I know this isn't environmental, but it's policy related, and I find the health insurance policy debate interesting, and I think market-based policy can go a long way toward solving environmental issues and health care, and just because it's not environmental hasn't stopped us from posting about it before, and it's my blog and I'll blog if I want to.
Let me start by reminding everyone that despite the current administration's hard push to make me liberal, I really want to be conservative (at least economically conservative). For the most part that means I want to let markets do their thing, because I think markets do a lot better job of allocating scarce resources than any person, or government or agency can.
But that doesn't mean there is no role for government in a conservative economist's world. Just as the role of a mechanic is to fix an engine when it isn't running properly, a role of the government is to fix a market when it isn't running properly. So how should the government fix health insurance?
The problem with health insurance markets is not that there is too much government intervention, the problem is that in the absence of regulation, the market can't function the way it should due to something called adverse selection. Adverse selection occurs when a buyer or a seller knows more about what they are buying or selling than their selling or buying counterpart knows.
In the case of health insurance, a buyer may know the likelihood that they are going to need medical care, but the seller has to guess at that likelihood (that's the job of actuaries). So if a seller offers insurance at a certain price, buyers have to decide whether that price is worth it, and that depends on the likelihood of getting sick (which the buyer is likely to have a better estimate of than the seller), and the buyers tolerance for taking risk.
But the risk piece is less of an issue because most buyers know that the government will act as an insurer of last resort. If a sick person shows up at the emergency room, they will not be denied service (damned Hippocratic Oath). So even without insurance, health care is available for those who choose to take the risk.
Who then will choose to not buy health insurance? Those who have a low expectation of getting sick, and those who can't afford coverage. Let's take a look at a simple example to see why this is a problem.
Suppose there are three people in our health insurance market (I told you it was simple). We are going to call these three people H,M and L for high, medium, and low risk of getting sick.
For each of these three people there are three possible health outcomes in a given year, and we can represent these outcomes by the cost of treatment. The three possible outcomes are $0, $1,000, or $10,000. The table below lays out the probabilities of each person getting sick in a particular year:
|$ -||$ 1,000.00||$10,000.00|
The high risk person has a 90% chance of incurring at least some health expenses this year, a 60% chance of incurring the worst loss ($10,000) and only a 10% chance of no loss. If we multiply the probability of each loss by the expense if that loss occurs, we find that a seller of insurance would expect person H to cost us $6,300 in an average year. As the seller, in order to maintain a viable insurance plan, I will have to charge at least $525 per month ($6,300/12) to H to make sure I bring in enough money to pay the expected medical expenses year after year.*
Now suppose we add the medium risk person, M, to our market. The medium risk person still has a 90% chance of getting sick this year, but the expected cost is only $3,600 (0.6*1,000+0.3*10,000). As the seller, if we have both H and M buying insurance, our expected expenses in any particular year are $4,950 per person (the average of $6,300 for H and $3,600 for M). So now I can charge premiums of $412.50 per month to both H and M and still break even**.
But at $412.50, person M is paying $4,950 per year in premiums, but M's expected loss in any particular year is only $3,600. So M has to decide whether to roll the dice on taking a big loss ($10,000) without insurance with 30% probability, or buy insurance and take an expected loss each year but have coverage each year in the event of the worst outcome. This will depend on M's tolerance for risk.
But what happens if we are able to get the low risk person to buy insurance too? The low risk person has a 90% chance of no loss at all, and an expected loss of 'only' $550 each year. If the seller can get the low risk person to buy insurance along with H and M, the expected expenses to the seller now drops to $3,483 per year, or $290 per month.
The insurer can now charge premiums of $290 per month to L, M and H and expect to break even in a given year. This also means that insurance is much more attractive to person M, because now the annual premium of $3,483 is below the expected loss of $3,600.
But why in the world would person L ever buy insurance? L has to pay $3,480 in premiums per year but only expects to have $550 in health expenses. It makes no sense for person L to buy insurance unless L is oddly and incredibly risk averse. But if person L drops out of the insurance market, the seller goes back to a situation where they have to charge premiums of $412.50 per month to expect to break even, Without L in the pool, premiums rise.
To make matters worse, there seems to be an inverse correlation between income and health outcomes. That means H is likely to be lower income than L. So as L drops out of the insurance market, the premiums rise for H and M and because of this negative correlation between income and health, H and M are less likely to be able to afford insurance. This is a primary reason why health insurance markets don't really work like we would expect.
So how can we make sure affordable insurance is available for everyone?
Require everyone to have insurance (whether you want it or not). This makes sure that all of the L people are included in the risk pool and it should lower premiums overall.
This is the much talked about Affordable Care Act insurance mandate.
The problem with the ACA mandate is that the penalty for not participating was set way too low. Suppose in our example above, we put an insurance mandate in place, requiring H, M and L to either buy insurance for $290 per month or pay a $1,000 fine each year. H and M will still buy insurance because $290 is below their expected losses from health expenses, but L will likely opt to pay the fine because $1,000 plus the expected $550 health costs are still lower than the annual premium. So L still drops out of the insurance market, and the sellers have to raise their premiums to cover H and M, otherwise the the seller will stop offering insurance because they can't make a profit. But the higher premiums might make M drop out of the market too which raises the premiums for H.
How then do we fix the ACA?
Scrapping the mandate is not the efficient answer. Scrapping the mandate will cause premiums to go up because low risk people (young and/or rich) will choose to not insure, roll the dice, and rely on savings, the government and bankruptcy to cover expenses if they crap out.
To me the fixes to the ACA are straightforward:
- Maintain (and even strengthen) the mandate: Everyone is required to have health insurance.
- Crank up the penalty for choosing to not have insurance to ridiculous amounts $10k, $20k...whatever, to make sure that it's not worth it to opt out of the insurance market.
- Make sure that insurance markets are truly competitive.
- Once the markets are working efficiently, ensure that those who still can't afford insurance receive lump sum tax rebates to buy insurance, or expand Medicare/Medicaid coverage to those who can't afford to participate in the market.
It's piece 3 where I think conservatives can really rally. Right now there are heavy restrictions on who can sell insurance to whom. Most insurance markets are at the state level. Loosen these regulations. Make sure that insurers have to compete for business nationwide. Competition lowers prices.
More competition combined with everyone being in the pool will lower insurance premiums.
Alternatively we could talk about single payer.
But that wouldn't be very conservative of me.
*The way it really works is that we have a lot of H people, and we charge them all the expected expense. Some will get sick and some won't so on average we break even. This is called risk pooling.
**Obviously I would charge a little more than that to cover administrative expenses and earn a profit, but you get the point.