The Casualty Actuarial Society (CAS) has developed a series of papers examining the issue of race and insurance pricing and seeking to contribute constructively to the policy discussion around it.
“Insurance pricing is a high-wire act,” CAS says. Actuaries have to quantify and differentiate among a massive variety of risk variables while avoiding unfair discrimination. “As regulation and society’s understanding of discrimination evolve, however, it is necessary for us to keep abreast of changes in the manner in which discrimination is defined and adjudicated.”
The CAS research has generated four papers – two published this week, two more to be published on March 31 – that define, quantify, and propose methods for addressing unfair discrimination where it is found to exist.
Confusion around insurance rating is understandable, given the complex predictive models being used today, which can lead to inappropriate comparisons and inaccurate conclusions. Algorithms and machine learning hold great promise for helping to ensure equitable pricing. However, research has shown these tools also can amplify biases that manage to creep into their programming.
Recent Colorado legislation requires insurers to show that their use of external data and complex algorithms don’t discriminate against protected classes, as well as other state and federal efforts to address perceived bias in pricing.
The actuarial discipline and the insurance industry are well positioned to continue helping policymakers and corporate decisionmakers understand and address these inequities.
Local governments in the United States in recent years have begun adopting “Vision Zero” policies, which aim at cutting roadway fatalities to zero. Such policies – which have demonstrated success abroad – have drawn even more interest since the onset of the pandemic, during which traffic fatalities and injuries have surged.
The Vision Zero Network is a nonprofit focused on helping local governments implement the Vision Zero plan. First implemented in Sweden in 1997, that country has seen its traffic fatalities halved, inspiring other governments to adopt similar measures. Vision Zero is also becoming an initiative for the entire European Union.
More than 40 communities across the United States have adopted these policies, including major metropolitan areas like New York City, Los Angeles, and Portland, Ore. In Portland, several data points are helping government officials better understand how to reduce traffic fatalities and injuries, including a high percentage of pedestrian crashes occurring because of long distances between marked crossings. Portland has taken the initiative, building “a system to protect pedestrians includes frequent safe crossings, street lighting, a cultural acceptance of slower speeds and people educated about how to interact safely on the streets.”
Success in Hoboken, NJ
Hoboken, a city of about 54,000 people across the Hudson River from New York City, has experienced zero traffic deaths for three years as of 2021. Instrumental in this has been Mayor Ravi Bhalla’s Vision Zero program. Mayor Bhalla’s 2019 executive order has resulted in the city extending its bike-lane network 38 percent in 2019 and 2020, with its total on-street network of 16.3 miles now nearly half of the city’s 33 miles of streets.
The city also has put in curb extensions at intersections, marked wider crosswalks, and timed traffic signals to give pedestrians a seven-second head start. When it’s warmer, major commercial areas of the city are closed to cars entirely or assigned as “slow streets” with decreased traffic and velocities.
“While we’ve made major progress in the past three years, having no pedestrian fatalities and a reduction in pedestrian injuries, we are striving to create even safer streets in the years ahead,” said Mayor Bhalla. “With the adoption of the Vision Zero Action Plan, we’ll be able to take even more actionable steps to reach our goal of all traffic-related deaths and injuries by 2030, one of the most ambitious Vision Zero goals in the entire country.”
With these steps being implemented nationwide, entire communities are becoming safer. Additionally, insurers could potentially pass the savings produced by lower accident rates onto consumers, as they did earlier in the pandemic.
Now the U.S. federal government has announced its own version of Vision Zero. In late January, federal transportation officials released a plan to reduce the tens of thousands of road deaths that occur every year.
Personal auto insurance premium rates have returned to pre-pandemic levels, but several trends are likely to sustain upward pressure on rates, according to a new Triple-I Issues Brief.
At the start of the pandemic, auto insurers – anticipating fewer accidents amid the economic lockdown – gave back approximately $14 billion to policyholders in the form of cash refunds and account credits. But while miles driven declined and accident frequency initially dropped, frequency and severity quickly started increasing again. Traffic fatalities also increased, after decades of steady declines.
While insurers’ personal auto loss ratios fell briefly and sharply in 2020, they have since climbed steadily to exceed pre-pandemic levels. With more drivers on the road and replacement parts climbing, this loss trend is expected to continue.
