Healthcare reformers can't leave employer-sponsored insurance untouched

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

Presidential candidates and other Democrats who want to patch or build on the ACA rather than create a substantially new system generally neglect, to varying degrees, an attendant problem: employer-sponsored insurance (ESI) is also in dire need of reform.

A Kaiser Family Foundation's report on large employer coverage released this month found that health spending by families in large employer plans has grown twice as fast as wages over the past decade. The employer share of costs in family coverage (premium plus out-of-pocket) has downticked only slightly:  it was 68% in 2008 and 66% in 2018.  But families and employers alike are tapped out: total costs increased 67% over those ten years. The total cost of large group coverage for a family of four now averages $22,000 per year. Reducing healthcare costs is among voters' top priorities.

Beyond the steady-but-high yearly increases recorded by Kaiser, yesterday NYT reporters Reed Abelson and Katie Thomas spotlighted a recent development that could be destabilizing: the proliferation of specialty drugs for rare diseases, which can carry price tags in the millions per patient per year. As Abelson and Thomas note, about 10% of the population is afflicted with rare diseases,  and more than half the drugs approved in 2018 targeted such diseases.

Who pays for astronomically expensive orphan drugs? Some questions prompted by the NYT report

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

The Times' Reed Abelson and Katie Thomas have a major story exploring a key factor in rising healthcare costs: specialty drugs for rare diseases, which increasingly can carry price tags in the millions per patient per year. The trend and the impact are laid out succinctly:

Rare diseases, however, aren’t all that rare. There are an estimated 7,000 of them, and about 30 million Americans have one — roughly the same number of people in the United States with diabetes. And although there are no treatments for most rare diseases, new therapies are coming on the market nearly every month,  with some reaching beyond $2 million a year for a single treatment. Of 59 new drugs approved in 2018, more than half, or 34, were for rare diseases. 

Leaving aside the broad question of how the U.S. might contain costs for these drugs without choking off their development, the story prompted a few thoughts and questions about how these costs are currently distributed, and how that might change. Some factors to consider:

Surprise! New Jersey ACA marketplace outperforms the national market

When CMS reported enrollment totals in the ACA marketplace at the end of Open Enrollment last December, the news for New Jersey was disappointing.

Despite (or per below, partly because of) a 9% average reduction in premiums last year, fruit of a new reinsurance program and a state-enacted individual mandate, on-exchange enrollment dropped 7.1% -- a performance that lagged the national marketplace (down 2.7%) and the 39 states that rely on the federal HealthCare.gov platform (down 3.8%) -- though right at the median for HealthCare.gov states that have expanded Medicaid (and so get less juice from silver loading)*.

When the state released off-exchange numbers in June, the results provided a measure of consolation. Off-exchange enrollment was up 3%, shrinking the total individual market enrollment loss to 4% from Q1 2018 to Q1 2019.

The main cause of the contrary results is not hard to find. Three quarters of on-exchange enrollment in New Jersey is subsidized.  In David Anderson's pithy formulation, "Reinsurance, all else being equal, helps the unsubsidized and hurts the subsidized." That's true of any measure or market force that reduces average premiums. When the benchmark (second cheapest silver) plan premium goes down, subsidies are reduced accordingly, and price spreads between the benchmark and cheaper plans tend to shrink, reducing discounts.

The retentive ACA marketplace, revisited

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

In my last post, I noted that enrollment drops in the ACA marketplace recorded in each year of the Trump era at the end of Open Enrollment more or less evaporate in yearly comparisons of average monthly enrollment, or end-of-year enrollment.

That is, it seems that fewer people in the last two years drop out without paying, and perhaps a higher percentage remain enrolled all year (many people in the ACA marketplace do have good reasons not to remain enrolled all year -- one of the marketplace's vital roles is as a stopgap). That's congruent with another change recorded in 2019: new enrollments down (-15.7%), re-enrollments up (2.3%), as of the end of Open Enrollment.

Why have apparent enrollment drops as of the end of OE in each of the last three years either shrunk or eroded entirely over the course of the year? A few possibilities:

CMS subtext: The ACA marketplace is in recovery

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

The CMS report on long-term enrollment trends in the ACA marketplace has a stark lede:

During two successive years of declining enrollment, from 2016 to 2018, unsubsidized enrollment declined by 2.5 million people, representing a 40 percent drop nationally. 

