The National Highway Traffic Safety Administration (NHTSA) recently found that 2015 witnessed a 7.2% increase in road accident fatalities, the highest increase since 1965. This is a great concern for insurers, as the report suggests that private insurers pay approximately 50% of all motor vehicle crash costs. As fatalities on the road increase, it is critical for automobile insurers to find ways to manage risk and consider factors leading to fatalities when pricing insurance products.

Currently, auto insurers base rates on what they know about the driver, the driver’s residential area and the vehicle. Other factors considered in insurance pricing and risk management include gender, age, employment, demographics (primarily rural vs. urban and occasionally economic factors), vehicle make, model and age, and driving history. The increase in fatalities points towards the fact that the current insurance pricing is not determinant enough to stop the increase in road fatalities.

From a carrier perspective, this is bad news. Both the amount of fatal accidents and the severity of claims have trended higher, adversely affecting auto insurers. Anticipating upsurges in claim frequency and severity can help in adjusting rates to absorb them. The substantial rise in fatal crashes makes it clear that additional factors must be considered. Data from the NHTSA study can provide answers on the possible drivers of on-road fatalities and link them with insurance premiums. In this whitepaper we examine the NHTSA data to find out:

  • The major drivers of fatalities by testing various assumption and hypothesis assumptions in light of the NHTSA data

  • The factors that auto insurers consider or should be considering in pricing insurance products

  • Factors insurers should consider to mitigate risk and improve underwriting efficiency

Some Numbers Based on NHTSA Data

Data Processing and Insights

EXL used descriptive analytics on the NHTSA data from between 2005 to 2015 to find patterns. There are three types of NHTSA data available - accident data, person data and vehicle data. These are rolled up in the form of the National Highway Accident Master File, which contains information about each highway crash that occurred including location, roadway geometry and features, driving conditions, driver age and gender, driver responsibility for the crash and vehicle information.

Accident Data

  • Contains total number of fatal crashes that occurred in a particular year, as well as details including the number of person involved, weather conditions, location of crash and time of crash. It captures the impact of weather on driver abilities through visibility impairments, precipitation, high winds, and temperature extremes to affect driver capabilities, vehicle performance pavement friction and roadway infrastructure.

Person Data

  • Total number of persons involved in the accidents including people in and outside the motor vehicles. This includes age, gender, seat position, drug abuse, alcohol impairment and other factors. These types of variables can help in identifying and prioritizing variables related to increased accident risk such as age, drug and drinking related detriments in functional capability.

Vehicle Data

  • Contains total number of vehicles involved in the crashes. It also includes the type of vehicle, make and model, license status of the driver, road conditions, speed limit and other factors. This helps in tracking the accident rate of different vehicles and decisions regarding vehicle recalls.

  • The Accident Master Data was prepared by rolling up the person and vehicle data to the accident level and merging all the three files. We used the above mentioned data files, restricting the time period to 6 years (2010-2015) and effectively using 5 years data (2010-2014) to explain the increase in fatalities in 2015

Insight 1: A booming economy doesn’t lead to an increase in insurance profits

On the surface, as the economy is booming, the employment rate and number of cars on the road have gone up. Owing to this, more people travel to work daily, accelerating the number of miles driven in 2015. The number of miles driven for the 12 months leading to September 2015 increased 3.4%, which was the fastest growth in 15 years. This should ideally lead to an upsurge in insurance company profits. Yet, the results of our analysis supported the fact that insurers sometimes miscalculate premium pricing, resulting in a lack of profits. The ineffectiveness of road safety policies is one of the major reasons the insurance industry loses revenue.

We matched our analysis with the results of various insurance companies to see if the booming economy with more cars driving more miles is really working to their advantage. The answer: NO, it isn’t. As a testimony to the fact that increased number of miles driven doesn’t necessarily result in increased profits for insurance companies, one insurer reported in August 2016 that its earnings were down 38% from the year before due to the increased frequency and severity of auto accidents.

