All posts by James Lynch

Updated 2018 Atlantic Hurricane Season Outlook: Cooler Atlantic Temperatures Could Lead to Below-Average to Near-Average Hurricane Season

Special to the Triple-I Blog

by Philip Klotzbach,Ph.D,
Research Scientist, Department of Atmospheric Science, Colorado State University and I.I.I. Nonresident Scholar

Colorado State University (CSU) has just updated their outlook for the 2018 Atlantic hurricane season, and is now calling for a near-average season with a total of 14 named storms, six hurricanes and two major hurricanes (maximum sustained winds of 111 miles per hour or greater; Category 3-5 on the Saffir-Simpson Wind Scale) (Figure 1).  This prediction is a slight lowering from their initial outlook in early April which called for 14 named storms, seven hurricanes and three major hurricanes.  Accumulated Cyclone Energy (ACE) and Net Tropical Cyclone (NTC) activity are integrated metrics that take into account the frequency, intensity and duration of storms.

Figure 1: May 31, 2018 outlook for the upcoming Atlantic hurricane season

CSU’s meteorological team uses a statistical model as one of its primary outlook tools.  This methodology applies historical oceanic and atmospheric data to find predictors that were effective in forecasting previous years’ hurricane activity. Based on data dating back to 1982, this model has shown consistent accuracy. (Figure 2)  Statistical forecast for 2018 is calling for a below-average season.

Figure 2: Accuracy of June statistical forecast model at predicting historical Atlantic hurricane activity (since 1982)

CSU also employs an analog approach, which uses historical data from past years with  conditions that are most similar to those currently observed (as of May 31, 2018).  The team also forecasts projected conditions during 2018 peak hurricane season (August-October) by looking at historical data from years with similar August-October conditions.

This approach yields a similar outlook of below-average to near-average sea surface temperatures (SSTs) in the tropical Atlantic and near-average sea surface temperatures in the eastern and central Pacific.  The average of the four analog seasons calls for a near-average season. (Figure 3)

Figure 3: Analog predictors used in the May 31, 2018 seasonal forecast

CSU does not anticipate a significant El Niño event for the peak of the Atlantic hurricane season.  At this point, the meteorological team believes that the most likely outcome is neutral conditions for the next several months.  El Niños tend to reduce Atlantic hurricane activity through increases in upper-level winds that tear apart hurricanes as they are trying to develop.  Most of the dynamical and statistical model guidance agrees with this assessment and calls for neutral conditions for the next several months. (Figure 4)

Figure 4: Statistical and dynamical model guidance for El Niño

Most models are calling for neutral conditions for August-October, as highlighted by the black arrow. (Figure courtesy of International Research Institute for Climate and Society.)

The primary reason for a reduced seasonal forecast (compared with earlier 2018 outlook), is due to anomalous cooling of the tropical Atlantic over the past couple of months.  As shown in Figure 5. most of the Atlantic right now is quite a bit cooler than usual. In addition to providing less fuel for storms, a cooler tropical Atlantic is also associated with a more stable and drier atmosphere as well as higher pressure—all conditions that tend to suppress Atlantic hurricane activity.

Figure 5: Current SST anomalies in the North Atlantic.  SSTs are much cooler than normal across the entire tropical Atlantic

The most important thing to note with all seasonal forecasts is that they predict basinwide activity and not individual landfall events.  However, regardless of what the seasonal forecast says, it only takes one storm near you to make it an active season.  Therefore, coastal residents are urged to have a plan in place now before the hurricane season ramps up over the next couple of months.

Extra: If you live in a hurricane-prone region, your homeowners insurance policy may have a separate hurricane deductible. This infographic explains what you need to know.

The Ellicott City Flood: Rebuilding Begins with Resilience

By Sean Kevelighan, CEO, Insurance Information Institute

On May 27, for the second time in three years, Ellicott City, Maryland was ravaged by what meteorologists term a “1,000-year flood”—this while some businesses were still celebrating the one-year anniversary of their reopening after the August 2016 catastrophe.

As affected households and businesses assess the damage and pledge to rebuild (or to relocate) after this deadly event, one fact looms largest: that 1,000- or 100-year floods now seem to strike with numbing regularity. The time has come, then, for communities and individuals to accept this paradigm shift by embracing resilience.

Local, state and federal governments have a wide range of tools at their disposal to effectuate resilience, including public policy solutions and rebuilding/retooling critical infrastructure to withstand greater stresses. However, for business owners, homeowners, and renters, the most important step they can take is to close the “coverage gaps” that expose them to massive uninsured losses that can delay or prevent recovery. And for regulators and insurers, this creates an excellent opportunity for public/private solutions to meet this growing challenge head-on.

Lloyd’s loosens its tie (rules)

Had a different post planned for today, but this article about Lloyd’s of London, from behind the Financial Times paywall is the perfect segue into the Memorial Day Weekend:

… over the past few months the insurance market has quietly started to relax its strict tie policy. While it has not yet formally repealed the rule, it is no longer enforcing it strictly.

