Participation, Pricing, and Enrollment in a Health Insurance “Public Option”: Evidence From Washington State's Cascade Care Program

Abstract

Policy Points.

• Policymakers considering introduction of a health insurance “public option” to lower health spending and reduce the number of uninsured can learn from Washington State, which offered the nation's first public option (“Cascade Care”) through its state exchange in 2021.

• This article examines insurer participation, pricing, and enrollment in the Washington public option. The public option was the lowest‐premium standard silver plan in 9 of the 19 counties in which it was offered.

• Cascade Care is available solely through private insurers. Voluntary participation of these insurers and uncertainty about the willingness of providers to participate may have hindered greater premium reductions and enrollment in the public option's first year.

Context

State and federal policymakers considering introduction of a health insurance “public option” can learn from Washington State, which established the nation's first public option, with coverage beginning in January 2021. Public option plans were offered voluntarily by private insurers through the Washington Health Benefit Exchange and were subject to state‐mandated plan design and payment requirements.

Methods

We used plan data from the Washington Health Benefit Exchange, linked to data from the US Census Bureau, the American Hospital Association, and InterStudy. We compared geographic availability and premiums of, and enrollment in, public option and non–public option plans, as well as characteristics of counties where the public option was available and counties where the public option was the lowest‐premium plan.

Findings

At least one public option plan was available in 19 of 39 counties and was the lowest‐premium option in 9 of the 19 counties where it was available. Five insurers offered public option plans, including one new entrant to the state and one new entrant to the Exchange. While public option availability was more common in counties where the Exchange was bigger and more competitive, public option plans had the lowest premium in smaller, less competitive counties. In the first year, 1% of enrollees selected the public option, in part due to automatic reenrollment of the majority of returning enrollees in their 2020 plan.

Conclusions

Public option plans offered a low‐premium choice in counties that otherwise had fewer affordable plans, but voluntary participation of insurers and providers and accompanying uncertainty about participation hindered widespread and substantial premium reductions. States should consider tying public option participation by insurers and providers to other state programs and using decision support tools to promote active enrollment. Federal policymakers can support state efforts while considering establishment of a national public option.

Introduction of a health insurance “public option” as a means of providing affordable coverage has risen to the top of the health care policy discussion at both state and national levels. The goal of a public option is to offer lower premiums by reducing payments for hospital and physician services below rates negotiated by private insurers (e.g., by setting a payment cap). Unlike a single‐payer system, consumers would have the choice to pick a public option plan or another health insurance plan, for example through the health insurance exchanges established by the Affordable Care Act or as a “buy in” to an existing program such as Medicare.

Proponents argue that a public option could provide immediate access to lower‐cost insurance, thereby reducing the uninsured rate, as well as potentially disrupting health insurance markets and increasing competition by offering a lower‐price alternative. Opponents are concerned that increased government involvement could threaten the private insurance market. To build on the role of private insurers, a number of states are considering a public option approach in which private insurers are permitted to offer a public option plan with state rules governing the benefit package and payment rates. This strategy is distinct from initiatives that use either the Medicaid or Medicare program as the basis for a public option and from the “co‐ops” that were established in the Affordable Care Act.

Washington State is implementing the country's first public health insurance option, with coverage that began on January 1, 2021. The Washington public option is state regulated, but is offered on a voluntary basis by private insurers through the existing health insurance exchange known as the Washington Health Benefit Exchange. Thus, insurers in the state have the choice of whether to offer a public option plan or not, similar to their choice of offering a Medicare Advantage or Medicaid managed care plan. ¹ Public option plans must meet certain requirements, including a cap on aggregate payments to hospitals and physicians of 160% of what Medicare would pay for the same services. This cap is below the estimated baseline statewide average of 174% of Medicare rates, with the objective of lowering premiums. ² Just as insurers are not required to offer a public option plan, providers in the state are not required to participate in public option plan networks, raising concerns about network adequacy and access to care for public option enrollees.

The Washington public option is part of a 2019 law establishing Cascade Care, which created two new coverage options: first, plans with standard benefit design set by the state Exchange including deductibles, cost‐sharing, and services available before meeting the deductible (“standard plans”); and second, public option plans, which must meet all the requirements of the standard plans as well as the payment cap described earlier. In the 2021 plan year, any insurer offering a plan on the Exchange is required to offer at least one standard plan. There are no requirements to offer public option plans. Insurers may also offer nonstandard plans on the Exchange, which are not required to meet standard benefit design parameters.