That's bad. But the report includes data that leads to a clear conclusion: The ACA-compliant individual market has stabilized. It is rife with problems and there are many enrollees whom it does not serve well, but it's passed through a crisis. Consider the following points (some derived from the report, others not):

1. Subsidized enrollment was higher in 2018 than in 2016, generally understood to be the enrollment peak. In fact, total on-exchange enrollment was higher in December 2018 than in 2016. (Off-exchange enrollment, where the bulk of unsubsidized enrollment occurs, did drop sharply in those years.)


2. Subsidized enrollment is higher in 2019 than in 2018, as of February of each year, when all enrollees have had at least one payment due (see CMS's  (2018 and 2019 Effectuated Enrollment Snapshots). 

Is subsidized enrollment in the ACA marketplace really up since 2016?

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

CMS's report on long-term enrollment trends in the ACA marketplace released this week emphasized the 40% drop in unsubsidized enrollment from 2016 to 2018. But the counterpoint came as something of a surprise: subsidized enrollment, according to CMS, is up since 2016.

If you look at the most often cited enrollment numbers for each year -- total plan selections reported by CMS annually at the end of the open enrollment season  -- total subsidized enrollment is down substantially -- 7.1% from 2016 to 2018, and 7.7% from 2016 to 2019. But average monthly enrollment was higher in 2018 than in 2016 -- and probably will be slightly higher this year.

It might appear that retention has improved -- more people stay in their plans for more of the year. But that's not clear, as average monthly enrollment in 2018 is not quite the same measure as in years prior.  Let's look at the numbers:

CSR attrition vs. silver loading stimulus, 2017-2019

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

In my last post, I noted that in the Trump era, ACA marketplace enrollment has shifted toward the middle of the income distribution. While enrollment has cratered for the unsubsidized, and eroded by 16% since 2016 among low income enrollees (100-200% FPL, in HealthCare.gov states), it's stayed basically flat at the upper end of the subsidy-eligible income scale, 201-400% FPL.

In brief, the market got much worse for the unsubsidized, as premiums spiked 21% in 2017 and 26% in 2018, and marginally worse for low income enrollees, as out of pocket costs rose yearly in plans with fixed actuarial values. It got better, however, for many enrollees in the 200-400% FPL income range, since a) rising premiums made more people at the upper end of the range subsidy eligible, and b) Trump's cutoff of direct CSR reimbursement for insurers in fall 2017 induced "silver loading" that created discounts in bronze and gold plans (see note below if you're unfamiliar with silver loading).

ACA marketplace enrollment is down at low incomes as well as at high ones

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

[Update, 8/15: CMS data on average monthly enrollment in each year appears to suggest that apparent year-over-year enrollment drops among the subsidized as of the end of Open Enrollment may have evaporated (in 2018, and perhaps 2019) as the year progressed. See this post.]

CMS is out with a report highlighting, not to say gloating over, a steep decline in unsubsidized enrollment in the ACA-compliant individual market from 2016 to 2018.  Nationally, while subsidized enrollment was up 1.3% in that period, unsubsidized enrollment crashed 40%, from an average monthly enrollment of 6.27 million in 2016 to 3.77 million in 2018.

That's not surprising, given premium increases averaging 21% in 2017 and 26% in 2018, and the obvious fact that the ACA marketplace was under-subsidized from the start and essentially unaffordable for many people with incomes modestly above the 400% FPL subsidy cutoff, as Urban Institute scholars Linda Blumberg and John Holahan pointed out in 2015:

Here I'd like to add a corollary: Enrollment since 2016 has declined not only at the top of the income scale, among those with incomes above 400% FPL, but at the bottom, among those with incomes up 200% FPL. It's shifted toward the upper end of subsidy eligibility, into the 200-400% FPL income range. The percentage of enrollees with incomes in that range has risen from 29.5% in 2016 to 34% in 2019 in HealthCare.gov states, and from 40.4% to 44.1% in California.

New Jersey individual market premiums set to rise, bucking national trend

Surprise...while individual market premiums for 2020 are up an average of just 0.7% in the 40-odd states tracked so far* by Charles Gaba, requested rates are up a weighted average of 8.6% in New Jersey -- after a 9% drop last year, the product mainly of a new reinsurance program and a state-enacted individual mandate.

AmeriHealth, which gained market share this year and currently enrolls 41% of the total NJ individual market, is requesting increases averaging 11%. Horizon Blue Cross, with 62% market share, is requesting increases averaging 6.2%. Oscar, with 4.4% market share, is requesting a 16.3% average increase.

