The trusted source of unique, data-driven insights on insurance to inform and empower consumers. Insurance Information Institute

I.I.I. satellite media tour: Peak hurricane season is upon us

 

By Lynne McChristian, I.I.I. Media Spokesperson and Non-resident Scholar 

 

 

The Insurance Information Institute, along with Colorado State University’s atmospheric research scientist Dr. Phil Klotzbach, will be conducting a satellite media tour on Tuesday, August 13 in Miami to talk about what may lie ahead for the remainder of the hurricane season. Nearly 20 media outlets have signed up to participate. We will be talking with news organizations throughout the U.S. about the rising frequency and severity of natural disasters and about what you can do to understand your risks and be ready to face them – both physically and financially.

Hurricane Season ebbs and flows. And, our collective attention spans mostly ebb. Granted, it’s hard to remain engaged with an event that lasts half a year. From June 1 to November 30, you hear the “Get Ready” message from multiple sources, and the human tendency to wait until a storm threat is on the doorstep makes preparedness more frenetic than necessary. Being hurricane ready is a smart goal, and now is an excellent time to put hurricane preparedness on the front burner. Peak hurricane season has arrived.

Mid-August through the end of October historically is when most hurricanes form. Remember HIM? Hurricanes Harvey, Irma and Maria made landfall between mid-August and the end of September in 2017. Two Category 4 storms and one Category 5 occurred in a matter of weeks in a season predicted to be slightly above average. Last year’s Hurricane Michael hit in late October. While this year’s latest prediction is for a “near-average season,” hurricane researchers will tell you that it only takes one hurricane making landfall near you to break the law of averages.

Frankly, natural disasters do not ebb and flow; there is no rhythmic pattern to them. Forecasting science is constantly improving, and that means we have a better idea of where hurricanes may be heading. But estimating intensity remains a challenge. You can’t “go with the flow.” Rather, be ready for whatever the wind brings.

The following stations will be broadcasting live interviews (times are Eastern Standard):

8:10 AM: Norfolk, VA, Radio 47  WXGM-FM
8:20 AM: Minneapolis, MN, Radio 15  KWLM-AM
8:40 AM: Raleigh, NC, TV 25 WRAZ FOX
9:10 AM: Albuquerque, NM, Radio 47 KDAZ-FM
9:15 AM: Myrtle Beach, NC, TV 95 WPBF ABC
9:45 AM: Dallas, TX, Radio 5 KKVI-AM/FM
10:25 AM: Norfolk, VA, TV 44 WTKR CBS
12:00 PM: San Diego, CA, Radio 29 KOGO-AM
12:10 PM: Los Angeles, CA, Radio 2 KMET- AM
12:40 PM: St. Louis, MO, TV 21 KTVI FO

These stations are taping segments and should air them over the next few days:

Austin, TX,  TV 40 KEYE CBS
Birmingham, AL,  TV WBRC FOX
Charleston, SC, TV 94  WCIV ABC
Chattanooga, TN, TV 83  WRCB NBC
Chicago, IL, Radio 3  WSRB-FM
Columbus, OH, Radio 34  WSNY-AM
Fort Myers, FL, TV 55  WFTX FOX
Jacksonville, FL, WTLV NBC
Mobile-Pensacola, AL, TV 58  WPMI NBC
Myrtle Beach, SC, TV 95 WPDE ABC
Roanoke, VA, TV 67  WFXR FOX
Roanoke, VA, TV 68  WSET ABC
Savannah, GA, TV 93  WSAV NBC
San Antonio, TX, TV 31  WOAI NBC
Seattle, WA, Radio 12  KORE-FM
Tampa, FL, TV 11 WTSP CBS

The roster of stations is subject to change. If you’re in one of the cities listed, please tune in!

 

NICB: Top 5 states for hail claims

The National Insurance Crime Bureau (NICB) recently released a three-year analysis of insurance claims associated with hail storms in the United States.  According to the NICB review of claims data from ISO ClaimSearch®, there were a total of 2.9 million hail loss claims in the United States from 2016 through 2018.

