Insurance 101

I read an article written by several highly respected people in economics and health care published in The Wall Street Journal today on several aspects of the health care reform that I have thought a fair amount about. I quite agree with what they wrote and recommend reading it. I want to expand on some of the things addressed there and talk about insurance principles that are pertinent to the discussion on health care reform.

I realize this is kind of a long post. It has taken me several hours to write it. I think if you read through the whole thing it will be worth your while. First I’ll talk about how insurance works (really basically); then I’ll talk about policies that are currently being debated and some of their implications. Hopefully that doesn’t make you stop reading now.

How Insurance Works

No matter what kind of insurance you are talking about, you are talking about protection from some kind of risk. Insurance companies want to sell you protection from that risk for some price. The way they do that is by pooling lots of people with risk together and then doing a lot of statistics to calculate how much you have to charge people to cover the costs. I could talk about this generally, but I think it will be easier to follow with a more concrete (though very simplified) example. If you hate numbers, just skim over the details in the next few paragraphs and get the basic idea.

Let’s say that there is a company that wants to insure against fire damage for houses. They expect to sell insurance to 10,000 people. Now for simplicity, let’s assume that every house is worth $200,000 so we’re talking about $2 trillion dollars in property. The company does a whole lot of research and determines that on average in a given year 5 houses catch on fire: 3 totally burn down and have to be replaced (at a cost of $200,000) and 2 only burn partway and it takes $100,000 to repair them. This means that they expect to pay $800,000 every year. In addition to that, they have to pay the actuary who did all the calculations and the salesman that went around selling. Let’s say that adds up to be $200,000. So their yearly expenses are expected to be $1 million. They take this number and divide it by the number of people that buy insurance to get the premium:

So they decide to charge $100 per year to each person they sell insurance to. Then as the year goes on they use all of the insurance premiums to pay for the houses that catch on fire.

Now let’s say they do some more research. They find out that houses with fireplaces catch fire way more often than those without. In fact, of the $800,000 they pay for houses that catch fire, $600,000 of it comes from houses with fireplaces. So they start adding a question on their paperwork and ask people if they have a fireplace. They find out that half of them have fireplaces, so they run some calculations:

Based on those calculations, they start charging those with fireplaces $140 and those without $60 in premiums. Then they still get the same amount of money and are still able to pay for all of the houses that catch fire.

Now you probably wonder why the company bothers to charge two different rates: they do it so that they are competitive. Suppose they didn’t but their competition did. Then all those without fireplaces would go get insurance elsewhere because they can get a better rate. That means that most of the people getting insurance from this company are going to have fireplaces, which means they are going to have more houses catch fire, so they are going to have to charge higher premiums. Eventually they will be charging more than the $140 that their competition is charging because they will have fewer clients so they can’t spread the fixed costs out across their pool of clients as well. Then those with fireplaces will go elsewhere as well and the company will go out of business.

There is also another effect of this that is quite positive: people buying houses make better decisions. When they find out that if they buy a house with a fireplace then they have to pay more in insurance they decide whether it is worth the cost. Those that think it is worth it get a fireplace and pay the extra; those that don’t think so don’t get a fireplace. And everyone is happy. Now if this wasn’t the case—if everyone was paying the same amount regardless of whether they have a fireplace—then those without fireplaces would be paying extra: they would be footing the extra insurance bill from everyone else having a fireplace without getting any of the benefits. If they knew this was the case they would start complaining.

Now most of you have probably assumed that the people that buy fire insurance in this example have houses that are not currently on fire. That’s pretty reasonable, but let’s think about what would happen if that’s not the case. Suddenly things are a lot different. The insurance company then wants to start asking the important question “Is your house currently on fire?” when people apply for insurance. If so, the odds that the insurance company has to pay for fire damage is a little higher than if it’s not on fire. So they do the same thing they did with fireplaces: they try and separate things out and calculate probabilities. They find out that if a house is currently on fire, then 60% of the time they are going to have to pay to rebuild it ($200,000) and 40% of the time they are going to have to simply repair it ($100,000). So then the insurance company says, “Your house is on fire right now? Sure we’ll insure it. But we are going to charge you a premium of $160,000 to do so.” That way they don’t have to charge those whose house is not on fire a different amount.