Here are the rate requests (from https://ratereview.healthcare.gov/). Enrollment is negligible in Horizon Health of New Jersey, Horizon's HMO. AmeriHealth HMO accounts for 11.4% of AmeriHealth's Q1 2019 enrollment. Oxford Health enrollment is negligible.

Some context, presented without attempt at causal explanation:

Triage: A moderate healthcare reform proposal

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

Prescript (8/7/19): it occurs to me that this rather kludgy patch to our current system boils down to a simple rule: no one pays more than 8% of income for less than 80% AV insurance.  Paid for in part by expanding the footprint of Medicare payment rates.

*          *          *

I want to float here a path to healthcare system reform that starts a gear shift or two below the Medicare for America bill, which creates a strong public option that anyone can buy into at an income-adjusted price. I am mindful of David Anderson's warning that a Democratic president who goes for sweeping healthcare reform (assuming at best a narrow Senate majority and abolishment of the filibuster) will have bandwidth for little else -- and I think other imperatives, like attacking global warming, should come first.

First, a set of working assumptions (and background on current bills) that undergird where I land (skip to the subhead, where the proposal starts, if you're so inclined).

1. The U.S. healthcare system is outrageously expensive, unjust, and inefficient, distorted by two related structural flaws:

Should the U.S. ditch private insurance?

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

Elizabeth Warren has gotten tight with Ady Barkan, a brave ALS patient and advocate for Medicare for All. In the July 30 debate, she told his tale:

The guy who has ALS, this isn’t this is somebody who has health insurance and dying. Every month he has about $9,000 in medical bills that his insurance company won’t cover. His wife Rachel is on the phone for hours and hours and hours begging the insurance company please cover what the doctors say he needs.

Barkan, in his turn, tweeted during the next night's debate: 

Every candidate on stage who doesn’t want to get rid of private insurance should have to explain what a company like Aetna adds to our health care system. How do they make patients’ lives better? #DemDebate

— Ady Barkan🔥🌹 (@AdyBarkan) August 1, 2019

Fair enough. Whether to preserve private insurance, and how to reform it if so, are the questions that divide Democratic candidates' approaches to healthcare reform, at least theoretically (what each of them would truly aim to accomplish via legislation introduced in 2021, and how they would balance their healthcare goals with other priorities, are questions they should be asked). 

In any case, let's look at the rap sheet against insurance companies -- and set it against their claims to add value to our healthcare system.

The raps

• Divided and conquered: All but uniquely among wealthy countries, the U.S. leaves private insurers to each separately negotiate payment rates with healthcare providers. The fragmented payer universe gives providers outsized bargaining leverage, particularly in areas with dominant hospital systems or provider shortages.  Accordingly, commercial insurers pay hospitals about 189% of Medicare rates, according to a Congressional Budget Office estimate, and 241% Medicare according to a more recent RAND study. Commercial rates for doctors are about 128% Medicare, according to MEDPAC. 

Bernie Sanders is the Pied Piper of Healthcare Hamlin

Political battles can be semantic, but important. Such is the case with Democrats wrestling with the meaning of Medicare for all.

This week Kamala Harris came out with a healthcare reform plan that seems designed to resolve her past flip-flops as to whether private insurance should be phased out entirely. In brief, she proposed a 10-year path to "Medicare for all" that includes Medicare Advantage -- private plans reimbursed by the federal government and conforming to strict coverage rules. Employers could also offer "Medicare Advantage" plans.  Lots of question marks, but the intent to preserve the public/private hybrid of Medicare as we know it is clear.

Bernie's camp lit into the plan, claiming in effect that the his bill's title (Medicare for All) is politically trademarked. “Call it anything you want, but you can’t call this plan Medicare for All," Sanders' campaign manager Faiz Shakir said in a statement. Pramila Jayapal, lead sponsor of the House version, tweeted, "as lead sponsor of #MedicareForAll, I find it misleading when my fellow Democrats use the #M4A name to describe proposals that are NOT #MedicareForAll."

That's a straight trademark play: we own the hashtag, we own the name. The subtext is that the One True Path to Medicare for All is Sanders and Jayapal's Big Rock Candy Mountain in which a single government entity provides 100% coverage of everything for everyone, funded entirely by over $3 trillion per year in new taxes ($1.5 trillion according to Bernie).

