COBRA Costs

[Guest ptpnthr]

Does anyone have any data on how much COBRA costs non-COBRA participants? By how much would an active participant's premiums (per participant) increase as a result of people going on COBRA? Assume about 100 employees, nobody is now on COBRA, we hire 10 employees and at the same time 10 employees are terminated, elect COBRA and stay on COBRA for one year. Assume further that premiums for the participant now cost $100 per month total. After one year, on average, how much would a participant's premiums increase as a result of those 10 participants being on COBRA?

I thought there would be little harm in asking, even though I may not get a good answer.

[mroberts]

Probably very little. First of all, if 10 employees will be terminated, only 2.3 of them are going to be electing COBRA (based on a recent survey). The main reason 77% of ex-employees do not elect COBRA is due to cost. And a secondary reason is that healthy COBRA eligibles will take a chance that nothing will come up while they are looking for another job.

Secondly, even if COBRA participants utilize benefits 50% more than regular employees, in your scenario, that would mean about 3 people out of 103 would be utilizing benefits at a higher level. Therefore, by using a simple example of averaging 100 100's and 3 150's, what do you come up with? 101. Therefore, ex-employees on COBRA do not really impact claims that adversely.

Obviously this is a simple example since there's no saying that one of your COBRA electors isn't going to get cancer. But all in all, it's a true tale.

[Guest ptpnthr]

Thank you so much for you response. So, using your example, if all 10 do elect COBRA, we would have an average of about 104.5?

[mroberts]

With a group your size, I wouldn't hold fast to the example I gave. My numbers were generalized. On average, COBRA participants utilize benefits 50% more than regular employees. If all 10 COBRA eligibles elect the coverage, there could be no claims experience. Additionally, while your active employees will have coverage for 12 months, COBRA participants may initially elect coverage and only have it for a month or two since they would most likely pick it up at their new job. The point I was trying to make was that all in all, it impacts your claims experience very little. Now, if you had 100 employees and had to lay off 50, that would be a different story. COBRA participants could damage your claims experience.

[Kirk Maldonado]

Here's the text of an article that I found on this topic. I am not vouching for the truthfulness of the statements contained in it; I don't know one way or the other. Unfortunately, I don't even recall where I got the article. With those caveats, here it is:

THINK COBRA DOESN’T COST EMPLOYERS ANYTHING? THINK AGAIN.

The COBRA subsidy provision contained in the Senate’s Trade Adjustment Assistance bill (S.1209) places

significant financial burdens on precisely those employers that are already suffering from significant economic

distress. Indeed, COBRA continuation coverage is an expensive proposition for most employers. In

addition to the costs associated with enrolling individuals in and administering COBRA continuation coverage,

employers face significant additional claims costs from COBRA enrollees due to adverse selection. This is the

idea that “low risk” or healthier individuals who utilize fewer health care services do not believe they need to

spend money on health insurance and are therefore far less likely to elect COBRA continuation coverage.

Each of the case studies below serves to debunk the myth that COBRA continuation coverage is a “nocost”

item for employers. The case studies are based on actual data recently collected from members of The

ERISA Industry Committee (ERIC) and represent a broad cross-section of our membership.

Case Study #1: A large employer with 200,000 employees had an average of 1,135 individuals enrolled in

COBRA continuation coverage each month during the period between May 2001 and December 2001. Each of

these individuals was charged a premium of $166.25 per month for COBRA coverage during that time, but

incurred, on average, $412.25 in claims and administrative costs each month. The company was required to pay

out just over $2.2 million during the eight month period noted above to cover the additional administrative and

claims costs attributable to these COBRA enrollees. Additionally, the company paid $2.7 million in claims and

administrative costs during the same period for individuals who elected COBRA continuation coverage for their

dependents as well. Thus, this company paid almost $5 million in additional claims and administrative costs on

behalf of its COBRA enrollees between May 2001 and December 2001.

Case Study #2: A large employer with over 50,000 employees paid nearly $4.5 million in medical, dental, and

prescription drug claims for COBRA participants in 2001. The company had an average of 1,675 COBRA

enrollees each month and collected just over $2.8 million in premiums from these individuals. Thus, this

company was forced to pay an additional $1.6 million during 2001 to cover the claims and administrative costs

attributable to COBRA enrollees.

Case Study #3: The premiums charged for COBRA continuation coverage by a large employer with over

300,000 employees only cover about one-third of the claims and administrative costs incurred by COBRA

enrollees. That means that for every $100 in premiums paid by a COBRA enrollee, the company in this

example must pay $200 in claims and administrative costs. During 2001, this company paid a whopping $5.7

million in claims and administrative costs and collected just $2.8 million in premiums – leaving the company

with a bill of just under $3 million for the COBRA continuation coverage it provided.

Case Study #4: A large employer with 38,000 employees and operations throughout the United States charges a

premium for family COBRA continuation coverage that covers just a fraction of total cost. More specifically,

for every $1.00 paid by an employee in COBRA continuation coverage premiums, the emp

[mroberts]

These are kind of stupid examples to use since the poster indicated that his or her company had 100 employees. Obviously anytime you give examples of companies with thousands of employees on COBRA, the dollars are going to look high. In order to really say these numbers mean anything, we need to compare them to aggregate claims for the group. Based on my experience dealing with claims, per 100 employees, on average there's going to be about $300,000 in claims.

Example #1 - 7,500,000/(2000)(300,000) = 1.25% (Annualized)

Example #2 - 1,600,000/(500)(300,000) = 1%

Example #3 - 5,700,000/(3000)(300,000) = 0.6%

As you can see, the 1% that I gave in my previous example seems to be in line. For a company of 100 lives, of course, $1.6 million is going to seem like a lot of money. For a company of 50,000 employees, I won't go as far as to say that $1.6 million is pocket change, but take a look at their annual report and you tell me what $1.6 million means to them.

[Guest ptpnthr]

I know this getting off topic, but if your experience is that, on average, 100 employees have $300,000 in claims (I assume annualized?), then why are premiums more than double the claims? $300,000/100=$3,000 per person in claims per year. My premiums are more than double that. Does group health insurance have a more than 100% markup, not to mention individual insurance?

[mroberts]

The $300K is an average number. This doesn't mean your group isn't running at $500K per year in claims. Think how much one cancer claim or one premature baby costs. Additionally, for smaller groups, carriers aren't going to look for the same loss ratio when determining premium rates. The smaller the group, the lower the loss ratio a carrier looks for simply because one shock claim could take 5 years of additional premium to recover from. The larger the group, the more credible the experience.

[Kirk Maldonado]

MRoberts:

You are right in that the examples in the article I posted are of extremely little relevance to the facts involved with the original poster.

My only excuse for posting it is that I thought that the article might be of some (limited) interest to our general readers.