Insight 2: Higher purchasing power and more cars on the road may be a leading cause of increase in fatalities

One popular notion among actuaries is that higher purchasing power leads to an increase in traffic accidents. However, this happens only when the inflation rate has not caught up with the corresponding increase in income. The NTSHA data reveals that the household income in the

U.S. went up by 14.3%, from $49,445 in 2010 to $56,516 in 2015. Owing to higher disposable income, this period was also characterized by low inflation.

As gasoline prices have declined, automobile sales have gone up. The U.S. has noticed an increase in the total miles driven over the past few years, rising from 2,966 billion miles in 2010 to 3,147 billion miles in 2015, an increase of 6.1%. During the same period, the fatality count has increased from 32,999 in 2010 to 35,092 in 2015, an increase of 6.3%.

Insight 3: Distracted driving is a major contributor to the rise in fatalities

Distracted driving is any activity that could divert a person’s attention away from driving. All distractions endanger driver, passenger, and bystander safety. Laws exempt the use of mobile phone when the vehicle is off the road, but the unfortunate reality is that very few people pull aside for calls and texts. The NSC estimates that texting while driving raises the likelihood of a crash by eight times, and that crashes involving texting or talking on a cellphone (hands-free or handheld) account for 27% of all accidents. While it is difficult to define the exact number of fatalities caused by using cell phones, it has been observed that a significant proportion of pedestrian fatalities are due to cell phone use by pedestrians and drivers. These observations in turn give rise to a very important question: What role can insurers play in deterring distracted driving?

Insight 4: State highways and roads connecting rural areas experience higher levels of fatalities

The rise of fatalities on state highways can be attributed to the high speed limits set by states on such highways. Speed limits have been on the rise since 1995, as states struggle to support the flow of increased vehicles on state highways.

In our study period (2010-2015), there was a downward trend in the proportion of rural fatalities, with the number declining from 54% in 2010 to 47% in 2015. But still this is a huge population considering that less than 20% of people in the United States live in rural areas. The majority of crashes classified as rural occur on state highways, while majority of urban crashes occur on municipal streets. Despite the fact that the proportion of urban accidents has gradually increased over time, drivers continue to face a significant threat on roads connecting rural areas.

According to highway safety officials and activists, this can be attributed to the risky behavior of people on rural roads and a lack of safety precautions, such as driving fast without using seat belts or driving under the influence of alcohol. Also, rural roads receive much less maintenance than urban roads and may not receive timely emergency medical care, further contributing to their hazardous conditions.

Insight 5: Urban residents are charged higher premiums in spite of rural areas witnessing more fatal accidents

Busy cities have busier roads, which increases the likelihood of accidents, theft, and other incidents that may cause higher insurance claims. Population density has a greater impact than the risk of fatalities in determining insurance rates. Insurance rates are lower in rural areas, even though there is a higher incidence of fatal automobile accidents in rural areas from regularly travelling longer distances at higher speeds while on less safe roadways. While serious and costly, fatal accidents still make up a small portion and cost of total accident claims. Being located in a more densely populated area still has great weight in determining overall insurance rates. This is the reason why urban city drivers face significantly higher premiums because of high population density, leading to a higher probability of having an accident.

Because of the population density factor, rates differ between metropolitan areas and suburbs. Rates also differ based on census data between metropolitan areas, as well as between suburban areas and between various rural towns. Our research based on NHTSA data indicates that the disparity in insurance premiums is going to increase further, as the proportion of urban fatalities is rising as compared to rural fatalities.

Insight 6: The pedestrian population is the more vulnerable to fatalities

Based on the NHTSA data, pedestrian fatalities in year 2015 increased on an aggregate level by 10% to 6,387, higher than the overall increase of 7.2%. In the very same year, the fatality rate among the pedestrians was 92%, implying an almost certain death of the victim.

The hit-and-run fatalities were unusually high in 2015 and accounted for nearly 5.04% of the total fatalities. The numbers increased by almost 12% from 2014 to 2015, which was the highest increase in the past 6 years. Hit-and runs have a strong correlation with pedestrian fatalities. The data reveals that one in five of all pedestrian fatalities are hit-and-runs, and 60% of hit-and-run fatalities have pedestrians as victims. Driving under the influence of alcohol has proved to be a major part of this problem. Older pedestrians are particularly vulnerable. According to the study, 54% of the pedestrian fatalities occurred with people aged above 46 years. Fatalities involving pedestrians belonging to the age group of 56-65 have shown an increase of more than

52%, the highest increase since 2010. In 2015, pedestrian fatalities for the age group 46-55 have increased the most at nearly 18%.

