Exchange Enrollment: Number Previously Uninsured

VII. Key Issues: Regulation & Reform >> C. Health Reform >> Affordable Care Act (ACA) >> ACA and Government >> ACA and States >> ACA Health Exchanges >> Exchange Enrollment: Number Previously Uninsured (last updated June 22, 2014)

Overview

Lack of Administrative Data. Regularly reported statistics do not show how many enrolled through the Exchanges were previously uninsured. According to Wall Street Journal 3.17.14, an insurance company official said that carriers would have information on who has paid insurance premiums, but they rely on the government’s application form to provide them with information about new enrollees characteristics and hence would not have information on who previously lacked insurance.

Baseline Characteristics of Non-Group Market Prior to Exchanges.

2011. According to Wall Street Journal (1.17.14), “about 66% of people buying new individual health plans in early 2011 were covered by employer-backed plans in late 2010, according to a Kaiser Family Foundation analysis of federal survey data prepared for The Wall Street Journal. About 20% of enrollees in early 2011 were previously uninsured, the analysis found.”
2014 Exchange-Eligible Market. McKinsey & Co. has conducted an on-going survey of the Exchange eligible population since October 2014: “The total pool of Qualified Health Plan (QHP) eligible from which we drew our adult survey respondents is approximately 24 million uninsured and 10 million individually insured, as well as those who were insured by their employer in 2013 and are now QHP eligible (QHP eligible defined as over 138 percent of the federal poverty level (FPL) in Medicaid expansion states and over 100 percent FPL in non-Medicaid expansion states, among those 18 to 64 years old). We refer to this as the addressable 2014 individual market.”
- Using exact figures reported in Exhibit 3 (24 million uninsured, 9.7 million insured), the uninsured constitute 71.2% of this market; this would be the expected uninsured share if both groups were equally likely to enroll.
- In the latest released McKinsey survey (using February survey results), there were 1,013 previously uninsured respondents and 1,083 previously insured respondents (Exhibit 2). If the uninsured had enrolled at the same rate as previously insured, the uninsured share would have been 48.3%.

Estimates of Uninsured Share of Enrollments

Because of the potential importance of paid enrollments, some analyses/estimates have sought to differentiate whether uninsured estimates relate to just paid enrollments or total enrollments. As well, some available uninsured estimates relate to the entire non-group marketplace, including both Exchange enrollments and off-Exchange enrollments. There are widely varying estimates of off-Exchange enrollments. Since these latter figures are used in some estimates of the uninsured share of enrollments, we start with a summary of such estimates.

Off-Exchange Share of Non-Group Market Enrollments

20% (Expert Opinion/WellPoint). Scott Gottlieb reports (3.17.14) that one Wall Street analyst has indicated that off-Exchange enrollments may be as high as 20% of total non-group market enrollments. He also asserts WellPoint WLP +0.16% has also said that about 20% of its applications came off exchanges, from customers going directly to the insurer.
31% (Highmark). Gottlieb also reports (3.17.14) The one data point we have on “off exchange” enrollment trends comes from Highmark, which recently announced that 31% of the 133,000 people it has enrolled across the three states where it operates (Pennsylvania, West Virginia, and Delaware) signed up off exchanges.
36.2% (Compilation). Charles Gaba, at acasignups.net, has estimated the number of off-Exchange enrollments from the states and insurance companies that have reported this info; these range from a low of 14% in Wisconsin to a high of 69% in Washington. Blogger Uri Manor has used these figures to calculate a weighted average of 36.2%. This compilation includes the WellPoint and Highmark figures cited above.
>51% (McKinsey Survey). The latest McKinsey survey conducted in February shows 60% of the Exchange eligible population renewed old plans (n=601) and 40% switched plans (n=289) or newly enrolled (n=106). By definition, renewals must have occurred off-Exchange. If the results are weighted to offset the undersampling of uninsured individuals (there were 247 uninsured for every 100 people with prior coverage in the sampling frame, but only 94 per 100 in the reported sample, so re-weighting requires multiplying each uninsured respondent by 247/94), the corrected off-Exchange share is 51%. However, this is a lower bound since new enrollments also could have occured through off-Exchange plans.
60% (RAND Survey). LA Times reports (3.31.14) that RAND surveys show 9 million off-Exchange enrollments (“the vast majority of these people were previously insured”) and 6 million on Exchanges.

