Yesterday, Drs. Petra Rasmussen, and Paul Shafer co-authored a public comment with me regarding the proposed rule for the ACA’s Notice for Benefits and Payment Parameters (NBPP-2022). NBPP is the ACA exchange rule book that lays out the general policy contours of the implementation of the ACA. I’ve co-authored with Petra for the first time this year. She has specific expertise in quality ratings and dominated plan choice. We’ve written together on dominated plan choice (paper just accepted before the holiday) and are planning to again write together on the matter of dominated plan choices and learning. Paul and I have an ongoing writing partnership. In 2019, we wrote about the change in elite messaging and operations during the end of the 2017 OEP leading to a loss of enrollment. In 2020, we published a study on the impact of advertising on enrollment. In 2021, we have an accepted manuscript that examines the impact of monopolies on insurer advertising levels. We have another ACA related paper under review and two non-ACA related papers cooking at the moment. He has also written extensively on Medicaid and preventative care. This year, I’ve also published research on zero premium plans, transaction costs, and quality ratings.
The Center for Medicare and Medicaid Services wants to do a lot of things in NBPP 2022. Some of the changes include display and collection of quality ratings, special enrollment periods after income shocks, lower exchange fees and therefore limits on outreach and advertising budgets and massive changes to the nature of what an exchange marketplace must be. Dr. Rasmussen, Dr. Shafer and I have written about these subjects. The three of us wrote a comment about what we know.
Commenting on executive agency rule-making is a form of active citizenship as I laid out last February:
First, it is a way to have some democratic accountability of the executive branch. Interested stakeholders can petition for redress of their grievances and bring about their particular and peculiar expertise to a subject that the government may or may not have as much in-depth knowledge. Sometimes comments can point out “DOH” moments to the agencies and allow for course correction.
Secondly, notice and comment is a core component of the Administrative Procedure Act (APA). The APA governs most of the rule-making regulatory state. A core APA tenet is rule-making entities can only implement final rules when the rules are neither capricious nor arbitrary. If comments identify a significant consequence or assumption of the rule that is not addressed in the final rule response to comments, that is good evidence that the rule may be flawed.
Our comment is below the fold:
We write to offer our comments on the proposed Notice for Benefit and Payment Parameters 2022 (NBPP-2022).
We offer our comments on several proposals including the display of quality ratings, open enrollment period and error correction special enrollment periods, exchange user fees and the proposal to allow for direct enrollment marketplaces to replace either a state based marketplace (SBM) or the federally facilitated marketplace (FFM).
Mr. Anderson and other colleagues have recently published two studies that examined insurer characteristics associated with effectiveness of care and behavioral health quality ratings. (1,2) These studies found that non-profit insurers were more likely to have above average quality ratings among the 185 insurer-product dyads in the CMS quality reporting public use file. Insurers that are primarily Medicaid Managed Care Organizations (MMCO) were more likely to have low quality ratings.
We are concerned that these ratings will merely increase the complexity of consumer choice without providing useful, actionable information for two reasons. First, in response to the COVID pandemic, CMS wisely decided to minimize data collection for non-essential services in the spring of 2021. Quality reporting was one of the many data collection efforts that were paused. Data is currently being displayed for the 2021 plan year merely as a repetition for 2020. We are concerned that the continuation of the pandemic will make any data that can be collected in time for the 2022 Open Enrollment Period (OEP) be haphazard or stale as insurers are likely to still be re-orientating their business processes and priorities towards the care and financial shock of the COVID pandemic. Secondly, the best available evidence strongly suggests that marginal buyers are primarily net premium price sensitive.(3,4) Third, forthcoming work by Dr. Rasmussen has found that consumers often select inferior coverage when the quality star ratings of the available plans are similar and do not provide meaningful information about differences between plans.
