Young Adult Enrollment Numbers

Get the Facts on Young Adult Sign Ups Under the ACA

ObamaCare enrollment numbers for young adults are about 2.6 million under the age of 26. The 18-34 demographic, the young adult demographic, disproportionately qualifies for cost assistance and Medicaid, this led to high enrollment numbers (28% of all enrollments) and an expectation of high enrollment numbers during next open enrollment. Last minute sign ups surged in the young adult demographic putting to rest the fear that not enough healthy young people would “enroll in ObamaCare”. Initial reports from insurers show that a large number of young adults signed up outside of the marketplaces, these numbers haven’t been confirmed yet, but point toward better than expected enrollments under the ACA.

Below we will take a look at how the ACA affected signups in the young adult demographic during ObamaCare’s first enrollment period, who signed up for Medicaid and CHIP, how each state faired, and what we can expect moving forward for young adult enrollment numbers.

The following article specifically focuses on the 18-26 age group (the “sub26ers” who could stay on their parents plan). Total young people enrollment numbers include all types of enrollment inside and outside the marketplace for the 18-34 demographic (which is typically considered the young adult demographic).

ObamaCare Enrollments and Young Adults

by Paul Mullen

A major concern has been the large number of young people who lack health insurance. Historically 37% of people from age 19-25 and 25% of 26-35 are uninsured, far higher than any other age group. Before the ACA employer family insurance typically covered children up to 18, dropping them on their 19th birthday unless they were a full-time student. C.H.I.P. which covers many children in lower income families, also cut off at a child’s 19th birthday.

These young people trying to establish themselves in the employment market, have a high unemployment rate, often work in low-paying jobs, and change jobs more frequently. This makes them much less likely to be covered by employer insurance. Public insurance programs such as Medicaid or Medicare are usually not available to them unless they are disabled. While individual insurers were keen to recruit them, as they generally have very low health costs, only a small proportion took this option up, either because they regarded themselves as “Young Invincibles” or because their money was tight and they could think of better things to spend it on (like beer and cigarettes). The Commonwealth Fund have produced detailed reports on these young people.

The ACA allows children to stay on their parent’s or guardian’s health plan until their 26th birthday. To be eligible for this coverage, children do not need to be financially dependent on you for support, claimed as dependents on your tax return, residents of your household, enrolled as students or unmarried. Spouses and any grandchildren however are not included. If you have a grandfathered group health plan, it didn’t have to cover a child who has access to group coverage elsewhere until 2014. The law also applies to parent’s individual plans that include family coverage. Plans that offer employee only coverage, and do not offer family coverage would not have to cover children.

This provision came into effect on September 23rd 2010 but in most cases newly eligible children could not be added until open enrollment for the next plan year starting after that date. Some employers however with an open enrollment period earlier in the year, started covering older children earlier.

There are several surveys that provide evidence of the number of young people that gained cover through this provision. However it is not as simple as looking at the % uninsured rate for this age group because over the same period (and longer) rates of employer insurance have been going down, and also younger people have been particularly badly hit by the recession. Hence we would expect that if it were not for this ACA provision, we would have seen a rise in the rate of uninsured among the 19-25 group. The best way to measure the impact of this measure is to compare the uninsured rate in the 19 to 25 year olds with the 26-34 year old group.

First (simply because it is the source most quoted by the government) the CDC National Center for Health Statistics, with the help of the Census Bureau, runs the National Health Interview Survey, which interviews about 75,000 individuals a year in their homes. It is collected throughout the year and preliminary results are released on a quarterly basis, about 6 months after the end of the quarter. Thus the most recent results are for January to September 2013, released in March 2014. Unfortunately the standard NHIS age groups are 18-24 and 25-34, which is not much use for our purpose where we want to compare how 19-25 year olds are doing. Although 19-25 data is given in an appendix, there are no figures for 26-29 or 26-34 to use as a control group.

The widely quoted figure of 3.1 million young people who newly gained insurance through this ACA provision comes from an analysis of NHIS data by Sommers. This uses a special tabulation of data from the NHIS for years 2008 to 2011.

The data used by Sommers came from the 2011 early release data, published the same month. This interestingly is the only publication in this quarterly series that includes a Table 8 which shows both the 19-25 age group and the 26-35 age group that provides the closest available comparison. It is also the only publication in the series to give this data on a quarterly basis.

