Results are in for the survey on freelancers and health insurance! Back on October 30, I asked people to provide numbers on how their health insurance premiums have evolved over the years—and how their costs will be affected by the Affordable Care Act (ACA). It was an effort-intensive survey, asking people to go back through their bills from the last 10-12 years, but we still managed to garner 23 responses (including mine).
The numbers confirm what most of us probably already knew: the insurance market has been a nasty ride for most freelancers over the last few years.
I’ll say more about those key results in a minute, but first some basics. Respondents ranged in age from 29 to 63 years (majority, late 30s to late 50s) and were roughly 80% female. The majority (57%) relied on individually-purchased health insurance, 19% relied on a spouse or partner’s insurance (generally provided by an employer), and 24% obtained insurance from other sources, including group coverage from the Freelancers’ Union or chamber of commerce.
I didn’t subject this data to any sophisticated statistical analyses; I’m no Nate Silver, and the number of respondents is too small to justify much analysis. But I’ve at least made some common-sense attempts to address major factors that would otherwise distort the results. I’ll summarize those tweaks at the bottom of this post, but first, the results. They’re summarized in the graph below:
From 2000 to 2013, the average monthly health insurance premium climbed from $273 to $643—a nearly 2.4-fold increase. This isn’t as bad as the 5.9-fold increase in monthly premiums that I experienced with my own individual insurance between 2002 and 2011 (starting $99 and ending at $587)—but it does reveal the same general pattern of unsustainable cost increases. If you’ve been buying your own health insurance, then chances are it’s been a decade from hell.
The above results are based on the 13 people, with health insurance of any kind (including spouse/employer-provided), who had usable time-series data. If you include only the 10 people who purchased health insurance on the individual market, the results are virtually identical. Average premiums increased from $260 in 2000 to $616 in 2013—also a 2.4-fold increase. Median premiums increased from $260 to $492.
I also wanted to look at how estimated 2014 premiums under the ACA health insurance exchanges would compare with what freelancers had previously been paying (I only included the 12 respondents who reported insurance premiums for 2011 or later). Based on estimated premiums for similar health plans on the exchanges, the average premium will drop from $694 in 2013 to approximately $586 in 2014, and the median premium will drop from $604 to $516 under the ACA—in either case, a decrease of about 15%. These numbers may under-estimate the cost decreases under ACA plans, since they generally didn’t consider the impact of subsidies.
Not everyone or every region will be affected the same way by the ACA—in fact, it seems likely to produce both big winners and big losers. The winners: survey respondents whose predicted monthly premiums fell from $550 to $330, from $677 to $250, from $587 to $380, and from $1,083 to $750. The losers: survey respondents with premium increases from $477 to $560, from $370 to $600, and from $670 to between $800 and $1,200.
Much more data would be needed to understand what factors contribute to increased versus decreased premiums. An October 23, 2013 New York Times report suggests that health insurance premiums are failing to fall in rural areas—a result, in part, of minimal price competition in areas served by only one or two insurers .
The roughly 15% average decrease in monthly premiums seen so far under the ACA may be appreciated by many—but in some sense represent little relief from the 140% average price increase seen since 2000. What will ultimately determine the success or failure of ACA for freelancers is how their rates continue to evolve in the coming years—whether annual premium increases of 15-40% are still seen in the years ahead, as they were in recent years, or whether rates actually plateau or even decline.
A quick note on methods for those interested. With only 22 respondents (not all of whom provided numbers), the potential for one individual’s aberrant results to produce large spikes and dips for the entire group average was significant. I censored some data in order to prevent this. In one case, a family’s monthly premium more than quadrupled over a 3-year period, due in large part to the birth of two children during that period. I removed this data in order to avoid the large spike that it introduced into the group averages. I also excluded individuals with premiums listed for only 1 year, since the purpose of this analysis was to reflect trends that occur within each person’s time series.
Finally, an individual whose monthly premium had grown from $350 to $1,375 between 2002 and 2009 chose in 2010 to switch to a plan that covered only catastrophic care—resulting in a substantial artificial fall in prices from 2009 to 2010. However, the price of the catastrophic plan also increased rapidly, from $400 in 2010 to $725 in 2013. Based on this continued increase in the price of the catastrophic health insurance, I chose to scale these numbers up and normalize them to the previous baseline of $1,375. I did so conservatively, assuming that no price increase happened between 2009 and 2010.
One cannot draw detailed conclusions using data from only a couple dozen people; however, even these limited numbers provide a surprisingly compelling view of a health insurance market that has been out of control for freelancers.