NEW: @USCBO scores #ACA 2.0 provisions of the #BuildBackBetter Act

CBO Logo

Regular readers may have noticed that I haven't checked in on the status of #BidenCare (aka #ACA 2.0) since way back in early August. There's several reasons for this: At first I was swamped with my COVID vaccination/case/death rate project; then, more recently, I had to scramble to get my annual Rate Change project up to date.

Both of these have indeed monopolized my time, but the main reason is simply that President Biden's Build Back Better (BBB) plan (which ACA 2.0 is a part of) has gotten bogged down over the past few months due to infighting and intransigence by a handful of "moderate/centrist" Democrats in the House but especially by two Democratic Senators in particular (Joe Manchin of West Virginia and Kyrsten Sinema of Arizona).

(Needless to say, I'm not even mentioning any Republicans in either the House or Senate, since not a single one of them is willing to support the bill.)

It seemed a bit pointless to write up a lengthy, in-depth analysis of the latest developments on the bill when what does or doesn't have a serious shot at making the final cut kept changing every few days.

Today, however, it feels like a good point to check in on where things stand. My COVID tracking has become pretty routine. I've pretty much wrapped up the Rate Change project a couple of weeks ahead of the 2022 Open Enrollment Period. Finally, and most significantly, the Congressional Budget Office just released a partial score of the Build Back Better which happens to cover the ACA-related healthcare provisions.

It's important to note that this analysis doesn't include other major BBB healthcare provisions such as expanding Medicare to cover dental, hearing & vision care, or allowing the federal government to negotiate prescription drug prices. This is purely about the following portions:

  • Permanently upgrading the ACA's premium subsidy formula (currently upgraded for 2021 - 2022 only)
  • Permanently eliminating the ACA's 400% FPL "subsidy cliff" (currently eliminated for 2021 - 2022 only)
  • Expand ACA subsidy eligibility to those earning less than 100% FPL for 3 years (i.e., a temporary fix for the Medicaid Gap), followed by...
  • Creating a federal Medicaid program (call it "Fedicaid?") to permanently cover those caught in the Medicaid Gap;
  • Extend the American Rescue Plan's (ARP) ACA Unemployment Benefit for another 4 years (through the end of 2025);
  • Permanently lower the Employer-Sponsored Coverage Affordability threshold for ACA subsidy eligibility from ~9.8% to 8.5%.

It's the first four bullets above which comprise the vast bulk of these programs, both in terms of cost and impact. Let's review:


Here's how the CBO summarizes this provision:

Under current law, people with a modified adjusted gross income between 100 percent and 400 percent of the FPL who are lawfully present in the United States are eligible for premium tax credits if they are not eligible for public coverage (through Medicaid or CHIP, for example) and if they do not have an affordable offer of employment-based coverage. For 2021 and 2022, however, the American Rescue Plan Act of 2021—enacted in March 2021—expanded eligibility for the tax credits to include people whose income is above 400 percent of the FPL.

Under current law, people can use those credits to lower their monthly out-of-pocket costs for premiums. The amount is calculated as the difference between the benchmark premium for health insurance (that is, the premium for the second-lowest-cost silver plan available in the region) and a specified maximum contribution, expressed as a percentage of income.

For most people, a silver plan pays about 70 percent of the total cost of covered benefits. (That “actuarial value” of the plan would require enrollees to pay out-of-pocket costs of about 30 percent, on average). Cost-sharing reductions (CSRs) effectively increase the actuarial value of silver plans for people whose income is between 100 and 250 percent of the FPL, as follows:

  • Between 100 percent and 150 percent of the FPL, the actuarial value increases to 94 percent;
  • Between 150 percent and 200 percent of the FPL, the actuarial value increases to 87 percent; and
  • Between 200 percent and 250 percent of the FPL, the actuarial value increases to 73 percent.

Because there is no appropriation under current law to pay for CSRs, most insurers use “silver loading”—they charge higher premiums for silver plans offered through the marketplaces.