The top five states for hail loss claims were:

  • Texas (811,381)
  • Colorado (395,025)
  • Nebraska (163,336)
  • Missouri (153,403)
  • Kansas (146,206).

The top five cities for hail loss claims during that period were:

  • San Antonio, Texas (75,187)
  • Colorado Springs, Colorado (67,920)
  • Omaha, Nebraska (52,803)
  • Denver, Colorado (48,357)
  • Plano, Texas (42,659).

Over the three years covered by the report, May had the highest monthly average for hail loss claims with 203,296. June was next with 178,881. April (164,232), March (153,716) and July (96,947) round out the top five.

Of the five policy types providing hail loss coverage, Personal Property-Homeowners was the most affected with 1,657,663 claims or 57 percent of the three-year total. It was followed by Personal Auto with 898, 500 claims and Personal Property – Farm with 149,215 claims.

“Hail damage fluctuates year-to-year, but what seems to be consistent is the number of unscrupulous contractors ready to swoop in promising a quick fix, which is why NICB encourages policyholders to use caution when selecting a contractor or other workers to help repair your property or replace your windshield following a storm,” said Brooke Kelley, NICB vice president of communications. “Always check first with your insurance company or agent before signing any documents presented by a contractor whom you did not request to appear. It’s why we say, “If you didn’t request it, reject it.”

The following tips are also helpful:

  • Get more than one estimate
  • Don’t be pushed into signing a contract right away
  • Get everything in writing
  • Require references and check them out
  • Ask to see the contractor’s driver’s license and write down the number and the license plate on his or her vehicle

The I.I.I. has facts & statistics about hail here and here.

 

Researchers continue to predict a near-normal 2019 Atlantic hurricane season

On Monday August 5th Colorado State University (CSU) hurricane researchers issued a news release in which they continue to predict a near-average 2019 Atlantic hurricane season.

The CSU team is predicting a total of 12 additional named storms to form after August 1st . Of those, six are expected to become hurricanes and two to reach major hurricane strength (Saffir/Simpson category 3-4-5) with sustained winds of 111 miles per hour or greater. These forecast numbers do not include Subtropical Storm Andrea and Hurricane Barry which formed prior to August 1.

The scientists, led by I.I.I. non-resident scholar, Dr. Phil Klotzbach, cite both near-average sea surface temperatures in the tropical Atlantic and a weakening El Niño event in the tropical Pacific as the primary reasons for the near-average prediction.

El Niño tends to increase upper-level westerly winds across the Caribbean into the tropical Atlantic, tearing apart hurricanes as they try to form. While El Niño has weakened over the past several months, they anticipate that lingering warming in the central tropical Pacific should be a slight inhibiting factor for the remainder of the hurricane season.

The tropical Atlantic currently has near average sea surface temperatures. A warmer tropical Atlantic provides more fuel for developing tropical cyclones. Increased tropical Atlantic warmth is also associated with moister air and a more unstable atmosphere, both of which foster organized thunderstorm activity necessary for hurricane development. Vertical wind shear was slightly stronger than normal across the Caribbean in July. This tends to be associated with quieter Atlantic hurricane seasons.

The team based this forecast on 40 years of historical data that include Atlantic sea surface temperatures, sea level pressures, vertical wind shear levels (the change in wind direction and speed with height in the atmosphere), El Niño (warming of waters in the central and eastern tropical Pacific), and other factors.

So far, the 2019 hurricane season is exhibiting characteristics similar to 1990, 1992, 2012 and 2014. “1992 and 2014 had below-average Atlantic hurricane activity, 1990 had near-average hurricane activity, and 2012 had above-average Atlantic hurricane activity.” said Dr. Klotzbach.

The full forecast report is available here.

Most popular blog posts: July 2019

It’s always interesting to know which of our blog posts get the most views. Here are the top five most viewed in July:

  1. Insurance ratings variables: a closer look
  2. How many homes are insured how many are uninsured?
  3. A letter to college graduates from I.I.I. CEO Sean Kevelighan
  4. How insurance supports the American farmer
  5. I.I.I./ICM presents recruitment and retention: Best practices and paths not taken

 

 

The Future of Social Security?

