My co-authors, Paul Shafer, Laura Baum, Erika Franklin-Fowler, Sarah Gollust, and I have a new article in the American Journal of Managed Care. We look into how private insurers changed their broadcast TV advertising volume in counties served by Healthcare.gov when there was a change in the level of competition from other insurers between the 2015-2019 Open Enrollment Periods (Fall 2014-Fall 2018 advertising volumes). Our previous research had found that there was no association between enrollment and private advertising volume but an association between state advertising and enrollment. We were curious to see how insurers responded when they were the only insurer in the region — would they increase advertising to pick up market share (we assume more ads leads to more enrollment) or would they decrease advertising to save money?
Going from multiple insurers to a single insurer resulted in 465 fewer private ads aired within a county during open enrollment (P < .01), a 17% to 38% reduction. Losing monopoly status is associated with a drop in advertising of 452 airings per insurer (P < .01), and becoming a monopolist is associated with 293 more airings per insurer (P < .01).
Total ad volume on a population weighed basis decreases when insurers leave a county but the ads per insurer increase. So we see a bit of a positive dose response when competition goes down. We did not see changes in advertising volume when a county went from one to not-one insurers.
Why does this matter?
Enrolling in health insurance on the ACA marketplace requires many steps for a new enrollee. One of the critical early steps is awareness that the markets exist and that they sell health insurance. Many people are tripped up by this step. Millions of Americans don’t know that the ACA markets exist or if they do exist that they offer highly subsidized coverage. Federal and state agencies can fill in some of these informational gaps with paid broadcast advertising of a general nature. However, advertising is mostly a private affair. We have also learned that enrollment tends to drop in a county when an insurer leaves the county. Advertising is a very plausible source of information for people to become aware of their options. If total advertising volume decreases in a county that is losing insurers, then this is a plausible mechanism to partially cause some of the enrollment loss that occurs when an insurer exits the marketplace.
We also think that this is useful in the ongoing debate about the approval or disapproval of the Georgia Access Model (GAM) 1332 Waiver. Georgia wants to get completely off of Healthcare.gov and not run an exchange. Instead, the GAM will count on brokers and insurers to do their own outreach and enrollment. We think that our research is evidence that insurers are not advertising and recruiting to maximize enrollment. Instead, they may be advertising for other objectives. And that makes perfect sense for an insurer — they have a different set of objectives and duties to their internal stakeholders than an exchange board has for the citizens and residents of the state or country. However, the 1332 waiver process requires that at least as many people are covered, so less advertising in minimally competitive markets is likely to occur while increases in competition may not have a similar but opposite effect on the volume of ads that potential enrollees see.
Brad F
If a region goes through two cycles, say competitive to monopoly, then competitive to monopoly again, what your research implies is the end result will be overall less advertising at time Y over time X (a chronologic decay in ads).
However, if this is so, how does the overall quantity of ads build at time X? Some inertia or phenomena causes ad volume to go up at the outset. Whatever that secret sauce is getting the process going and I am wondering what that sauce might be and whether it is random (or not)?
Brad
Mary G
How do you know how much a given company advertised? Is there a database somewhere? Is it just broadcast TV?
There are a lot of other places that carry ads like the evil borg that is Facebook Also in games apps. I play some mindless Candy Crush types when my hands are OK. They are “free to play” but have in-app purchases to get extra moves on a level or items to make winning easier.
If you don’t want to cough up cash, you can watch ads instead. They run between 5-30 seconds. I am also listening to an audio book at the same time so I mute the ads. But sometimes I don’t get around to it or want to see what that whacky thing is. Aetna just started one in heavy rotation with a very fit woman talking about she started with Aetna all the way back to the dark ages of 2014 and will never change. Covered California jumps in a couple of weeks before open enrollment starts, though of course they are not an insurance company themselves.
David Anderson
@Mary G: We use the Kantar Media Group broadcast TV advertising database that the Wesleyan Media Project (Franklin-Fowler, Baum and Gollust) organize/manage. We know it is limited as we don’t get cable ads nor internet ads, but it is the best currently available data out there.