The Washington PostDemocracy Dies in Darkness

Obamacare enrollment numbers: what we know and what we don’t know

Senate Majority Leader Harry Reid (D-Nev.) said more than 10 million Americans are newly insured - a number that's misleading. Here's a look at other politicians who have gotten the Obamacare insurance numbers wrong. (Video: Julie Percha/The Washington Post)

“The Affordable Care Act. To date over 10 million newly insured Americans benefiting from the law are now in effect, now have the benefit of that. And there’s millions of more who have changed their insurance because of this legislation.”

— Sen. Harry Reid (D-Nev.), on the Senate floor, March 31, 2014

For months, politicians have rushed out with either wildly inflated — or deflated — numbers about the Affordable Care Act, a.k.a. Obamacare. (See the Truth Teller video above.) Just on Monday, Senate Majority Leader Harry Reid claimed that 10 million people now are “newly insured.”

Here’s what Sen. Richard J. Durbin (D-Ill.) said on Feb. 9:

“The bottom line is this: 10 million Americans have health insurance today who would not have had it without the Affordable Care Act. Ten million.”

Actually, the best estimate back then was about 4 million who were newly insured. He earned Four Pinocchios.

And then there is House Speaker John A. Boehner (R-Ohio) just two weeks ago:

“When you look at the 6 million Americans who have lost their policies and some — they claim 4.2 million people who have signed up — I don’t know how many have actually paid for it — that would indicate to me a net loss of people with health insurance.”

This nonsensical math also earned Four Pinocchios. By any measure, millions of people have gained insurance.

With the enrollment period ended on March 31, it’s time to do the numbers. What do we actually know about Obamacare enrollment numbers — and what is missing? 

The Facts

First, let’s take a look at Reid’s assertion of “over 10 million newly insured Americans.” What is the source of that statistic?

Reid’s staff pointed The Fact Checker to an article in the Los Angeles Times, which claimed that “at least 9.5 million previously uninsured people have gained coverage.” A big part of that figure was “at least 4.5 million previously uninsured adults have signed up for state Medicaid programs,” based on unpublished data from the RAND Corporation. Reid then rounded up from that figure on the assumption that the recent surge of enrollments meant the numbers had increased.

But RAND economist Katherine Grace Carman said RAND’s data, which are based on a survey of 2,600 people, simply note the percentage change in the insured; the Times then did its own calculations to translate those numbers into millions of people. Because the RAND data are based on a survey, they have a margin of error that could significantly change the results.

As of March 22, the survey found that the percentage of Americans on Medicaid had increased from 7.7 percent in 2013 to 10 percent in 2014; with the survey’s margin of error of 2 percentage points, the percentage of Medicaid recipients in 2014 could be anywhere between 8 and 12 percent. Carmen said the Times calculated that the 2.3 percent increase would translate into 4.5 million people, but that’s not RAND’s calculation. With the margin of error, the numbers could swing dramatically. As an economist, Carmen said, she felt more comfortable sticking to just percentages.

Moreover, one cannot assume that the increase in the Medicaid population reflected in the survey consisted only of people who had never been insured. A significant percentage may have been previously insured but now qualified for Medicaid because of the law’s expansion of the program.

In terms of numbers, a 4.5 million increase in Medicaid recipients would be a setback for the Affordable Care Act because the Congressional Budget Office estimated that 8 million people would be added to Medicaid in 2014. (Table B-2.) CBO also estimated that the number of uninsured would be reduced by 13 million, and the RAND survey indicates that estimate would also fall well short. “It is quite a bit lower than these other numbers,” Carmen said, referring to the difference between RAND’s calculations and the CBO estimate.

(Note: the CBO’s estimates are for all of calendar year 2014. So that means someone who signed up at the end of the enrollment period would count as two-thirds a person in the calendar year. Thus the administration did not necessarily exceed the CBO’s 6 million estimate by signing up 7 million people because it will only be known by year’s end how whether all of the enrollments added up to 6 million calendar-person years.)

Reid, in other words, fell into the trap of grabbing from a news report the highest-estimate numbers he saw – -and then inflating them a bit. (He’s done this before.)

By contrast, President Obama was relatively careful in his news conference on Tuesday. Let’s look at his specific claims.