The design of the public option program in Washington, specifically the voluntary participation of both insurers and providers and the design of the payment cap, raises a number of questions that are relevant for the future of Cascade Care as well as other state and federal policymakers considering the introduction of a public option: (1) Will existing insurers choose to participate in the public option? (2) Will new insurers enter the market? (3) Do public option plans offer lower premiums than other exchange plans? (4) What are the characteristics of counties with a public option entrant? (5) What are the characteristics of counties where the public option plan is the lowest‐premium option? (6) How does enrollment in public option plans compare to other plans offered on the exchange? In this article, we used administrative data from the Washington Health Benefit Exchange to address these questions. We also considered how experiences in the first year of Cascade Care can inform bigger‐picture policy questions, such as whether and how a public option administered by private insurers can serve the public and how policy in this setting can be designed to align insurer and state government incentives and goals.

We found that five insurers offered a public option plan in Washington in 2020 for the 2021 coverage year, including one new entrant from Oregon and one Washington insurer that was a new entrant into the Exchange. These new entrants were the most aggressive, offering the public option in more counties than the three other insurers who were already participating in the Exchange and continued to offer non–public option plans in addition to public option plans. Public option plans were available in 19 of Washington's 39 counties.

To compare premiums holding benefit design and actuarial coverage fixed, our main analysis focused on premiums for standard (i.e., Cascade Care public option and non–public option standard) silver plans. A public option plan was the lowest‐premium standard silver plan in nine of the 19 counties where a public option plan was available, and was the overall lowest‐premium silver plan (including standard and nonstandard plans) in one county. We examined characteristics of counties where insurers offered a public option and found that public option entry was more common in counties where the Exchange had more enrollees and was more competitive, but that public option plans were more likely to be the lowest‐premium standard plan in counties where the Exchange was smaller and less competitive (i.e., with fewer enrollees and fewer non–public option plans available).

Understanding patterns of public option participation, pricing, and enrollment is important for Washington as state policymakers design the next phase of Cascade Care, for other states considering introduction of a public option, and for payers, providers, and consumers who are considering participation in this type of approach. Market dynamics and consumer decisions will determine whether a public option approach has the potential to accomplish the goal of broadening access to affordable health insurance coverage. The results may also have relevance to the national debate regarding a public option.

Data and Methods

We used plan‐county data from the Washington Health Benefit Exchange on actuarial coverage level (gold, silver, bronze), geographic availability (i.e., county), and premiums of all plans (nonstandard, non–public option standard, and public option) offered through the Exchange. For our main analyses of premiums, we examined rates for a 40‐year‐old non‐smoking individual across silver‐tier public option and non–public option standard plans. This approach allowed us to hold actuarial coverage level and benefit design fixed (since all standard plans are subject to the same benefit design requirements) when comparing premiums. In secondary analysis, we included nonstandard silver plans.

We linked plan data to county characteristics that might influence insurer entry and pricing. We examined three categories of county characteristics: county demographics, characteristics of the Exchange in that county, and characteristics of county provider markets. We used US Census data to measure county demographic characteristics that may influence demand for and pricing of public option plans, including the percentage of the population that is uninsured and the share of population currently enrolled in Medicaid.

To capture characteristics of the Exchange in each county, we obtained the share of each county's population enrolled in Exchange plans from the Exchange website, since the size of the potential market is likely to impact insurer decisions to offer a plan and the price of the plan. We examined participation in the Exchange of insurers that also offer Medicaid managed care plans. In general, these insurers offered Medicaid managed care but did not offer commercial insurance prior to participation in the Exchange. National evidence suggests that Medicaid managed care organizations (MCOs) have been dominant in many state exchanges, offering the lowest‐premium plan in close to 60% of regions in which they participate. ³ Therefore, we expect that the presence of these already low premium plans in the Exchange in a given county may impact willingness to enter the market in that county and pricing decisions regarding the public option. We used 2019 data from DRG InterStudy to identify MCOs participating in the Exchange by county. We used the same InterStudy data to measure insurer concentration, which has been shown to affect premiums, both overall and specifically within the Exchange. ⁴

Finally, in light of the public option cap on provider reimbursement, we expected that local provider payment rates would influence insurers’ decision to enter the market and where they set their premiums. The public option must pay an average of 160% of Medicare rates across providers. While the state average is 174%, the percentage varies by county. In a high‐priced county, insurers may be less able to comply with the cap and may therefore choose not to offer a public option plan.

We used two measures related to local payment rates in our analysis. First, given evidence that highly concentrated provider markets have higher prices, we calculated county‐level hospital market concentration with the Herfindahl–Hirschman Index (HHI) using 2017 data from the American Hospital Association Annual Survey. ⁵ Second, we directly measured hospital payment rates in terms of the ratio of private‐to‐Medicare payment rates. We used 2018 data (the most recent available) from the Centers for Medicare and Medicaid Services Hospital Cost Report Information System to calculate the ratio between average commercial and Medicare payments for a “discharge equivalent” (incorporating both inpatient and outpatient services) at the hospital level for all Washington hospitals. We then constructed a county‐level average of hospital ratios (weighted by discharge equivalents). ⁶

We compared these county Exchange, provider market, and demographic characteristics across (1) counties where a public option plan was available (19 counties) versus counties where no public option was available (20 counties) and (2) counties where the public option was the lowest‐premium standard silver option (9 counties) versus counties where the public option was available, but was not the lowest‐premium standard silver plan (10 counties).

