Thursday, October 11, 2012

OR Study - Take It With a Grain of Salt

It's that time of year again, or I should say, every other year again.

The 2012 Oregon Department of Consumer and Business Services (DCBS) “Premium Rate Ranking Summary” was released on Wednesday and of course the survey declared winners and losers in the race to be cheapest, and reactions are predictable with winners boasting and losers bemoaning.

Oregon has been publishing its study comparing average premium rates for all 50 states and the District of Columbia since 1994.

Many folks take the survey as gospel, in particular politicians, who like anything that can convey a complex message in simple terms regardless of accuracy or context ("See, in my line of work you got to keep repeating things over and over and over again for the truth to sink in, to kind of catapult the propaganda."—G. W. Bush, May 24, 2005).

One criticism of the survey is that states' economies and mix of hazards are very different, so drawing comparisons among states would not be an accurate representation. But Oregon officials have said in the past that their survey controls for the differences in industry mix among the states by comparing premiums for identical classification codes.

What the survey can not control for, however, is the relative cost of living among states, and the huge impact payroll (i.e. the economy) has on rates, which translates to premiums.

In the summary DCBS advises that, "Rates vary by classification and insurer in each state, and actual cost to an employer can be adjusted by the employer’s experience rating, premium discount, retrospective rating, and dividends. Nevada’s index rate dropped significantly since the 2010 study, due in part to inclusion of a payroll cap adjustment in 2012"

The last sentence is very informative of how the data can be manipulated by an economic force that is not a direct class code to class code comparison - an artificial limitation on payroll adjustment put Nevada's rate much lower than 2 years ago, even though Nevada's economy remains stubbornly stuck with recessionary artifact.

In addition, the rankings are weighted against class codes that officials in Oregon have determined to be most important to that state: "Of approximately 450 active classes in Oregon, 50 were selected based on relative importance as measured by share of losses in Oregon."

What if the ranking were determined by ranking classes based on "relative importance" as the share of losses in California, or Oklahoma, or Nevada...?

I'm not saying that the Oregon survey is meaningless. Far from that. In fact the Oregon study is the only study that I know of that at least attempts to rank states on their relative workers' compensation costs. This can be good and this can be bad depending upon what argument you want to make.

Each state has a different set of workers' compensation laws that presumably are reflective of each state's different personalities, different cultures, different industries and economies (and politics). And high cost states are high cost in many other respects - e.g. Alaska just plain costs more to live in than Oregon. Same with other top ranking states: California, New York, etc.

One thing that has me most curious is that the 2012 median premium cost per $100 of payroll was $1.88, which is a drop of 8% from the $2.04 median in the 2010 study. While the high cost states get more costly, apparently that inflation is offset by the low cost states getting cheaper - at least that's my interpretation.

Even more interesting though is that the spread among the highest and lowest cost states has thinned considerably.

Mike Manley, a research coordinator for the Oregon Department of Consumer and Business Services (DCBS) and co-author of the 2012 study, said in 2004 average premiums in the most expensive state were six times what was being charged in the cheapest state.

“One thing that this tighter distribution does is make state rank values more volatile from one study to the next, particularly for the states near the middle,” he said in an email to WorkCompCentral. “There are 20 states in the latest study that are within plus or minus 10% of the median. Fairly minor differences can bounce rankings around by several places.”

What would be a very interesting study would be to match the Oregon survey numbers to carrier profitability - my guess is that we would see some very different results from state to state with not much correlation between costs and profits.

The Oregon survey is good - it just has to be taken with the proverbial grain of salt.

No comments:

Post a Comment