Friday, December 5, 2008

What Exactly is Variable Annuity Hedging?

I was planning on writing up a summary of some of the variable annuity hedging effectiveness reports, but I received some questions asking to give a little background on what exactly hedging is. Therefore, before I dive into hedge effectiveness, I will dedicate today's blog to describing hedging basics. I will keep this at a high level, but if you are really dying to learn more or have any questions please let me know.

Q: In one sentence or less, what is variable annuity hedging?
A: It is a risk management technique that allows insurance companies to transfer the capital market risk (i.e. stock prices dropping) involved in issuing annuity guarantees to other parties for a fixed cost.

Q: Why do insurers need to hedge?
A: When an insurance company sells a variable annuity guarantee to a policyholder, it is taking on capital market risk. Typically the big risks for these products are if stocks or interest rates drop or if market volatility increases. Additionally, some of the richer benefits also offer guaranteed growth (roll up) rates, and there is a risk that these rollup rates outpace the underlying fund growth.

While insurance companies are in the business of taking on risk, they are most comfortable in risk that can be pooled. For example, in car insurance, insurers have a pretty good idea, that for every 1,000 16 year old boys they insure, x% will get into an accident that on average will result in say $5,000 in claims. It's not fortune telling, it's just using statistics to know that if you have a large enough group of people with similar characteristics, the insurer can estimate the total cost of claims for the group and then charge each person in the group appropriately to have a high likelihood of covering all claims. This is how insurance works most of the time. However, a guarantee on the stock market is not a poolable risk because generally speaking all policyholders have a guarantee on the same stock market. If the S&P goes down, then everyone has claims. There is no law of averages.

Fortunately, hedging techniques can be used to successfully manage this risk.

Q: Do insurers try to make a profit with hedging?
A: No. It is common to confuse the concept of hedging with hedge funds. These are totally different concepts. Hedging is a technique used by insurance companies to mitigate risk. In contrast hedge fund managers use fancy investment strategies and try to get filthy rich.

Q: In layman's terms, how does hedging work?
A: Hedging is complex, no doubt about it.

First, lets start with a product to hedge, so you can picture the risk. The simplest (and most boring) of the guarantees is a Guaranteed Minimum Accumulation Benefit (GMAB). Let's say there is a 10 year deferral period and very basic return of premium guarantee. This means the policyholder invests a premium - say $100,000. This money goes into mutual fund type investments. The insurer will guarantee the policyholder that at the end of 10 years, they will receive at a minimum their initial $100,000 investment or their account value at year 10, if it has appreciated (I don't think a product this boring is actually sold by any major insurers, but it's easy to understand).

So what is the risk to the insurance company? It is the risk that at the end of 10 years, the investor's account is worth less than $100,000. So if the value drops to $90,000 the insurer would have to pay a claim in year 10 of $10,000. Simple enough, right? Now if the insurer didn't hedge, they would have an unknown claim amount when they issued the policy. Or they could transfer the risk to the derivatives market and lock in a fixed cost that would cover claims regardless of what happens in the future. Still with me? Let me use an analogy to illustrate what I'm talking about.

Lets say in January of this past year, you walk into a gas station and they are selling gift certificates that if you pay $2, the gift certificate allows you to receive 1 gallon of gasoline at a point in time in the future. Over this past year, we've seen gas prices rise and fall, but if you owned this certificate, you would be able to get that gallon of gas for the $2 you paid, regardless of what the price of gas is actually at. Therefore you locked in the future price of gas at $2 and transferred the risk of gas prices changing to whomever sold you the certificate. In other words, you hedged against the future price of gas changing... making something that was uncertain now certain.

The same concept is applied with hedging guarantees, only a bit more complicated than buying a gift certificate at a gas station (or else I'd be out of a job)... The insurer uses fancy computers and sophisticated financial modeling techniques to figure out which hedge assets they should invest in, so when the markets move, the change in asset value will offset the change in their expected claim (liability) value. Using this technique, regardless of market fluctuations they can then lock in a fixed cost of issuing the guarantee.

