Pro tip: When you invite 200 people to your wedding, don't be surprised when 200 people show up.

Over at Politico this morning, Rachana Pradhan has an excellent article about the real-world impact of the ACA's Medicaid enrollment expansion program on, well, the expansion of Medicaid enrollment. The gist of it is that, as I've been noting for months now, additional enrollment in the Medicaid and/or CHIP programs have far exceeded "expectations" to date, with a net increase of over 12.6 million people since the ACA was passed into law (over 11.7 million of which has happened since the expansion provision officially went into effect a year and a half ago).

As far as I can tell, there are four reasons why enrollment in Medicaid/CHIP is higher than "expected":

  • 1. Underestimates of how many people were already eligible for Medicaid (pre-ACA rules)
  • 2. Underestimates of how many of that group of people would go ahead and enroll (the "Woodworker" effect)
  • 3. Underestimates of how many people would be eligible for Medicaid expansion (post-ACA rules)
  • 4. Underestimates of how many of that group would enroll (or at least how quickly they would do so).

There's a bunch of examples of individual states included, but I'll focus on Michigan in particular because I've been tracking MI's numbers so carefully and have noted and speculated about the high enrollment numbers here repeatedly.

The Politico story says:

In Michigan, where the first-year enrollment projection was 323,000 people, sign-ups hit 605,000 before falling back to 582,000 earlier this month.

It's important to note the distinction between the total eligible for the expansion program, which ranged somewhere between 477K - 500K, the total expected to enroll the first year (323K) and the number who have actually enrolled already.

The line above makes it sound like MI enrollment is 80% higher than expected, and technically that's true...except that since the state is paying 0% of the program the first three years, from a state budgeting perspective, what's relevant is how many people are enrolled starting in 2017 and beyond (when the states have to start picking up a small percentage of the cost) vs. how many people they were expecting to do so at that point. In those terms, the actual increase is likely to be far less dramatic: Since hitting a high water mark of 605K a month or so ago the number has stabilized at between 580K - 600K. Assuming it sticks at that point for the foreseeable future, that's "only" about 24% more than expected.

Here's the thing, however: When you invite 200 people to your wedding, you have to budget for the possibility that all 200 will show up. Sure, there's a few who you know won't come (that 3rd cousin who lives overseas who you only invited because your mother insisted), but beyond that, you can't just say "oh, 25% of them probably won't show up so we'll only contract with the caterer for 150 meals". That doesn't mean that you should pay for 50 chicken dinners which have to be thrown out; order, say, 170 and make some sort of contingency plan just in case the other 30 show up as well..and for the love of God, at the very least make sure that the banquet hall can hold 200.

The difference is that unlike wedding invitations where you can have people RSVP before you commit to a number of dinners, you have no way of being sure how many potential Medicaid enrollees might take you up on the offer until after the fact.

In the case of Medicaid expansion, none of this should have been surprising, at least not for the past year or so. I've been tracking the estimated number of woodworker enrollees for quite some time now, and it appears to be roughly 3.4 million people to date, or roughly 25% of the total nationally (remember, this includes through early May, whereas the 12.6M number only runs through the end of February).

The political side of this is primarily about who made the enrollment projections, when and with whom they used those projections and why they used them. If the actuaries/analysts simply misread the data in their projections, that's one thing. If the powers that be deliberately lowballed the projections in order to "trick" the state legislators/governors into expanding Medicaid, that's a different issue.

When I make projections, I generally try to include reasonable caveats. Then again, I'm not a professional analyst/actuary, and I'm doing this voluntarily anyway. If anything, I tend to err on the overly cautious side of things (although I do mess up on the high side now and then, like I did with my 2015 Open Enrollment QHP projection of 12.5 million...which, by the way, we appear to finally be approaching, just 3 months late...)

The fascinating thing to me about this particular budgeting "crisis" is that conservatives/Republicans and liberals/Democrats have two very different attitudes towards resolving it.

  • The conservative perspective is essentially, "See! More of these lazy, able-bodied poor/working class people are enrolling in Medicaid than predicted!! Plus, they can't find a doctor to take care of them since Medicaid reimbursement rates are so low!! Obamacare's a failure, we need to kick them off the program or cut services even further to save money!!"

This also raises an interesting contradiction: One of the biggest talking points attacking Medicaid is that "doctors/hospitals refuse to see Medicaid patients anyway because of the low reimbursement rates". OK, if that's true then yes, it's a problem.

However, if that's the case, then why are Medicaid costs increasing in the first place? According to the Kaiser Family Foundation, the average cost for Medicaid per enrollee for non-disabled adults is around $3,300 apiece. If that's true, and if these 12 million+ new enrollees are supposedly "not able to find a doctor willing to see them"...then where the hell is all that money going? Obviously there's some amount of overhead in running the Medicaid program itself, but I find it hard to believe that plugging a name into a database and issuing them a Medicaid card costs $3,300 per year.

Both Larry Levitt and Margot Sanger-Katz have noted that the average cost of the expansion crowd is likely to be somewhat different from the pre-ACA enrollees (different demographics, likely different healthcare statuses), but my point still stands: Even if it's "only" $2,000 apiece (or if it's higher, perhaps $4,000 apiece), you can't complain that "Medicaid costs a fortune" and complain that "none of these people can find a doctor to see them" at the same time...unless what you're actually arguing is that, say, half the new enrollees are finding treatment (but it's costing twice as much) while the other half can't find a doctor at all (thus costing effectively zero).

Pradheep Shanker, a conservative-but-reasonable MD, notes that the money "could be going to ER/urgent care" and that the "capacity for primary doctors is filling rapidly/maxed out for existing doctors." While this does confirm the "doctor shortage" claim and the "Medicaid expenses racking up" claim, it's the exact opposite of the "doctors are refusing to treat new Medicaid patients" claim.

As far as I can tell, what seems to have happened is this: Millions of people have been suffering with all sorts of ailments for years, even decades in many cases, because they couldn't afford treatment. Their condition has worsened over the years, to the point that treating them now will, yes, cost a lot more than it would have to treat them 5, 10, 15 years ago in the earlier stages of their problems.

In this sense, this is exactly the same as our national highway/bridge infrastructure problem. We've slashed budgets for road/bridge repair funding for years, to the point that actually repairing them now would cost several times more than it would have to just maintain them properly in the first place.

The difference, of course, is that a crumbling road or bridge doesn't feel pain while it's falling apart. You can make a cold-hearted budget decision to either repair/replace it or not, and accept the consequences of your decision. In the case of Medicaid, however, we're talking about human beings.

And that's the reality that a lot of people don't seem to be willing to face: Medical care for someone with long-standing ailments can be expensive. Poor/working-class people generally can't afford to pay for those treatments. That leaves you with two choices: Either tell them to continue to suffer & die (which is what we've done until now in many cases)...or ask those who have the money to help foot the bill.