Now if the prices of insurance for a house that is already on fire versus one that is not is not separated out then we see the same effect as we did before when we didn’t separate out those with fireplaces (though it looks a little different). Suppose the insurance company just groups everyone together regardless of whether their house is on fire and they simply charge a uniform rate of $100 to each person. First of all, the insurance company will get an influx of people wanting to buy insurance when their house is on fire, so they have to raise their rates to compensate and stay in business. Even without raising their rates people will be inclined to simply wait until their house catches fire before they buy insurance. That way they can save money when their house isn’t on fire. But this means the insurance company has less people in the pool who are paying premiums and more people in the pool with houses that are burning down. They won’t have enough money to cover the burning houses so they will have to raise their rates again to compensate. Then more whose houses aren’t burning will stop paying insurance and the problem worsens until you only have people whose houses are burning left buying insurance. Thus, it’s pretty important that insurance companies are able to ask questions and segment the insurance pools. If they don’t then large groups of people will not buy the insurance because the premiums are too costly.

Applying this to Health Insurance

Now let’s talk about what this has to do with health insurance. You are probably already making the connections. Health insurance works precisely the same way as the fire insurance example (taking into account, of course, the gross oversimplifications of the example).

Here is a list of a few of the prominent points being debated, most of which I took directly out of President Obama’s speech to congress a couple weeks ago. There are a lot of other points, some of which I would like to address—possibly in a future post. For now, let’s look at these specific ones:

  • Make it illegal to deny coverage based on a preexisting condition
  • Introduce a “community rating” system so that those that are sicker or have preexisting conditions couldn’t be charged more for coverage
  • Make it illegal to drop coverage when you get sick
  • Remove per year or per lifetime caps on benefits
  • Place limits on the amount of out-of-pocket expenses that can be charged
  • Force insurance plans to include checkups at no additional charge
  • Make it illegal to not have insurance

Now many of these could have very good results. What I want to do here is discuss some of the implications of these in terms of the insurance industry. Once we know what the change in the insurance industry will be from changing these things then we are more able to decide whether it is worth the extra benefit to make these changes.

First, let’s look at what happens when you make it illegal to deny coverage for a preexisting condition. This is very analogous to the fire insurance company having to insure people whose homes were already on fire. The insurance company is going to have to pay for those preexisting conditions, so they are going to have to take in more money from premiums to offset that higher expense. Insurance companies’ first inclination will be to do the same thing we saw in our fire insurance example: they will incorporate as much information about those it has to insure as possible into their probabilities and charge those with preexisting conditions higher rates.

This isn’t a very good outcome for those with preexisting conditions. This means that if you have cancer, the company is going to factor that piece of information in when calculating your premium and it’s not going to be cheap. To keep this from happening then, the insurance companies would be mandated to use a “community rating” system, so those with preexisting conditions couldn’t be charged more than those without. That’s like telling a fire insurance company it can’t charge a higher premium to those whose houses are already on fire. The money to pay for the houses that are already on fire has to come from someplace: it will come from higher premiums charged to the entire pool. Similarly, if those with preexisting conditions start having preexisting conditions covered, and insurance companies can’t legally segment the pools to include that information when charging premiums then they are going to have to start charging higher premiums to the whole pool. This means healthy people are going to be paying for those that have preexisting conditions through their higher premiums.

This same thing happens when talking about those that are sick that are already covered. The “community rating” system would mean that those that are already sick couldn’t be charged more than those that are not sick. This includes those that are older: older people tend to have more health expenses than those that are younger. If the insurance company can’t segment these groups apart then those that are sick, older, and/or have preexisting conditions will all be charged relatively the same amount as a younger, healthier person.

Insurance companies also cut costs by creating loopholes allowing them to drop your coverage or placing limits on the amounts that they pay in a year or over a lifetime. This is really bad for those affected; consequently, it is proposed that this be made illegal. This will make the costs seen by insurance companies go up—they have to pay for these new expenses that they were previously getting out of. Once again, they have to pay for this somehow: insurance premiums will have to go up to compensate. Since they can’t segment the insurance pools the premiums have to go up for everyone in the pool to pay for those that are very sick (those exceeding current limits).

Statistically, this could have a huge impact because of the magnitude of outliers. Suppose an insurance company is insuring 100,000 people and they expect to pay out $75 million dollars each year for those 100,000 people. That works out to be $750 per year per person in premiums. Now suppose that those numbers take into account a limit of $1 million in benefits per person and also suppose that limit is removed. Then there is one person in that 100,000 that needs to have a series of surgeries, transplants, medications, and therapy that totals $10 million—I don’t think this is unreasonable when dealing with healthcare. This makes the payout for the year $85 million which raises the premium to $850—all because of that cap was removed. That’s a big increase because of one person. As the pool size grows then the effect per person is less; but as the pool size grows, the odds that an outlier could cost significantly more than $10 million dollars also grows. The amount that an outlier will require cannot be easily calculated statistically and it doesn’t take very many serious outliers to cause devastating financial problems for an insurance company.