Auto premium rates reflect a range of factors that contribute to an insurer’s loss experience. In a world of perfect information, rate changes would correlate perfectly with changes in loss experience. As the chart below shows, until the pandemic these two metrics for the overall industry tracked quite closely. The disruptions of 2020 led to volatility for both, and losses have proved more volatile than pricing.
Barely profitable
To remain viable, insurers have to set premiums at levels appropriate to the risks they cover. Insurers’ underwriting profitability is measured by a “combined ratio”, which is calculated by dividing the sum of claim-related losses and all expenses by earned premium. A combined ratio under 100 percent indicates a profit. A ratio above 100 percent indicates a loss.
As the chart above shows, personal auto insurance has been a barely profitable line for the industry for years. If recent accident and replacement-cost trends persist, upward pressure on premium rates is likely to continue.
As with so many other goods and services, insurance for commercial trucks has become more costly since the pandemic – but a closer look at the numbers shows that this trend pre-dates COVID-19’s economic and supply-chain disruption.
“Despite reductions in insurance coverage, rising deductibles, and improved safety, almost all motor carriers experienced substantial increases in insurance costs from 2018 to 2020,” according to a recent report by the American Transportation Research Institute (ATRI). And, while frequency and severity have been on the rise from 2009 to 2018, the report shows the rate of insurance cost increases during the period far exceeding the crash rate increase.
ATRI’s observations are consistent with findings in a recent study by Triple-I and the Casualty Actuarial Society (CAS) that the phenomenon known as “social inflation” accounted for $20 billion in commercial auto liability claims between 2010 and 2019.
“External factors that go well beyond carrier safety force commercial trucking insurance costs to increase,” says Triple-I Chief Insurance Officer Dale Porfilio. “The higher premiums ultimately tend to be passed along to consumers in the form of higher prices for goods and services.”
ATRI recognizes three key areas of influence on premiums beyond crash history and policy components:
Economic impacts on the insurance industry,
Carrier-specific factors, and
Social inflation.
External economic conditions, including general inflation and rising health-care costs, contribute to increased insurance premium rates.
“Medical advances help save lives, but these treatments directly contribute to higher medical costs,” ATRI points out. “Similarly, technological advances in motor vehicles contribute to increasing costs associated with repairing them; electronics now make up 40 percent of the cost of a new vehicle.”
These higher costs affect premiums through larger claims and losses that have to be incorporated into pricing.
Premium rates also are affected by carrier-specific considerations like operational sectors, cargo values, states or regions of operation, company growth, and commitment to safety culture and technologies.
“Carriers demonstrating consistent year-over-year improvements in safety technology adoption, safe driver hiring and training practices, and crash history can potentially lower their premium costs, despite the current adverse environment,” ATRI said.
“Social inflation” refers to the impact of litigation and government policy trends on insurance claims and, ultimately, costs to policyholders. Social attitudes and behaviors affect insurance payouts through changes in laws and propensity to litigate, and jury awards don’t necessarily reflect logical conclusions or precedents. Jury decisions can be influenced by emotions, state and local laws or procedures, and plaintiff bar tactics. In recent years, practices like third-party litigation funding – investment by hedge funds and other third parties in lawsuits in return for a share in the awards – have played an increasing role in social inflation.
After years of steady declines, traffic fatalities in the United States are on the rise, contributing to increasing auto insurance rates. This comes despite declines in the average number of miles driven due to the pandemic. In 2020, 38,680 deaths occurred on U.S. roads, the most since 2007.
In late January, federal transportation officials released a plan to reduce the tens of thousands of road deaths that occur every year, an issue that has become more significant since the beginning of the coronavirus pandemic.
“We cannot and should not accept these fatalities as simply a part of everyday life in America,” said Transportation Secretary Pete Buttigieg. “No one will accomplish this alone. It will take all levels of government, industries, advocates, engineers and communities across the country working together toward the day when family members no longer have to say good-bye to loved ones because of a traffic crash.”
Pandemic’s impact
Roadway safety in the United States had increased for decades before the pandemic, primarily due to enforcement of seat belt laws and vehicle safety features, such as airbags, improved braking, and stability control. Yet, the first year of the pandemic saw a 7.2 percent rise in U.S. roadway deaths from 2019. Some experts saw this rise in reckless driving as due, in part, to the isolation associated with the pandemic lockdowns.
“You’ve been cooped up, locked down, and have restrictions you chafe at,” said Frank Farley, a professor of psychology at Temple University in Philadelphia.