A spokesperson for Lloyd’s said that the new policy was “in keeping with the norms of business dress in the City”.

One underwriter who works at Lloyd’s welcomed the move. “It’s the right thing to do,” he said. “If you had walked around without a tie 10 years ago it would have been the same as wearing a yellow mankini but this is part of general modernisation.”

… could do without the mankini reference, though.

The “After Glow” of Tax Reform Politics Too Good to Pass Up for Anti-Insurance Crowd

By Sean Kevelighan, CEO, ‎Insurance Information Institute

After the Tax Cut and Jobs Act of 2017 passed late last year, the Insurance Information Institute received numerous queries about the impact on property/casualty insurers. Given our mission at I.I.I. is not rooted in direct lobbying advocacy, we consciously refrained from engaging in what was sure to be (and was, in fact) a political battleground in some areas during the legislative process. That said, the industry deserves credit for coming together in many ways to ensure insurance receives fair treatment — a lesson learned from 1986 when the industry was sidelined.

While the anti-insurance crowd (most often misleading themselves as “pro consumer” groups) has been quick to add political rhetoric in the form of baseless and wildly exaggerated claims the industry will receive a “windfall” of income, the I.I.I. will, once again, adhere to facts that are based on actuarial and economic soundness.

Objectively, the I.I.I. sees the overall benefits to tax reform for the insurance industry to be well under 1 cent for every premium dollar.

How do we get that estimate?

Equity analysts at J.P. Morgan estimate tax reform would be about 5 percent of industry earnings, which seems reasonable based on what we know. In 2016 – 2017 industrywide results aren’t out yet – net income was $42.6 billion. Five percent of that would be a bit over $2 billion – more than I have in my pocket, but only about one-third of 1 percent of the $600 billion the industry wrote that year.

Here are a couple of other things to consider about insurers and taxes:

  • Insurance companies pay a wide variety of rates. They pay one rate on underwriting profits, another on dividends from preferred stock, another on bond payments and yet another on municipal bond payments which are almost, but not quite, tax-free. The headline rate fell considerably, but many of the other rates didn’t change at all.
  • Some companies may get a tax increase. Foreign-based groups that have historically ceded a portion of their U.S. business to an offshore affiliate based outside the U.S. are now subject to the Base Erosion and Anti-Abuse Tax – call it BEAT. However, the reduction in the overall tax rate may offset the other changes, depending on each company’s circumstances.

It is important to understand that insurance costs will quickly adjust to the new tax reality. Insurers in the largest lines – personal auto and homeowners – adjust their rates annually – sometimes more frequently. The rate – by law – explicitly reflects every cost an insurer incurs, including taxes. When the tax law changes, insurers build the new rate into their models.

Much like any business in America, insurance will use some of the benefits to invest — in its employees, products and services — so as to improve and grow. Given the industry is the second largest financial services contributor to our economy (2.8% of GDP), employing nearly 3 million Americans, it is critical that insurers make their own decisions.  If not, then where does the line get drawn? Next, the anti-business crowd would (or perhaps already has) call on other industries to make uneconomic pricing decisions.

Update: This blog post has been changed to clarify information regarding the BEAT tax.

Commentary by I.I.I. chief actuary published in law review

Commentary on workers compensation insurance by I.I.I Chief Actuary James Lynch is being published in the upcoming issue of the Rutgers University Law Review.

Lynch’s piece, “Comment to Economic Incentives in Workers’ Compensation: A Holistic International Perspective,” was written in response to an article by Stanford Professor Alison Morantz and others. The two spoke at a 2015 conference about the Grand Bargain in workers compensation.

Speakers discussed whether recent events were jeopardizing the so-called ‘Grand Bargain’ – workers forfeit their right to sue for on-the-job injuries in exchange for predictable benefits from a no-fault system. Professor Morantz’s article compared social safety net programs (including workers comp) worldwide and “discusses several mounting pressures that are jeopardizing the capacity of the U.S. workers’ compensation system to carry out its intended goals.”

In his response, Lynch noted that much of the current research in U.S. workers compensation “involves finding ways to reduce incentives that drive costs higher with no discernable benefit to the worker.” One example cited: When New York doctors book surgery for comp patients in New Jersey, they charge an average of $4,954, or 266 percent more than if the same surgery with the same surgeon took place in New York. (The I.I.I Blog featured this research two years ago.

Neither Morantz’s nor Lynch’s article was online as of February 16, but both should become available at the law review website soon.

