The Washington PostDemocracy Dies in Darkness

A guide to understanding Obamacare’s sign-up numbers

Each time the Obama administration publishes new health law enrollment numbers, they seem to create more questions than answers. Counting the number of people who have signed up for Obamacare turns out to be vexingly difficult -- and, after four months of open enrollment, now seems like an excellent time to break down what we do and don't know about the exact number of people gaining coverage through the Affordable Care Act.

So how many people have signed up for Obamacare?

There actually isn't a great answer to this question, for a couple of reasons we can explain later. What we can say is 3.3 million people have signed up for private insurance plans through the Affordable Care Act's exchanges. We know that another 6 million people have signed up for Medicaid since they could start enrolling in the health law's expansion of the public program -- but that number includes people who would have signed up for Medicaid regardless of the health-care law, alongside those who are newly eligible.

Why is it so hard to count Obamacare enrollment?

Mostly because there's no national database of who is uninsured, who has coverage, and, if they do have coverage, where they get it. When you log onto to shop, the Web site doesn't ask you whether or not you currently have insurance coverage. Likewise, when states are reporting Medicaid enrollment data to the federal government, they aren't currently breaking down the number between a "gained Medicaid because of Obamacare" population and a "would have gotten Medicaid anyway" group. They just send off data on the total number of people who signed up.

Well, do we have any good guesses? Clearly this is an area that people are interested in!

On private insurance enrollment, I've only seen two attempts to figure out how many people on the exchange are newly insured (meaning were uninsured prior to buying exchange coverage). We're not, in other words, working with a ton of data points, but here's a guide to what we've got. There's a frequently-cited study from the McKinsey Center for U.S. Health Reform that finds only 11 percent of those buying coverage through the exchange were previously uninsured. The rest -- 89 percent were replacing a plan that they already had. It's hard to know if this is representative of the overall exchange population, because they're working with a pretty small sample size of 389 exchange enrollees -- the people in the bottom right-hand corner of the chart below.

It's possible, too, that the early enrollees would tilt more toward those with a previous plan -- these are people who valued insurance coverage enough prior to Obamacare to buy on the individual market, so they might be the most adverse to a gap in coverage. It's possible that's not the case -- with the data we have right now, we just don't know.

New York has also tried to gauge how many of its enrollees previously had insurance coverage and estimates that 66 percent were previously uninsured -- although this number applies to their entire Medicaid and private insurance enrollment, and it's near certainly true that many of the previously uninsured went into the Medicaid program.

How many people signing up are paying their first month's premium? 

This is another metric that is frustratingly difficult to track because -- and most state-based exchanges -- do not currently handle the first month's premium payment. When someone signs up for coverage online, that enrollment typically gets bounced over to the health plan responsible for collecting the money (there are a few states, like Washington and Rhode Island, that are exceptions and do collect premium payments).

Right now, the best data on this comes from the health insurance plans themselves. At the J.P. Morgan Health Care Conference last month in San Francisco, executives from large insurance companies like Aetna and Wellpoint estimated that about 70 percent of the people signing up for coverage were paying their first month's premium. Washington, one of the few states that does collect premium payments, says that just over half of people who have signed up for private insurance (90,000 Washingtonians) have paid for their coverage and slightly fewer than half (85,000) have not yet submitted a premium.

That's all private insurance. What about Medicaid enrollment?

Medicaid enrollment is just as difficult to count as private insurance enrollment. Sorry. Here, brace yourself with this video of a penguin chasing a zookeeper.

Okay, that was very nice. I'm ready. What's the deal with Medicaid?

We don't know how many of those people got Medicaid because of the Affordable Care Act. There are lots of people who were eligible for Medicaid prior to the Affordable Care Act but didn't enroll in the program. Maybe they never got around to signing up or found the paperwork too difficult. Maybe they hadn't heard about Medicaid -- but did start hearing about it in the fall, when there was lots of focus on the health law's insurance expansion. When states send the federal government information about the number of people signing up for coverage, they don't specify whether it's "Obamacare Medicaid" or "normal Medicaid." They just tell the federal government that somebody signed up for Medicaid.

Do we have guesses on how many people have enrolled because of Obamacare?

We do! Avalere, one of the best health research companies in town, has looked at trends in Medicaid enrollment prior to Obamacare, and compared them to the last three months of 2013, to estimate that between 1.1 and 1.8 million of the new sign-ups in Medicaid are due to the Medicaid expansion. That would be about 20 to 30 percent of enrollment coming from the health-care law.

The reason you see some impact of the expansion in states that are not expanding Medicaid is something that health wonks call the "woodwork effect": with more publicity surrounding the health-care law, people who were previously eligible but unenrolled come out of the woodwork to sign up.

A handful of states do also report the number of people gaining coverage because of Obamacare, although this is the exception rather than the norm. Washington, for example, counts a total of  381,000 people who have signed up for Medicaid since October. Of those, about one third -- 134,000 people -- are newly eligible for the program. An additional 63,000 were eligible already but are now signing up for the first time. And, last, 183,000 people were previously covered under Medicaid and were re-determined eligible to stay on the program.

Similarly, in Maryland, Charles Gaba's analysis suggests that about one-third of Medicaid sign-ups were people renewing policies.

Will we ever know the true number of Obamacare's Medicaid sign-ups?

Actually, yes -- and we only  have to wait a few months. The states that expanded Medicaid will have to tell the federal government how many people are getting signed up through the Obamacare at the end of the first quarter of 2014. They have to do this for reimbursement purposes: The federal government pays the full cost of anyone gaining coverage through the Medicaid expansion, but only a part of the price for people signing up for the traditional program (the state pays the rest of the bill). So by some time this spring, we should actually know how many people gained Medicaid because of the Affordable Care Act.