Results

Availability and Selection of Nonstandard, Standard, and Public Option Plans

For the 2021 coverage year, a total of 115 plans were offered through the Washington Health Benefit Exchange. Of these, 64 were nonstandard plans (not subject to Cascade Care benefit design requirements), 36 were standard plans (subject to Cascade Care benefit design requirements), and 15 were public option plans (subject to Cascade Care benefit design requirements and payment cap) (Table 1). A total of 13 insurers participated in the Exchange, of which five offered public option plans. While nonstandard and standard plans were available in all 39 counties, public option plans were available in 19 counties (Table 1 and Figure 1). Approximately two‐thirds of the state population resides in these 19 counties.

Table 1.

Plans and Enrollees in the Washington Health Benefit Exchange by Plan Type in 2021 ^a

		Plan Type
Exchange Characteristics	Total	Nonstandard	Standard (Non–Public Option)	Public Option
Number of plans (%)	115	64 (56)	36 (31)	15 (13)
Number of insurers offering plans	13	10	11	5
Counties available	39	39	39	19
Number of enrollees (%)	222,731	187,717 (84)	33,142 (15)	1,872 (1)
% Enrollees by plan metal tier
Bronze	49.5	51.2	40.8	33.3
Silver	37.6	37.9	36.2	38.7
Gold	11.9	9.7	23.0	28.0
% Enrollees who were new to the Exchange	19.7	14.0	49.6	59.5
% Enrollees by age group
<18	7.4	7.4	7.3	9.4
18‐34	24.8	24.1	28.8	28.3
35‐54	35.8	35.6	36.7	35.3
>55	32.0	32.9	27.3	27.0

^a Analysis of Washington Health Benefit Exchange data. ²¹

Figure 1.

Map of Public Option Availability^a [Colour figure can be viewed at wileyonlinelibrary.com]

^a Analysis of Washington Health Benefit Exchange data. ²¹

A total of 222,731 plan selections had been made as of January 15, 2021. Of these, 84% were selections of nonstandard plans, 15% were standard plan selections, and 1% were public option plan selections. The fact that the most selections were of nonstandard plans reflects a policy decision to automatically reenroll existing Exchange customers, who make up 80% of enrollees, in their plan from the previous year. This automatic reenrollment is important for maintaining coverage but makes it less likely that returning customers will shop for new plans, including the public option. Consistent with this policy, public option enrollees were much more likely to be new Exchange enrollees (60%) and were more likely to be under age 35 than those in nonstandard plans (Table 1).

Public Option Availability, Insurers, and Premiums

As noted, five insurers offered public option plans in 2020. Details of these insurers and their plans are shown in Table 2. Two insurers, Community Health Network of Washington and United Healthcare of Oregon, were new entrants to the Exchange and offered only public option plans. United Healthcare of Oregon was a new entrant to the state and does not participate in other insurance segments, while Community Health Network of Washington offers Medicaid and Medicare managed care plans. The other three insurers (BridgeSpan Health Company, Coordinated Care Corporation, and LifeWise Health Plan) were existing Exchange participants and continued to offer non–public option plans on the Exchange.

Table 2.

Details of Insurers Participating in the Washington Public Option ^a

Insurer Characteristics	Community Health Network of Washington	United HealthCare of Oregon	BridgeSpan Health Company	Coordinated Care Corporation	LifeWise Health Plan
New entrant to Exchange?	Yes	Yes	No	No	No
Public option enrollees	380	1,129	<10	<10	359
Standard plan enrollees	0	0	336	3,070	4,029
Nonstandard plan enrollees	0	0	1,778	35,862	28,953
Counties offering a public option plan	9	10	1	1	3
Counties offering non–public option standard silver plans on Exchange	0	0	15	27	33
Counties offering non–public option, nonstandard silver plans on Exchange	0	0	14	27	33
Offering non–public option plan in same county as public option plan?	NA	NA	No	Yes	Yes (all 3 counties)
Counties where insurer public option plan is the lowest‐premium standard silver plan	6	2	0	0	1
Counties where insurer public option plan is the lowest premium silver plan (nonstandard included)	0	1	0	0	0
Average silver public option monthly premium (range)	$413 ($392, $451)	$440 ($403, $463)	$490	$431	$441 ($414, $481)
Average non–public option standard silver monthly premium	NA	NA	$496	$422	$467
Average monthly premium difference between public option and lowest‐cost standard silver plan in counties where carrier public option is not lowest‐cost silver plan	$28.01	$48.81	$90.21	$6.81	$36.23

Abbreviation: NA, not applicable.