By Dr. Steven Weisbart, Chief Economist, Insurance Information Institute

Beginning in 2020, the Social Security fund for retirees will be paying out more than it is taking in. This means that if there are no significant changes, in about 2034 the fund will exhaust the surplus it had built up since 1983. In that case, income to Social Security (from FICA taxes) will only be able to fund about 75 percent of benefits payable. It is for this reason that surveys show that many people under age 50 believe that Social Security won’t be available to provide retirement income for them.

Since Social Security income will be an important part of virtually everyone’s retirement, and since 2035 isn’t very far off (in financial planning terms), we should all be mindful of what might happen, and what we can do now to cope with adverse scenarios.

The government currently has no plan for what to do when the money runs short. One possibility is that everyone’s check in 2035 will be for 75 percent of what it was in 2034. Another possibility is that all those who received checks in 2034 will get the same amount in 2035 and new recipients will have benefits trimmed to fit the remaining funds. A third possibility is that those who are entitled to the highest dollar benefits will get nothing (on the presumption that they had high incomes and so likely have other sources of retirement income) so that those with smaller benefits can be paid their whole entitlement. And other possibilities exist, too.

It’s also possible that Congress will act to change the program so that none of these possibilities take place. Indeed, earlier this year H.R. 860 (The Social Security 2100 Act) was introduced in the U. S. House of Representatives to do just that. The House Ways & Means Committee held a hearing on this bill on July 25, 2019. As of July 30, the bill had 211 co-sponsors—nearly enough for the full House to pass the bill and send it on to the Senate.

There are essentially seven major provisions in H.R. 860. Two of them raise payroll taxes to help fund Social Security benefits. Oddly, other provisions raise Social Security benefits. The two that raise payroll taxes are:

  • Payroll subject to taxation. Currently, Social Security payroll tax (on employee and employer) currently stops at $132,900 (indexed by increases in the average wage). H.R. 860 would create a new payroll tax beginning at $400,000 without cap. The $400,000 would be frozen (not indexed), so that over time, an increasing number of people would be affected by it.
  • Payroll tax rate increase. Currently the payroll tax is 6.2 percent on employer and employee. H.R. 860 would raise it by 0.05 percentage points per year over 24 years (beginning in 2020) up to 7.4 percent (in 2043) each on employer and employee. Note that this higher rate would apply to payroll income up to $132,900 (indexed) and payroll income of $400,000 and over (not indexed). Note that if average wages grow at 2 percent per year, the $132,900 in 2019 would become $213,800 in 2043 and keep climbing after that.

The provisions that raise Social Security benefits are mostly focused on low- and moderate-income earners:

  • There would be a small increase in the formula for the lowest “tier” for computing benefits. This would affect everyone receiving benefits. The percent effect on checks would depend on the base amount but because this change affects only the lowest tier, it would have the greatest effect on those whose average career wage was low. One actuary estimated the dollar increase to be $28.
  • Cost of living adjustment (COLA) change. Currently, the COLA for Social Security is the CPI-W (the cost of living for wage earners). Since 1982 the Bureau of Labor Statistics has been computing a cost-of-living index for elderly consumers (62 and over)—designated the CPI-E—which H.R. 860 would substitute for the CPI-W in the Social Security COLA formula. Because the CPI-E weighs spending on medical care and housing more heavily than does the CPI-W, and because prices in these categories have been rising faster than other categories, it is estimated that if past trends continue, this change could increase the COLA by 0.2 percent per year.
  • Alternative minimum benefits. For individuals who worked for more than 10 years, the bill creates an alternative minimum benefit. A qualifying beneficiary would receive that alternative minimum if it is higher than the standard calculated benefit amount.
  • Income taxation of Social Security benefits. The thresholds for income taxation of Social Security income currently are expressed in frozen dollar amounts but H.R. 860 would double these amounts. This would lower the income to the Social Security reserve funds but would make Social Security income-tax-free for more people.
  • Earnings-related benefits. New (but tiny) additional benefits for retirees whose average earnings were $400,000 and above to recognize the new payroll taxes they’ll pay while working.

