COBRA premiums for self-insured

[Guest cosmo01]

A TPA firm is being told by a client that the applicable premium for COBRA purpposes must be determined on an actuarial basis for each possible COBRA election scenario. For instance, suppose an employee has family coverage, but has a qualifying event and elects COBRA for only the child for whatever reason. The TPA firm is being told that it must determine what the cost is of providing coverage to that child on an actuarial basis. The TPA firm currently just provides that the cost will be that of individual coverage. However, based on this information, they will have to determine the actuarial cost of providing coverage to a child. Back to the above example, if individual coverage under the plan is only $50, but actuarially, the cost to the plan of providing coverage to a child is $30, COBRA is based on $30... any thoughts? I know that 4980B(f)(4) provides that for self-insured plans, the COBRA premium is based on actuarial assumptions, but I have never seen this before.

[oriecat]

I am certainly no expert on this, but it sounds very weird to me. COBRA premiums have to be determined for each 12 month period of time, in advance, so I don't see how you could determine the premium each time a circumstance arises, unless I am misunderstanding.

[papogi]

In theory, you definitely could come up with COBRA premiums like this. Arriving at COBRA premiums is ultimately a guess no matter how you break down the coverage levels. You could look at all claims over the past year or two or three for individuals who are not employees or spouses, and use those claims numbers to come up with COBRA rates. That would certainly be a complicated undertaking, but it is possible, depending on the reporting capabilities of the TPA’s claims system. If I were the TPA (I do work for one), I would satisfy the client by arriving at COBRA rates the usual way (a single rate, employee+spouse, employee or spouse +child, employee or spouse +children, family, or what have you). Then I would look at the ratio of dependent claims on a PMPM basis to employee claims on a PMPM basis over the same time period. I would multiply that factor times the single COBRA rate to arrive at a single child COBRA rate. Something along those lines should give something that the client wants.