^a Analysis of Washington Health Benefit Exchange data. Plan selections as of January 15, 2021. ²¹

The two new entrants offered the broadest coverage and enrolled most of the public option participants. Community Health Network and United offered public option plans in 9 and 10 counties, respectively. In total, 1,509 out of 1,872 public option enrollees (81%) selected public option plans offered by Community Health Network and United. Existing insurers offered a public option plan in fewer counties with limited enrollment, though they offered non–public option standard plans and nonstandard plans in many more counties and had substantial enrollment in these plans (Table 2).

In terms of pricing, a public option plan was the lowest‐premium standard silver plan in nine counties where it was offered and the lowest‐premium silver plan overall (including standard and nonstandard plans) in one county (Figure 2). Among all silver plans available (including nonstandard plans that may have benefit designs that differ from public option plans), public option plan premiums were on average just over 11% higher than the lowest‐premium silver plan across counties (Figure 3).

Figure 2.

Map of Public Option Relative Premium^a [Colour figure can be viewed at wileyonlinelibrary.com]

^a Analysis of Washington Health Benefit Exchange data. ²¹

Figure 3.

Monthly Premiums of Lowest‐Premium and Public Option Silver Plans by County^a

^aAnalysis of Washington Health Benefit Exchange data. ²¹ Lowest silver premium includes both standard and nonstandard plans. Clallam is the only county where the public option plan is the lowest‐premium silver plan.

Characteristics of Counties With Public Option Availability

We compared Exchange and county characteristics across counties where a public option plan was available (19 counties) versus not available (20 counties) and where the public option was the lowest‐premium standard silver option (9 counties) versus counties where the public option was available but not the lowest‐premium standard silver plan (10 counties) (Table 3).

Table 3.

Exchange and County Characteristics by Public Option Availability and Premium ^a

Exchange and County Characteristics	Counties Where Public Option Is Available (N = 19)	Counties Where Public Option Is Not Available (N = 20)	Counties Where Public Option Is Lowest Premium Plan b (N = 9)	Counties Where Public Option Is Not Lowest Premium Plan (N = 10)
2020 Exchange Characteristics
Average number of insurers offering a silver plan (SD)	2.95 (1.61)	2.55 (1.43)	2.00 (0.71)	3.80 (1.75)
Average number of silver plans (SD)	8.47 (4.11)	5.95 (4.43)	7.33 (3.64)	9.50 (4.43)
Average monthly silver premium c (SD)	$420.98 (31.06)	$449.56 (49.02)	$426.83 (31.11)	$415.71 (31.69)
2021 Exchange Characteristics
Average number of insurers offering standard silver plan (SD)	4.84 (2.11)	3.70 (1.70)	1.44 (0.72)	1.10 (0.32)
Average number of non–public option standard silver plans (SD)	3.79 (1.90)	3.75 (1.83)	2.50 (1.01)	4.90 (1.85)
Average monthly premium for non–public option standard silver plans c (SD)	$457.83 (45.00)	$472.03 (50.80)	$462.68 (28.76)	$465.56 (51.00)
Number of counties with Medicaid MCO participation in Exchange (%)	18 (94.74)	12 (60.00)	8 (88.89)	10 (100.00)
Average number of Medicaid MCOs offering Exchange plans (SD)	1.73 (0.80)	0.85 (0.81)	1.55 (0.73)	1.90 (0.88)
Number of counties with multiple public option insurers (%)	4 (21.05)	—	3 (33.33)	1 (10.00)
Number of counties with a new public option insurer entrant (%)	16 (84.21)	—	9 (100.00)	7 (70.00)
County Characteristics
Average % of Exchange enrollees as share of county population in 2020 d (SD)	3.15 (1.60)	3.57 (2.33)	2.69 (1.24)	3.57 (1.84)
Average % Medicaid enrollees as share of county population in 2020 d (SD)	30.15 (8.70)	29.16 (6.38)	33.25 (10.57)	27.36 (5.84)
Average share of under‐65 population uninsured in 2018, %	9.59	8.45	11.45	7.92
Average 2017 hospital HHI (SD)	6,502 (2,495)	8,207 (2,065)	6,375 (2,062)	6,867 (2,887)
Average overall e insurer HHI in 2019 (SD)	1,564 (372)	1,454 (296)	1,797 (294)	1,353 (310)
Average Exchange insurer HHI in 2019 (SD)	6,266 (2,652)	6,986 (3,096)	7,274 (2,567)	5,359 (2,506)
Average 2018 private insurance hospital payment as a percent of Medicare, % (SD)	171.98 (0.38)	186.54 (0.45)	159.59 (0.35)	181.90 (0.39)

Abbreviations: HHI, Herfindahl–Hirschman Index; MCO, managed care organization; SD, standard deviation.