Dodge Charger tops HLDI’s list of most likely to be stolen vehicles

The Dodge Charger HEMI and the Dodge Challenger SRT Hellcat are at the top of the Highway Loss Data Institute’s (HLDI) most-stolen vehicles list this year. Both cars have theft rates that are more than five times the average for 2016-18 models, with the same as the Infiniti Q50, a midsize luxury sedan.

HLDI released its most likely to be stolen list for 2016-2018 models today, and almost all 20 models with the highest theft rates either have big engines or are luxury vehicles or pickups.

At the top of the least stolen list is the two-wheel-drive BMW 3 series, a midsize luxury sedan. It had just one claim for whole-vehicle theft in 104,901 insured vehicle years (an insured vehicle year is one vehicle insured for one year).

The Tesla Model S and Model X are also on the least-stolen list. A 2018 HLDI report showed that electric vehicles from a variety of manufacturers have lower theft claim rates than comparable vehicles. Their low theft rate may be due to the fact that they are usually parked in garages or close to a house to be near a power supply.

The Cadillac Escalade, which previously dominated HLDI’s rankings of most-stolen vehicles, is notably absent from this year’s list. Part of the reason is that there are more large luxury SUVs for thieves to choose from but also because Cadillac added enhanced security features beginning with the 2015 model year.

“The models most likely to be stolen tend to be powerful, pricey or pickups, but vehicle theft is also a crime of opportunity,” says HLDI Senior Vice President Matt Moore. “Better security features on all vehicles would be the best way to address the problem.”

 

How underwriters can prepare for child sexual abuse claims

Seventeen states and Washington, D.C. have laws taking effect in 2019 that either abolish or extend statutes of limitations for victims of child sexual abuse to sue or seek criminal charges against their abusers. A recent A.M. Best report compares child sexual-abuse claims to asbestos liability because the claims can affect decades-old insurance policies and the settlement amounts can be hard to predict.

In a recent blog post, Carey Quigley, a Gen Re treaty account underwriter, discusses what the new laws mean for underwriters that handle commercial “child custodial care” risks. These risks encompass schools, churches, sports, camps, day care and any other organized activities involving minors.

Quigley notes that unless their policies were written on a claims-made basis, the liability of these organizations for the past conduct of employees and volunteers does not typically affect their exposure under current insurance policies. Nevertheless, he recommends that underwriters take the following three steps in reviewing guidelines and policy forms:

Build a hazard scale: The degree of risk increases with the length of the activity, so a boarding school would be on the far end of the spectrum. Since parents are now more involved with their children’s activities, local groups and gatherings would present a lower risk.

Review insurance forms: Most general commercial writers may have a local dance school or a small church in their portfolio. For these types of policyholders insurers have developed Sexual Abuse and Molestation (SAM) endorsements offering critical but not unlimited protection.

Quigley recommends that insurers include language in their SAM endorsement to: Move all coverage into the policy when the abuse first began; treat all abuse by a single perpetrator as a single claim; treat all related or interrelated abuse as a single claim, without further qualification; and provide coverage on a claims-made basis.

Decide exclusions and check wording: When writing exclusions it’s important to determine whether they will extend to all types of physical abuse, or only sexual abuse. Often these terms are defined to prevent overlap with the GL policy and stacked limits from the endorsement and base policy. If a lawsuit alleges sexual abuse with false imprisonment or battery for instance, the insurer probably intends that all such allegations trigger only the SAM endorsement.

In conclusion Quigley says that underwriters should monitor court decisions to learn how policy language is interpreted by courts and check forms filed by other insurers to see how they address stacking issues.

Insurance Rating Variables: A Closer Look

Figuring out what the cost of an insurance premium should be is quite complicated — so complicated that insurance companies employ entire actuarial departments to do just that. Rating variables are an indispensable tool for setting the cost of insurance.

