Volvo City Safety loss experience by vehicle age

Highway Loss Data Institute Bulletin Vol., No. : April 5 Volvo City Safety loss experience by vehicle age Summary City Safety technology was first introduced by Volvo to the U.S. market with the XC6 as standard equipment. The Highway Loss Data Institute (HLDI) published bulletins in and presenting the relationship of this technology and insurance losses. The results showed that the City Safety system was associated with reductions in insurance losses. The purpose of this report is to see if the benefits of City Safety persist over time. To examine this, claim frequencies for City Safety-equipped vehicles and comparison vehicles were evaluated by vehicle age. There is some variability in the results by vehicle age but there is not an indication that there is a pattern of diminished benefits as the vehicles age. 4..5..5..5.5 PDL claim frequencies for City Safety-equipped vehicles and comparison vehicles by vehicle age City Safety vehicles comparison vehicles..8.6.4. Ratio of PDL claim frequencies for City Safety-equipped vehicles to comparison vehicles by vehicle age

Introduction This Highway Loss Data Institute (HLDI) bulletin provides a look at the impact of vehicle age on the insurance loss benefits of Volvo City Safety technology previously presented by HLDI. For this bulletin, the loss experience for Volvo XC6s and S6s equipped with City Safety were compared with losses for comparable vehicles without the system by vehicle age. City Safety is a low-speed collision avoidance system first released as standard equipment on the Volvo XC6, a midsize luxury SUV, and on the S6, a midsize luxury car. The system was developed by Volvo to reduce lowspeed front-to-rear crashes, which commonly occur in urban traffic, by assisting the driver in braking. According to a Volvo news release, 75 percent of all crashes occur at speeds up to 9 mph, and half of these occur in city traffic (Volvo, 8). The City Safety system has an infrared laser sensor built into the windshield that detects other vehicles traveling in the same direction up to 8 feet in front of the vehicle. The system initially reacts to slowing or stopped vehicles by pre-charging the brakes. The vehicle will brake automatically if forward collision risk is detected and the driver does not react in time, but only at travel speeds up to 9 mph. If the relative speed difference is less than 9 mph, a collision can be avoided entirely. If the speed difference is between 9 and 9 mph, the speed will be reduced to lessen the collision severity. The City Safety system studied in this analysis operates at speeds up to 9 mph but a newer version of the system operates at speeds up to mph. City Safety is automatically activated when the vehicle ignition is turned on but can be manually deactivated by the driver. When examining the effect of City Safety on insurance losses, it is important to consider that the system is not designed to mitigate all types of crashes and that many factors can limit the system s ability to perform its intended function. City Safety works equally well during the day and at night, but fog, heavy rain, or snow may limit the ability of the system s infrared laser to detect vehicles. The driver is advised if the sensor becomes blocked by dirt, ice, or snow. Method Vehicles In the initial analysis, loss results for the XC6 were compared with other midsize luxury SUVs, while loss results for the S6 were compared with other midsize luxury cars. For the results presented here, the XC6 and S6 were pooled, as were the comparison vehicles. Sales of the Volvo XC6 began in February 9, when other brands still were marketing 9 models. Consequently, the control populations for the XC6 analyses included vehicles starting in model year 9. The total study population for the XC6 was model years during calendar years 9 4 with control vehicle model years of 9 4. The loss experience of the model year 9 vehicles in calendar year 8 was excluded because no XC6s were on the road during this time period. City Safety was added as standard equipment to the Volvo S6 in model year. The analyses considered model years for the S6 and its control vehicles during calendar years 4. Calendar year was not included in the S6 analysis because of the very small number of model year S6s insured that year. HLDI Bulletin Vol, No. : April 5

