Data from Georgia’s Insurance Commissioner indicates that insurers received 221,604 applications through the marketplace through the end of March. Premiums had only been received for 107,581 of those policies by March 31st, covering 149,465 lives. Given that more than twenty percent of the total enrollment occurred in the last two weeks of March it is likely that the final number of covered lives is likely to be higher. About two thirds of paid enrollees through the exchange are receiving subsidies.
When all is said and done the numbers of individuals covered through the exchange reflect about 40% of Georgia’s individual insurance market. Especially in Georgia however many of those who had coverage in the individual market continued with that coverage purchased outside of the marketplace.
What is not yet know is how many of those who purchased coverage in the marketplace were previously uninsured. The Rand Corporation’s survey completed in mid-March estimated that nationwide about 35 percent of the marketplace enrollees came from the ranks of the uninsured. That survey was completed before the surge in Marketplace enrollment in the last two weeks of March. It is likely that the percentage of uninsured in that surge was at least slightly higher in the last two weeks than previously as people moved to avoid the non-coverage penalties. Definitive information on changes in insurance coverage will have to wait until public data sources are made available.
One of those sources of data, the Census Bureau’s Current Population Survey, has been at the center of some pretty silly conspiracy theories over the last couple of days. The issue is that the survey is being changed as a response to a long history of researcher complaints about its methodology. The change affects the survey that was conducted last month so some have argued that the change reflects the Obama attempts to “cook the books”. These comments reflect a complete lack of understanding of the survey and how it’s conducted.
In March of each year the Annual Social and Economic Supplement to the Census Bureau’s Current Population Survey (CPS) is conducted. The Census Bureau surveys approximately 89,000 families across the nation, receiving information on almost 210,000 individuals. One section of the survey solicits information on the source of health insurance coverage for each individual in the household. Respondents are asked about their health insurance coverage for the preceding year, so the data collected last month reports insurance coverage for 2013.
The problem has been that the survey as previously designed consistently underreports the number of individuals enrolled in Medicaid and thus overstates the number of uninsured. This issue has been recognized for over a decade and many users of CPS data attempt to adjust for these errors.
The Census Bureau has redesigned the questions starting with the 2014 survey to provide more accurate estimates of coverage in 2013. Since the major changes in coverage are occurring now in 2014 the newly redesigned survey will provide a more accurate baseline of coverage estimates before the Affordable Care Act’s most important provisions take affect. The 2015 survey which will have estimates of coverage for 2014 will then provide us with more complete data on the impact of the Affordable Care Act. Far from “cooking the books” these more accurate estimates are likely to provide estimates that have a smaller number of people gaining coverage than the older survey would have produced.
For an accurate assessment of the impact of the ACA this is the perfect time to make these changes in the Annual Social and Economic Supplement to the Census Bureau’s Current Population Survey. If those changes had been delayed a year or more it would created a discontinuity in coverage estimates at the exact time health insurance coverage undergoing major change. By doing it this year we have a more accurate picture of coverage before the ACA and during its implementation.