So let's put this all together now in terms of the GMAB product discussed above. The insurer calculates the hedge cost (expected claims if hedged) they would expect to pay on a policy, let's say this is $1,000 at issue. They can then lock in the cost of issuing the guarantee by purchasing hedge assets that would insulate the insurer from any changes in the markets. So if after they lock in the hedge say that the stock market drops and by the time year 10 comes around, the policyholder's account value has dropped to 90,000. This means the insurer must pay them $10,000. Fortunately, the insurer has this money. They set aside $1,000 for the hedge cost. In addition, the hedge assets they "purchased" have increased by $9,000. Adding these together provides the insurer with $10,000 to pay the claim. Similarly, had the policyholder's account value dropped to 80,000, the insurer would owe 20,000. This 20,000 would be funded with the $1,000 they set aside for the hedge cost plus the $19,000 increase in the hedge assets. Thus, by hedging, insurers transfer the market risk to the derivatives market.

At a high level this is how hedging works. The uncertain risk is transferred for a more certain set of cashflows by entering into hedge assets whose payoff help to offset the change in liability cashflows that are owed to the policyholder. Hedging isn't perfect, and there will be some mismatch between the asset and liability cashflows, but in my next post, we'll take a look at how effective hedging has been in this recent period of declining markets and high volatility.

Wednesday, December 3, 2008

The Full Story on Annuities

Annuities and their issuers have been all over the news lately - getting a lot of bad press. It's one thing to report the truth about the industry, but its another thing to talk negatively about a subject an author doesn't understand and put out misguided information to scare investors (there is a time and place for this sort of behavior, but it requires running for a political office).

One misconception that I hear a lot involves the cost of variable annuity guarantees. Some say they are too expensive. Others say they don't charge enough. Well, which one is it?

Also, lately I've heard a lot of negative comments that hedging doesn't work. Insurers don't understand the risks they are taking and don't know how to manage them. Similar comments have been made addressing the capital infusions insurers are requiring.

Well, here is the full story: annuities are complex products and pricing and hedging annuities is even more complex. In order to draw conclusions on pricing and hedging these products, you need to understand how they work and what the goal is. In addition, the capital requirements and accounting rules that va sellers must abide to have inefficiencies, that weren't fully realized until current markets, and as a result they have penalized insurers in unintended ways. Discussions are currently underway to correct these inefficiencies. This is not to say that insurers aren't hurting right now and insurers don't need more capital, but there's more than meets the eye.

For financial advisors and individual investors, it is extremely important that you understand the whole story on these issues in order to form your own opinion on the strengths of these companies and their products. My goal is to deliver you the full picture on a number of these "hot topics" over the next few weeks.

The first topic I will dive into is the effectiveness of companies managing annuity guarantee risk. I'll set the scene and provide some (optional) background reading on this issue below, but to keep this blog at a manageable length, I will put the meat in my next blog.

There were two reports put out earlier this year that have investigated the effectiveness of hedging, one was by Moody's (August 2008 - Link to summary)and the other by Milliman (May 2008 - Link). They both concluded that hedging works in periods of "market turmoil' as stated in the title of Moody's report.

Well "market turmoil" is a relative term, or else what what do you call the last 3 months since that report was published. Milliman just published a new report (link), looking at how hedge programs have performed in September and October, during the period of "really really really bad market turmoil" (I had to answer my own question above). It is a bit technical, but a great report that provides a lot of incite. I would encourage a read (and please let me know if you have any questions).

While, many have heard of Moody's, one of the top rating agencies, I doubt many have heard of Milliman, so let me tell you a little bit about them.

Milliman is one of the top independent insurance consulting companies in the world. They are the global leaders in variable annuity financial risk management and have a market share of about 70% (which is huge). I would say two things about these guys. They understand hedging and its effectiveness better than anyone else. They also are critiquing their own work in a way, which some may say adds bias. However they use real, factual data from insurer hedging programs in their report to draw their conclusions, and I believe everything they say to be true. In addition, Moody's previous report drew similar conclusions to Milliman's May report. However, I would also say, that in a report like this, it takes a trained eye to dissect how they say things to read in between the lines, which I will add commentary on.

For full disclosure, I would add that Milliman is my former employer, and while I have much respect for the people there, I do not have any vested interests in the company.

Please read my next blog entry (above) to get the full story on hedging variable annuity guarantees.