APTC Hacks: A comment for meaningful difference improvement

Last night I submitted a fairly long comment (below the fold) to the Center for Medicare and Medicaid Services (CMS). My comment is on the initial parameters for the 2018 Exchange year. I and several colleagues think that CMS needs to pay more attention to meaningful difference regulation because the current set of rules allows for too much gaming of the system and minimization of the membership pool via the Silver Spamming strategy. The comment identifies the problem and proposes a viable solution.

I do not know what my expectations are for this comment. I know it will be read. I know some low level analyst will bucket the letter into two or three categories and there may be a response. Beyond that, I do not know if CMS will alter their policy at all on this issue. I hope they do as I think getting to a tighter definition of meaningful difference will boost enrollment while stabilizing the risk pool which is an objective that CMS shares. But I don’t know.

Now I’m going to get on Tim F’s beat — if you see something that the Federal government is doing that either intrigues you, pisses you off or touches on an area of specialized knowledge that you have, comment on the rule making. This letter took me three or four hours to write and revise with the other signers. It will be read, and it will be remembered by CMS technical staff as it engages them and it is not a mindless screaming of rage (from what I’ve heard, the EPA gets the best rage-grams.) Thoughtful, relevant responses that point to possible ways of accomplishing the mission of the agency in question will be read and they will be remembered. So please, do so especially as the comment submission process through Regulations.Gov is much easier than I thought it would have been.

The comment is below the fold:

We write a public comment in regards to the draft Patient Protection and Affordable Care Act; HHS Notice of Benefit and Payment Parameters for 2018 (CMS-9934-P). The proposed rules make several important steps to improve access, lower decision making fatigue for consumers and otherwise strengthen the PPACA marketplace. However, a noticeable deficiency exists in the rules. There is no strengthening of the definition of “meaningful difference.” The current rule as defined in the 2017 Letter to Issuers in the Federally-facilitated Marketplaces which was sent to carrier and other interested parties on February 29, 2016, allows for significant manipulation of plan design that does not provide practical differences for consumers while lowering total enrollment and increasing consumer costs. This letter will propose enhancements to the meaningful difference rule that will lead to clearer choices for consumers as well as higher enrollment levels of eligible individuals.

The current rule is defined in the February 29, 2016 2017 Letter to Issuers in the Federally-facilitated Marketplaces on Section 12, page 48.

“CMS removed the following criteria in assessing whether a reasonable consumer would be able to identify one or more material differences between a plan and other plan offerings: Health Savings Account eligibility, self-only plan offering and non-self-only plan offering….

CMS will review each subgroup to determine whether the potential QHPs in that subgroup differ from each other based on the criteria of one or more material differences in cost sharing, provider networks, and covered benefits.”

Further specifications of material differences in cost sharing and covered benefits are provided. However these differences within a metal band for a given geographic offering region still allow for aggressive exploitation of benefit configurations that lead to lower Advanced Premium Tax Credits and thus lower levels of subsidized enrollment.

Ambetter from Peach State Health Plan in Georgia (HIOS ID 70893) uses the loose current definition of meaningful difference to create nearly identical Silver plans . These plans are tightly clustered in price with the least expensive and the benchmark setting second least expensive Silver plans barely priced differently These plans all share a common plan type (HMO) and network ID. From the point of view of the Silver buying pool, minor differences in deductible levels are distributional issues that do not hide the fact that the pool as a whole must pay the same level of medical costs in a given policy year.

Isomorphic plan offerings that have minor cost sharing differences have two significant failings of public concern. The first is lower total enrollment with a significantly adverse risk pool. The second is that these cloned plans lead to significant buyer confusion and sub-optimal choices.

In 2016, an individual, a forty year old single non-smoker who earns $18,000 would have a personal contribution to their premium for the second lowest Silver plan of $63 per month. The Advanced Premium Tax Credit (APTC) covers the difference between the unsubsidized premium and the individual’s contribution. The APTC is a constant dollar sum. If an individual chooses to buy either the least expensive Silver plan or a low cost Bronze plan, the APTC covers a higher percentage of the total cost of the plan. However if there is a de minimis price difference (in some markets such as Chicago and Pittsburgh the least expensive and the second least expensive Silver plans have either no difference or a $1 dollar difference), APTC subsidies do not significantly reduce the net cost of an On-Exchange Qualified Health Plan (QHP)for low income buyers. The least expensive Silver plan could cost the individual $62 per month.

Healthy individuals who qualify for significant APTC subsidies due to comparatively low income may rationally decide that on-Exchange, subsidized health insurance is not a good value proposition if their choices are to buy Bronze plans that cost $20 to $30 per month with $6,000 or more deductibles or to buy Silver plans with Cost Sharing Reduction (CSR) subsidies at a price point of at least $62 per month. This decision to not enroll could occur at two points. The first point would be during the annual open enrollment period where the individual elects to not purchase a QHP. The second decision point is every month as the premium becomes due for the next month. Individuals who either are comparatively ill or anticipate that they will be comparatively costly will enroll and continue to stay enrolled at a high rate because the value proposition is positive.