In a new paper, Insurance Rating Variables: What they are and why they matter, the Insurance Information Institute (I.I.I.) and the Casualty Actuarial Society (CAS) explain why actuaries apply variables when setting rates – for example using a driver’s age and gender, accident history and vehicle model year to calculate the premium on an auto policy.

Rating variables are basically the characteristics of individual policyholders that can help approximate the cost of their risks. Insurance companies have been using rating variables to help set rates (and thereby price their policies) for decades.

However, the use of some rating variables has recently generated discussion within the United States. Some states have even passed legislation controlling the use of certain variables, such as gender.  “Variables are designed to make insurance affordable and available to everyone,” said Ken Williams, FCAS, CAS staff actuary. “When a variable is removed from rate setting, the consumer stands to lose the most because lower-risk individuals will end up subsidizing higher-risk individuals; or, insurance companies may choose to accept fewer applications from consumers who might cause them to lose money.”

The paper explains that when regulators restrict the use of a particular variable, actuaries may replace it with another variable as a “proxy,” which might not help them price policies as well.

“Imagine that male drivers have higher accident costs and are more likely to drive pickup trucks,” Williams explained. “If gender is restricted, the proxy for gender could become pickup trucks. In this scenario, rates for pickup trucks may increase while rates for other types of vehicles may decrease.”

It is important to note that all rating variables, including proxies, are regulated in every state. For example, rating variables and proxies cannot directly or indirectly impact groups based on certain characteristics, such as race.

The paper notes that the use of rating variables has resulted in a drastic reduction in the number of consumers seeking coverage in state-supported auto risk pools. Since the increased use of rating variables, the number of consumers in assigned risk pools has decreased almost 90 percent.

James Lynch, FCAS, chief actuary and vice president of research and education at the I.I.I., added, “Rating variables are regulated by state and federal authorities and they meet a variety of important criteria: they are credible, objective, and verifiable. They are an essential tool for setting accurate prices that are lower for low risk customers, higher for high risk customers, yet sufficient to cover an insurer’s costs.”

Here are a few key points from the report:

  • Insurance companies use rating variables to develop premiums that better reflect the risks that consumers face.
    • Rating variables are characteristics of individual consumers that can help approximate the cost of their risks, like vehicle model year in auto insurance or the age of a building in homeowners insurance.
    • Rating variables help ensure that less risky consumers pay lower rates than consumers who are at greater risk.
  • Rating variables are studied rigorously by actuaries for their effectiveness and impact on the societal goal of keeping insurance available and affordable. They are also closely regulated.
    • Actuaries are very careful to make sure that each variable is effective and is subject to a wide range of criteria, including being credible, objective, verifiable, and inexpensive to administer.
    • Actuaries also make sure variables are legal. Variables are subject to regulation, and state and federal laws prohibit using rating variables that either directly or indirectly impact groups based on characteristics such as race, nationality, religion, or income.
    • Almost every state in the U.S. has the regulatory authority to reject a rating variable that it feels does not meet state requirements.
  • Widespread use of rating variables has given consumers more choice and more fairness in the insurance marketplace.
    • The ability to set accurate prices is a cornerstone to setting actuarially sound premiums that are lower for low risk customers, higher for high risk customers, yet overall sufficient to cover all the insurer’s costs.
    • Better and more available data for use in rating variables means increased ability to determine a person’s exact risk profile.
  • Restriction of rating variables that do their job well can lead to potential unintended consequences for the consumer.
    • Restrictions on some variables may result in lower-risk consumers effectively subsidizing higher-risk consumers.
    • Restricting rating variables affects consumers much more than it impacts insurance companies.

Click here to read the paper.

 

From the I.I.I. Daily: Our most popular content, July 11 to July 18

Here are the 5 most clicked on articles from this week’s I.I.I. Daily newsletter.

To subscribe to the I.I.I. Daily email daily@iii.org.

 

 

Latest research and analysis