The Department of Health and Human Services just released enrollment figures for Obamacare in October. How you assess them depends entirely upon your baseline and expectations.

Here’s the quick summary:

According to HHS calculations, 846,852 people have used the site to complete applications. That means they have created accounts and submitted information to see whether they are eligible for federal programs or tax credits. Those applications include people applying for households with multiple members. In total, it represents 1,509,883 people. The federal government has processed applications for the vast majority of them—98 percent, or 1,477,853 people. Of those, about a third have actually selected a health plan or been deemed eligible for a program like Medicaid. That’s 502,466.

How does that half million break down? About four out of five (396,261) are in Medicaid. The rest (106,185) of them have picked private insurance plans. These numbers include both those who enrolled through the website that the federal government is maintaining ( and those who enrolled through sites that states like California, Kentucky, and Connecticut are running on their own. The majority (three-fourths) of the people getting private insurance have done so through state sites. Just a quarter, or 26,794, have enrolled through the federal site.

The 28-page report HHS released breaks down these numbers into greater detail. And the numbers are probably a little conservative, because some states haven’t submitted final figures and people who registered directly with state offices handling Medicaid or Children’s Health Insurance Program enrollment don’t count toward the total.

But most people will focus on the topline numbers, out today. The question is how to judge them. And that comes down to how you think about two separate issues:

1) What counts as an enrollment? HHS counts somebody as enrolled once they have selected an insurance plan. Critics have said enrollment should only count for somebody who has written a check. The administration notes that doesn’t actually have a payment option. You have to get a bill from the insurer. Most people aren’t going to pay now for a policy that begins January 1, they say, adding that many insurers don’t even ask for payment until December. Critics say it doesn’t matter. Enrollment figures shouldn’t count if no money has changed hands.

2) What’s a realistic baseline?  You’re going to hear a lot about an internal HHS projection, later obtained by the Associated Press, suggesting half a million people would sign up for private plans in the first month. The real number is obviously way, way short of that figure—and surely has something to do with problems on the website, which have made the enrollment process slow and cumbersome. But that HHS projection is not gospel. For another baseline, it’s helpful to look exclusively at Massachusetts. I’ve described the situation previously:

Of the 36,167 people who eventually enrolled in premium-charging plans from Commonwealth Care, 123 signed up in the first month. That’s right—one hundred and twenty-three, or about 0.3 percent. Over the first two months, the number was a bit larger—2,289. But that’s still just 6.3 percent.  

The analogy to Obamacare is far from perfect, in that Commonwealth Care didn’t include wealthier people who didn't qualify for subsidies. (In the Massachusetts scheme, they essentially had a separate exchange—and enrollment there began half a year later.) Also, the Massachusetts open enrollment period was twice as long. So it’s reasonable to expect that, with a fully functional website, early enrollment in Obamacare private plans would be higher than those numbers above suggest. But the general point stands. Very few people sign up for insurance in the first few months. Most wait until much later in the game. 

If you do the math, you’ll see the October Obamacare numbers compare favorably if you compare them to the first month in Massachusetts, but unfavorably if you compare them to the first two months. The low enrollment in the federal website is further confirmation that has worked very poorly.

These numbers are likely to be a Rorschach test of sorts. People who like Obamacare will note the high enrollment in Medicaid—that’s nearly a half million people who just got insurance. They’ll point to the large number of people with eligibility determinations in hand, ready to buy insurance. People who dislike Obamacare will say the actual enrollment numbers are way below the official projection—and that a lot of those people with eligibility determinations may be waiting to buy because they don’t like what they see.

Here's what Larry Levitt, senior vice president at the Kaiser Family Foundation, told Greg Sargent of the Washington Post:

...a million people have been determined eligible. That means if the website had been working well, and a million people had gotten to the end of the process, we’d be looking at a very different trajectory now. We heard about the surge in traffic when went live. This suggests there is in fact a lot of interest. ... Assuming the website gets fixed, I would assume a surge of enrollment in December, and another surge in March. ... We’ve heard all about the negatives of the Affordable Care Act. This is the first tangible sign of what the ACA set out to do, which is to cover people.

Me? Barring unforeseen developments or wrinkles in this report, I’m sticking with my story. The October numbers are low, which was to be expected given the website problems and tendency of people not to buy insurance right away. But what matters isn’t the figures for October or even November. It’s December and the months that follow—particulalry into next year, as the prospect of paying fines for uninsurance start to hit people in the face. "It's too early to say anything useful," says Jonathan Gruber, professor of economics at MIT. "And, really, I don't think we can draw any significant conclusions about effectiveness of the law until March, because any firm conclusion requires effects of individual mandate to be felt."

Updated: I added the quotes from Larry Levitt. The full item, by Greg Sargent, is worth reading.