MAJOR UPDATES: CHIP & Puerto Rico funding are on the way! With some catches, of course...

UPDATE: The Center on Budget and Policy Priorities has issued their own more detailed analysis which adds substantially to my concerns.

Three days after allowing Children's Health Insurance Program funding to run out for 9 million kids across America, House Republicans are supposedly working on a bill to lock in 5 full years of funding for the program, along with a substantial initial funding infusion to help out Puerto Rico after the devastation of Hurricane Maria:

Republicans on a leading House health-care committee are proposing to send $1 billion in extra Medicaid funding to Puerto Rico as it deals with severe hurricane damage, as part of a five-year plan to fund the federal health insurance program for children.

That's the good news...and if that was the end of the story, I'd honestly cheer the Congressional GOP for (finally) doing the right thing on both fronts, even if the funding for Puerto Rico seems suspiciously small given the massive need for help there.

UPDATE: The CBPP analysis confirms my concerns: The CHIP funding is fine, but the Puerto Rico funding is only a small fraction of what's needed, and it doesn't do bupkis for the US Virgin Islands, which were also hit hard by Maria.

Of course, the GOP being the GOP, they couldn't leave well enough alone:

The proposal from the House Energy and Commerce Committee, provided Monday night to The Washington Post, would be paid for with a bucket of items, including raising Medicare rates for wealthier seniors, redirecting dollars from the Affordable Care Act’s prevention fund and shortening a grace period for enrollees who don’t pay their premiums.

...The funding, which would be provided to Puerto Rico over a two-year period, would be part of a package to fund the Children’s Health Insurance Program, community health centers and other health-care extenders. Congress officially missed the deadline on Oct. 1 to extend funding for the programs, although most states have a few more months before their dollars run out.

OK, so Community Health Center funding would also be included; fine. So, what's the catch?

The Energy and Commerce bill proposes the following funding mechanisms:

  • Charging higher Medicare premiums to seniors earning more than $500,000: $6 billion.
  • Allow states to dis-enroll lottery winners from Medicaid: $400 million.
  • Shorten the grace period for ACA enrollees who don’t pay their marketplace premiums from 90 days to 30 days: $5 billion.
  • Redirect money from the ACA’s prevention and public health fund to community health centers: $5.5 billion.
  • Strengthen Medicaid’s third-party liability policy by making it easier for state programs to avoid some medical costs if they’re already covered by private plans or other government programs: nearly $4 billion.


Let's start with the obvious: The fourth item simply amounts to stealing money from the Centers for Disease Control in order to provide funding for the ACA's Community Health Centers. This is literally robbing Peter to pay Paul.

The fifth item ("making it easier for state programs to avoid medical costs covered by private plans/other government plans") is murkier. It sounds like they're saying is that if some procedure/service is already paid for by, say, the Indian Health Service or the Veteran's Administration (which I think are both 100% federally funded), it shouldn't have to be paid for by Medicaid (which is partly paid for by the states). On the surface this sounds reasonable, but I can't fathom how this would generate $4 billion in federal savings, since it would supposedly result in the federal government paying 100% of the cost of that service instead of only part of it. I guess they assume that the bulk of the cost shift would be over to private policies, but I'd have to see the math on that one; it sounds pretty squirrelly to me. I'd have to learn more about the details of this one to pass judgment on it.

UPDATE: The CBPP analysis seems to suggest that yes, this probably would save the federal government quite a bit of money after all...but would also likely cause some pregnant women and/or parents of young children to forego some types of treatment.

As for the first three items...

To be honest, the first two items on this list don't bother me all that much, depending on the specifics. I'm a bit surprised to see Republicans pushing for wealthy people to pay higher premiums for Medicare...but as long as it's not exhorbant, this might not be such a bad thing. Then again, this could be meant as a means of getting wealthy people to turn against Medicare in general.

UPDATE: The CBPP analysis gives the details on this: Anyone earning over half a million dollars would have to pay 100% of the cost of Medicare parts B and D, which pretty much shoots down a major reason why Medicare is so popular in the first place. While I'm not crying in my beer over rich people having to pay more, that should really be done via the tax code, not Medicare premiums. CBPP points out this could cause a lot of wealthy older people to drop Medicare coverage altogether. which would likely hurt the risk pool. And of course the GOP is also trying to slash taxes for the wealthy at the same time, so they can spare me the "fiscal responsibility" argument.

