Wow! Over two years since I added anything here. Getting reemployed is a good thing, but some items do suffer.
April is Distracted Driving Awareness Month, and by not paying attention it’s now two-thirds over. As I’ve been working in the property catastrophe insurance space the last ten years, I’ve missed some of the current (and not so current) research on distracted driving. I saw the news reports about various states passing anti-texting while driving laws, and here in Massachusetts junior operators may not use cell phones while driving. I also see the problem most days while driving as other drivers suddenly drop from 65-70 mph to 55 or less when they answer their phone. I also know that it is very tempting, especially at a red light, to check my phone while I wait for the light. What I didn’t realize was how much research has been done on the cognitive side of the issue.
When I first heard about the issue of driving while distracted, the first thing that popped into my head was an old episode of CHiPs where a man was pulled over for shaving while driving. How hard can this be, really? Adjusting the radio, drinking coffee, talking with someone in the car, none of these seem particularly dangerous, but cell phone use seems to be, even with a hands-free device. The study from Carnegie Mellon University in 2007 found that the activity in the part of the brain that performs spacial processing declined 37% when the subject was listening to sentences. I think the same thing happens while folks listen to talk radio, but there is a difference. The radio can be ignored or shut off when driving becomes more complicated. A conversation with another passenger isn’t as distracting since the passenger will be more forgiving that the full attention isn’t being put to the conversation. On the phone, though, the usual expectation by the person not in the vehicle is that you are giving the their full attention just as if you were at your desk speaking to them instead of hurtling down the highway or driving past a school.
Fun research that will likely show you how easily it is to miss things can be found at The Invisible Gorilla. Take a look at some of these videos and then think about whether you really can phone and drive without missing important dangers while driving: the signal turning red, a child in a group not getting across the street as fast as his friends, or a parked car pulling into traffic mid-block. For one who thought he was pretty observant, failing to see what is more than obvious (if you expect it) is humbling.
My question for further thought is, “How do we segment this inattentive behavior in the Auto Liability insurance class plan?”. Back in the 1990s and earlier, one may have asked if you owned a cell phone or car phone and assumed that it was being used while driving, but now there are few who don’t own a mobile device.. How do you separate those that understand the danger and use technology safely from those who need to “Hang up and drive!”.
I’m still catching up on recent issues of The Economist. In the March 19 edition, there was an article discussing the impact of the Tohoku earthquake on the reinsurance market. The sense of the article is that this event may not be enough to turn the market given that nuclear risks aren’t in the private market and shake risk will mostly land in the Japan Earthquake Reinsurance Company. This is consistent with the remarks by Bob Hartwig of III at the CANE meeting last week where he said that for global reinsurers, the Tohoku earthquake will be an earnings event rather than a capital event. (See slides 7 and following in his presentation.)
In the next edition of The Economist, there was an article that hit closer to (my childhood) home. The Alaskan Way Viaduct was built in the 1950s and severely damaged during the Nisqually earthquake just over ten years ago. Local officials are currently betting the safety of commuters (110,000 vehicles/day) against the economic issues of gridlock if it is closed before its replacement is ready with completion scheduled for 2016. The fact that the replacement is a tunnel (think Big Dig) makes me wonder how long it will really take.
All this got me thinking about nuclear insurance in the US. Any insurance or reinsurance contract I’ve ever dealt with excludes nuclear risk entirely. When I was studying for actuarial exams I learned that there was coverage provided by the federal government, but would this coverage be enough? The Diablo Canyon Power Plant in California is built directly over a fault line. The decommissioned Trojan Nuclear Power Plant in Oregon along the Columbia River could have experienced earthquake followed by tsunami had a major event occurred along the Cascadia Subduction Zone.
The Price-Anderson act became law in 1957 and was designed to make sure that adequate funds would be available to pay the claims of the public in the event of a commercial nuclear accident. About $71 million has been paid for the Three Mile Island event in 1979. The act was extended in 2005 through the end of 2025. Current limits are a bit less than $13 billion set up in two tiers: private insurance through American Nuclear Insurers (a pool) for $375 million, and an additional $12.6 billion assessable to the commercial nuclear power industry. What do you think? Would the second layer be collectible in the event of a major meltdown? Is $375 million enough?
I was getting caught up on my reading at the gym earlier this week, and I saw this (A boy-racer’s dream?) article in the 2011-03-05 edition of The Economist. It seems the European Court of Justice ruled that effective for insurance contracts entered into after 2012-12-21, sex (gender) may not be used as a rating variable. Here in the US, most states still allow sex as a rating variable, and the arguments tend to fall to whether using it is “unfairly discriminatory”. In automobile insurance, young male drivers tend to have worse loss experience than young female drivers, and young drivers of both sexes tend to be worse than middle aged drivers. When I first moved to Massachusetts, the state mandated class plan was gender and age blind (except for a mandatory discount for the elderly). Rates for “inexperienced” drivers were higher than for the “experienced”, thus a 35 year old just learning to drive got the same rate as the new driver half his age.
I’m wondering, what are acceptable criteria or factors to use in pricing insurance? Most would say that skin color is absolutely unacceptable, but what if there are measurable correlations between skin color and risk? Clearly correlation is not causation, but to what extent should an insurer be barred from using a rating factor? When a rating factor gets banned, are rating factors that are highly correlated to the banned factor also banned, or do they become the new appropriate way to classify and rate?