• Premiums should not be going up as fast

    Premiums always seem to go up, a lot. Yet, I’m not convinced they should be doing so right now, not unless the current recovery from a recession is different from past recoveries. I’m not the only one questioning big premium hikes. Robert Pear reminds us that the Affordable Care Act includes provisions that will lead to greater scrutiny of health insurance premium increases.

    In some states like New Hampshire, groups of more than 20 workers have experienced premium increases of around 20 percent this year, while smaller groups have seen increases of 40 percent or more. At the same time, insurance agents say, coverage is shrinking as deductibles have increased and insurers limit the choice of hospitals.

    To ensure that “consumers get value for their dollars,” the new health care law required annual reviews of “unreasonable increases in premiums.”

    Under the new rule, federal and state officials will review rates in the individual and small-group insurance markets. In effect, the administration said, large employers can take care of themselves, as they are more sophisticated purchasers and have more leverage in negotiating with insurers.

    Perhaps large employers don’t need government assistance to obtain better deals. Maybe their market power cancels that of large insurers, returning the market to something like a competitive one. Jared Bernstein explained that insurers are claiming that premiums must rise to protect against future higher claims due to demand for care that was unmet during the recession. However, this shouldn’t happen in a competitive market, he writes.

    In competitive markets, sellers can’t typically set today’s prices based on where they expect demand to be in the future.  If one of them did so, others would capture their market share by pricing based on current supply and demand conditions.

    Actually, it probably shouldn’t be happening even if the market is not that competitive, at least not according to recent history. As the following chart from PriceWaterhouseCoopers shows, in the years following unemployment peaks, the rate of increase in employers’ costs of health benefits fell and fell dramatically.

    If employers’ health care costs rise more slowly after peak unemployment, so should premiums. Are we not in a period following peak unemployment? Should group-market premiums be increasing rapidly now? Not according to recent history.

    Why might health care costs rise more slowly after peak unemployment? I don’t really know, but I can speculate. Here are some possibilities that come to mind:

    • Relatively sicker individuals disproportionately leave or are forced from the workforce during periods of high unemployment. If you’re sick and possibly can’t work as hard, maybe you’re more likely to leave or be laid off, and you may never come back even when the economy begins to improve.
    • Similarly, firms preferentially hire healthier (or younger) individuals as the economy improves. Younger workers are cheaper to hire, so this is also an incentive.
    • Concerned about retaining their jobs, people take less time to visit the doctor or are less likely to take sick or disability leave. If your colleagues had just been laid off, wouldn’t you work extra hard to keep your job, possibly forgoing care?
    • Periods of rising unemployment are bad for health. (Actually, I’ve seen just the opposite reported, so I’m not sure I buy this one.)
    • Worried about losing one’s job, and health benefits, in a time of rising unemployment, people use more care while they have insurance. So, it isn’t that health costs fail to rise as fast after peak unemployment so much as the rate of increase is boosted as unemployment grows. The peaks are high, the troughs aren’t low.

    What am I missing?

    UPDATE: Colleague and occasional contributor here, Steve Pizer, suggested to me that employers may preferentially hire low-wage workers after a recession. Those workers may be less likely to be offered or to take up insurance. That makes employer health costs lower if averaged over all employees, and it wouldn’t necessarily translate into lower premiums for those who do take up insurance. See also this prior post on low wage workers and health insurance.

    UPDATE 2: Read the comments for other good suggestions from readers. More on the underwriting cycle (with graphs!) herehere (to which there was a follow-up here).

    • Maybe it’s the $64 billion in excess cash that the five biggest for-profit insurance companies moved off their balance sheets from 2003 to 2010 by repurchasing their own stock. That drives up share prices for big holders of the shares, such as the CEOs. That helps drive up premiums too. Or maybe it’s the nonprofit insurance companies building up excess risk-based capital reserves — up to 10 times greater than regulators require.

    • When unemployment is falling the economy is growing and insurers have huge amounts of ‘cash’ on hand, called reserves. When their investments do well, they will tolerate a higher MLR, which, if they choose (and if there is competition) will be manifested as lower premiums–or a lower rate of increase. In other woeds, the more profits they make with investments the less they need to take from premiums.

    • Your last possibility rings true for what we have seen with our employees. I always talk with local business owners when they come in and they say it is true for them also. Steve’s idea may be supported by Mulligan’s piece noting that job losses were highly concentrated in the lower income brackets.



    • Your reference to the underwriting cycle was a bit cryptic. My read of the graph you provided is that this time is very different in that the medical benefits cost trend was at relative highs during or soon before the unemployment peak in the past, but was at historically low levels when this recession started. In short, we started out this unemployment peak lower than average over the last thirty years, so it isn’t much of a mystery why we would stay go up this time.

      But the benefits cycle is not the same thing as the underwriting cycle. The underwriting cycle is not the health benefits cost trend itself, but the overpricing or underpricing around this trend by insurers. It’s an epicycle. That epicycle has been diminishing in amplitude ever since the 70s, as your 2nd update link to an earlier graph shows.

      One other contributing factor to accelerating costs now is ACA. I believe it added 1-2% to premiums in the last year due to added benefits. This is a one time change, not a permanent increase in trend.

    • Great post Austin. I do have a question, more that I’m asking for an opinion. Given the big insurance companies have reported on average 30% greater earnings than analysts predicted in the last few years, we know that there are reserves of cash, and costs have dropped due to less going to the doctor after a recession. Is there any merit to some views that the insurance companies are raising premiums to offset what they know will be difficult times to do so in the future? Points on that being the government mandate in September that insurance companies must justify increases over 10% (even if the gov can’t block), and the potential fact that the new health care system should demand a drop in costs. Are the insurance companies getting while the getting is good? Thoughts?