“7.1 million Americans have now signed up for private insurance plans through these marketplaces….”

Note the phrase “signed up.” As we have repeatedly pointed out, a key test is how many people actually pay the first month’s premium. No one really knows what the trend is, but as a rough guide, The Fact Checker is using an 85 percent payment rate. That is the experience in California, and the Golden State accounts for about one-fifth of enrollments. An 85 percent payment rate would translate into about 6 million actually paying premiums.

(Why would people sign up but then decide not to make payments? One reader, Yossi Gestetner, sent The Fact Checker his account of the choices involved. While he qualifies for subsidies, his plan also comes with deductibles and co-pays that he thinks are not affordable if he or his wife have major medical problems. But his experience may not be representative. A UC Berkeley study released April 2 documented that there is a fair amount of churn in the insurance market, so people may not make a payment because they found other insurance, such as through a job.) 

As a result of the payment-rate question, some readers tweeted that The Fact Checker should brand this White House tweet as a lie:

But, wait, there’s more. In every state (except Vermont and in the District of Columbia), you can buy insurance outside the exchanges — by going directly to insurance brokers, agents or companies.  At least 550,000 people have already purchased qualified health policies “off exchange” and the real number is almost certainly double that. (Update, April 3: Blue Cross Blue Shield said they sold 1.7 million ACA-compliant policies off the exchanges between October 1 and March 1.) So the net result is at least 7 million, though it should be noted that CBO, in its estimates, assigned such policies to a different category than exchanges. It is not entirely clear in CBO’s tables but essentially the agency estimated that about 8 million people in 2014 would obtain insurance directly from insurance companies.

What percentage of the 7 million are “newly insured?” No one really knows at this point, though a reasonable, conservative estimate would be about one-third. That’s a little over 2 million people.

“…on top of the more than 3 million young adults who have gained insurance under this law by staying on their family’s plan…”

This is a two-year-old figure, based on a Health and Human Services Department estimate, and so it is a bit dubious. The Fact Checker dug into the more recent reports and found that it was based on a single quarterly figure that was not sustained in later quarters. However, the overall increase in the percentage of insured young adults suggests a gain of about 2.8 million from the third quarter of 2010 to the first half of 2013. Given quarterly fluctuations, a more conservative approach would be to use the average figure for 2010 as the base; that would reduce the increase to 2.2 million people. We would prefer if HHS updated its report, but 2 million is still a pretty good number.

Update, April 3: Avik Roy raises some further good questions about this data and estimates the true figure is at best under 1 million. This is worthy of further inquiry, but it demonstrates that administration’s claims about the under-26 enrollment should be treated with extreme caution.

“…on top of the millions more who have gained access through Medicaid expansion and the Children’s Health Insurance Program.”

The president, perhaps mindful of the Four Pinocchio rating he previously received when he cited a specific figure, uses an appropriately fuzzy figure of “millions.”  Through January, the health consulting firm Avalere estimates that at least 2.4 million people got insurance for the first time through the Medicaid expansion, and of course that figure is going to increase. As we noted, CBO expects an additional 8 million people to sign up for Medicaid. But until the data can be closely examined, specifics on the Medicaid sign-ups are best avoided.

Note also that Obama, unlike Reid or Durbin, made no sweeping claim that all of these people were “newly insured.”

The Bottom Line

We’re not giving Pinocchios today because even Reid’s claim, with his reliance on high-end estimates, might actually turn out to be correct, by sheer luck, when the dust settles and the data can be fully examined. The real Medicaid numbers, in particular, are a mystery.

In general, politicians need to be cautious in how they talk about the ACA enrollments. You can’t just add up figures from various reports, with different methodologies, without getting in trouble. Obama’s more recent language, with his transparent explanation of the various numbers, is in the ballpark.

(About our rating scale)

Send us facts to check by filling out this form

Follow The Fact Checker on Twitter and friend us on Facebook

Senate Majority Leader Harry Reid (D-Nev.) said more than 10 million Americans are newly insured - a number that's misleading. Here's a look at other politicians who have gotten the Obamacare insurance numbers wrong. (Video: Julie Percha/The Washington Post)