Cohen also uses a similar slide in a conference presentation  but goes one further and shows the matching rise in private health insurance among the 19-25 year olds.

We can now see clearly that private insurance went on declining and the uninsured rate rose among the 26-35 year old group while private insurance climbed remarkably among the 19-25 year olds. It took almost a year after the 23 September start date for the full effect to take place, presumably due to the timing of open enrollment periods when children could be added to a parents plan.

Critics have claimed that the Sommers study “cherry picked” data. In fact when it came out it used the latest data available. They suggest that results would have been worse if later data had been used, and that was why ASPE had not updated the study. In fact Sommers left ASPE to take up a post at Harvard University but his work was continued at the CDC’s National Center for Health Statistics. The 2012 NHIS results were published in June 2013 and contained similar data in Table 8.

Looking at the quarterly data (chart below) we see a lot of variation (each quarterly figure has a standard error of around 1.5). With the exception of an odd value in Q2 2012, the % of 19-25 year olds with private insurance has declined steadily until 2011 and then leveled out in 2012. There is a corresponding increase in uninsured for that age range.

However when you look at the 19-25 year olds, they start by having a consistently lower rate of private insurance and this worsens until 2010 Q3. After that, at just the time this provision of the ACA started to take effect, their rate of private insurance starts to increase, though the full effect takes until the middle of 2012 after which both age groups have essentially the same rate for private insurance, and also a slightly lower rate for % uninsured.

Because we cannot determine if the variations from quarter to quarter are due to sampling variation or some seasonal factor, it appears more reasonable to look at year on year changes. Since the effect of the change in law in September 2010 took time to work though the various different plan enrollment dates, only 2012 shows the full effect of the law. But which year should we compare it to? 2008, or 2009, or perhaps the first 3 quarters of 2010?

Comparing full year data for 2008 to full year 2012 we see that uninsured rate for ages 19-25 fell from 30.9% to 26.4%, while uninsured rate for ages 26-34 rose from 25.5% to 27.7%. That is a 6.7% change in the difference between the two age groups. Similarly the percentage with private insurance among 19-25 year olds rose from 55.7% to 57.2% while for 26-34 it fell from 63.7% to 58.7%, which is a relative improvement (from -8.0% to -1.5%) of 6.5%.

A better way of looking at the 2008 data is to note that as uninsured levels rise, so too does the difference between these age groups. So it is more reasonable to suppose that without ACA the uninsured among ages 19-25 would always be about 22.6% higher (the average for years 2008 to 2010). This suggests that by 2012 without ACA the uninsured % for ages 19-25 would have been 32.4%, so the improvement is 8.0%

Using 2009 as the base year, uninsured rate for 19-25 fell from 32.7% to 26.4% while uninsured rate for 26-34 rose steadily from 25.5% to 27.7%. So there was a relative improvement (from -7.2 to +1.3%) of 8.5%. Similarly the % privately insured for 19-25 year olds improved from 52.6% in 2009 to 57.2% while for 26-34 year olds it fell from 61.3% to 58.7% So the relative improvement (from -8.7% to -1.5%) was 7.2%

If we use 2010 as the base year, we see a 7.5% drop in uninsured for 19-25 year olds at a time when uninsured rate of ages 26-34 remained fairly constant (increasing by an insignificant 0.2% So that is a 7.7% improvement. Similarly Private insurance for ages 19-25 rose 6.2% while for ages 26-34 it fell by 0.8%, making a relative improvement of 6%.

Finally we can compare the difference between the average for 2008 to 2010 and 2012. Uninsured fell by 6.1% for ages 19-25 and rose by 1.2% for ages 26-34. Private insurance rose by 4.1% for 19-25 and fell by 2.8% for ages 26-34. So relative changes are uninsured -7.3%, privately insured +6.9%.

So, 6.7%, 6.5%, 8%, 8.5%, 7.2%, 7.7%, 6%, 6.9% or 7.3%?