Total exposure measured as insured vehicle years and the total number of claims for the XC6 and S6 are shown by insurance coverage type in Table. Because previous HLDI analyses have shown them to have different loss patterns, hybrids, convertibles, and two-door vehicles were excluded from the control groups. Table : Exposure and claims by coverage type XC6 S6 Coverage Claims Exposure Claims Exposure Property damage liability,6,8,89 76,87 Bodily injury liability 98 5,54 59 6,9 Collision 6,88,8 5,597 76,87 Medical payment 7 8,5 66,95 Personal injury protection 49 57,44 46 7,565 The study and control vehicles in this analysis can also be equipped with optional collision avoidance features that have been shown to affect claim frequency in other studies by HLDI. It should be noted that this analysis does not account for their presence or absence because the information needed to identify the vehicles with the optional features is not available. Furthermore, the take rate for these features is thought to be low. Insurance Data Automobile insurance covers damages to vehicles and property as well as injuries to people involved in crashes. Different insurance coverages pay for vehicle damage versus injuries, and different coverages may apply depending on who is at fault. The current study is based on property damage liability, collision, bodily injury liability, personal injury protection, and medical payment coverages. Data are supplied to HLDI by its member companies. Collision coverage insures against vehicle damage to an at-fault driver s vehicle sustained in a crash with an object or other vehicle; this coverage is common to all 5 states. Property damage liability (PDL) coverage insures against vehicle damage that at-fault drivers cause to other people s vehicle and property in crashes; this coverage exists in all states except Michigan, where vehicle damage is covered on a no-fault basis (each insured vehicle pays for its own damage in a crash, regardless of who is at fault). Coverage of injuries is more complex. Bodily injury (BI) liability coverage insures against medical, hospital, and other expenses for injuries that at-fault drivers inflict on occupants of other vehicles or others on the road; although motorists in most states may have BI coverage, this information is analyzed only in states where the at-fault driver has first obligation to pay for injuries ( states with traditional tort insurance systems). Medical payment (MedPay) coverage, also sold in the states with traditional tort insurance systems, covers injuries to insured drivers and the passengers in their vehicles, but not injuries to people in other vehicles involved in the crash. Seventeen other states employ no-fault injury systems (personal injury protection coverage, or PIP) that pay up to a specified amount for injuries to occupants of involved-insured vehicles, regardless of who is at fault in a collision. The District of Columbia has a hybrid insurance system for injuries and is excluded from the injury analysis. HLDI Bulletin Vol, No. : April 5

Statistical methods Regression analysis was used to model claim frequency per insured vehicle year while controlling for various covariates. Claim frequency was modeled using a Poisson distribution. This model used a logarithmic link function. The covariates included calendar year, model year, garaging state, vehicle density (number of registered vehicles per square mile), rated driver age, rated driver gender, marital status, collision deductible, and risk. A separate regression was performed for each vehicle age, study vehicle (XC6 and S6), and coverage type for a total of 5 models. To illustrate the analysis, the Appendix contains full model results for collision claim frequency for the XC6 at vehicle age. Vehicle age in this study is defined as vehicles aged to. For example, a model year vehicle in calendar year would have a vehicle age of, while a vehicle in the same calendar year would be aged. The results of this study are presented as ratios of claim frequency of City Safety-equipped vehicles to claim frequency of comparison vehicles. If the ratio is, there is no difference between the two types of vehicles being compared. If the ratio is less than, it means there are fewer claims made on vehicles with City Safety technology than those without. The estimated effects of City Safety by age were calculated separately for the XC6 and S6, along with their respective standard errors. A combined, or pooled estimate was calculated as a weighted average of the two estimates, with weights proportional to the inverse variance (i.e. weight =/SE). Thus, if an estimate for the XC6 had a smaller standard error (tighter confidence interval) compared to that of the S6 estimate, it would have more influence on the combined estimate. The combined standard error for the weighted average was calculated based on the same assumptions. Results Results for this study are broken down by coverage type. Figure shows the PDL claim frequencies for City Safetyequipped vehicles and comparison vehicles by vehicle age. Figure shows the ratio of PDL claim frequencies for City Safety-equipped vehicles to comparison vehicles by vehicle age. At each vehicle age, the PDL claim frequency is significantly lower for City Safety-equipped vehicles than for comparison vehicles. Estimated effects range from 7 to 6 percent benefits for the City Safety-equipped vehicles. Vehicles from ages to indicate an increased benefit while age is slightly less benefical than ages -. 4..5 Figure : PDL claim frequencies for City Safety-equipped vehicles and comparison vehicles by vehicle age City Safety vehicles comparison vehicles..5..5.5 HLDI Bulletin Vol, No. : April 5 4