The second one (banning lottery winners from Medicaid) actually seems pretty reasonable depending on what the "lottery winner" threshold is set to. Somoene winning $200 million in the PowerBall is quite different from someone winning $200 one time in a scratch-off game, after all. Also, I'd like to know how they came up with the $400 million figure...are they saying that there's 400 low-income people who win $1 million per year? Or 4,000 who win $100K apiece? If so, OK, sure.

I should also note that the GOP has a rather strange obsession with making sure that lottery winners don't stay on Medicaid. Remember last March when they devoted a whopping seven full pages of the AHCA repeal/replacement bill to this issue? Again, I agree that big-money winners should be kicked off of the program, but they just seem oddly fixated on it.

UPDATE: The CBPP analysis notes that while the lottery winner provision may seem innocuous enough on the surface (it turns out the minimum winnings threshold is $80,000, which sounds reasonable), it would likely be a royal pain in the ass to administer. Here's the details:

...[winners] would lose Medicaid eligibility for a period of time based on the size of the lump sum. For example, a person who received $95,000 would be ineligible for three months.

The bill would have a significant impact on the streamlined enrollment process designed to simplify the process of determining Medicaid eligibility and coordinating eligibility for Medicaid with eligibility for marketplace subsidies. States would have to add new questions to the Medicaid application and track lottery winnings and other lump sums for what would likely be a limited return. For example, Michigan’s Medicaid expansion waiver allows the state to garnish state tax refunds and lottery winnings to recoup unpaid premiums and cost-sharing from participants; yet the state collected just $3,622 in 2015 and 2016 combined from 36 lottery winners, suggesting that most lottery winnings are too small to trigger a change in eligibility. States would likely spend more to collect these funds than they would save from making people ineligible for Medicaid.

This is basically similar to the failed "drug testing welfare recipients" experiment which many states have tried over the past few years: The state invariably ends up spending far more for the drug tests themselves (hundreds of millions of dollars in some cases) than they ever end up "saving" from denying welfare benefits to a handful of applicants who end up testing positive for illegal drugs. The scenario the CBPP is describing sounds like it would be less lopsided, but still would cost the states far more than they'd ever save.

Even then, it would only save them about 1/3 of that, since the other 2/3 or so would be "saved" by the federal government...which is, after all, the entire point of this provision in the first place. Remember, the federal government pays anywhere from 50-75% of the cost of Medicaid depending on the state (90% for expansion enrollees). Assuming an average of around 67%, that means in order for the feds to save $400 million, the total Medicaid savings from lottery/other lump sum winners would have to exceed $600 million.

The third one (shortening the ACA grace period from 90 to 30 days) is kind of interesting for two reasons: First, because I'm actually on the record for supporting shortening the grace period for ACA exchange enrollees somewhat in order to help cut down on people gaming the system, which has supposedly been causing problems with the risk pool as people enroll for only 9 months per year instead of 12. However, 30 days sounds too short to me; I'd prefer splitting the difference at 60 days. Again, I'd be very interested in seeing how they came up with $5 billion per year from shortening the grace period by 2 months.

The annual individual mandate penalty is $695/year or 2.5% of household income, and it's broken out by month, so (using the $695 amount) that's $58/person per month. The average paid so far is usually a bit higher, so let's call it $70/month on average. Assuming the $5 billion is over the 5 year period, I suppose the idea is that 7.1 million people will have to pay an extra 2 months worth of mandate penalties, adding up to $1 billion per year.

The problem with this thinking is that the point of the penalty is not to generate income; it's to encourage people to enroll and stay enrolled year round. Let's say that cracking down on the grace period causes half of those people to stay enrolled for an extra two months per year. That means only $500 million/year would be generated...while APTC subsidies would have to be paid out by the federal government to around 85% (3 million) of the other 3.5 million people. The average APTC subsidy this year is around $370/month or so for those receiving it. Multiply that by two months and it's an extra $740 per APTC enrollee. Multiply that by ~3 million and you get about $2.2 billion.

Yes, that's right: Depending on how it plays out, cutting down the grace period from 90 days to 30 could conceivably end up costing the federal government around $1.7 billion per year.