New colors in the spectrum of Democratic healthcare reform plans

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

For some time, Democratic next-gen healthcare reform proposals have fallen on a spectrum  that I've thought as anchored by three main proposal types. Here they are, in ascending order of cost and degree of system change:

• ACA 2.0 - Improve the ACA mainly by a) bulking up the subsidies and extending them farther up the income scale, and/or b) introducing a national public option within the ACA marketplace framework. Chief exemplars: Elizabeth Warren's ACA 2.0 bill; Senators Bennet and Kaine's Medicare-X. 


• Medicare for all who want it: Introduce a strong public option that anyone can buy into on an income-adjusted basis, even if they have access to other affordable insurance, e.g., insurance offered by an employer. Chief current exemplars: The Center for American Progress's Medicare Extra, which is the basis of Reps. DeLauro and Schakowsky's Medicare for America bill.


• Medicare for All (that covers everything for everyone at no cost to anyone except via taxes). Private insurance essentially ends. Chief exemplars: - Bernie Sander's Medicare for All bill, and Rep. Jayapal's House variant.

The Ten Counts of Obstruction (with apologies to Lin-Manuel...)

It's been said no one's read Mueller's 448-page report. Well but, the 8-page summaries of Parts 1 and 2 are pretty direct. Well but, that's still a lot of text.

So let's try it Lin-Manuel Miranda style (soundtrack here):

The Ten Counts of Obstruction

For the nation's instruction,
The ten counts of obstruction:

Number 1: 

Russian hacks electrify the nation.
Call it fake news while you seek more information.

Number 2

Summon Comey to the throne like you're royalty.
Tell him you demand unconditional loyalty.

Pelosi and co. on impeachment: Not if, when?

In some ways, the press conference held by Pelosi, Cummings, Nadler and Schiff in the wake of Mueller's testimony today was exquisitely frustrating. Their dual message made no sense on its face: the president manifestly deserves impeachment but we need more evidence. Nadler said the president committed crimes. Pelosi also said that "crimes were committed against our Constitution," albeit in passive voice. Schiff said Trump was disloyal to the country, citing his negotiating the Moscow Trump Tower deal while seeking the presidency without telling the public. So how can they not open an impeachment inquiry?

At the same time, it seemed to me that these committee chairs and Pelosi in particular may have turned a corner. This was the first time I've heard Pelosi speak about impeachment without actively undercutting it. They were stalling, but they were implicitly promising to get there.

This promise -- too strong a word -- was almost explicit at the end. Schiff, all but acknowledging that the Senate would not convict, also declared that that was not the point. My emphasis (and rough transcript) below:

New enrollment drop in California: Probably the mandate repeal

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

Covered California, the state's ACA exchange, recently released its March Active Member Profile, the year's first detailed breakout of enrollment data for all enrollees (repeated quarterly). It follows a prior breakout for new enrollees. The Active Member numbers show some market stability -- they're broadly similar to last year's. They may also support CoveredCA's previous explanation of a drop in new enrollment.

On January 30, CoveredCA put out an analysis  as of the end of open enrollment, emphasizing that while enrollment was virtually flat overall compared to 2018 (down 0.5%), new enrollments had tumbled 23.7%. CoveredCA attributed this worrisome drop (offset by a 7.5% increase in re-enrollments) to repeal of the individual mandate penalty, noting that the new enrollment drop was "relatively evenly spread across demographics." California has since enacted measures intended to reverse the damage, instituting a state individual mandate and state-funded premium subsidies to supplement the federal subsidies.

I wondered whether part of the drop in new enrollment may have been caused by an overall 8.7% average weighted increase in premiums in 2019 and/or by changes in the discounts in gold and bronze plans generated by silver loading (see note at bottom for an explanation). The answer appears to be no -- it was the mandate repeal. Ready for a null-result report?

Gold plans are pretty cheap in Florida's ACA marketplace. They should be way cheaper

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

Florida's ACA marketplace has benefitted from silver loading. But not as much as it should.

("Silver loading" refers to the pricing of Cost Sharing Reduction subsidies into silver plans only, creating discounts in bronze and gold plans. The practice began in 2018 after Trump cut off direct reimbursement to insurers for CSR. See note at bottom for a fuller explanation.)

Florida has more marketplace enrollees than any state -- and has marginally increased enrollment since 2016 (up 2%) while enrollment has dropped 10% nationally and 14% in the 39 states on the federal exchange, HealthCare.gov. There are several reasons for that:

Biden Plan: An ACA 2.0/Medicare for America hybrid

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

Joe Biden released a healthcare reform plan outline today. It looks like a cross between the Medicare for America bill, which would create a strong national public option, and various "ACA 2.0" bills, which would bolster ACA premium subsidies and reduce enrollees' out-of-pocket costs.