^a Analysis of Washington Health Benefit Exchange, US Census Bureau, American Hospital Association, and InterStudy data.

^b Lowest‐premium standard silver plan.

^c Premium for 40‐year‐old non‐smoker.

^d Enrollment reported in February 2020.

^e Overall includes commercial, Medicare Advantage, and Medicaid managed care.

To understand Exchange characteristics associated with entry of a public option insurer, we examined these characteristics in 2020 (prior to the introduction of a public option). We found that counties where a public option plan was available in 2021 were lower priced and had more health plans participating prior to introduction of the public option. Specifically, in 2020, these counties had more insurers offering a silver plan (2.95 vs. 2.55), more silver plans available (8.47 vs. 5.95), and lower average silver premiums ($421 vs. $450). Patterns were similar in these counties in 2021. These findings suggest that public option plans were offered in counties where the Exchange was more competitive and had lower prices, perhaps reflective of the insurer perception that the payment cap requirements may be easier to meet in these counties.

Consistent with the hypothesis that insurers offered the public option in lower‐priced counties due to the requirement to stay within the payment cap, we found that counties with lower hospital concentration (hospital HHI of 6,502 vs. 8,207) and lower private insurance payment rates for hospital services (172% vs. 187% of Medicare) were more likely to have a public option entrant compared to counties without a public option. These lower baseline payment rates may have indicated to insurers that it would be easier to establish a provider network in compliance with the public option payment cap in these counties.

Characteristics of Counties Where the Public Option Was the Lowest‐Premium Standard Silver Plan

In nine of the counties where a public option plan was offered, it was the lowest‐premium standard silver plan available. These nine counties include about 14% of the state's population. Counties where a public option plan was the lowest‐premium standard silver plan available tended to have a smaller percentage of the population enrolled in Exchange plans, fewer insurers offering silver plans (in 2020), fewer silver plans, and higher monthly premiums relative to the ten counties where the public option was available but was not the lowest‐premium standard silver plan. These county features indicate that public option plans offered a relatively low‐premium option in counties where the Exchange was less competitive and had higher premiums, potentially meeting an important consumer need for more affordable health insurance options.

Counties where the public option plan was the lowest‐premium standard silver plan all had a new entrant insurer offering the public option plan and were more likely to have multiple public option insurers (three of nine counties) than counties where the public option was not the lowest‐premium plan (one of ten counties). Consistent with relatively low public option plan premiums, these counties also had lower private payment rates, 160% of Medicare payments on average, compared to 182% of Medicare in counties where the public option was not the lowest‐premium standard silver option. Finally, counties where the public option was the lowest‐premium standard silver plan had slightly fewer Medicaid MCOs present in the Exchange in 2021 (1.55 vs. 1.90). This finding supports the idea that public option plans may be better able to compete on price in counties where they do not have to compete with MCOs, which are likely to have preexisting low‐priced networks from their Medicaid managed care products to draw on for their Exchange product.

Discussion

The Washington public option is the first such program in the country, and its experience in the first year offers insights for policymakers in other states and at the national level, where a federal public option is a key component of the Biden administration's health policy strategy. In its first year, a public option plan was available in 19 of Washington's 39 counties, was the lowest‐premium standard silver plan available in 9 of the 19 counties, and was the lowest‐premium silver plan of any kind (i.e., including nonstandard plans) in one county. We found that the majority of public option plans were offered by the two new insurance companies participating in the Exchange, with limited participation by the three returning Exchange insurers.

Insurers offered public option plans in counties where the Exchange was more competitive and had lower premiums. Both lower premiums and public option entry in these counties were likely driven by lower prices for hospital and physician services, though this is speculative given the early stage of the program. We observed that counties with a public option plan available were characterized by less concentrated hospital markets and lower commercial hospital prices relative to Medicare compared to counties with no public option plans available. This finding is unsurprising given the requirement for insurers to meet a cap on provider reimbursement but highlights a potential unintended consequence of such a cap combined with voluntary insurer and provider participation. The goal of the public option is to increase access to affordable care, which suggests it is most needed in counties where premiums are relatively high; the structure of the program, however, resulted in public option plans being more likely to be available in already low‐priced, relatively more affordable counties. It may be necessary to compel hospitals to participate if there are going to be public options in relatively higher‐cost areas.