Here's a table which summarizes what the ACA subsidy formula looked like prior to the ARP and under the ARP (for 2021 & 2022 only). As you can see, it dramatically reduced the net cost for a benchmark Silver plan for nearly everyone:

Prior to the ARP, a single adult who earns $19,000/year would have to pay around $65/month in premiums; under the ARP, they pay nothing in premiums and only a nominal amount in deductibles/co-pays. A family of four earning $78,000/year would have to pay up to $630/month; under the ARP, they pay around $380. And an older couple earning just over $70,000/year (just over the 400% FPL threshold) might have to pay as much as a jaw-dropping $1,900/month...versus no more than $495/month under the ARP.

Thanks to the expanded formula under the ARP, an additional 2.8 million Americans signed up for ACA coverage during the 2021 Special Enrollment Period. It's safe to assume that even more will do so for 2022 thanks to the ARP's enhanced financial assistance continuing for a second full year.

The CBO estimates that federal deficits would increase by around $209.5 billion over the 10-year period:

CBO and JCT estimate that section 137501 would increase federal deficits by $209.5 billion over the 2022-2031 period as the result of increased direct spending of $119.7 billion and revenue reductions of $89.8 billion. Those net effects primarily reflect a $259.0 billion increase in premium tax credits for health insurance obtained through the marketplaces partially offset by higher revenues. Those revenues would increase because taxable wages would increase as employment-based coverage declines. CBO and JCT estimate that about 10 percent of the estimated increase in premium tax credits would stem from the enrollment of people whose income is above 700 percent of the FPL.

CBO and JCT expect that section 137501 would have a twofold effect on health insurance coverage obtained through the marketplaces. First, most enrollees who have subsidies under current law would be eligible for enhanced subsidies that would lower their out-of-pocket costs for premiums. Second, subsidies would be extended to include some people who will lose eligibility after 2022 under current law. CBO and JCT anticipate that, in addition to reducing current enrollees’ out-of-pocket premium costs, the enhanced subsidies would attract more enrollees to the marketplaces. CBO and JCT estimate that those additional enrollees would account for $167.2 billion of the increase in premium tax credits and that current-law enrollees would account for the remaining $91.8 billion.

CBO and JCT estimate that enacting section 137501 would increase the number of people who have coverage through the marketplaces by 3.4 million, on average, over the 2022-2031 period. The agencies also estimate that the income of 65 percent of those who would not have enrolled without that provision would be above 400 percent of the FPL. For people whose income is more than 600 percent and 700 percent of the FPL, those estimates are 20 percent and 10 percent, respectively.

In other words, they figure that of the ~3.4 million additional ACA exchange enrollees, the income breakout would be roughly:

  • 35% under 400% FPL
  • 35% between 400 - 600% FPL
  • 20% between 600 - 700% FPL
  • 10% over 700% FPL.

The estimated increase in marketplace enrollment consists of 1.4 million fewer uninsured people, 600,000 fewer people with nongroup coverage purchased outside of the marketplaces, and 1.6 million fewer people with employment-based coverage. The estimated reduction in employment-based coverage is primarily driven by a reduction in offers as a response to the increased subsidies for coverage through the marketplaces. CBO and JCT estimate that 200,000 people would enroll in coverage through Medicaid and CHIP as a result of that reduction in offers of employment-based coverage.

In other words, 1.4 million currently uninsured would sign up for subsidized on-exchange coverage; 600K currently enrolled in off-exchange plans would move on-exchange for the subsidies; and 1.6 million people would shift to ACA exchange coverage due to small business employers (under 50 employees) stopping offering small group coverage.

Personally I suspect the first two numbers would be larger and the third one smaller, but (shrug) who knows?


As noted above, this woud be a 2-stage process. The first would temporarily remove the other subsidy cliff at the lower end of the income spectrum (100% FPL). This would make the ~2.2 million Americans caught in the Medicaid Gap (across 12 GOP-controlled states which have refused to expand Medicaid) eligible for full ACA subsidies and enhanced Cost Sharing Reduction (CSR) assistance, effectively giving them ACA market coverage at virtually no net cost to them:

Beginning in 2022, the bill would extend subsidized coverage to people whose income is below 100 percent of the FPL who otherwise meet eligibility requirements.