In the early months of the COVID pandemic, insurers were giving rebates for personal auto policies, spurred by reductions in miles driven and anticipation of fewer accidents. However, it quickly became clear that reduced miles driven didn’t automatically lead to fewer deadly accidents. Instead, reckless driving – and fatalities – increased.
The end of pandemic shutdowns hasn’t helped either, with the U.S. Department of Transportation’s National Highway Traffic Safety Administration (NHTSA) estimating that 31,720 people died in motor vehicle traffic crashes for the first nine months of 2021, rising about 12 percent from the 28,325 fatalities projected for the first nine months of 2020. In Q1 of 2020, traffic fatalities were 1.12 per 100 million miles driven. By the end of Q3 of 2021, this number had spiked to 1.41 per 100 million miles driven.
What can be done?
The federal infrastructure deal promises to spend more on new safety measures, with the goal of eliminating road deaths. With this in mind, the Department of Transportation is implementing a Safe System Approach, under the premise that fatal accidents can be avoided if individuals understand the need for safe driving and accept that crashes can be avoided. The aim is zero traffic deaths.
Indeed, the Safe System Approach, which has been adopted across several countries in Europe, has seen remarkably positive results. Traffic fatalities fell 50 percent In Sweden and the Netherlands between 1994 and 2015.
“There are communities that have gotten to [zero traffic fatalities] already,” added Buttigieg. “And I’m not just talking about Oslo,” which experienced zero pedestrian deaths in 2019, “but a place like Hoboken, N.J., in the U.S. has seen multiple years with zero deaths.”
Auto insurance premium rates are affected by many factors, and accident and fatality trends are a major ones. Reckless driving trends – combined with increasing auto repair costs associated with safety, efficiency, and comfort – can only continue to put upward pressure on rates. Individual behavior and government policies must converge in the direction of improving responsibility and safety for all drivers.
The phenomenon known as “social inflation” accounted for $20 billion in commercial auto liability claims between 2010 and 2019, a new study by Triple-I and the Casualty Actuarial Society (CAS) finds.
Social inflation isn’t a new term. Warren Buffett used it in the 1970s to describe “a broadening definition by society and juries of what is covered by insurance policies.” It has since become common parlance among insurers and risk managers for a range of factors causing losses in certain lines to rise faster than general inflation would predict. These include:
Class-action lawsuits;
Growing awards from sympathetic juries;
Third-party litigation funding, in which investors finance lawsuits against large companies in return for a share in the settlement; and
Rollbacks of tort reforms that were intended to control costs in the wake of the 1980s “liability crisis”.
Hard to measure, important to understand
Reliably quantifying social inflation for rating and reserving purposes is hard because it’s just one of many factors pressuring pricing. The paper, authored by actuaries James Lynch and David Moore, uses “standard actuarial metrics and visualizations to demonstrate how actuarial insights can be presented to an interested lay audience, such as lawmakers, regulators, the news media, and the public.”
This is an important contribution to the public policy discussion because actuaries are well positioned to spot shifts in loss severity.
Separately, Triple-I has published an “Issues Brief” that succinctly describes the drivers of social inflation, as well as its potential impact on insurers, policyholders, and the economy and society.
“More frequent suits and bigger awards can lead to increased insurance costs as rates are adjusted to reflect the changing risk profile – or even to insurers ceasing to write particular forms of coverage,” the brief says. “Higher premiums tend to be passed along to consumers in the form of higher prices and, in extreme cases, can ripple through the entire economy, creating conditions analogous to the 1980s liability crisis.”
In the 1980s, liability claims were pushing the U.S. insurance industry to the brink of collapse. Tort reforms – ranging from capping non-economic damages and limiting contingency fees to specifying statutes of limitations and eliminating “joint and several” liability – were enacted, and losses declined. It has been argued that legislative efforts to roll back these reforms in many states have contributed to social inflation, but the research is not conclusive.
Louisiana, Florida, and Michigan remain the least affordable states for auto insurance, while Iowa remains the most affordable, according to a new study from the Insurance Research Council (IRC).
The report, Auto Insurance Affordability: Countrywide Trends and State Comparisons, looks at auto insurance expenditures as a share of median household income. The IRC affordability index ranges from a low of 1.02 percent in Iowa to a high of 3.09 percent in Louisiana. A higher ratio indicates less affordable insurance in the state.