Of course, if this type of gaming really is having that much of a negative impact on the individual market risk pool, it's possible that cracking down on it would help lower premiums to the point that APTC payouts would be reduced substantially for everyone, cancelling out this increase...but there's no way of knowing how much of an impact it would have (and that wouldn't show up until at least 2019 - 2020 anyway).

The other reason why the "grace period crackdown" item is so surreal is that the GOP has spent seven years...and especially the past 9 months...desperately trying to REPEAL THE INDIVIDUAL MANDATE, along with the rest of the ACA itself! Hell, there's even a GOP subcommittee trying to push a bill which would prevent the IRS from enforcing that very mandate.

Yet now they want to rely on the single most hated part of the ACA to fund CHIP and Puerto Rico assistance. Huh.

Put yet another way, the Republican Party is effectively calling for the ACA Individual Mandate penalty to be increased by 22% (11 months vs. 9 months). That's not quite how it would work in practice, but it's an amusing way of looking at it...

UPDATE: Here's the "best" (as I presume is interpreted by the House GOP) and "worst" case scenarios of cutting the grace period down from 90 to 30 days:

  • "Best" case scenario (from GOP POV): ~7.million people have to pay an extra ~$140 in mandate penalties to cover the 2 "additional" months they weren't covered. Alleged result: ~$1 billion/year in additional federal revenue.
  • "Middle" scenario: ~3.5 million follow "best" scenario above, but the other 3.5 million are inspired by the crackdown to actually stay covered for at least 11 months of the year. Result: ~$500 million/year in additional federal revenue...more than cancelled out by ~$2.2 billion in additional APTC subsidies = Net loss of ~$1.7 billion per year to the federal government. That's around $8.5 billion in additional expense over five years to the government. UPSIDE: 3.5 million people will have compliant coverage for 2 more months per year!
  • "Worst" scenario: All 7 million are inspired by the crackdown to stay covered year-round. Result: $0 in additional federal revenue...and an additional $4.4 billion in additional APTC assistance paid out, or around $17 billion over five years. UPSIDE: 7 million people will have compliant coverage for 2 more months per year!

UPDATE: Oddly, the CBPP analysis doesn't even mention the "grace period" provision of the House GOP bill at all. I'm not sure if this is because it wasn't included in the version of the bill they looked at or what, but it seems like a pretty important one to consider.

UPDATE: OK, strike that; they do address this in a separate article which actually explains quite a bit about how the 90-day "grace period" actually works which even I didn't really understand before:

People who are eligible for and receive an advance premium tax credit for the insurance they buy in state or federal marketplaces have a three-month grace period when they miss a payment to catch up before they lose coverage.

If a person doesn’t catch up on all overdue premiums by the end of the grace period, his or her coverage ends retroactively to the end of the first month of the grace period. The enrollee (1) must repay the premium tax credit that the insurer received for the first month of the grace period, (2) owes the insurer the outstanding premium for that month, (3) is responsible for the full cost for any medical bills incurred in months two and three, and (4) may owe the Affordable Care Act’s financial penalty for not having insurance for the second and third months and any subsequent months he or she was uninsured. That’s far from a free ride for an enrollee losing coverage for nonpayment.

In other words, shortening the grace period may or may not be a good idea for other reasons (again, I favor cutting it from 3 months to 2 myself), but it's even less likely to generate a billion dollars per year than I already thought.

UPDATE: (sigh) OK, Aviva Aron-Dine of CPBB seems to think that a) the $5 billion generated would be over ten years, not five ($500 million/year), and that the GOP assumes that "savings" would actually come not from additional mandate penalty revenue, but from the government saving $500M/year in APTC subsidies which they're currently paying out to enrollees. Again, not only does this completely misunderstand how the grace period works, even if it didn't, it would only work if every single person who is currently playing "catch-up" with 90 days to do so would be unable to play catch-up with only 30 days...which I'm sure is true for some folks, but what happens if most of these people are inspired to scramble and catch up after all? Goodbye some/most of that "savings".

The bottom line is that there may be other good reasons for shortening the mandate grace period (from 90 to 60 days, but not to 30)...but "this will save/generate big bucks for the federal government!" is NOT among them.