The Kaiser Family Foundation and the Peterson Center on Healthcare tracks cost sharing and premiums for employer sponsored insurance (ESI). I am using their Table 1 data to look at the distributional consequences of benefit design changes.
The blue line is the actuarial value. It has hovered at about 85% over the decade of data. Insurers pay roughly 85% of the incurred claims. 15% of the claims are paid by individuals through cost sharing. The implied challenge is that medical costs are growing significantly faster than wages so total spending (both premiums and cost sharing) are going up quickly for people.
The orange line is the proportion of the cost sharing that is taken up by the deductible. That has changed dramatically. It went from roughly 30% of cost sharing was borne by deductible payments in 2006 to slightly more than 50% of cost sharing was paid through the deductible in 2016. The other two types of cost-sharing are coinsurance which is a percentage payment based on the contracted amount and co-pays which are a fixed per unit fee.
Deductible market share went up 23% points, coinsurance increased its share by a smidge, ~2% points while co-pays lost 25% of share over the decade.
If we hold actuarial value constant, deductible heavy cost sharing arrangements are good for people with very high expenses and bad for people with low but non-zero expenses. Co-pay designs are good for people who use few services.
In 2018, I have had a $144 urgent care visit and a $6 generic prescription. This is a very common archetype of low but non-zero utilization. A friend of mine has cancer. She has routine $10,000 treatment months in 2018. That is a common profile of a very high spender.
Under a deductible heavy scheme, I am kicking in $150 to the cost sharing pool. My friend is maxing out her deductible in the first month of treatment. If urgent care has a $50 co-pay and generic drugs are no cost-sharing, I kick in $50 instead of the $150. That other $100 has to be made up somewhere and that means it will be made up from people with frequent interactions with the healthcare system which means it will increase cost sharing for very sick individuals.
I wonder if part of the political anger and angst on healthcare is the transition from a cost sharing regime that is heavy on co-pays and light on deductibles to the opposite. The original baseline is pretty good for the bottom 50% to 60% of the healthcare spending distribution. Their out of pocket costs were fairly low the few times that they touched the system in a year. A $20 or $40 charge was what they saw. Now, under a deductible heavy, co-pay light system, they are seeing their cost sharing increase significantly so they don’t see the value of their health insurance when they pay $144 for the urgent care visit. Holding actuarial value constant, it is a change in regimes that is good for very expensive individuals but those are the folks who are not high in number nor particularly price sensitive as their treatment choices are pay or die.