The index uses median household income data from the U.S. Census Bureau and auto insurance expenditure data published by the National Association of Insurance Commissioners (NAIC). The rankings are based on 2018 data (the most recent available). Since 2018, Michigan has enacted reforms aimed at lowering auto insurance expenditures for Michigan’s drivers.
Some affordability studies estimate insurance costs by gathering quotes for minimum coverage. The NAIC measure, by contrast, provides an estimate of what consumers actually spend per insured vehicle. The index isn’t intended to serve as an absolute threshold for when auto insurance becomes affordable. This would be entirely subjective, as different parties can reasonably disagree about what constitutes affordable insurance. Rather, it’s a tool to compare auto insurance affordability over time and across jurisdictions.
Underserved communities not directly addressed
The index also does not address the important issue of affordability among underserved populations, which would require more granular data than used for this analysis. It is important to note that affordability for traditionally underserved consumers is determined by underlying costs, just as it is for the overall population.
A recent analysis of NAIC data showed that the higher premiums in lower-income ZIP codes were in line with the higher claim costs in those areas. Efforts to improve auto insurance affordability in those areas must address these higher costs.
While state-level data cannot directly address affordability among these populations, collaborative efforts to reduce the following key cost drivers can improve affordability for all consumers:
Accident frequency related to traffic density, road conditions, and other factors that lead to more frequent accidents in some states.
Repair costs, which vary widely by state.
Tendency to file injury claims, which tends to be higher in less affordable states.
Injury claim costs.
Attorney involvement, which is associated with higher claim costs and delays in settlement.
Claim abuse – Insurance fraud is a factor in the high cost of insurance.
In a letter responding to a federal request for information, Triple-I earlier this year said U.S. auto insurers accurately price their policies by using a wide variety of rating factors. All these factors must conform to the laws and regulations of the state in which the auto insurance policies are sold.
“Lower-risk drivers should pay less for auto insurance, and premiums have closely tracked broader U.S. economic trends for decades,” Triple-I told the U.S. Treasury Department’s Federal Insurance Office (FIO) in its letter.
The letter also said the rating factors U.S. auto insurers use to price their policies not only serve their purpose but are constantly retested to ensure their accuracy and reliability.
Lower-risk drivers should pay less for auto insurance, and premiums have closely tracked broader U.S. economic trends for decades, Triple-I told the U.S. Treasury Department’s Federal Insurance Office (FIO) this week.
In a letter responding to a federal Request for Information, Triple-I said U.S. auto insurers accurately price their policies by using a wide variety of rating factors. All these factors must conform to the laws and regulations of the state in which the auto insurance policies are sold.
“There is no credible evidence that insurers charge more than they should, either across the broad market or in specific subsegments, such as neighborhood, race, income, education or occupation,” the Triple-I stated. The letter also said the rating factors U.S. auto insurers use to price their policies not only serve their purpose but are constantly retested to ensure their accuracy and reliability.
“If rating factors do their job well, they make insurance relatively inexpensive for some people and quite expensive for others,” the letter said. “In both cases, the assessment is correct. Drivers who present less risk pay less for coverage.”
The response to FIO’s information request highlighted how the appropriate price for an insurance policy varies greatly from customer to customer and from state to state. Insurance is regulated by state governments.
“Insurance companies and their actuaries have focused on finding factors that make sure every customer pays the appropriate rate,” the Triple-I said. Rates are based on historical loss experience for similar risks. Premiums constitute the price customers pay for insurance coverage.
Critics of U.S. auto insurer pricing practices have expressed concerns that certain rating factors, such as credit-based insurance scores and the geographic location of the customer’s residence, discriminate against lower-income drivers and minority groups. Triple-I explained that eliminating any rating factor – for whatever reason – forces those with less risk to overpay for auto insurance and allows those with greater risk to pay less than they should for auto insurance.
Interventions can backfire
“Eliminating factors does not affect the truth that they reveal, and if factors reveal that costs need to be high for a customer, banning them does nothing to change the underlying costs that are the reason the rate is high,” the Triple-I stated.
Regulators occasionally intervene in the rating process to make insurance less expensive for certain groups, citing the need to make insurance “affordable.”
“These interventions, however well-intentioned, can backfire in a spectacular way,” the Triple-I letter says, “raising the overall costs and severely reducing availability, as well as impeding innovations that could address the issue.”
Real problems need real solutions
Real solutions exist to make insurance more affordable, Triple-I says: “These solutions come not from tinkering with how insurers set prices but by addressing the costs that insurance covers.”