For each year from 2022 to 2024, sections 137504 and 137505 would:

  • Expand access to subsidized coverage through the marketplaces by extending eligibility for premium tax credits and CSRs to people whose income is below 100 percent of the FPL;
  • Expand eligibility for premium tax credits and CSRs to people whose income is below 138 percent of the FPL who have access to an offer of employment-based coverage that is considered affordable under the ACA;
  • Modify the subsidy recapture and tax-filing requirements for people whose income is below 138 percent of the FPL; and
  • Appropriate funds for outreach and education.

For 2023 and 2024, section 137505 also would increase CSRs for eligible enrollees whose income is below 138 percent of the FPL from the current-law actuarial value of 94 percent to 99 percent. Because funding for CSRs has not been appropriated under current law, most insurers use silver loading to cover those costs. Under section 137505, the federal government would directly reimburse insurers for a portion of the cost of CSRs for eligible people whose income was below 138 percent of the FPL in 2023 and 2024. CBO and JCT expect that most insurers would continue to use silver loading to finance the remaining costs.

For 2024 only, section 137505 would provide marketplace enrollees whose income was under 138 percent of the FPL with additional benefits, such as subsidies for transportation to medical appointments, that currently are covered by state Medicaid programs but not required for marketplace plans.

As an aside, here's my simple explainer of what "Silver Loading" is and how it works. It's worth noting that the phrase "Silver Loading" has now been accepted as an official, U.S. government-recognized term. I take some amusement by the knowlege that while I'm not the one who first came up with the concept (others did so years earlier), it's entirely possible that I'm the one who first coined the phrase in terms of ACA health insurance premium pricing (the earliest usage I can find is from a tweet of mine on July 31st, 2017) might also have been David Anderson or someone else, however; as he noted, a lot of healthcare wonks were discussing the concept throughout the summer of 2017.

Expanding ACA subsidies to those earning less than 100% FPL (in non-expansion states) would be a stopgap, however; it would only be in place long enough to allow the federal government to establish and ramp up the permanent solution to the Medicaid Gap starting in 2025:

Starting in 2025, section 30701 would establish a federal Medicaid program to provide coverage to adults whose income is up to 138 percent of the FPL and who reside in a state that has not expanded its program. The Secretary of the Department of Health and Human Services would be required to administer the program through third-party entities and under contracts with Medicaid managed care organizations. The federal program would be required to provide health care services and enrollee protections that are consistent with the services and protections provided to adults residing in states with programs as expanded under the ACA.

As of 2018, 69% of all Medicaid enrollees nationally were already enrolled via a state-contracted Managed Care Organization (MCO), which basically means private insurance carriers who administer the program, so it's not terribly surprising that a brand-new federal version of the program (some have already started calling it "Fedicaid") woud contract all of their enrollees out to MCOs as well.

But what's to prevent the other 38 states which have expanded Medicaid under the ACA from allowing this to lapse? After all, under ACA Medicaid expansion they have to pay 10% of the cost (which can amount to hundreds of millions of dollars), whereas the states which have refused to expand the program are apparently getting off scott free. Well, there's a resolution for that as well:

In addition, the section would require states to maintain their Medicaid expansions or pay the federal government an amount approximately equal to the expenditures associated with maintaining expansions. That requirement would apply to states that had expanded their Medicaid programs as of January 1, 2022, but subsequently terminate those expansions. CBO expects that such a requirement would cause most states to maintain their expansion programs rather than have the new federal program cover their adult residents. As a result, CBO estimates that over the 2025-2031 period, states that continued their expansion programs would spend $86.6 billion to operate those programs; states that terminated their expansion programs would pay the federal government $3.6 billion.

I'd be fascinated to find out whether the CBO has any specific states in mind. FWIW, $3.6 billion over 7 years woud average around $514 million per year; for comparison, in 2018, the combined state share of Medicaid expansion coverage of AK, ND, VT, RI, DC, H, MT, HI, DE & WV came to around $350 million.

