Thursday, October 17, 2013

Health Insurance Premia, Competition, and Endogeneity

An interesting article ran in our local paper today; the original was from Vermont Digger which provides news for Vermont.

The article looks at health insurance premia in Vermont versus the rest of the country; Vermont turns out to have the 5th highest rates out of 48 states.  These rates are for 2014 plans, offered on the exchanges, and are pre-subsidy.

The article mentions the lack of competition as one reason for high rates in Vermont, and it correctly notes that competition is relevant at two levels:  that at the supplier (hospital) level and that at the insurer level.

Vermont turns out to have little competition at both levels.  There are only two insurers offering policies on the Vermont exchange, and there is only one large hospital system in the state, Fletcher Allen.  Dartmouth Hitchcock, based in NH, would be the second largest supplier, with many Green Mountain folks driving across the Connecticut River for their care.

The only problem with the article's analysis is the problem of endogeneity.  Why does Vermont have only two insurers offering policies?  No doubt it is to a great extent because of Vermont's low population.  States with large populations tend to have more insurers, states with small populations tend to have fewer insurers.  Larger populations allow for economies of scale in insurance operations, and by itself will lead to lower rates.  So is it just the low population of Vermont that drives up rates, or is there an independent effect of little competition?  I suspect it is both.

The same is true on the health care supplier side. Why only one (relatively small) hospital system in Vermont?  There is just not enough market for more than one supplier of even close-to-efficient size. Again, economies of scale are limited by the extent of the market.  This alone will drive up health costs and hence insurance rates.  But it also limits competition, and that has an independent effect on rates.

Is small really so beautiful?  If nothing else, it comes at a price.

Monday, October 14, 2013

Delaying the Individual Mandate is Difficult

There is still talk about delaying the individual mandate, and there are some reasonable arguments in favor of that position -- not the least of which are the ongoing difficulties with the Federal health exchanges.

But, the fact is that health insurers have already posted health insurance policies with prices, available to buy.  The prices (premia) of those policies reflect expectations about who will sign up (premia have to cover expected health costs of the enrollees).  Importantly, who signs up depends on whether or not the individual mandate is in force or not.

The tax for not signing up is $95 or 1% of income, whichever is greatest (with a cap equal to the average cost of  a bronze plan).  This is not insignificant.  Dropping the mandate/tax will definitely induce some individuals to go without insurance, and there will certainly be adverse selection in that choice -- the healthiest individuals will tend to not buy insurance, the least healthy will tend the other way.  This will distort the pool of insured people from what the insurers would have expected when they posted their premia for the 2014 year.

I do recognize per my earlier post that there is risk-sharing on the exchanges.

That said, delaying the individual mandate seems quite unfair and dare I say in violation of principles that libertarian-oriented folks would generally respect.  How can the government induce businesses to offer contracts at binding prices and then significantly change the rules so as to increase the cost of fulfilling the contracts?

I have not seen anyone offer the insurers the chance to re-price their policies if the individual mandate were delayed. 


Thursday, October 10, 2013

Why insurance premia on the new exchanges are not comparable to existing policies

Are prices for health insurance on the new insurance exchanges lower than what was available before? A number of articles have suggested that prices are at least lower than expected -- see for example here and here.

What I have not seen noted anywhere is that the Affordable Care Act creates extensive loss-protection subsidies for insurers on the exchanges.  Because of these loss-protection measures, it is extremely difficult if not impossible to compare prices on the exchanges to prices that existed before.  

Indeed, if prices were not lower, it would be very surprising.   

There are three loss-protection measures for insurers offering policies on the exchanges; two are temporary (two years) and one is permanent.

The one that I think is most significant in regard to pricing is known as Risk Corridors.  Insurers compare their premium income less administrative cost to their actual claims payments.  If that difference is negative (claims exceed premiums less admin costs) the government (aka you and me) bears up to 80% of the loss.  Symmetry prevails, so the government will also tax any "excess" profits.

With problems of adverse selection and with expected stickiness of consumer choices, this policy has to induce lowball pricing for the two years that it will be in force.  Why not price low and lock in consumers?

The second major risk reduction policy for the exchanges is known as Reinsurance.  Insurers that have high-claim individuals will be eligible for reimbursement of losses.  Who pays for this?  Self-funded plans (again, aka you and me) pay a tax to fund this reinsurance scheme.  This policy also obviously encourages low prices for insurance policies while the reinsurance scheme is in effect.

The third risk reduction scheme is permanent and is known as Risk Adjustment.  Insurers compare the risk of their insured populations; insurers with higher risk individuals receive payments from insurers with lower risk individuals.  The technical details of this are very important and I expect some good careers for previous bankers who can figure out how to game the risk-adjustment formulae.  I agree that in principle risk adjustment is good, as it reduces the problem of adverse selection on the exchanges.  Also, this policy does not have any net inflow of government funds, so if the policy reduces prices on average there can be little argument that the policy is efficient.

Where else has anyone read about the risk reduction policies built into ACA and how those policies invalidate the debate over whether prices are lower on the exchanges relative to prior individual health insurance policies?