Insight 7: Hostile weather increases the chances of fatalities

Rough weather conditions make it more difficult to control a vehicle and avoid accidents, resulting in a great risk to motor vehicle drivers and pedestrians. This can be easily verified from the NHTSA fatality statistics.

The proportion of fatalities in cloudy weather increased from 56.3% in 2010 to 61.2% in 2015. The year 2015 had the highest amount of precipitation with rain as the second most prominent reason for weather related crashes, accounting for 28% of accidents in 2015. The overall proportion of fatalities under adverse weather conditions versus clear weather increased from 23.9% in 2010 to 28.8% in 2015.

The weather data reveals that the years following 2010, especially the year 2015, were significantly warmer. More driving is done when the weather is warmer, leading to increased fatalities. This can be another justification for nearly 10% increase in fatalities in 2015 as compared to the overall average of 7.2%.

Insight 8: Lack of restraint measures indicate an increase in fatalities.

The number of accidents which did not involve use of any restraint measure continued to decline since 2010 before increasing again in 2015. Surprisingly, while such road accidents declined over the years, the fatality rate continued to increase. In 2015, out of the 17,905 people who were not using any kind of restraint, 68.6% died. This is reportedly the highest fatality rate since 2010. Over the years insurers have jumped to action to intervene and motivate drivers to follow safety measure and use restraints such as seat belts, insurance companies offer discounts and inflict a penalty for disregarding safety rules. For instance, one of the national carriers offers a passive restraint system discount to cars with airbags, which reduces the risk of fatality in the event of an accident.

The number of states assessing points for seatbelt violation has also increased. If a state assesses points for seatbelt violations, these points determine whether an individual’s insurance prices are likely to increase. Insurance companies use these points as one of the indicators of the risk customers pose for the company.

Insight 9: Driving under the influence of alcohol contribute towards increase in road fatalities

Driving under the influence of alcohol renders the driver incapable of operating a motor vehicle safely, and has been a major contributor towards road fatalities.  There was a decrease in the number and proportion of drivers found driving under the influence of drugs or alcohol influence since 2010, which can be attributed to strict traffic laws and harsh penalties in place for drivers. But despite the decreasing trend, these numbers are still significant. Drivers in the

26-35 age group account for the greatest proportion of road accidents under the influence of drugs or alcohol, followed by drivers in the 17-25 age group.

To control driving under the influence of alcohol, certain car manufacturers are trying to include drunk driving prevention systems. For example, Nissan is introducing a hi-sensitivity alcohol sensor built into the transmission shift knob, which will be able to detect the presence of alcohol in the perspiration of a driver’s palm.

Insight 10: Drivers with previous violations are more likely to be involved in an accident even though they face higher insurance premiums

Previous violations by a driver include any accident, suspension, DWI conviction, speeding conviction or any other conviction occurring in the past 3 years. This category accounted for nearly 51% of total fatalities, increasing 11% from 2014 – the highest increase in 6 years. Moreover, 28% of such drivers in 2015 were previously involved in accidents in the same year.

According to the National Association of Insurance Commissioners (NAIC), auto insurers take into account an individual’s driving record when setting the cost of insurance. Traffic violations, such as speeding tickets, accidents and suspensions are considered as a part of these driving records. But with an increase in the number of fatalities involving individuals with previous violations, insurance companies need to be even more cautious and think about ways to deter.

Insight 11: Drivers with invalid licenses are more likely to be involved in crashes

In the year 2015, 12% (5,879) of the total drivers involved in crashes had invalid licenses. There was a 5% increase in fatalities involving drivers with invalid licenses in 2015, reaching an all-time high of 6,620 deaths since 2010.

An interesting point was that 53% of these drivers were not owners of the vehicle. Some of these might be working as drivers for private owners or companies, implying that there is a need for implementing proper background checks and verifications, before hiring drivers. Insurers manage the lack of compliance in accordance to the prevailing state regulations.