Although entry was more likely in counties where the Exchange was bigger and more competitive, our findings suggest that public option plans offered a lower‐premium option where the market appears to be relatively less competitive. Public option plans tended to be the lowest‐premium choice in counties where the Exchange was smaller and had fewer plans. Consistent with the broader literature on the competitive effect of multiple insurers in a market on premiums, counties where the public option plan was the lowest‐premium plan tended to have more than one public option insurer and a new entrant. ⁷

Another notable pattern is the role of Medicaid MCOs. Nationally, MCOs have performed well in terms of offering low‐premium plans through the exchanges; our results from Washington State are consistent with this and offer insights into how public option plans and plans offered by MCOs may interact on a state exchange. ³ Public option plans were likely to be the lowest‐premium option in counties with less MCO presence. In counties where a public option plan was not the lowest premium plan, the lowest premium plan was offered by an existing Medicaid MCO or a limited‐network insurer such as Kaiser. For example, in the most populous county in Washington State, King County, where Seattle is located, the lowest‐premium plan was offered by Kaiser, and there were multiple plans offered by Medicaid MCOs Molina and Coordinated Care, all with lower premiums than the available public option plan, which was offered by United. Patterns were similar in other large, urban counties such as Pierce County, where Tacoma is located. In these types of markets, public option insurers may have had difficulty competing with insurers such as Kaiser and Molina, which are likely to have preestablished, low‐priced, limited‐network plans.

It is important to emphasize that this study is reporting on the first enrollment period of the Washington public option program. Given the early stage of the program, as well as the voluntary participation of both insurers and providers, a key metric for success in the first year is whether insurers and providers participated and whether consumers enrolled. Cascade Care succeeded on this front. Over time, metrics of program success are likely to evolve to meet the needs of consumers and balance demands across insurers, providers, and enrollees.

In terms of enrollment, substantial literature has shown that plan‐switching rates are relatively low, particularly in the presence of auto reenrollment. ⁸ Therefore, public option enrollment is most likely to occur among new enrollees, who make up only 20% of Exchange enrollment in Washington. It is significant that amongst this group, 40% selected a Cascade Care plan. Low enrollment could also be hampered by the subsidy structure of the exchanges. Premium reductions in the Affordable Care Act exchanges largely benefit the subset of enrollees who do not receive subsidies, since these consumers face the full premium while others are buffered from the full premium. Thus, the main population likely to benefit from a public option is limited without changes to auto reenrollment and the subsidy structure (e.g., tying a subsidy to a public option plan). For new enrollees, a public option appears to be a viable lower‐priced option.

A public option program such as Cascade Care also has broader effects on health care markets, including introducing new entrants into state insurance markets (e.g., United Healthcare of Oregon entered Washington to offer the public option), which can have longer‐term benefits for increasing competition and lowering premiums. Use of the payment cap also introduces a new strategy for lowering health care spending. While premiums did not differ substantially between public option and other plans in the Exchange in 2021, the payment cap may be increasingly effective at lowering premiums over time if health care prices continue to rise and premiums of other plans grow substantially. Plans and beneficiaries will learn if this is a viable option to increase insurance coverage by lowering premiums.

Further efforts by the state to control hospital rates, potentially paired with stronger participation incentives, would likely give public option plans a greater premium advantage. Overall, participation by insurers and providers may change in future years given experience with the program and changing regulations, with potential implications for geographic availability, premiums, and enrollment. It will be important to monitor the program as it develops in order to continue to inform policy efforts.

Lessons From Washington for Other States Considering a Public Option

Multiple states and the federal government are interested in exploring a public health insurance option. Whether public option initiatives have the intended effects of creating lower‐premium options for consumers will depend on the details of their design. There are a number of ways to structure a public option, with probable trade‐offs between provider participation and premiums. ⁹ If the state chooses to rely on private insurers to offer the public option, as in Washington (i.e., rather than pursuing an approach such as a Medicaid or public employee plan buy‐in), the state will have to consider whether or not to make insurer participation mandatory or voluntary, and, if voluntary, how to encourage private insurer participation. A separate but related decision is whether provider participation in the public option will be voluntary or mandatory and what rate to pay providers, which impacts consumer premiums. In the following paragraphs we discuss how early experiences from Washington State can inform other state policies as well as the national policy debate on these decisions.

Insurer Structure and Participation. As described, Washington's public option program was offered on a voluntary basis by private insurers through the state exchange with state‐established benefit design requirements and a provider payment cap intended to lower premiums. Colorado and Nevada recently enacted legislation establishing public option programs with a similar proposed structure (i.e., offered by private insurers), with Colorado's program effective on January 1, 2023, and Nevada's on January 1, 2026. ¹⁰ , ¹¹ , ¹² A second structure that has been proposed is a “buy‐in” to an existing state program such as Medicaid (e.g., New Mexico, Delaware) or the state employee health plan (e.g., Connecticut).

