to be uninsured by close to 4X by changing the nature of the penalty so that it becomes a penalty for a lack of continuous coverage1
  • Streamlining the enrollment process to be opt-out to encourage uptake by ~15% of the uninsured through access to a plan with premiums within only 2% of their household income (zero premium for many)2
  • Widening the age curve from 3:1 to 5:1 to reduce premiums for younger adults by an average of 16% (this would likely increase premiums by an average of 14% for older adults, who are often less price-sensitive)
  • Attracting close to 15% of the uninsured by introducing a lower 50% AV high-deductible plan that would be available to all3

1. Assumes that continuous coverage is defined as maintaining enrollment in a QHP; defines the penalty for lack of continuous coverage as exposure to full costs associated with pre-existing conditions for the first year, as well as a premium surcharge.
2. Assumes the “unreachable” uninsured would opt-out.
3. Assumes the plan would be available to all QHP-eligible and that subsidies could be applied toward the premium. 

These initiatives collectively could increase enrollment up to ~20%, since many of the currently uninsured would have incentives to sign up for coverage. The resulting impact could potentially lower average PMPM claims costs by up to ~5%.

Reduce cost through coverage redesign and payment innovation24

Current QHPs cover a wide range of medical expenses—including low-dollar routine costs, expenses related to discretionary and purely elective procedures (those not dependent on medical necessity), and expenses for management of chronic conditions and acute events related to exacerbations of those conditions, as well as expenses related to unpredictable, infrequent, catastrophic events. Deductibles and other out-of-pocket mechanisms available today to enable value consciousness are largely blunt instruments that do not recognize the difference in consumers’ ability to control or absorb expenses. As a result, under our current model, consumers are simultaneously over-insured (leading to inadequate personal responsibility for controlling costs) and under-insured (leading to insufficient financial protection).

Individual market insurance plans could be redesigned to alleviate this situation by focusing on fuller protection for unforeseen catastrophic care costs; carving out coverage for routine, discretionary, and purely elective care; and varying coverage levels for chronic condition-related expenses based on adherence with treatment regimens and health outcomes achieved. Health savings accounts (HSAs) could be expanded to allow consumers to use savings to cover the cost of routine care and those discretionary services they opt to have, with a potential subsidizing mechanism (e.g., pre-funded HSAs with current subsidy dollars) put in place for those with low incomes. This enhanced consumer-protection and personal-responsibility approach could be coupled with evidence-based payment innovations to align care delivery incentives and further lower costs.

This type of benefit redesign holds the potential for more efficient healthcare consumption, greater consumer financial security, and lower overall costs. Some principles of this approach currently underlie the structure of Singapore’s healthcare system25 and could be tested more broadly. The net result of this type of benefit redesign would likely be a better alignment of the incentives for health insurers, providers, and consumers. Properly aligned incentives are crucial if healthcare cost growth is to be kept from outpacing overall economic growth.

Initiatives that could be taken to support a redesign of health benefits and spur a more consumer-driven retail market include:

  • Remove routine (not preventive) care from covered benefits
  • Remove discretionary care for purely elective procedures from covered benefits
  • Add a savings vehicle (e.g., tax-advantaged HSAs independent of a plan’s design) to encourage savings to cover out-of-pocket expenses for the above two categories
  • Enable use of value-based insurance design and wellness incentives to tie the level of coverage for chronic care to personal responsibility for treatment adherence and health outcomes achieved (“chronic care” would include both the management of chronic conditions and treatment for acute events due to exacerbations of those conditions)
  • Fully cover (with low cost-sharing) unforeseen catastrophic expenses
  • Encourage the adoption of population-based payment models that reward effective management of total cost of care, as well as episode-based payment models that reward effective specialty care

We estimate that redesigning health benefits in this way could lower average PMPM claims costs by up to ~35% (Exhibit 3).26 If translated to lower premiums, the savings could likely increase enrollment by up to ~10%27 (limited by the price insensitivity of the subsidy-eligible uninsured, as described above). In addition, the claims reductions, if translated to lower premiums, could substantially reduce government outlays for federal subsidies, which might create room to fund mechanisms to stabilize the market (as described previously). Furthermore, increased consumer value consciousness could potentially further lower costs by fostering competition among providers for routine, discretionary, and purely elective services, as has already occurred for non-reimbursable medical procedures such as Lasik eye surgery.28