In some insurance carrier don’t have authority to cancel a customer’s policy in the middle of a policy term without a very good reason such as nonpayment, fraud or misrepresentation. However, there are certain states that allow mid-term cancellations based on the drivers’ license being suspended or revoked.

Insight 12: There exist strong geographic variations for fatality trends across the country

NHTSA data reveals that the top three populated states of the U.S. - California, Florida and Texas - have the highest total number of fatalities for the past six years. Oregon, Nebraska and Nevada have seen the highest percent change between 2010 and 2015. Alaska, Texas and Montana have the highest fatality rate per square mile for 2015. The increase in miles travelled increased by 8% but the increase in fatalities was more than 20% for Oregon, which can be attributed to some of the factors outlined above.

As per reports, Montana has the highest insurance premium rates. While there are a number of factors that have led to this, but one of the biggest factors is the high fatality rate. According to the Insurance Institute for Highway Safety, Montana had one of the highest vehicle accident fatality rate in the country at 22.6 deaths per 100,000 people – twice the national average.

Conclusion

This whitepaper reemphasizes on the importance of insures being up-to-date with changing driver behaviors, traffic trends, meteorological conditions, economic factors, demographics, geographic factors, and modern technology to gauge and analyze these variables. Doing so enables better risk management and profitable pricing strategies for automobile insurance providers.

As a study relevels that a large number of fatalities occur in adverse weather conditions, insurers should partner with education institutions and governments need to educate motorists on inclement weather safe driving practices in implementing technologies that improve road safety in such weather. Some insurers are already doing this though a number of mechanisms, such as combining new and existing data sources, and analyzing this data using new analytical platforms. These initiatives include applications that convert geospatial data from mobile devices into metrics that describe travel patterns in the vicinity of fatal crashes, access fatality analysis reporting data, information on crashes along a commuter route, crash avoidance technologies and other capabilities.

To augment profits, insurers need to come up with innovative solutions to address challenges posed by a lack of road safety compliance and increasing traffic. One such solution could be offering highly personalized premium rates and engaging less risky populations with expected premium reductions, as this population tends to subsidize the high-risk population.

For instance, some companies offer usage based insurance programs that monitor driving behavior in real time using telematics. These programs use either a device installed in the vehicle or a driver’s smartphone to collect data on when, how and where they drive. After these devices collect enough data, then customers receive a personalized rate for their insurance. This encourages safer driving behavior, reducing risk for insurers and cutting rates for customers. In the long term, this would also lead to decreased road fatalities and a lower number of claims.

References

1. http://www.insure.com/car-insurance/carinsurance-rates.html

2. https://www.aaafoundation.org/sites/default/files/weather_report_2016.pdf

3. https://books.google.co.in/books?id=71wDS5bH6nMC&pg=PT45&lpg=PT45&dq=NHTSA+marketing+insurance&source=bl&ots=fl0H5A-Pax&sig=9mcVWI7a531tegnVa2wHfD9fpKY&hl=en&sa=X&ved=0ahUKEwiZ94rt7rSAhXBO48KHQY9CeoQ6AEIRTAI#v=onepage&q&f=false

4. http://www.cnbc.com/2016/05/27/autoinsurance-rates-rising-at-fastest-rate-in-almost-13-years.html

5. http://www.businessinsider.in/It-hasnt-beenthis-bad-for-the-car-insurance-industry-in-15-years-and-its-all-the-strong-economys-fault/articleshow/50613684.cms

6. http://www.oregonlive.com/commuting/index.ssf/2015/12/oregon_traffic_deaths_top_400.html

7. http://www.cccis.com/miles-driven-trafficfatalities-rise/

8. https://www.dolmanlaw.com/auto-accidentscommon-rural-urban-areas/

9. http://www.iihs.org/iihs/topics/t/generalstatistics/fatalityfacts/state-by-state-overview

10. https://obamawhitehouse.archives.gov/blog/2016/08/29/2015-traffic-fatalities-data-hasjust-been-released-call-action-download-andanalyze