The only points I haven’t hit yet is placing limits on the amount of out-of-pocket expenses you can be charged for health care and forcing insurance companies to provide preventative checkups at no extra cost. From the discussion so far, I think it should be clear that this also increases costs seen by insurance companies (because now they’re paying for things they weren’t) so premiums will also be affected. True, those insured may then be able to get checkups at “no extra cost” but the costs will simply be built into the premiums which will be higher than they would otherwise be because of those extra “free” services.

Now if you are healthy and your premiums start going up because of all of the foregoing then you have more incentive to not get insurance. Why do you need it? The insurance companies have to cover your preexisting conditions when you finally decide to get it. So why pay premiums when you aren’t sick? Thus, healthy people drop out of the insurance pool, which causes medical costs per person to go up in the pool, which causes insurance companies to raise their rates further. Then more people have an incentive to drop out. Eventually, no one has insurance unless they are sick and insurance premiums become astronomical.

The solution to this new problem then is to require everyone to have insurance. Making it illegal to be uninsured would keep all the healthy people in the pool and allows insurance companies to function with non-astronomical rates while still covering preexisting conditions, the elderly, and the very sick. But then you have the situation that we talked about with the fireplaces in houses. If those with fireplaces are grouped with those that don’t have them then those that don’t have a fireplace are paying for the risk incurred by those that do have them. Same thing here: those that are healthy are paying for those that are not through higher premiums.

There would be a loophole to this, though (at least for some). People that don’t want insurance (and who don’t have an employer offering insurance) can pay an additional excise tax and not get insurance. If insurance rates are high enough this would be worthwhile: you could pay the cheaper tax until you get sick. If you do get sick, then you get insurance that legally has to cover your preexisting condition. Again, this would cause insurance rates on the pool to go up because healthier people are choosing to stay out of the pool until they get sick.

One positive thing to note, however, is that if that loophole is not very big so that overall more people have insurance than do currently, then it is possible (and probably likely) that this will bring premiums down from what they otherwise would be. This is because more healthy people would be brought into the pool to further distribute the costs incurred by the sicker people. Though I doubt this effect would be anywhere near significant enough to offset all of the other increases that I have discussed.

A Few Conclusions

When taken together then, these changes mean that a lot of people will be better off: those with preexisting conditions, the elderly, those that are sicker than average, and those that are really, really sick. All of these people are currently facing limitation on what benefits they get and are seeing higher premiums because of their current health. These are really serious problems that need to be addressed.

But these benefits also come at a cost: it means that those that are younger and not sick are worse off. Under this plan they would be subsidizing the health expenses of those that are sick, and the subsidies have the potential of being pretty large. Now that may not be entirely bad—in fact it might be worth it. But those who are deciding need to be aware of the costs that are being incurred because of the benefits that are being received.

Those that are pushing for these changes (like President Obama) are putting a lot of emphasis on the benefits to the elderly and the sick because of these changes. But there is not a lot of emphasis on the costs that are incurred in exchange for these benefits. Under this plan health insurance premiums would go up for the majority of people—particularly those that are the healthiest. This would make wages and income for those people fall significantly.

What we are talking about here is a method of income redistribution. Income will be redistributed from those that are younger and healthier to those that are older and sicker. The redistribution is simply hidden in insurance premiums so that most people are not aware that is what is being proposed. And President Obama and Congress are certainly not trying to let people know that is the case.

I think it is worth quoting here from the article I mentioned when I started:

[Because of these proposed policies] younger, healthier, lower-income earners would be forced to subsidize older, sicker, higher-income earners. And because these subsidies are buried within health-insurance premiums, the massive income redistribution is hidden from public view and not debated.

If Congress goes down this road, health insurance premiums will increase dramatically for the overwhelming majority of people. Even if Congress mandates that everyone have health insurance, many will choose to go without and pay the tax penalty. If you think people are dissatisfied with health care now, wait until they understand that Congress voted to mandate hidden premium increases and lower wages.

If you read this whole thing, hopefully you find it helpful to understanding (at least part of) the economic implications of the debate currently going on. Please comment—I’d like to hear what you think about this topic and also what I have said. Hopefully I’ll tackle another topic again soon.

Now that I finished writing this, I can’t help but wonder: Why am I working as a lowly support engineer at a software company?