We applaud the proposed new Special Enrollment Period (SEP) for individuals to switch to lower premium plans in the scenario where they are no longer eligible for premium tax credits. We propose that a separate and new SEP should be established for technical corrections of selected but not effectuated plans during the Open Enrollment Period (OEP). A recently published working paper that examines the zero premium effect on the Connect for Health Colorado state based marketplace.(5) In that work, we found notable duration effects with individuals who were exposed to zero premium plans having a higher probability of effectuating enrollment of plans first selected by December 15th on January 1st of the following year than those who had to pay a premium. We hypothesize that this difference is due to the reduction in transaction costs for individuals that allowed for plans to be instantly effectuated at the moment of selection if the person chose a zero premium plan but significant transaction costs led to some individuals with non-zero premium plans selected in December to fail to effectuate for January 1 start dates. However, individuals still had the opportunity to correct any errors and effectuate their desired policy on February 1. We propose a new SEP for individuals who select a plan during the OEP but fail to effectuate the plan by January 1 of the contract year. The error correction SEP would last for 15 days after the failure to effectuate and would effectuate coverage for the first of the following month (e.g. if an individual fails to effectuate their coverage by December 15th, they would have until December 30th to effectuate, with coverage starting on January 1st). Alternatively, an extended open enrollment period through January 15th would serve a similar purpose.
CMS has also proposed to lower exchange user fees which may limit funds available for advertising and outreach. We note that federally sponsored television advertising serves a distinctly different role in enrollment than private advertising.(6) Recent research by Dr. Shafer, Mr. Anderson, and others has found that private advertising has not been an effective replacement of marketplace outreach.(6,7) Forthcoming research has found that insurers advertising primarily serves insurers’ competitive goals and not the public goal of maximizing enrollment in affordable health insurance. We propose that CMS maintains the current exchange user fee and use the increased revenue to improve consumer navigation, assistance, and marketing and/or offset the costs of lengthening the OEP and creating additional SEP eligibility.
The proposed rule also envisions a significant expansion of the role of direct enrollment entities by allowing for the creation of state based direct enrollment marketplaces and federally facilitated direct enrollment marketplaces by 2022 and 2023 plan years respectively. This is an extension of the recently approved Georgia 1332 waiver application. We are deeply concerned that this proposal will make the consumer choice environment far more fragmented and confusing, creating less standardization and less uniformity in the marketplace experience across states. We believe that transaction and cognitive costs are a meaningful barrier to enrollment for consumers. Crucially, Direct Enrollment entities would not be required to equally display all plans offered by all insurers, going against the very goal of open competition and choice among insurers and plans that underlies the exchanges. Instead preferred insurers could be over-emphasized for purported positive reasons (superior customer service, reputation, better networks) or for venal reasons such as superior commissions paid to agents and brokers. Inferior plan choice has a persistent and long lasting effect of increased consumer premiums paid and lower satisfaction, with potential downstream effects of greater underinsurance and financial hesitancy in using care which could have real effects on health.(8)
We are also concerned that direct enrollment marketplaces could be a source of selection that may not be appropriately balanced by risk adjustment. Curated lists of selected insurers and products may lead to some objectively superior products, from the point of view of the consumer, to not be offered or to be made readily and easily known. The proposal on P. 7863 155.220(c)(3)(iii)(A) is inadequate as it forces consumers to engage in more clicks and transfers which increases their transaction and attention costs.
As it is, the ACA choice environment is complex. Consumers face many choices. Recently accepted research by Dr. Rasmussen and Mr. Anderson has shown that dominated plan choices were common on Covered California during the 2018 policy year as the policy shock of silver loading disrupted normally effective heuristics. Dominated choices were made in a fairly simple choice environment with full and transparent equal display of information. We believe that this proposal will increase the probability that individuals, likely those with low incomes (9), enroll in plans that are dominated (10–13).
Furthermore, we believe that the rationale offered in the approval of the Georgia 1332 waiver, that direct enrollment entities will be more aggressive and effective in enrolling marginally likely to enroll individuals is logically incoherent. Under current rules, direct enrollment entities are able to aggressively pursue business. The presence or absence of a centralized state or federally facilitated marketplace does not change their fundamental profit seeking incentive. A trusted, third party source of information such as a centralized exchange like Healthcare.gov may lower transaction and trust costs and increase net enrollment.
