• Breakdown of insurer administrative cost

    I’m reading Mark Pauly’s Health Reform without Side Effects (ungated). Here are a few useful tidbits I’ve encountered so far about insurer profits and other administrative costs.

    The time trend in the loading percentage [the “Total” column above] has been modestly upward across the board; the average loading in private insurance has probably increased by 1 to 1.5% of premiums over the last 20 years. This is largely due to the spread of managed care, which is somewhat more costly to administer (per dollar of benefits or premium) than was old-style insurance, which just paid claims but did not select panels of providers, preapprove coverage, or attempt to manage care.

    It’s worth keeping in mind that managed care may reduce spending by negotiating lower prices and imposing some checks and controls on utilization, but that does not come without costs.

    There is, however, more competition among insurers than simple market-share measures would suggest. For one thing, most large groups self-insure, purchasing administrative services only (or some reinsurance) from the private insurers. These large buyers would not tolerate prices (for coverage or administrative services) that yield high insurer profits because they can turn both to administrative-services-only firms (such as benefits consultants) and plans in other states to provide those services. Smaller employers and individuals are more vulnerable. Even here, however, there is potential entry into the pure insurance business should any existing insurer try to exercise serious market power.

    It’s an interesting rebuttal to the claim that high insurance market power is to blame for high and increasing health care costs. Given the results of the table above, which suggests relatively modest profits (circa year 2000), there likely are sources of competition generally overlooked.

    Still, one should not be terribly happy about the high loadings for small group and individual market products. If, by rearrangement, we can avoid those costs, we ought to.

    • Hey Austin. I wish all of my responses didn’t seem so curmudgeonly. I think the following additions to the chart above would give a more comprehensive picture of the breakdown of costs in private insurance market as a whole:

      1) The inclusion of the self-insured firms (i.e. those with an Administrative Services Only contract). These firms are usually (and almost by definition) extremely large; they act as their own insurer. They pay nominal fees to private insurers for network access, and their administrative costs would likely mirror the larger groups.

      2) a weighted average (or just Totals) of the above chart, which would likely show a greater pull towards the very large groups.

      Also I note that a) there are some big missing pieces in the above chart (groups of size 100 – 500?) and b) that the admin costs are represented as a % claims. Doesn’t it seem more natural (though would weaken your case) to express all of these costs as a % of the premium dollar from which they came? In many years of reviewing similar data, I don’t know that I’ve ever seen admin costs represented as a % of claims. Is there a real reason to do so? Typically one wants to represent all components of premium as a % of premium so that they would add back up to 100% and you have a full picture of your costs.

      • @love actuary – Isn’t it clear I lifted the chart from the Pauly document I cite? Maybe not. So, be as curmudgeonly as you like. Just know your curmudgeon is not reaching the person you intend it for.

        Also, I’ve posted on loadings and MLRs before and have included figures, consistent with these, that address many of your concerns. You can’t have it all tied up in a neat bow the way you like in one post!

        Oh, sorry, I got a little curmudgeonly there myself. 🙂

    • No worries of course; it was clear that the chart was lifted from elsewhere, so I can reserve my bone-to-pick with Mr. Pauly.

      I would make another note, too, that may be more optimistic: as health care claim costs trend on the order of, say, 3 x’s inflation, and remaining admin costs trend at more like 1 x’s inflation, then as a % of premium over time, admin costs should (under this simple model) decrease.

      I don’t know whether or not we’ve seen that over the long run … thoughts?

    • Pauly’s numbers agree fairly well with this paper claiming that small businesses pay about 18% more for equivalent insurance. (Paper covers tax credits in the ACA and how they will help small businesses.)


      @love actuary- If true, and this has been happening for a long time, then we would expect private insurance costs to show a modest long term decrease in the rate of increase. I certainly dont see any such pattern, or if it is there, it is too small to see. Last time I looked at MLRs I did not see a significant change.