Monthly Archives: February 2012

Testing 1… 2… 3…

Got a smartphone and going mobile.


The Week in a Minute, Feb. 24, 2012

  • Accuweather predicts an above-average tornado season this year, according to the cat bond experts at Artemis. January got off to a rousing start with 95 twisters, more than double the average. The storms are set to infest the South in March, then move northward a month later.
  • Good news for AIG, which earned $19.8B in Q4. Of course, $17.7B came from reinstituting a tax break. Still, core p/c business Chartis posted a $348M profit.
  • Stickier situation for the Hartford, where key investor (and hedge fund manager) John Paulson has called for the company to split its life and p/c units, detailing the plan to unlock the synergies of breakup in an SEC filing. Actuary/investment analyst Todd Bault glanced at the matter quickly and conjectured that Hartford won’t split because the life company could be more or less worthless. Here’s the company’s statement.
  • Tim Zawacki at SNL blogs (paywall, folks) the quote of the week: “If we were a normal insurance company, we’d be belly up right now.” That from an unidentified governor of Florida’s Citizens Property Insurance. He had just heard a presentation on the impact of sinkhole claims on 2011 results.  46% of the state-run insurer’s losses last year came from sinkholes.

Civil War pensioners still collecting benefits

Further exploring the boundary between genealogy and actuarial science:

Despite the fact that the Civil War ended April 9, 1865 (53,630 days ago, for reference), the government is still paying out veterans’ pensions.

Records from the Department of Veterans’ Affairs show that two children of Civil War veterans, as of September, are receiving pensions from their fathers’ service.

Thanks to Ancestral Archaeologist for the link.

According to U.S. News and World Report, “it’s likely that the children of the Civil War veterans, who have wished to remain anonymous, both had illnesses that prevented them from ever becoming self-sufficient.”

BTW, John Tyler’s grandchildren are still alive.


The Week in a Minute, Feb. 17, 2012

  • The Thai floods cost Lloyd’s insurers $2.2B, the third-largest loss in the syndicate’s 324 years. (Katrina’s No. 1, and 9/11 is No. 2.) The Thai government is contemplating covering floods up to $16.2B in a single treaty, which would be some sort of public-private partnership.
  • Penn State puts its tab on the Sandusky scandal at $3.2M – so far. And it’s suing PMA for refusing to provide general liability cover. The university set up a web site as a pledge to openness at the institution.
  • European windstorm Andrea cost insurers $350M, PERILS estimates.
  • Zurich Financial plans to change its name to Zurich Insurance Group, the rest of that financial stuff haven’t not worked out so well.

Does bootstrapping understate reserve variability?

Over at Guy Carpenter land, actuary Jessica Leong posits that the bootstrapping method of estimating reserve variability understates that variability.

Bootstrapping, of course, is a method of using the company’s own data to drive a random sampling process; said process re-estimates the reserve. Sample enough times, keep track of the results, and you’ve got a distribution of results. (Better descriptions are here.)

But the method doesn’t seem to work as advertised.

Leong uses company Schedule P homeowners data. Schedule P gives a triangle of paid losses and also ultimate losses. And homeowners has a short tail but produces 10 years of results. The result 10 years out is an excellent estimate of ultimate.

So you have an a priori distribution of losses. And you know, 10 years out, what the actual losses turned out to be. So you can pinpoint where in your original distribution the actual losses came in. In her post, Leong walks through an example, which happens to come in at the 91st percentile.

If you did this across a bunch of companies, across a bunch of years, the actual losses should occur uniformly, at least if bootstrapping works as advertised. So 10% of outcomes would fall within the bottom tenth percentile, another 10% would fall within the top tenth percentile, and so on.

The actual results are here:

The chart shows that

.  .  . around 20 percent of the time, the actual reserve is above the 90th percentile of the bootstrapped distribution, and 30 percent of the time the actual reserve is below the 10th percentile of the distribution.

When you tell management the 90th percentile of your reserves, this is a number they expect to be above 10 percent of the time. In reality, we find that companies have exceeded this number  20 percent of the time. The bootstrap model is under-estimating the probability of extreme reserve movements, by a factor that is clearly material for the purposes of capital modelling and therefore Solvency II.

Hers is the first post in a series to play out over the coming days. In the meantime, I’m a bit curious:

  • Do the outliers have much in common with each other? Are they smaller companies or larger ones?
  • The study works from net data. Would you get the same results gross of reinsurance? Schedule P triangles are net of reinsurance, of course, but you can construct a gross triangle from successive annual statements. Net triangles could be skewed by the presence of reinsurance, especially excessive of loss or catastrophe cover.
  • Does the data set include reinsurers? Reinsurers don’t always receive – and rarely record – losses by accident year. Usually everything is recorded on an underwriting year. So the accident year loss payments and ultimates in Schedule P homeowners are estimates arrived at by allocating underwriting results across accident years. And, believe me, the allocation can be shot full of holes.
  • Do other lines of business exhibit the same phenomena? Homeowners is short-tailed, but the presence of catastrophes skews development and results. Would you see the same phenomenon in, say, private passenger auto?
Tagged ,

A new wrinkle in IBNR

Behind its firewall, SNL writes about reinsurers establishing cat IBNR reserves that aren’t tied to an event:

Typically, reinsurers establish IBNR reserves to cover future payments on individual losses that have occurred but have yet to be reported or reported losses that have not been recorded in full.