We encourage CMS to adopt rules that lower transaction costs for consumers. Choosing insurance is a complex task and the rule-making objective should be to simplify the choice environment as much as feasibly possible in order to enable the highest probability of individuals achieving high quality choice.
1. Anderson D, Cai S-T, Abraham J, Drake C. Association Between Effectiveness of Care Quality Ratings and Insurer Characteristics in the Health Insurance Marketplaces. J Gen Intern Med [Internet]. 2020 Sep 28 [cited 2020 Dec 28]; Available from: https://doi.org/10.1007/s11606-020-06248-5
2. Abraham JM, Cai S-T, Anderson D, Drake C. Behavioral Health Care Quality Among Marketplace Insurers in 2019. Psychiatr Serv [Internet]. 2020 Dec 18 [cited 2020 Dec 28];appi.ps.202000115. Available from: https://ps.psychiatryonline.org/doi/full/10.1176/appi.ps.202000115
3. Marquis MS, Buntin MB, Escarce JJ, Kapur K. The role of product design in consumers’ choices in the individual insurance market. Health Serv Res. 2007 Dec;42(6 Pt 1):2194–223; discussion 2294-2323.
4. Ericson K, Layton T, McIntyre A, Sacarny A. Frictions vs. Premiums Impeding Take-up of Subsidized Health Insurance Coverage: Evidence from a Field Experiment. In Preparation;
5. Drake C, Cai S-T, Anderson D, Sacks DW. Zero-Price Effects in Health Insurance: Evidence from Colorado [Internet]. Rochester, NY: Social Science Research Network; 2020 Dec [cited 2020 Dec 28]. Report No.: ID 3743009. Available from: https://papers.ssrn.com/abstract=3743009
6. Shafer P, Franklin Fowler E, Baum L, Gollust S. Television Advertising and Health Insurance Marketplace Consumer Engagement in Kentucky: A Natural Experiment. J Med Internet Res [Internet]. 2018 Oct 25 [cited 2019 Dec 11]; Available from: https://www.jmir.org/2018/10/e10872/
7. Shafer PR, Anderson DM, Aquino SM, Baum LM, Fowler EF, Gollust SE. Competing Public and Private Television Advertising Campaigns and Marketplace Enrollment for 2015 to 2018. RSF Russell Sage Found J Soc Sci [Internet]. 2020 Jul 1 [cited 2020 Dec 28];6(2):85–112. Available from: https://www.rsfjournal.org/content/6/2/85
8. The impact of consumer inattention on insurer pricing in the Medicare Part D program – Ho – 2017 – The RAND Journal of Economics – Wiley Online Library [Internet]. [cited 2019 Dec 12]. Available from: https://onlinelibrary.wiley.com/doi/full/10.1111/1756-2171.12207
9. Drake C, Abraham JM. Individual market health plan affordability after cost-sharing reduction subsidy cuts. Health Serv Res [Internet]. 2019 [cited 2019 Nov 18];54(4):730–8. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1111/1475-6773.13190
10. Afendulis C, Sinaiko A, Frank R. Dominated Choices and Medicare Advantage Enrollment [Internet]. National Bureau of Economic Research; 2014 May [cited 2019 Dec 11]. Report No.: 20181. Available from: http://www.nber.org/papers/w20181
11. Bhargava S, Loewenstein G, Sydnor J. Choose to Lose: Health Plan Choices from a Menu with Dominated Option. Q J Econ [Internet]. 2017 Aug 1 [cited 2020 Jan 13];132(3):1319–72. Available from: https://academic.oup.com/qje/article/132/3/1319/3769420
12. Liu C, Sydnor JR. Dominated Options in Health-Insurance Plans [Internet]. National Bureau of Economic Research; 2018 Mar [cited 2019 Dec 11]. Report No.: 24392. Available from: http://www.nber.org/papers/w24392
13. Sinaiko AD, Hirth RA. Consumers, health insurance and dominated choices. J Health Econ [Internet]. 2011 Mar 1 [cited 2020 Feb 6];30(2):450–7. Available from: http://www.sciencedirect.com/science/article/pii/S016762961100004X