The last few “Ask An Economist” questions sent my way have dealt with the topic of money and banking, so I was happy to get a different sort of question from Heath this week. He asks,
“Should ‘health insurance’ actually be called insurance anymore? My understanding is that insurance is where you pay a small premium to the insurance company so they take on the risk of something catastrophic unexpectedly happening. Currently "health insurance" covers lots of things that are known and predictable (physicals, screenings, etc)” ... “I hear from a lot of people is if preventative care isn't covered under insurance then people won't pay for it which would then lead to higher expenses down the road. I am not sure if I agree with that counter argument since people get ‘preventative care’ all the time for their vehicles since they do not want to have to pay for more expensive repairs in the future.”
Heath brings up an interesting question about the nature of insurance. Before I directly address his question, I think it’s best we analyze what a “pure” version of insurance looks like.
A Fair Bet
Insurance is, primarily, not a pathway to lower risk for society. As economist Armen Alchian and co-author William Allen wisely observe,
“insurance divides, spreads, and diversifies the economic losses of disasters over a large number of people, reducing each person’s risk of a large loss. The total loss by the occasional disaster is shared among the insured people, not eliminated.”
To understand this point, consider the possibility of house fires. Let’s imagine there is a one in one hundred chance that any given house will catch on fire in a year. Obviously, this is way higher than a realistic probability. One percent of all homes don’t burn in a given year. But to make the math easy, let’s assume that’s true.
Second, let’s assume the average house fire causes $50,000 in damage. Finally let’s say “Extinguisher Insurance Company” sells house fire insurance in a town of seven thousand homeowners.
If every homeowner buys insurance, what is the expected number of insured homes that will go up in flames in a given year? Well, if one in one hundred burns, and there are seven thousand insured houses, we would expect 70 houses (7,000/100) to burn each year.
At this point someone may note that just because there is a one in one hundred chance of a house burning down, that doesn’t mean one house will actually burn down for every hundred. This is true, but as the number of cases gets larger, it’s more likely that the odds will actually manifest in reality.
Consider flipping a coin. If you flip a coin twice, how likely are you to get heads once and tails once? It can happen, but it wouldn’t be crazy to get heads both times. What if you flip it 100 times though? It’s extremely unlikely you’ll get heads every time if you flip 100 times. The more times you flip the coin, the more likely it is to average to a 50-50 split between heads and tails.
So now that we can reasonably expect 70 houses to burn a year, we can also calculate the expected damage. If 70 houses burn at a cost of $50,000 per house, the total cost over a year is $3.5 million.
If the company offering insurance wants to just break even on the insurance payouts, they’ll need to divide that $3.5 million among the seven thousand policy-holders for a total yearly premium of $500 each.
When the expected payout of an insurance company is equal to the expected total premium they collect, economists call this a fair bet.
But of course, this ignores the fact that insurance companies need to pay all sorts of costs, such as paying laborers to administer the program. In order to break even, these administrative costs must be divided up among policyholders as well (for simplicity we will ignore other administrative expenses—the logic is the same).
So if the company paid salaries of $700,000 on top of the expected total cost of $3.5 million the total cost of $4.2 million is divided among the seven thousand policy-holders for a total yearly premium of $600 each.
In its purest form then, insurance is a method of spreading over a whole group the total expected cost of some unforeseen catastrophe plus the administrative costs of doing so.
Why Not Insure Everything?
People insure against some bad events but not every bad event. This brings up an interesting question—why insure some things and not others?
To understand why, consider the example of a small island which does not do business with the rest of the world. Let’s say the island is sufficiently small such that a tsunami is enough to destroy every house on the island. Would we expect a tsunami insurance market to pop up?
Absolutely not. If all houses are damaged equally by the Tsunami, spreading out the total cost of the damage among all the property-holders would just mean each property-holder pays for their own damage in full plus any administrative cost.
So insurance only makes sense if policy-holders can have different outcomes.
Consider another example. Imagine in our fire insurance example, all the citizens had psychic powers and would know for sure whether or not their house would catch fire in that year. Would an insurance market exist?
No again. Anyone whose house was going to catch fire would want insurance, but everyone whose house was going to be fire-free for the year would opt out. The result would be only those who were going to use insurance would buy it, and therefore, similar to our island example, insurance companies wouldn’t be able to do anything besides add administrative costs.
The primary insight to glean here is insurance only makes sense when there is uncertainty. This brings us back to Heath’s question. Our modern incarnation of medical insurance provides coverage for events unrelated to uncertainty. There’s nothing particularly uncertain about getting a well-check. So why would insurance cover it?
Insurance has some deleterious impacts on the behavior of the purchaser. If you know your car repair will be covered by someone else in the case of an accident, you’re more likely to drive recklessly.
When buying risk protection causes purchasers to engage in riskier behavior, economists call this a moral hazard. And while moral hazard doesn’t mean people totally neglect their health, it means they will neglect it more than if they had to pay for their own bad health all by themselves.
If you know health insurance will cover major surgeries, you may take worse care of your health. Maybe you’ll decide to skip paying for a yearly well-check. In that case, you allow high-cost preventable surgeries to occur because you know insurance will cover it.