WCRI conference: The annual deep dive into workers comp

If I were to pick out the hottest topics in workers compensation these days, these three would be near the top:
• Opioids.
• Marijuana.
• How technology will affect the industry.
All three will be prominently featured at the Workers Compensation Research Institute’s Annual Issues & Research Conference March 22 and 23 at the Westin Copley Place, Boston.
The research organization, known by its acronym, WCRI, has for more than three decades conducted deep, objective research into what makes the workers comp system tick. Its conference annually hosts about the deepest dive you can find into the intricacies of the largest commercial line of business. I attend every year, and am happy for the privilege.
This year attendees will hear the latest on how opioids affect the ability of workers to return to their job; how Americans deal with prescription drugs in the workplace; and how a major employer – United Airlines – addresses the issue.
And a University of Georgia researcher, Dr. David Bradford, will discuss his studies on the effect of medical marijuana programs on prescription drug spending.
And opening speaker Erica L. Groshen, former head of the U.S. Bureau of Labor Statistics, will look at how artificial intelligence, robots, driverless cars and such will affect the labor force. From the conference website:

She will argue that much of the hype about the future of work is either far too optimistic or to pessimistic. In addition, she will talk about how the official statistics are more important than ever.
They are the information infrastructure that we all need to see through the haze, so that we make good decisions for our companies, our communities and ourselves.

Conference details are here.

Flood insurance after the government shutdown

At 12:01 Saturday, the U.S. government shut down. Here is what that means for the National Flood Insurance Program, as taken from the NFIP’s web site (last updated Wednesday):

FEMA and Congress have never failed to honor the flood insurance contracts in place with NFIP policyholders. In the unlikely event the NFIP’s authorization lapses, FEMA would still have authority to ensure the payment of valid claims with available funds.

However, FEMA would stop selling and renewing policies for millions of properties in communities across the nation. Nationwide, the National Association of Realtors estimates that a lapse might impact approximately 40,000 home sale closings per month.

 

Actuarial science vs. Neuroscience

I get interviewed by a lot of newspapers, magazines and TV stations, but maybe the most interesting one came last year when I spoke to David Scharfenberg of the Boston Globe about neuroscience and actuarial science.

David’s article looks at the criminal justice system and suggests that people under the age of 25 should be classified and punished differently from people older than that. Their young-ish minds aren’t fully developed.

He points to scientific studies and programs, but he wanted to talk to me about insurance. The I.I.I., of course, has no opinion on criminal justice, but famously, auto insurers charge younger drivers more than older drivers, and the rates generally change about age 25.

From the article, here is what I said:

The insurance industry’s decades-old imposition of higher rates on young adult drivers is . . . rooted in hard numbers.

The data show a significant decline in the number of accidents for drivers over the age of 25, because they’re more experienced and more mature. And property casualty insurers — more than 2,000 in all — have to retest that proposition year after year, in order to justify the elevated rates to state regulators.

“It’s like, ‘OK, here we are in Arkansas — well, looks like we’re going to be drawing the line at 25, 26 again,’ ” Lynch says. “Now, we’re looking at Massachusetts — oh, there we are again.” The industry, he says, has known for decades what the white coats in the lab are now confirming.

“We were there,” he says, “long before the neuroscientists.”

Postscript: This article was actually published in November, but I only heard about it in mid-January when a prisoner at a correctional center in Massachusetts asked for more information. I sent him this link from our Facts and Statistics page.

NFIP taps reinsurance market again in 2018

The National Flood Insurance Program returned to the private reinsurance market for 2018, paying $235 million for $1.458 billion coverage from a single flood event.

The coverage limit is 40 percent more than what the NFIP purchased last year ($1.042 billion), and the premium is 56 percent higher than the $150 million NFIP paid last year. The 2017 treaty was the first significant foray for the government insurer into the private sector, and the government recovered the entire $1.042 billion from Hurricane Harvey’s floods.

The structure is a bit different this year. Last year reinsurers covered 26 percent of $4 billion in losses after NFIP retained $4 billion losses. Reinsurers will pay 18.6 percent of the first $2 billion of losses excess of $4 billion and will pay 54.3 percent of the $2 billion excess $6 billion.

Both last year and this, the NFIP gets no protection for the first $4 billion of any flood event – the $4 billion acts similar to a deductible on an insurance policy. After that, the worse the flood gets, the more NFIP recovers, and this year the maximum is $1.458 billion.

Examples:

  • A $5 billion flood would result in a recovery of $186 million – $5 billion minus $4 billion is $1 billion and 18.6 percent of that is $186 million.
  • A $7.5 billion flood would result in a recovery of $1.1865 billion:
    • For the first $6 billion, the recovery would be $372 million, being 18.6 percent of $2 billion (after the $4 billion “deductible.”)
    • For the $1.5 billion in losses above $6 billion, the recovery would be $814.5 million, being 54.3 percent of $1.5 billion.

NFIP explains the structure in a press release, with program Director Roy E. Wright adding his thoughts in a blog post.

Harvey was the third worst flood in NFIP’s 50 years, behind Hurricane Katrina in 2005 ($16.3 billion) and Superstorm Sandy in 2012 ($8.7 billion). Harvey has generated 91,514 claims through January 5, according to messaging from FEMA, and 90.9 percent of them have closed. The average payment has been $108,825.

The I.I.I. has more information on floods and flood insurance here.