Major drivers of change

  • The ~24% of expenses that are carved out (routine and elective) and passed along to consumers are likely to become more affordable in a competitive consumer market for these services (per the Lasik eye surgery example detailed above).
  • Other drivers of claims reductions are detailed in Exhibit 3 (i.e., better management of catastrophic and chronic care costs through a combination of narrowed networks, episode-based payments, and incentives for chronic condition management).
  • Further reductions in costs through value-based insurance designs and wellness incentives, as well as the secondary effect of the lower care-delivery prices resulting from a competitive consumer market for routine and discretionary services, are not included in the estimates.

* * *

The federal and state governments have a number of pathways they could take, some simultaneously, to implement changes in the individual market. Congress, for example, could take legislative action to change the market’s structure or use budget reconciliations to alter provisions affecting spending and revenue. (The latter approach would not require bipartisan support.) The new administration could use a combination of approaches, including regulations, sub-regulatory guidance, model waivers, or demonstration tools (e.g., the Center for Medicare and Medicaid Innovation), to make changes.29

The new administration could also work with individual states via 1332 and 1115 waivers to change some aspects of the individual market and the Medicaid program. This approach would allow each state to enact the changes it thinks best suit its local environment and its vision for the future. In addition, states could take independent action to alter their individual markets. For example, the state of Alaska recently funded its own reinsurance program.30

For optimal impact to be achieved, the federal and state governments, insurers, and providers will likely need to work closely together to achieve a well-functioning individual market and to ensure a smooth transition to a steady state. Although the federal and state governments can alter the market’s governance and funding, success will ultimately require innovation and greater efficiencies—areas the private sector is well placed to lead.

 


 

For questions about the methodology used for the calculations in this article, contact the authors:

Shubham Singhal (Shubham_Singhal@mckinsey.com), a senior partner, is head of McKinsey’s global Healthcare Practice. Erica Coe (Erica_Coe@mckinsey.com), a partner, is head of the McKinsey Center for US Health System Reform. Patrick Finn (Patrick_Finn@mckinsey.com) is a senior partner in the Healthcare Practice.

McKinsey will soon release interviews with two former HHS Deputy General Counsels about the opportunities with and limitations of 1332 and 1115 waivers. It is likely that the new administration will give states greater flexibility in how they can use these tools.

The authors would like to thank Jim Oatman, FSA, MAAA; Tara Bishop, MD; Stephanie Carlton; and Jenny Cordina for the expert insights they provided during the development of this paper. The authors would also like to thank Pavi Anand; Ananya Banerjee; Alex Luterek; Brock Mark; and Martina Miskufova for their assistance with the article’s development.