After accounting for the effects of section 137501, CBO and JCT estimate that enacting sections 137504, 137505, and 30701 would increase federal deficits by $323.1 billion over the 2022-2031 period: An increase in direct spending of $335.6 billion would be partially offset by an increase in revenues of $12.5 billion. Those effects reflect a $390.0 billion net increase in Medicaid outlays and $27.2 billion in administrative costs, partially offset by a $75.6 million net decrease in subsidies for health insurance obtained through the marketplaces along with other smaller effects.

I think this offset refers to those living in non-expansion states who are already enrolled in subsidized ACA exchange plans who earn between 100 - 138% FPL who woud be shifted over to Medicaid itself...except it seems to me that should be a much larger offset than a mere $76 million. Perhaps they're referring to something else?

CBO and JCT estimate that enacting sections 137504, 137505, and 30701 would increase the number of adults who enroll in Medicaid, on average, by 3.8 million annually over the 2022-2031 period. That increase would result, on average, in 2.3 million fewer uninsured people per year, 700,000 fewer people with nongroup coverage, and 900,000 fewer people with employment-based coverage. The estimated effect on the number of people with employment-based coverage is primarily driven by fewer people taking up an offer of health insurance coverage.

Again, the 700K presumably refers to those in the 100 - 138% FPL range who are currently receiving heavily subsidized ACA coverage. They'd be shifted over to the new Fedicaid program. Again, however, I'd think that would offset a lot more than $76 million. Huh.

CBO and JCT estimate that over the 2022-2024 period, during which eligibility for marketplace subsidies would be extended to people whose income was below 100 percent of the FPL, enrollment in nongroup coverage would increase by 2.3 million people annually, on average. The estimated increase consists of 1.7 million fewer uninsured people, 300,000 fewer people with employment-based coverage, and 200,000 fewer people enrolled in Medicaid.

Very interesting that they see 200K moving from Medicaid to subsidized ACA exchange plans for the first three years...before shifting back to "Fedicaid" after that?

After establishment of the federal Medicaid program, Medicaid enrollment would increase by 5.6 million, on average over the 2025-2031 period, CBO and JCT estimate. That projected increase consists of an estimated 6.4 million people enrolling in the federal Medicaid program established by section 30701, partially offset by a decrease of 800,000 people enrolled in state-expanded Medicaid programs. The estimated reduction is associated with CBO’s expectation that states that would have expanded after 2021 (according to the agency’s baseline projections) would not do so and that few states that already have expanded would terminate their expansions once the federal program was implemented. CBO and JCT expect that people in those states would instead enroll in the federal Medicaid program. According to CBO and JCT’s estimates, the net increase in Medicaid enrollment would result in 2.5 million fewer people being uninsured, 1.9 million fewer people having nongroup coverage, and 1.1 million fewer people with employment-based coverage.

The bit about states which "would have expanded after 2021" not doing so seems like a bit of a head-scratcher to me. Of the 12 remaining holdout states, I don't know of any which have made any official moves towards Medicaid expansion...I suppose the CBO is assuming one or two of them might have come around eventualy, but that seems like a hell of a leap of faith.

It's also interesting to see, again, the CBO assuming another massive chunk of small business employees shifting from group coverage to Medicaid. Combined with the ACA subsidy expansion provision, that's 2.7 million small group enrollees they foresee making such a move (actually, they put it at 2.8 million in their opening statement). According to a Commonwealth Fund study, the small group market was only around 13.1 million people in 2018 (a bit smaller than the Individual market), so seeing over 21% of its enrollees shift to either the Individual or Medicaid markets would be pretty dramatic.


Truth be told, I didn't even realize that this was part of the BBB Act. The ARP included an interesting provision which made any American who received unemployment benefits for even a single week in 2021 eligible for maximum ACA subsidies regardless of their actual income...even if they were caught in the Medicaid Gap. It did this by essentially declaring their annual income to be 133% FPL for purpses of subsidy eligibility. This means that theoretically someone could have received unemployment benefits for the first week of January, then gotten back on their feet, had a reasonably decent income for the rest of the year and still qualified for full ACA subsidies as if their income was much lower.

It's an interesting provision, and according to CMS, around 209,000 people have taken advantage of it in HealthCare.Gov states (probably closer to ~280,000 nationally). Of those, around 34,000 are in "Medicaid Gap" states. However, I personally assumed this was a one-time thing specific to the COVID pandemic scenario.

