How insurance companies work

Vast simplification: very carefully.

Okay, I’ll give you a little more detail than that, but still very simplified.

Let me start with a very simple product: term life insurance. You buy a term life contract, where you pay certain premiums periodically for a certain amount of time, and if you die within that time period, your beneficiaries get a specified amount of money.

To make sure the insurance company actually can pay on this potential claim, they set up reserves. There are lots of types of reserves they have to set up, but let me focus on statutory reserves. Insurance companies in the United States are regulated by the individual states, though the states have made for some uniformity via the NAIC. [I'm not going to get into the history of insurance regulation and why it's done by the states; it just is.]

There are two prongs of state insurance regulation: solvency and market conduct. Let’s focus on solvency first.

Insurance companies make promises through their contracts to pay money when certain things occur; some of these promises are very long-dated or involve a lot of money. Insurance contracts are worthless if you can’t count on the money being paid when you need it. Statutory reserves are a piece of this, where insurance companies have to calculate how much money they need to have on hand to pay expected claims for the life of the contract; there are all sorts of complications and bobbles in how the actuaries calculate this stuff, and the regulators make sure a certain amount of conservatism is layered in there so that the insurance companies are not low-balling the risk they’re taking on. The reserves are part of the general account of the insurance company, and is mainly invested in fixed income assets such as bonds and mortgages.

But wait, there’s more. It’s not enough to have the reserves – you also need surplus or capital. What if there’s a flu pandemic? Terrorist attack? Or just normal deviations you get from randomness and you get excessive claims one year? The regulators require even more money set aside, on top of reserves, called required risk-based capital [RBC]. Insurance companies can also mitigate these claims in excess of reserves by entering reinsurance treaties [reinsurance is insurance for the insurance companies].

By the way, RBC is calculated on a firm-wide basis, and includes needing more cash on hand if the reserves are being invested in riskier assets [which is why insurance companies don't have portfolios full of junk bonds]; RBC isn’t allocated on a per-policy basis. Contrast to this with reserves, which can be [and is] allocated to each policy [having this allocation is very important should the company go insolvent].

Insurance companies actually have quite a bit more money on hand than just reserves and required RBC – having that level of cash is the dividing line for being considered solvent by regulators. The rating agencies like A.M. Best, Moody’s, and S & P want to see even more than that to give insurance companies their best ratings.

I forgot to mention cash-flow testing, also required by regulators, which is yet another set of solvency regulations. You’ve got to model the assets and liabilities under specific interest rate scenarios [very low or very high, or a specific pattern... there is a set called the New York 7 that have to be run]. This started because of the high interest rate environment of the early 1980s, and due to even more complexity and volatility, regulations are trending to more stochastic testing. Look, I’m trying to keep this simple, so enough on that.

The point is: insurance companies are heavily regulated, seemingly more regulated than banks. Hmmm. Something on that another time.

But should an insurance company go insolvent in the U.S., what happens? There are a variety of remedies, but as a last resort, each state has a guaranty fund that will pay off claims [after the state has taken over the company, meaning it will have all the reserves and capital/surplus at its disposal as well].

So that’s one prong of insurance regulation. Net premiums are used to build up reserves and capital, and money from investors [if it's a stock company] provides even more capital. If you’re concerned about the eeeeeevil insurance companies making money off of death and destruction, then there are also mutual insurance companies, where the shareholders are the policyholders themselves. There are plenty of mutual out there, in life insurance as well as property/casualty. There are various fraternal societies, such as the Knights of Columbus, that also have insurance arrangements.

So what about this other prong of insurance regulation: market conduct.

The point of market conduct regulations is to protect people from being defrauded by insurance companies – either not paying legitimate claims [which is why if an insurance company is refusing to pay a claim to you, you should contact your state's department of insurance], or selling policies that are not appropriate [deferred annuities to a 90-year-old man, say], or unfair premiums.

Market conduct regulation is where things get really political.

Let me tell you a little story about Florida. You may have noticed that Florida has the Atlantic Ocean on one side and the Gulf of Mexico on the other. You may have also noticed that when there are hurricanes [been kind of quiet lately, but these things go in cycles] Florida is apt to be hit. People may have forgotten that the year of Katrina was also the year of Rita, Tammy, and Wilma . All four hurricanes [or tropical storms, they change classifications through their paths] took a visit by my grandfather’s house in central Florida. I don’t quite remember how his homeowner’s claim got sorted out there, but I do know he took damage from more than one storm and that became an issue with regards to the deductible [I believe his deductibles were on a per claim basis, and the insurance company saw two separate storms as two separate claims...which makes sense to me.]

As you can imagine, the insurance companies took a hit to their surplus [the claims were higher than average, duh], and the surplus built up over less-than-usual claims years was drawn down… but they had to build that up again immediately. Yes, premiums were going to be increased some.

Well, some politicians ranted about this of course: those evil evil insurance companies charging more for living on the coasts, where the risk of damage was higher. How dare they! And those climate change models predicting more hurricanes being used for justification — well, well, wait! Those models are intended to be used for us to soak the taxpayers, not be taken seriously for increasing insurance costs! [It will be interesting to see what the fallout of warmaquiddick will be in the actuarial world. They just announced a request for proposals on the impact of climate change...I sent them an email regarding making sure the models and data are fully disclosed]

In any case, the Florida DOI was rejecting the premiums being set by P/C insurers for homeowners coverage. And the insurance companies said “Screw you, we’re leaving”. Here’s a thread at the Actuarial Outpost about State Farm exiting [or, rather, trying to exit... it's complicated] the Florida homeowners market. Here’s a thread about the amount of catastrophic exposure Florida has re: hurricanes.

The thing is, yes, some of these companies are nationwide, but the premiums need to reflect the risks and the reserves and capital required for that particular line of business. Florida has been trying to nationalize these risks, basically by getting people in Nebraska paying higher premiums so that people in Florida can live in hurricane alley and underpay for the risks being covered. But an insurance company can operate within a single state, so if Florida costs are spread to Nebraska, and the Nebraska risks are a lot cheaper, a company can start in Nebraska for homeowners insurance with much lower premiums. And the P/C market is very driven by prices … it’s become very commoditized. If the net premiums collected can’t even cover reserves, much less required capital, then the companies are likely to lose money over the long term.

In any case, it’s interesting seeing the push and pull of the need for solvency versus the political drive to keep premiums below risk costs. Florida has been lucky the last couple years… it’s been pretty quiet hurricane-wise. But insurance companies can’t count on this – they’re going to have to pay if there’s another year of high hurricane numbers [and considering past patterns, even absent the whole AGW debate, it seems likely this will happen again.]

People look at the huge reserves insurance companies set up, conveniently forgetting that the companies can’t do anything with that money. It’s there to cover future claims by statute. Yes, if the claims never materialize, then it will eventually feed into profit or surplus [in the case of mutual companies]; but if the companies are always overcharging for premiums, they will be undercut by competition.

At another time, I’ll explain adverse selection, underwriting, and moral hazard. But this is just to give you a flavor of what’s going on.