A buy‐in to a public program such as Medicaid would likely have lower provider rates from already established provider networks, which translate into lower program costs and premiums. Additionally, programs like Medicaid are already up and running. These programs, however, could suffer from limited provider engagement if participation is voluntary in states with low Medicaid payment rates. These low rates can restrict access to certain physicians and hospitals for public option enrollees, making it less likely that people will enroll. A public option program operated by private insurers has the advantage of building on existing relationships between private insurers and hospitals and thus may have higher provider participation than a Medicaid buy‐in, though this would likely depend on payment rates, which may differ between the public option plan and the insurer's other plans.

Evidence from Washington discussed in this paper suggests that a public option offered on a voluntary basis by private insurers may have limited participation by existing insurers, at least initially. This is the first year of the program and insurers may be waiting to assess the system, enrollment, and hospital and physician participation before participating. Limited participation among existing insurers may be due to concerns that established provider networks may not comply with the payment cap. Perhaps reflective of this concern, we found more entry in less consolidated hospital markets, counties with lower commercial prices, and counties with lower premiums for non–public option plans, indicative of lower spending.

A related decision is how to address auto enrollment and the subsidy structure. Auto enrollment hinders enrollment in the public option, and it may be necessary to give everyone the choice to auto enroll or join the new public option. The public option can be the lowest‐premium option but, depending on how the subsidy structure operates, the person may not benefit from choosing the lowest‐premium public option. These decisions may impact insurer participation if voluntary given their anticipated implications for enrollment.

Setting Provider Payment Rates. States also are pursuing different options in terms of setting payment rates within a public option. In Washington, the state set an aggregate cap on provider and facility payment across inpatient, outpatient, and professional services. The insurers offering the public option must keep aggregate reimbursement at or below 160% of the “total amount Medicare would have reimbursed providers and facilities for the same or similar services.” The law has separate rules for payment of critical access hospitals (reimbursement at 100% of allowable costs as defined by the Centers for Medicare and Medicaid Services) and primary care services (a floor of 135% of Medicare payment). ¹³ With this design, insurers are held accountable for meeting the payment cap.

In the context of regulating public option payment rates, placing the responsibility for meeting the payment cap on insurers has the advantage of giving insurers the flexibility needed to build robust provider networks—for example, to offer relatively higher rates to certain hospitals in order to ensure participation in the network while offsetting with lower rates to other hospitals. This arrangement, however, may preserve existing dynamics in the hospital market, such as hospitals with substantial market power charging substantially higher prices. Indeed, we observe patterns consistent with this type of behavior in Washington. Specifically, we find that insurers are more likely to offer public option plans in areas with lower hospital consolidation and lower provider prices, which likely reflects greater ability to negotiate with hospitals in these markets compared to markets with relatively greater consolidation and resulting hospital market power.

Another approach is for the state to set maximum rates for hospital and physician services. Regulating the rates that providers receive from the public option plan allows the state to employ its regulatory power to lower rates directly, but requires the state to set payment rates for each provider and is likely to face political resistance. For example, Colorado initially proposed provider‐specific rate setting based on a provider‐specific base rate of 155% of Medicare, with add‐ons for certain hospital characteristics (independence, critical access, payer mix, and management of cost of care). ¹⁴ During legislative debates, this approach has been revised to apply only if participating insurers are unable to meet a premium reduction requirement that public option plan premiums be at least 6% less than premiums for individual and small group plans offered by that carrier. ¹¹

The state must determine whether provider participation is mandatory or voluntary. When physician and hospital participation is not mandatory (as in the current Cascade Care program in Washington), setting provider rates presents a trade‐off between rates that are below current rates paid by private insurers and low enough to result in savings to consumers (i.e., via lower premiums), but sufficient to induce providers to participate. Nevada's public option bill requires any provider participating in Medicaid or the public employee plan to participate in at least one public option network. ¹² The Colorado bill indicates that the state could mandate hospital participation if networks are inadequate. ¹¹

Policy Recommendations

Decisions regarding the design of a public option, such as those discussed in this paper, have implications for what types of insurers and providers participate and where rates are set, which, in turn, determine premiums faced by consumers and enrollment. These decisions are also likely to determine the political feasibility of passing legislation to establish (or, in the case of Washington, update) a public option program. A structure with mandatory provider and insurer participation and direct rate setting has more leverage and is more likely to result in lower premiums, but is likely to face substantial political pushback. ¹⁵

Indeed, the structure of Cascade Care was a product of the legislative process in Washington, during which the state experienced substantial opposition from insurers and providers. ¹⁶ An example of the political compromise engrained in the design of Cascade Care was the effort to finalize legislation related to the program's second year. The evolution of the bill through the legislative process reflects the trade‐offs that states face as they seek to balance the objectives of increasing enrollment in the public option (directly related to the ability to reduce premiums) while maintaining or increasing insurer and provider participation in the program.