The second failing of the current meaningful difference rule is that it promotes intense consumer confusion. Health insurance is a complex purchasing decision where individuals must perform several probabilistic forecasts of their lives for the next year. They must project income. They must project familial balance sheet capacity to absorb risk. They must project probabilistic health and health expense profiles. They must project life changing events (births, deaths, employment changes, moves across county and state lines).

Once individuals have made the numerous personal projections that lead to satisfactory purchasing decisions even if those decisions are not optimized, individuals must then navigate the Exchanges and the confusing array of terms, networks, benefit offerings and cost sharing configurations to choose a plan that suits their probable needs.

Evidence strongly suggests that most people have not been able to make optimal choices for their health insurance needs. Most individuals are unable to define basic health insurance terms such as deductible and coinsurance.,

Minor changes in cost-sharing as allowed by the current Meaningful Difference rule further increases confusion and leads to significant sub-optimal choices which leads to individuals spending more on their healthcare expenses (premiums and actual out of pocket expenses) than they would with an optimal decision. Significant research has shown that choice paralysis will lead to lower uptake and sub-optimal decision making.

A significant revision of the Meaningful Difference rule would address both the problem of underenrollment of health individuals into the risk pool and consumer choice overload. Currently five factors are taken into consideration for meaningful difference (plan type, network ID, cost sharing configuration, covered benefit and population subgroup). Any one difference is sufficient to create meaningful difference.

A better rule is possible.The first element would form the plan base. The plan base would be the population subgroup, plan type and network ID. A single cost sharing configuration would be allowed to be attached to an on-Exchange plan base. Ideally, this cost sharing configuration would be one of of the standardized Simple Choice cost sharing configurations. Added covered benefits would be allowed but they would not be considered as meaningfully different when calculation of the second least expensive Silver plan is made.

This new rule solves the current problems that are created by the current Meaningful Difference rule. Buyers will have far clearer choices when they actively shop on Healthcare.gov. The number of plans that can be offered for a geographic region and metal band will most likely be reduced. The reduction in plans will come from the removal of plans that are extremely similar to plans that will still be offered on Exchange. Finally, the removal of isomorphic clones will lead to higher enrollment and improvement of the risk pools as more individuals will find the post-subsidy premiums to be a better deal than paying the individual responsibility fee (individual mandate) penalty.
We have seen what very strong meaningful difference regulation has done in California. Covered California is an active purchaser model instead of a clearinghouse exchange so there are limits to the relevant lessons learned but one example provides key insight.

Andrew Sprung describes how strong meaningful difference regulation led to affordable options for individuals in California.

In Los Angeles, it’s almost entirely in Region 16 that Molina made its play and gained its share. Molina put up both the cheapest silver and the cheapest bronze plans in that region — massively undercutting the benchmark silver plan, and thus offering buyers a major discount on the Cost Sharing Reduction (CSR) subsidies available only with silver. For a 40 year-old earning $23,000, a shade under 200% of the Federal Poverty Level (FPL), Molina silver is just $74 per month in Region 16 in 2016, vs. the benchmark silver’s $119 (Health Net). Molina’s bronze plan at that age and income level is just $1 per month, vs. $55 for the nearest competitor.

Molina accordingly grabbed the largest share of new enrollees in that region 28.5%, or 21.010. Given the price spread, it’s perhaps surprising that Molina’s share wasn’t even bigger. There were, however, six competitors. In the much smaller Region 13, comprising Mono, Inyo and Imperial Counties, Molina got 78.3% of new enrollees (3,580). There, in much of the region — i.e., most or all of Imperial County, by far the largest of the three counties — Molina put up an insanely cheap silver plan, fully $136 per month cheaper than the benchmark for a 40 year-old, rendering it essentially free ($1 per month, as CA does not allow 0-premium plans) for our $23k earner. Its bronze plan was the same $1 per month for that customer.

Wider spreads between the least expensive Silver plan and the benchmark Silver plan lowers the net of APTC price of a Silver plan. For individuals who believe themselves to be likely to be high intensity healthcare service utilizers, this is irrelevant to their decision making process. However, for individuals who project that they will have minimal to no utilization of healthcare services, lowering the post-subsidy price of a Silver plan will increase the probability that they will enroll in a plan during open enrollment and maintain continuous enrollment during the course of the policy year.

The Silver plan gap is most relevant for individuals who earn less than 200% of the Federal Poverty Line (FPL) as the CSR assistance is strongest for these groups. Above 200% FPL as CSR assistance ramps down, the relative attractiveness of a low cost Bronze plan with significant subsidy versus the least expensive Silver will lead to more individuals to purchase and maintain continual enrollment in Bronze. Some proportion of the added Bronze uptake will be members who otherwise would have purchased Silver plans while an incremental number of individuals would have paid the shared responsibility fee instead of buying a QHP.

For these reasons, we believe it is in the public interest to revise the current meaningful difference regulation in order to promote higher enrollment and improved risk composition of the QHP markets.