Unallocated IBNR reserves are a slightly different animal. They are an additional reserve, on top of event-specific reserves or account-specific reserves, created to respond to any potential deterioration in losses across a number of events. One way to think of them is as “just-in-case” reserves when just-in-case losses really get out of hand. Ian Gutterman, an analyst with Adage Capital Advisors LLC, referred to it as “cookie jar IBNR.”

As an example, PartnerRe Ltd., in addition to recording charges in its fourth-quarter 2011 earnings for the flooding in Thailand, twin earthquakes in New Zealand and Tohoku earthquake in Japan, created an IBNR provision of $50 million for all the catastrophe events of 2011.

I’m not on the ground reserving this stuff, but I think the practice stems from the unusual number of poorly understood exposures that produced losses. Japan quake models didn’t contemplate tsunami. New Zealand models didn’t know about fault lines beneath Christchurch. Thai floods were unmodeled.

Add to that, as SNL notes, the fact that the companies reporting losses are slower than usual:

The advent of unallocated IBNR reserves comes at a time when reinsurers are struggling to pin down their ultimate losses from the string of large catastrophes that occurred last year around the globe. In the case of the Tohoku earthquake, loss estimation has been hindered by reporting delays from the large mutual insurer Zenkyoren. In New Zealand, the government decision to deem wide areas of Christchurch as uninhabitable is said to have slowed down claims settlement. The complexity associated with certain types of claims, such as contingent business interruption, has also been a factor. Altogether, the delays have led to significant increases to reinsurance IBNR reserves.

Meanwhile, no one truly knows what the ultimate losses would be for certain events. The Thai floods, for example, could eventually be a $12 billion industry loss event, or a $15 billion or $20 billion, said Validus’ Noonan. “I am starting to sound like Donald Rumsfeld,” he said, adding: “There are things that we don’t know that we don’t know.”

As a result, a couple of companies – PartnerRe and Platinum, perhaps some others – have been dinged by upwardly creeping loss estimates for catastrophes. In recent years, companies have been expected to reserve adequately at the first estimate with those estimates perhaps creeping down over time. Rising cat estimates disappoint Wall Street, so stock prices can fall if the estimates keep going up.

I’d be surprised if this became a standard practice. It’s unusual to have so many highly uncertain events to reserve against. And you can’t just throw some IBNR against the wall figuring you’ll need it someday.

Tagged , ,

Math problems

Alex Tabarrok at Marginal Revolution frets that as people age, a shrinking proportion answer $400,000 to the following question:

If five people all have the winning number in the lottery, and the prize is $2 million, how much will each of them get?

But of course the answer is not $400,000. As most big-ticket lottery stories point out, the prize is a nominal amount spread out over something like 20 years. And in most cases, the winners take the present value of the winnings. Which is less than $400,000 as long as interest rates stay above zero.

On top of that, there’s taxes to think about, which are always withheld from winnings that big.

So, contra Tabarrok, maybe people are getting smarter as they age.

File under emerging cat risk

Offshore windmills pose no environmental or socioeconomic risk, according to the Interior Department. But the department didn’t consider this:

In certain risky offshore regions off the Atlantic and Gulf Coasts, there is a high probability that at least one turbine would be destroyed by hurricanes within 20 years, and a smaller chance that half the turbines in a farm would be wiped out.

Current windmills are designed to withstand gales in the North Sea, which, while rough, are nothing next to a Cat 3 hurricane. So windmills could snap off.

It would help if windmills could spin to face the wind, called yawing. Many windmills yaw, but need electricity to do it. But in a big storm, power out to the windmill would be cut off.

The study calls for stronger windmills and battery backups to the yawing mechanism.


The latest in predictive modeling

I know this is a big story, because my Mom railed on about it this weekend. She’s not an Allstate customer, and she doesn’t live in Oklahoma.:

If you’re a lousy driver, are you also more likely to be a mishap-prone homeowner?

Allstate Corp.believes there’s a correlation.

The Northbrook-based company confirmed it has begun considering the driving records and auto-claims histories of people who apply for a new homeowners product called House & Home that Allstate plans to roll out to states beyond Oklahoma through 2014.

Of course, if your a good driver, your homeowners rate would go down.



Claim of the Week

Thanks, ClaimsJournal:

A judge has thrown out a New Jersey man’s lawsuit against a Greenwich Village pub over injuries he suffered following a beer pong game.

The man played for 3½ hours, then got hit by a car while crossing a highway in Manalapan, NJ, 45 miles away.

Berger claimed the bar should have been monitoring the game to make sure players weren’t getting visibly drunk.

However, the judge “said Berger ‘consumed alcohol to the point of diminished capacity.’ “