Instead, the BBB bill hopes to extend a modified version of this provision out by another four years:

Section 137507 would increase the amount of the premium tax credit for people who receive unemployment benefits for any length of time in a year between 2022 and 2025. Under that provision, people whose household income was above 100 percent of the FPL after excluding unemployment benefits, and who are otherwise eligible for premium tax credits, would receive the same credit available to them if their income was 150 percent of the FPL in the year they receive unemployment benefits.

After accounting for the effects of section 137501, CBO and JCT estimate that section 137507 would increase federal deficits by $10.6 billion over the 2022-2031 period as a result of an increase in outlays of $4.9 billion and a decrease in revenues of $5.7 billion. Those effects would stem primarily from the increase in premium tax credits for health insurance obtained through the marketplaces.

CBO and JCT estimate that 2.0 million people receiving unemployment compensation would be eligible for enhanced premium tax credits under section 137507 if they meet other eligibility requirements. The agencies estimate that, on average in each year from 2022 to 2025, roughly 500,000 people who already would be expected to enroll in marketplace coverage under section 137501 would receive an increased subsidy under section 137507. CBO and JCT estimate that, on average, about 500,000 people would newly enroll and receive a premium tax credit if section 137507 was enacted. The agencies estimate that most of those people would have otherwise been uninsured.


This is another smaller change which I wasn't expecting to be included in the BBB plan...though I'm glad to see it being thrown into the mix.

In short, the ACA says that if your employer offers you group coverage which meets minimum ACA requirements and you turn it down, you're only eligible for ACA subsidies if that group plan would have cost more than around ~9.8% of your household income (it's not exactly 9.8%...officially, it's an indexed 9.5%, which means the exact percent fluctuates slightly from year to year). Note that this is the same ~9.8% that can be seen in the "Premium Cap" subsidy eligibility table at the top of this blog entry.

The problem is that under the American Rescue Plan, that "indexed" 9.8% has been reduced to a flat 8.5%...but there was no corresponding change to the "indexed" 9.8% figure for the group plan affordability threshold section. That would be fixed by this BBB provision:

Section 137502 would modify the criteria used to determine an affordable offer of employer-sponsored health insurance for purposes of premium tax credit eligibility. Under current law, unaffordable offers are those that require employees to contribute more than 9.5 percent of their income (indexed annually for inflation) for self-only coverage. Section 137502 would modify that affordability threshold from an indexed 9.5 percent to a nonindexed 8.5 percent of income. If an employee’s contribution exceeded 8.5 percent of household income, they and their dependents would be able to purchase subsidized coverage through the marketplaces.

After accounting for the effects of section 137501, CBO and JCT estimate that enacting section 137502 would increase federal deficits by $10.8 billion over the 2022-2031 period as a result of an increase in outlays of $12.1 billion and an increase in revenues of $1.2 billion. Those effects would stem primarily from an increase in premium tax credits for health insurance obtained through the marketplaces, partially offset by higher revenues stemming from higher taxable wages that would result from a reduction in employment-based coverage.

CBO and JCT estimate that, on average over the 2022-2031 period, 300,000 more people would enroll in nongroup coverage under the section. That increase consists of estimated reductions of fewer than 100,000 people without insurance and fewer than 300,000 people with employment-based coverage. The estimate of the reduction in employment-based coverage is driven primarily by the expectation that fewer people would take up an employment-based offer. Those choosing to take up nongroup coverage instead would do so because the premium tax credits for plans available through the marketplaces would make those plans less expensive than employment-based plans.

I absolutely agree that this is a fix which should be made, but I was under the impression that it's something which could also be done via regulatory changes by the HHS Secretary, similar to how the Family Glitch is supposedly being dealt with...although come to think of it, I haven't heard a peep about that since April either...

When you add all of this up, the CBO projects a total net 10-year cost of around $553.2 billion for all of these provisions.

It's important to stress that, once again, any or all of these provisions could still be modified, weakened or thrown out entirely in the final version of the bill to reduce the official 10-year cost.

Stay tuned...