  1. Congressional Budget Office. The budget and economic outlook: 2017 to 2027. January 2017.
  2. McKinsey Proprietary Exchange Offering Database. McKinsey Center for U.S. Health System Reform, 2017 exchange market: Pricing trends, November 2016.
  3. Cordina J et al. Understanding consumer preferences can help capture value in the Individual market, McKinsey white paper, October 2016. McKinsey Center for U.S. Health System Reform, 2016 OEP: Consumer Health Insights survey findings.
  4. Adjusted for actuarial value means that claims have been normalized based on the estimated average plan benefit richness in each market.
  5. McKinsey Proprietary Payor Financial Database. McKinsey Center for U.S. Health System Reform, Exchanges three years in: Market variations and factors affecting performance, May 2016.
  6. McKinsey Center for U.S. Health System Reform, Exchanges three years in: Market variations and factors affecting performance, May 2016. Overall, 30% of the health insurers (which together covered close to 40% of individual market enrollees in 2014) earned a profit that year. However, the overall 2014 individual market suffered an aggregate loss of $2.7 billion (–5% post-tax margins). There was similar performance variability in the 2015 market.
  7. S&P Global Ratings, The ACA individual market: 2016 will be better than 2015, but achieving target profitability will take longer, December 2016.
  8. McKinsey Center for U.S. Health System Reform, 2017 exchange market: Carrier participation trends, November 2016. Coe E et al, The emerging story on new entrants to the individual health insurance exchanges, September 2015.
  9. Singhal S, Coe E, Navigating the uncertainty of potential ACA ‘repeal and replace’: A preliminary analysis, December 2016.
  10. Other proposals have been made that we did not assess, including the sale of insurance across state lines, age-based tax credits, eliminating the family glitch, and the public option.
  11. Note, however, that the examples of impact shown in Exhibit 2 cannot be added together.
  12. CMS recently tried to limit third-party payments but was blocked by a federal court injunction. Health Affairs, Reasoning In ESRD Case Could Pose Obstacle To Speedy Changes Under Trump, January 2017.
  13. Changes are assumed to be incremental to alterations to the SEP verification process already implemented in 2016 or being rolled out in 2017 (e.g., reducing the number of SEP-qualifying events or further tightening the documentation requirements). HHS notice of benefit and payment parameters for 2017, March 2016. CMS special enrollment confirmation process fact sheet, February 2016.  
  14. Current federal rules specify that for federally facilitated marketplace states, the first month’s premium payment is required within 30 days of the coverage effective date, and insurers can decide the exact payment date as long as it is not before the initial coverage effective date (meaning that they could require premium payment at the beginning of each month). However, state-based marketplaces have full discretion over this policy, leading to variation market by market.
  15. CMS, Pre-enrollment verification for special enrollment periods fact sheet. Urban Institute, Helping special enrollment periods work under the Affordable Care Act, June 2016.
  16. To date, several members of Congress and some state leaders have publicly stated their support for temporary continuation of cost-sharing reduction payments. (Politico, Republicans move to fund Obamacare subsidies they once sought to kill (e.g., Greg Walden (R-Ore.), Lamar Alexander (R-Tenn.), Utah Gov. Gary Herbert), January 2017.)
  17. Academy of Actuaries, An evaluation of the individual health insurance market and implications of potential changes, January 2017.
  18. Kaiser Family Foundation, A look at the private option in Arkansas, August 2015.
  19. ASPE Issue Brief, How many individuals might have marketplace coverage at the end of 2016?, October 2015.
  20. For example, those who do not support the law or do not believe they need health insurance. McKinsey 2016 Post-OEP Consumer Survey. McKinsey 2011 Consumer Healthcare Survey.
  21. Specifically, a decrease in gross premium could result in realized government savings from lower subsidy outlays, but no realized consumer savings for the many low-income consumers for whom the cost of the lowest-price plan exceeds the value of their premium cap.
  22. Potential implementation considerations for deploying effective auto-enrollment mechanisms, including identification of the eligible uninsured and handling payment to participating insurers, have not been taken into account in our sizing.
  23. Our impact sizing focuses on the potential for the uninsured to purchase coverage in response to such a plan. We assume subsidies could be applied to this new lower AV plan. We have not directly modeled the potential impact of buy-down among current enrollees, which could cannibalize industry revenues.
  24. Singhal S, Coe E, The next imperatives for US healthcare, December 2016.
  25. Brookings Institution, The Singapore healthcare system: An overview, July 2016.
  26. Singhal S, Coe E. The next imperatives for US healthcare, exhibit 6, December 2016.
  27. Estimated enrollment impact accounts only for lower premium; it does not account for anticipated consumer reaction to having access to close to fully covered catastrophic and chronic care, or to carve out of routine and discretionary care.
  28. Singhal S, Coe E. The next imperatives for US healthcare, exhibit 3, December 2016.
  29. The new administration does not have the overarching authority to take all described actions in this paper on its own, as many may require legislation, yet there are pathways possible to achieve these changes.
  30. Alaska has submitted a 1332 waiver application to request federal pass-through funding for APTC savings realized from its reinsurance program, to help fund the program going forward. December 29, 2016 Letter to Secretary Burwell from Governor Walker, Re: State of Alaska-Section 1332 State Innovation Waiver.