Improving the transportation environment and addressing societal issues that often force minorities and low- and moderate-income individuals to live and drive in circumstances where auto insurance costs the most are among the solutions suggested.
Extensive Triple-I research shows that rising claims costs have been the primary factor generating increased auto insurance rates.
It has been said the best way to eat an elephant is “one bite at a time.” Social inflation is an elephantine topic, so we’re launching a series of blog posts dedicated to each piece of it in turn, starting with litigation funding.
“Social inflation” refers to rising litigation costs and their impact on insurers’ claim payouts, loss ratios and, ultimately, how much policyholders pay for coverage. It’s an important issue to understand because – while the tactics associated with it typically affect businesses perceived as having “deep pockets” – social inflation has implications for individuals and for businesses of all sizes.
The insurance lines most affected are commercial auto, professional liability, product liability, and directors and officers liability. There also is evidence that private-passenger car insurance is beginning to be affected. As increased litigation costs drive up premiums, those increases tend to be passed along to consumers and can stifle investment in innovation that could create jobs and otherwise benefit the economy.
Much of what is discussed and published on the topic has been more anecdotal than data based. Reliably quantifying social inflation for rating and reserving purposes is hard because it’s just one of many factors pressuring pricing. We’ve found that the most meaningful way to think about social inflation and its components is to compare their impact on claims losses over time with growth in inflation measures like the Consumer Price Index (CPI).
Litigation Funding
It’s been said that the best way to eat an elephant is “one bite at a time.” Because of the diversity and complexity of social inflation’s causes and effects, we’re launching a series of blog posts dedicated to each one in turn. The first set of posts will look closely at litigation funding: the practice of third parties financing lawsuits in exchange for a share of any funds the plaintiffs might receive.
Litigation funding was once widely prohibited, but as bans have been eroded in recent decades, the practice has grown, spread, and become a contributor to social inflation.
Litigation funding seemed a good place to begin this series because it’s a distinct legal strategy with a clear history that doesn’t involve a lot of the sociological subtleties inherent in other aspects of social inflation. We’ll look the emergence of the practice, how it came to the United States from abroad, and track its evolution with that of social inflation. We’ll also discuss the current state of litigation finance, along with ethical concerns that have been raised around it within the legal community.
This series will be led by IRC Vice President David Corum with support from our partners at The Institutes and input from our members, as well as experts beyond the insurance industry. As befits any discussion of a complex topic, we look forward to your reactions and insights.
Connecticut this week became the latest state to legalize recreational use of marijuana, and more are expected to follow.
The increased marijuana use that accompanies legalization has raised concerns about road safety.
Researchers at Insurance Institute for Highway Safety (IIHS) and the Highway Loss Data Institute (HLDI) since 2014 have been examining how legalization has affected crash rates and insurance claims, and evidence is emerging that crash rates go up when states legalize recreational use and retail sales of marijuana.
The most recent of these studies, released on June 17 by the IIHS, shows that injury and fatal crash rates in California, Colorado, Nevada, Oregon, and Washington jumped in the months following relaxation of marijuana laws in each state. The five states experienced a 6 percent increase in injury crash rates and a 4 percent increase in fatal crash rates, compared with other Western states where recreational marijuana use was illegal during the study period.
Only the increase in injury crash rates was statistically significant.
“Our latest research makes it clear that legalizing marijuana for recreational use does increase overall crash rates,” says IIHS-HLDI President David Harkey. “That’s obviously something policymakers and safety professionals will need to address as more states move to liberalize their laws — even if the way marijuana affects crash risk for individual drivers remains uncertain.”
Insurance records show a similar increase in claims under collision coverage, which pays for damage to an at-fault, insured driver’s own vehicle, according to HLDI’s latest analysis. The legalization of retail sales in Colorado, Nevada, Oregon, and Washington was associated with a 4 percent increase in collision claim frequency compared with the other Western states from 2012 to 2019. That’s down slightly from the 6 percent increase HLDI identified in a previous study, which covered 2012 to 2018.
While the evidence that crash rates have increased in states that legalized marijuana is mounting, it appears that further study is needed to determine whether marijuana use alone is responsible. Preliminary data suggests people who use alcohol and marijuana together are accountable for most of the crashes.
Another factor may be that marijuana users in counties that do not allow retail sales are driving to counties that do. The increased travel could lead to more crashes, even if their crash risk per mile traveled is no higher than that of other drivers.