Uninsured Share of Paid Enrollments on Exchanges

18.4% Uninsured (McKinsey Survey). A February 2014 McKinsey survey of 20,096 Exchange-eligible consumers found that previously uninsured respondents accounted for 27 percent of the 395 respondents who reported having selected a new 2014 product (i.e., insured who switched and previously uninsured who enrolled). However, only 53% of previously uninsured had paid their first month’s premium, vs. 86% among the insured. Thus, among paid enrollments, only 18.4% were previously uninsured (paid uninsured = 53% x 106=56; paid insured = 86% x 289=249, from figures in Exhibit 4). This survey includes enrollments on and off the Exchanges, so this percentage may be only a rough estimate of the uninsured percentage among Exchange enrollees. As well, Exhibit 5 shows that 65% of those who had not yet enrolled but intended to enroll were uninsured (however, similar percentages were reported in November, December and January even though actual enrollments showed a far different pattern).
23% Uninsured (Estimated Assuming 20% Off-Exchange Enrollment). Blogger Uri Manor notes that the McKinsey figures include off-Exchange enrollments; thus 18.4% may not be an accurate estimate of paid enrollments on the Exchanges. Assuming 20% accurately depicts off-Exchange enrollments nationwide and that 100% of off-Exchange paid enrollments are from people who previously had insurance (this makes the ACA looks as good as possible), would imply 23% of paid Exchange enrollments were previously uninsured.
24% Uninsured (Estimated). LA Times reports (3.31.14) “about one third” of the 6 million Exchange enrollments were previously uninsured. This is based on “newer survey data from the nonprofit Rand Corp. and reports from marketplace officials in several states.” When combined with McKinsey estimates of payment rates (53% for uninsured and 89% for previously insured), the overall uninsured share of paid enrollments is 24%, as reported by Avik Roy.
25.0% (Goldman Sachs). A GS analysis estimates that as of March 11, 2014, Exchanges had gotten a net total of 4 million paid enrollees, of which only 1 million were previously uninsured. PunditFact reports (3.26.14) that the GS estimate was based on McKinsey Survey data. However, McKinsey actually reported that 27% of all enrollments were uninsured–paid and unpaid, hence one can impute that the uninsured share of paid enrollments is only 18.4% (see above)..
28.8% Uninsured (Estimated Assuming 36.2% Off-Exchange Enrollment). Assuming 36.2% accurately depicts off-Exchange enrollments nationwide and that 100% of off-Exchange paid enrollments are from people who previously had insurance (this makes the ACA looks as good as possible), would imply 28.8% of paid Exchange enrollments were previously uninsured.
43% Uninsured (RAND Compare Model). The RAND Compare micro-simulation model predicts that 43% of Exchange enrollment will be uninsured in 2016 (Table 1). It is not obvious whether that fraction would be higher or lower in 2014.
59% Uninsured (Estimated Assuming 60% Off-Exchange Enrollment and Correctly Weighted McKinsey Figures). If the McKinsey figures are correctly weighted to offset the undersampling of uninsured individuals, the corrected uninsured share of non-group enrollment (on and off the Exchange) is 49.1% (see 49% under Uninsured Share of Total Enrollments). If 53% of the uninsured pay, as per McKinsey survey, then there are 260 previously uninsured who have paid for every 1,000 enrollees on and off the Exchanges. If 89% of previously insured members pay and the chances of payment are the same on and off the Exchange, and all the formerly uninsured who pay are assumed to join Exchange plans, then the uninsured amount to 59% of paid enrollments on the Exchange.

Uninsured Share of Exchange Enrollments

These figures focus exclusively on enrollments through the Exchanges, i.e., excluding off-Exchange enrollments.

