Reporting Changes in Uninsured Due to Obamacare: Government Agencies Can Be Clearer

One of the themes of this blog is that the number of uninsured Americans is not decreasing as quickly or surely as Obamacare’s supporters would have us believe. Part of the problem estimating this is a confusing series of releases from federal agencies, which has led to inconsistent interpretation by scholars.

Last month, the Centers for Disease Control and Prevention (CDC) released the results of the National Health Insurance Survey (NHIS), which I discussed under the headline “Number of Uninsured Americans Aged 18-64 Down 2 Percentage Points.” What this was referring to was that the percentage of residents in that age group who were uninsured at the time of the interview had dropped from 20.4 percent in 2013 to 18.4 percent in the first quarter of 2014. That’s about 3.8 million people.

However, the proportion who were uninsured for at least part of the last year barely budged from 24.4 percent to 24.3 percent, and the proportion who were uninsured for more than a year dropped by 1.7 percentage points, from 15.7 percent to 14 percent. That’s about 3.2 million people. Obviously, the last group is a subset of the second group. The first group is also a subset of the second group, but it is not clear how to connect the first group and the third group.

The person responding that he is uninsured at the time of the interview might have lost coverage yesterday, or many years ago. This is an important question for public policy, because the short-term uninsured are likely short-term unemployed: Fix the job market and you fix their health insurance. Those who have not had health insurance for a long time likely have a different challenge. Indeed, they may be eligible for Medicaid but not be enrolled. However, Obamacare is changing this. Its expensive outreach efforts are leading to more Medicaid enrollment than enrollment in subsidized, private Obamacare plans.

Indeed, it looks like only 2.3 million people had signed up for subsidized Obamacare coverage through the end of June 2014, which is quite a comedown from HHS’ May estimate that 8.1 million people “selected” private coverage in exchanges.

Some scholars are even looking for an increase in the number of uninsured due to Obamacare. The theory is that subsidies will not increase as fast as premiums, so people will let their coverage lapse. This will become more likely if people think their risk is limited because they are likely to get a job with benefits in a short time. So, the question of short-term versus long-term uninsured is very important. However, government agencies are confusing us on this issue.

A day after releasing the NHIS, the CDC quietly released a comparison of the NHIS data with previously unreleased data from the Census Bureau’s Current Population Survey (CPS). This reported an uninsured rate of 13.8 percent from the CPS for January-April 2014. The Census Bureau itself had only released 2013 CPS data on the uninsured, which reported 13.4 percent. Hey Presto! It looks like the uninsured rate increased from 2013 to 2014. Joe Antos of the American Enterprise Institute discussed that conclusion after comparing the estimates.

At the Incidental Economist, Adrianna McIntyre clarified that the two agencies’ releases confused two measurements of uninsured: The Census Bureau’s 2013 estimate of 13.4 percent is the proportion of uninsured for the entire calendar year, whereas the CDC’s 13.8 percent for 2014 is the proportion uninsured at the time of the interview.

This confusion should not happen. Indeed, the Census Bureau changed the CPS last year in order to improve its definitions:

One of the main differences between traditional and redesigned and health insurance instruments. The traditional instrument asks about coverage at any time in the past calendar year. The redesigned instrument also captures this information but does so in a different way: it starts by asking about current coverage and then uses follow-up questions to find out (1) when that coverage began and (2) which months the individual had the coverage. If the individual does not have current coverage, the instrument asks about coverage during the previous calendar year through the present.

However, the September release did not report detailed outcomes of these questions. It only reported directly the estimate of people without insurance for the entire year, while releasing more recent information through a side door to its colleagues in the CDC.

2014 and 2015 are critical years to measure the effects of Obamacare. We rely on the professionals at these agencies to publish data that helps us find our way through the Administration’s misinformation. Let’s hope that it gets a little clearer as the rollout proceeds.

Print