In brief, here are the core features of these antecedents:

• ACA 2.0 bills and plans, including Elizabeth Warren's, are generally variants on an ACA enhancement plan published in August 2015 by Urban Institute scholars Linda Blumberg and John Holahan.  The core is simply to increase ACA premium subsidies and extend them to more people. Blumberg and Holahan proposed raising the ACA benchmark, against which subsidies are calculated, from "silver" to "gold" - -that is, from an actuarial value of 70% (paying that percentage of the average user's annual medical costs) to 80%. They further proposed raising AV higher than that income levels up to 300% FPL, improving the ACA's Cost Sharing Reduction subsidies at lower income levels; reducing the percentage of income paid for the benchmark at every income level; and capping premiums as a percentage of income at 8.5% for anyone at any income level who lacks access to employer insurance or other insurance -- removing the ACA's 400% FPL income cap on subsidies. At bottom, I've posted Blumberg and Holahan's proposed premium and AV schedule.


• The Medicare for America Act of 2019 would create a revamped Medicare available to people at all income levels, paying Medicare rates (adjusted modestly) to providers, and costing no one more than 8% of income. All Americans would be eligible for income-based premium subsidies, even if their employers offer affordable coverage. The plan would be free -- with no cost-sharing -- to people with incomes below 200% of the Federal Poverty Level (FPL). Newborns would be auto-enrolled. Employers could "buy in" by paying a payroll tax, or continue to offer coverage. Existing Medicare and Medicaid would be integrated into the new program, which would include long-term care insurance. Medicare would negotiate drug prices, and a price review board would be empowered to crack down on price gouging.

Ambiguity on key points

Biden's plan would establish "a public health insurance option like Medicare" that, "like Medicare...will reduce costs for patients by negotiating lower prices from hospitals and other health care providers."  Note that it's not entirely clear whether this option will simply adopt Medicare prices or negotiate separately -- in fact, the language implies the latter. The plan will be available to anyone, including those with access to employer insurance -- but here too the language is ambiguous; it's not clear whether those with access to employer insurance would be eligible for subsidies.

Sweeney's amendment to ACA exchange bill: Moar Medicaid enrollment integration?

Some clarification about what New Jersey state Senate president Stephen Sweeney was seeking when he held up passage of the bill to create a state ACA exchange -- other than leverage in bigger battles with Gov. Phil Murphy. 

After threatening not to allow a floor vote on the bill, Sweeney relented at the last minute after negotiating an amendment concerning integration of Medicaid processing in the platform. If the bill had not passed by the end of the legislative session, there would not have been time to meet federal requirements to get the exchange up and running in the fall of 2020 for enrollment in 2021.

Sweeney's amendment changed the language concerning Medicaid integration, directing the state to pursue federal funding for the integration. The text to be replaced is in brackets.

Mining the silver lode (or not): A tale of two blue states

Subscribe to xpostfactoid via box at top right (requires only an email address; you'll get 2-3 emails per week on average)

I have argued in recent posts that

1. State insurance regulators should take the advice of actuaries Greg Fann and Daniel Cruz and force the silver loading spring -- that is, pretty much mandate that insurers price on-exchange gold plans below or at least only slightly above on-exchange silver plans. Fann and Cruz recommend that regulators require insurers to price silver plans more or less as if their actuarial value is 87% or higher, as it is for enrollees with incomes up to 200% FPL. If they do so, no one at incomes above 200% FPL will buy silver, so the AV estimate will become a self-fulfilling prophecy. (See note at bottom for a brief explanation of silver loading, which began in 2018.)


2. New Jersey enrollment has suffered since 2017 from a lack of discounts in bronze and gold plans that silver loading has produced in many other states. 

States that have refused to expand Medicaid have a built-in silver loading advantage. Since eligibility for marketplace subsidies in nonexpansion states begins at 100% FPL rather than the 139% FPL threshold in expansion states,  the nonexpansion states have a high concentration of silver plan enrollees who obtain the highest level of CSR, which raises the actuarial value of a silver plan 94%.  Still, some expansion states have enjoyed pronounced silver loading effects, while others have seen almost none.

Below, I contrast the experience of two expansion states, New Jersey and California. All enrollment figures are derived from the 2019 state-level Public Use Files published by CMS, unless otherwise noted.  I am going to indulge in a bit of shorthand in this post and neglect to provide definitions and back story, excepting the note on silver loading at bottom.