A general problem with interviews is that many people don’t know what kind of insurance they have. For example they may have a state Medicaid program which, if they live in WA) is called “AppleCare”, or in an increasing number of states Medicaid may be administered by a big health insurer, whose name is on the insurance card. So even though it is still Medicare, they may report it as private insurance. So I prefer to stick with the numbers reported as uninsured, although the increase in private insurance at the same time provides confirmation that the reduction is due to the Under 26 rule in ACA.

It is obviously impossible to know how many 19-25 year olds would be uninsured in 2012 if the change had not happened. Using the proportionate change method appears to me to be the most reasonable approach (since it fits years 2008, 2009 and 2010) and using this for 2012 results in a 8% increase in insured for ages 19-25 with a 1% error margin. Since there are about 31 million aged 19-25 in 2012 we can assume that about 2.5 million young people have gained coverage due to the ACA.

There is however an additional effect to take into account. Young people seem to have much less job stability compared to their older workers. So those with employer insurance provided by their own job are much more likely to experience periods of being uninsured than if they are on a parent’s plan. As they lose one job and then gain another one, which may impose a waiting period before cover starts, those who were covered by employer provided insurance at the time of the survey may have been without insurance for much of the year.

Kirzinger, Cohen and Gindi discuss this in NCHS Data Brief #137 which was published in December 2013. First, looking only at young adults who were privately insured at the time of interview they asked how many had been uninsured for part of the preceding year. The big spike in Jan-July 2011 is obviously people now covered on their parent’s policy who had previously been uninsured. Apart from that though we can see that this represents the stability of private insurance for young people – they easily lose it if it is from their employer and they lose that job. Those aged 19-25 are much more likely to have been uninsured at some time in the past year than those aged 26-34. Yet if listed on their parent’s plan they are less likely to have experienced periods of insurance, presumably because parents (at least those with employer insurance) are more likely to have job stability.

The study also illustrates how many young people switched from their own policy to a parent’s plan. These people were not counted in the previous exercise since they still have private insurance but likely gained from more stability of cover, better cover and lower premiums. All these issues were analyzed by the Commonwealth Fund surveys.

The Gallup-Healthways poll goes back to January 2008 and interviews some 500 people every night. We will discuss more about it elsewhere. They most recently looked at young people in July 2013.

While their sample size if far larger than the NHIS (about 180,000 a year) , this is a telephone poll and much less detailed than the NHIS in-person interviews. Also their published tabulations do not use exactly the same age groups as we would like to see. Here they compare 18-25 year olds (though 18 year olds were not affected by the change) with the whole 26 to 64 age group. Before September 2010 the difference in uninsured rate varied between 12.6% and 9.4% while recently it has hovered around 4%. So there has been a drop of about 6.2% but since the change only affected 6/7ths of the age group, that works out at about 7.2% drop in uninsured for those affected. Comparison with the whole 26 to 64 age band, rather than the more comparable 26-34 age group will also have lessened the impact.

Next, the most detail on the reasons young people lack health insurance, and the effects it has on them, comes from the biennial surveys conducted by the Commonwealth Fund. Comparing the results of their 2010 and 2012 surveys, they found that 79% of 19-25 year olds were insured at the times of the 2012 survey, up from 69% in 2010. This represents 3.4 million fewer uninsured young adults. Meanwhile the uninsured rate for 26 to 49 year olds had increased by 2%. Sample sizes were small however (4432 responses in 2012).

This chart also illustrates the problem of frequent job changes for young people: many covered by insurance at the time of the survey had experienced periods without insurance during the last year.

13.7 million 19-25 were on parent’s [policies, 6.6 million as result of ACA Still 39% of young adults were without health insurance for all or part of 2011. 60% reported delaying care because of cost and half had problems with medical debt. Low income groups were less likely to benefit, presumably because parents did not have employer coverage or it was unaffordable to them. The Commonwealth Fund study also showed that young people were more likely to get better insurance: underinsurance was a major problem for those in low income groups.

My conclusion is that anywhere between 2.3 and 2.9 million young people gained insurance as a result of the ACA, with a best estimate of 2.6 million.

Young Adult Enrollment Number Resources:

We suggest checking official releases from HHS, the CBO, the US census, and popular polls like Gallup for the most current sign-up information (we used all of these sources in our breakdown). Another great resource is acasignsups.net which provides pretty detailed and unbiased looks at sign up numbers as well.