Now this by itself is no big problem for insurance companies. If they know everyone does this, they can just charge a higher premium to make up for the higher costs. In that case, policy-holders as a group will cover the cost of the moral hazard and insurance companies don’t need to prevent it.
But when you combine moral hazard with other issues, the need to avoid moral hazard becomes more important.
When insurance companies have customers cover the cost of moral hazard, the premiums on insurance will increase. Whenever the price of anything goes up, including insurance, the quantity of that good demanded falls, all else held constant. Economists call this the law of demand.
So if insurance premiums go up for everyone, who stops buying insurance? The people for whom insurance is the least necessary are likely to ditch it first. In health insurance, that’s probably healthy people.
If the healthiest people leave, that means the probability of a policy-holder needing medical intervention will go up. When this happens, premiums will rise further, causing more relatively healthy people to stop buying insurance. This positive feedback loop (with negative consequences) where all the healthy customers refuse to buy insurance is known as adverse selection.
How can this be avoided? One option would be to charge lower rates to customers, but insofar as insurance companies can’t do that either because they don’t have information about a person’s health or its illegal to charge different prices for different health factors (like pre-existing conditions), this won’t be an option.
In that case, it may make sense for insurance companies to provide coverage for well checks. If well checks are included in health insurance plans, people will have less incentive to take bad care of their health and more problems will be fixed sooner and more cheaply. This will mean lower premiums which will encourage some healthier customers to jump back into the market.
This isn’t the only fix for the issue caused by adverse selection and moral hazard, but it may be one way of lessening the problem.
So if insurance companies pay for things like well checks because they help diminish problems like adverse selection and allow the insurance market to be viable, I’d say this coverage is insurance. It’s true that it isn’t directly covering unpredictable problems, but it is enabling insurance companies to operate profitably to cover those problems.
In that sense it may be a necessary add-on.
When Add-ons Aren’t Real Insurance
So, theoretically, I think coverage of well-checks could be a legitimate function for an insurance company to take on, but, in reality, that’s not the full story in the United States.
With the passage of the Affordable Care Act (colloquially known as Obamacare), the government began requiring that insurance companies cover preventative cares like well-checks and vaccines.
Why do this? Well, when insurance companies face the moral hazard problem, they at least have the option of making customers pay more through one form or another (e.g. higher premiums and deductibles).
But at least part of Obamacare’s purpose is to insulate poorer individuals from high costs. There are basically two ways the government could theoretically do this.
If the government generates cost savings by encouraging more preventative care, it can pass the savings on to lower-income citizens.
The difficulty, though, is it’s unclear specifically how much preventative care will produce the lowest costs. Insofar as preventative care is curbing the unhealthy moral hazard created by health insurance, it will lower the total cost. But exactly how much preventive care provides the lowest total cost to society?
There’s no easy way to know. Companies handle the problem of minimizing cost when they try to maximize profit. A company with exorbitant costs goes bankrupt. But government does not earn profits and therefore does not have access to cost measurement in this way.
It could be the case that governments spend too much on preventative care. How could they do that? Consider cancer. There are various types of cancer, and there are several very reliable tests to catch them early. So why don’t we test every person for every cancer every year?
It would cost too much. Imagine the resources used in billions of cancer tests every year. It’s true we would catch some cancers earlier and those cancers would be less costly to treat. But cancer is rare enough that this cost savings may not make up for the cost of testing.
Insofar as government stimulates the market to take on too much preventative care, this is no longer a necessary add-on. It’s a transfer. When government subsidizes inefficient preventative care, it’s essentially taking resources (in the form of higher premiums) from everyone and transferring them to those who benefit most from the care.
When you take a step back from our insurance industry in the US with this in mind, you realize a lot of the insurance industry serves the function of facilitating government transfers rather than insuring against uncertainty.
Consider the existence of individual mandates. For some time, the national government penalized those who didn’t have health insurance. Some state governments still have these mandates. In the market for auto insurance, individual mandates are alive and well.
Notice the effect of the mandates. If healthy people or safe drivers don’t feel like insurance is worth it, they’ll choose to opt out. This will mean higher premiums for those who remain. Government solves this problem by forcing people who don’t believe they need insurance to buy it. Again, this is essentially a transfer from the healthy to the less healthy.
Another example is the requirement that insurance companies cover pre-existing conditions. There is no uncertainty about preexisting conditions. A normal insurance market would require those with pre-existing conditions to pay higher premiums to prevent those premium costs from spilling over to healthier customers who may decide to leave.
By banning companies from charging more for pre-existing conditions with no uncertainty, the government is transferring money from the other patients and taxpayers to those with pre-existing conditions. Regardless of whether someone supports or opposes this requirement, its function is not insurance. It is a wealth transfer.
So is our current health insurance regime really health insurance? Not completely. Insofar as funding preventative care enables insurance companies to provide coverage for uncertainty, I’d say it is still insurance in one sense. But insofar as it’s a system for enabling transfers, that’s a different story.
This article was adapted from an issue of the FEE Daily email newsletter. Click here to sign up and get free-market news and analysis like this in your inbox every weekday.