23% Uninsured (RAND Survey). The Mail Online reports (4.1.14) “The unpublished RAND study – only the Los Angeles Times has seen it – found that just 23 per cent of new enrollees had no insurance before signing up.”
Less Than 33% Uninsured (Wall Street Journal). Based on several surveys of enrollment through December and interviews, Wall Street Journal reports: “insurers, brokers and consultants estimate at least two-thirds of those consumers previously bought their own coverage or were enrolled in employer-backed plans.”
About 33% Uninsured (LA Times). RAND researchers have been surveying 3,300 people monthly since Fall 2013. LA Times reports (3.31.14) that “at least 6 million people have signed up for health coverage on the new marketplaces, about one-third of whom were previously uninsured.” This is based on “newer survey data from the nonprofit Rand Corp. and reports from marketplace officials in several states.” It appears to conflict with the report that the RAND survey data show only 23% uninsured (see above).
35% Uninsured (HealthMarkets, Inc.). HealthMarkets Inc., an insurance agency that enrolled around 7,500 people in exchange plans, said 65% of its enrollees had prior coverage.
42% Uninsured (Estimated Assuming 36.2% Off-Exchange Enrollment). Blogger Uri Manor has used the McKinsey figure of 27% uninsured among all enrollments (paid and unpaid), the 36.2% off-Exchange enrollment figure, and assumed that none of the uninsured enrolled off-Exchange to calculate that the uninsured represent 42% of Exchange enrollments.
“About Half” Uninsured (Expert Opinion). According to PunditFact, Robert Laszewski “has spoken to some of the largest insurance carriers who are selling through the marketplaces. From that, he guesses that about half of the people buying there were previously insured.”
59% Uninsured (NY). Charles Gaba reports (3.25.4) that 59% of Exchange enrollees in New York were previously uninsured.
66.7% Uninsured (Estimated Assuming 72.4% Off-Exchange Enrollments). Manor argues that 36.2% is a conservative figure since the evidence we have on off-Exchange enrollments is coming exclusively from companies that offer plans on and off the Exchange, thereby ignoring companies selling exclusively off-Exchange. If off-Exchange enrollment is double the 36.2% figure he estimated and the remaining assumptions above (see 42% Uninsured bullet) apply, then the uninsured would represent 66.7% of Exchange enrollments.
70% Uninsured (CA). According to Wall Street Journal (3.17.14), California is one of few states tracking the number of previously uninsured Exchange enrollees: 70% of those enrolled through the state’s Exchange site were previously uninsured. However, PunditFact reports that the NY figure “mixes together those who signed up for Medicaid and those who bought private insurance.”
75-80% Uninsured (KY). Charles Gaba reports (3.25.4) that 75% of Exchange enrollees in Kentucky were previously uninsured.PunditFact reports (3.26.14) “Kentucky’s Cabinet for Health and Family Services told PunditFact that 80 percent of the people who bought private insurance through their marketplace reported that they didn’t have insurance. But spokesperson Jill Midkiff said the intake form does not ask how long they have been without coverage. It could have been as short as one day or as long as a year or more.”
57% Uninsured. (Kaiser Family Foundation Survey of Non-Group Health Insurance Enrollees). This survey of 742 individuals from early April to early May finds that 57% of those who bought insurance on the exchanges were previously uninsured. Seven of every ten of this group purchased insurance because of the ACA’s mandates. (Kaiser Family Foundation, June, 2014)

Uninsured Share of Total Enrollments

25% Uninsured (Priority Health/Goldman Sachs). At Michigan-based Priority Health, only 25% of more than 1,000 enrollees surveyed in plans that comply with the law were previously uninsured.
27% Uninsured (McKinsey Survey). The February 2014 McKinsey survey found that previously uninsured respondents accounted for 27 percent of all enrollments (paid and unpaid). An earlier McKinsey survey had found that only 11% of those signing up on the Exchanges had been previously uninsured (based on a sampling of 4,563 Exchange-eligible consumers performed between November and January, of whom 389 had actually enrolled).
49% (Correctly Weighted McKinsey Survey Data). If the McKinsey figures are weighted to offset the undersampling of uninsured individuals (there were 247 uninsured for every 100 people with prior coverage in the sampling frame, but only 94 per 100 in the reported sample, so re-weighting requires multiplying each uninsured respondent by 247/94), the corrected uninsured share of non-group enrollment (on and off the Exchange) is 49.1%.

Lack of Interest Among Uninsured

McKinsey Survey. Of the Exchange-eligible population that is uninsured, the February survey showed that 56% have not shopped (compared to 7% among the insured) and another 34% have shopped but not enrolled (compared to 11% among the insured). Only 10% had actually selected a new plan (Exhibit 2). About 50% of respondents cite affordability as an issue inhibiting them from enrolling even though over 80% who cite concerns about affordability are eligible for subsidies (however, “66 percent of these consumers were not aware of their subsidy eligibility status or subsidy amount”).
Robert Wood Johnson/Urban Institute Survey. It was reported (3.24.14) that a survey by Robert Wood Johnson Foundation and the Urban Institute, produced data that “is similar to other recent findings about the exchanges showing that the uninsured continue to remain uneducated and confused about coverage under the law.” The new survey showed:
- “that 48 percent of uninsured respondents did not plan on looking for information on the exchanges or had not heard about them.
- “On the other hand, those who haven’t looked yet said they might if they knew more about the financial assistance available or knew where to go to explore their options or enroll in a plan.”
- “Also, 52 percent of uninsured respondents said they have looked into their options under the law — or at least are planning to before the March 31 deadline.”