The initial draft of the year 2 Cascade Care bill required that any hospital receiving payment from the public employee plan or Medicaid must contract with the public option plan at the plan's request. The draft also established a hospital reimbursement rate of 135% of the Medicare amount for services provided to public option enrollees (with add‐on payments for hospitals with a high Medicaid proportion of patients and those operating efficiently). After political compromise, the final bill has relatively weaker requirements, mandating that hospitals receiving payment from the public or school employees’ benefits program or Medicaid must participate in at least one 2023 public option network only if a public option plan is not available in all counties in 2022. The hospital reimbursement rate formula is excluded entirely, leaving the 160% aggregate reimbursement cap requirement of insurers in place as in year 1. ¹⁷ , ¹⁸

Evidence from the first year of the Washington State program suggests that it may be difficult to achieve substantially lower premiums with voluntary participation of both insurers and providers and a lack of direct rate setting. States interested in pursuing a public option might consider other approaches to increase participation of insurers, providers, and enrollees in a public option. One option would be to incentivize participation of Medicaid MCOs through financial incentives (e.g., federal grants to states), since these insurers have existing low‐priced provider networks and have been successful in enrolling the exchange population. The Colorado public option bill requires that insurers offer the public option in counties where the insurer currently operates small group or individual plans. ¹⁹ The Nevada public option bill requires participation of Medicaid MCOs as a condition of participation in the state's Medicaid program. ²⁰

To increase provider participation, states may want to tie participation in other public programs, such as the state employee health plan, to participation in the public option. This approach is in line with prior suggestions to tie participation in Medicare to participation in a public option. ¹ As discussed earlier, the initial legislation for Cascade Care year 2 included this type of provision, but it was not retained in the final bill except in cases where a public option plan was not available in every county. The Nevada bill ties hospital participation in the public option to participation in Medicaid and the state employee plan. ¹² , ¹⁵

States may also want to reassess use of decision support tools and auto enrollment procedures when introducing a public option. While auto reenrollment can be very effective for maintaining continuous enrollment, it may preclude substantial enrollment in a new public option. In Washington, 86% of enrollees in nonstandard plans were returning customers, many of whom were auto reenrolled. In contrast, 60% of public option enrollees were new to the Exchange. ²¹ To minimize loss of enrollees due to transaction costs, states may consider the use of active choice or enhanced decision support tools to encourage switching, or outreach to specific groups of consumers who are likely to benefit from public option plans due to reduced premium costs or more generous benefit design.

In terms of broader changes to improve plan selection given a public option, states may consider restricting the number of plans available (e.g., reducing the number of nonstandard plans as in the year 2 Cascade Care bill in Washington ¹⁷ ) to limit “choice overload” and reduce options that may have lower premiums but less generous coverage, thereby improving the competitiveness of the public option. ²² Other changes to “choice architecture” in exchanges may also be effective; for example, evidence shows that listing an appropriate public option plan first in a list of plan choices may encourage selection. ²³

At the federal level, policymakers are considering possible approaches for the structure of a national public option, including an option based on Medicare. At the same time, increased support for state efforts may be worthwhile. Short‐term actions could include grants to states to assess opportunities for a public option and increased availability of user‐friendly tools to allow state policymakers to calculate, set, and monitor provider rates. Setting provider rates (e.g., as a multiple of Medicare) is a key component of any public option initiative; however, many states currently have limited capacity to determine existing commercial and Medicare rates due to lack of timely and comprehensive data and lack of a clear methodology to calculate a Medicare benchmark. The federal government could make additional data and analytic tools available.

In the longer term, a key barrier for states considering a public option is a lack of comprehensive data on current commercial payment rates within their state. This type of data is needed, for example, to determine where to set payment rates for the public option. Even in states with robust all‐payer claims databases, these data do not include comprehensive pricing for services provided to individuals with self‐insured plans; this is due to the Employee Retirement Income Security Act of 1974 (ERISA), which exempts self‐insured plans from state requirements. Given that close to 70% of people with employer‐sponsored insurance were in self‐insured plans in 2020, ²⁴ any commercial data that does not include self‐insured plan payments are unlikely to be representative of the full commercial market and may not be a good basis for pricing. Thus, the federal government should advance efforts to amend ERISA to allow state policymakers to mandate self‐insured plans to participate in all‐payer claims databases.

Conclusion

The public option has the potential to improve access to affordable health insurance, especially if it grows over time in terms of enrollment and insurer participation. Assessment of public option provider networks and prices, enrollment, and effects on broader insurance market dynamics (i.e., pricing and entry responses from other insurers) will be important to allow state and federal policymakers to learn from Washington's experiences to optimize the program in that state, other interested states, and, potentially, nationally.

Funding/Support: This work was supported in part by Arnold Ventures. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication. This project was supported by grant number T32HS000029 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

Conflict of Interest Disclosures: All authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest. No conflicts were reported.