..8.6.4. Figure : Ratio of PDL claim frequencies for City Safety-equipped vehicles to comparison vehicles by vehicle age Figure shows the collision claim frequency ratios of City Safety-equipped vehicles to comparison vehicles by vehicle age, and all ratios are below. For the newest vehicles, collision claim frequencies are about percent lower for City Safety-equipped vehicles. For older vehicles (ages ), the collision claim frequency ratios are similar to each other, at around.8..9.8.7.6.5.4... Figure : Ratio of collision claim frequencies for City Safety-equipped vehicles to comparison vehicles by vehicle age HLDI Bulletin Vol, No. : April 5 5

Figures 4 6 show the BI, PIP, and MedPay claim frequency ratios of City Safety-equipped vehicles to their comparison vehicles by vehicle age, respectively. All ratios are below, indicating that there are fewer injury claims made for City Safety-equipped vehicles than comparison vehicles at every vehicle age, and there does not appear to be a relationship between vehicle age and injury claims. In general, there are fewer injury claims in the HLDI database compared with collision and PDL claims, creating larger confidence intervals..9.8.7.6.5.4....9.8.7.6.5.4....4..8.6.4. Figure 4: Ratio of bodily injury liability claim frequencies for City Safetyequipped vehicles to comparison vehicles by vehicle age Figure 5: Ratio of personal injury protection claim frequencies for City Safetyequipped vehicles to comparison vehicles by vehicle age Figure 6: Ratio of MedPay claim frequencies for City Safety-equipped vehicles to comparison vehicles by vehicle age HLDI Bulletin Vol, No. : April 5 6

Discussion This is the first opportunity for HLDI to evaluate if the benefits of crash avoidance systems degrade over time. This is because collision avoidance systems are relatively new technologies to the vehicle fleet. For example, for the model year, front crash prevention was available on about percent of vehicle series and will not reach 95 percent of the vehicle fleet until 48 (HLDI, 4). The results presented here for the Volvo City Safety system are promising as there does not appear to be a clear change in the benefits of the system by vehicle age. This suggests no significant degradation of the City Safety system during the 5 years it has been available. Because the City Safety system uses the vehicle s brakes to prevent crashes, tire wear or brake aging could affect the City Safety system. As these vehicles age, these possibilities do not appear to be negatively impacting the success of the system. Limitations All of the XC6s and S6s included in the current study were equipped with the City Safety technology, but there was no way to know whether any drivers in the crash-involved vehicles had manually turned off the system prior to the crash. Also, most of the vehicles in this study, including the XC6 and S6, can be equipped with a variety of collision avoidance features that might also affect claim frequencies, and it was not possible, based on data available to HLDI at the time of the study, to control for the presence of these other features. The study and control vehicles may have other collision avoidance features that could be influencing the results. To fully understand the benefits of City Safety, subsequent analysis will be required as additional loss data become available involving more and potentially different drivers. This analysis controlled for a variety of possible demographic differences (rated driver age, gender, marital status, and risk) between the study and control populations. It still is possible that rated drivers who chose to purchase vehicles with City Safety differ in other ways that could affect crash likelihood perhaps drivers who are more concerned about safety or who drive more cautiously because they have experienced prior collisions. HLDI Bulletin Vol, No. : April 5 7

References Highway Loss Data Institute.. Volvo City Safety loss experience: initial results. Loss Bulletin Vol. 8, No. 6 Arlington, VA. Highway Loss Data Institute.. Volvo City Safety loss experience: an update. Loss Bulletin Vol. 9, No. Arlington, VA. Highway Loss Data Institute. 4. Predicted availablity of safety features on registered vehicles - an update. Loss Bulletin Vol., No. 5 Arlington, VA. Volvo Car Corporation. 8. Volvo cars presents City Safety a unique system for avoiding collisions at low speeds. Press Information. Available: https://www.volvoclub.org.uk/press/pdf/city_safety_eng.pdf. Göteborg, Sweden. Appendix Parameter Appendix: Illustrative regression results - collision frequency for Volvo XC6 age Degrees of freedom Estimate Effect Standard error Wald 95% confidence limits Chi-square P-value Intercept -8.86-8.8788-8.755 76.6 < Calendar year 9 -.4.7% 5 -.56-949 69. < Vehicle make and series 84.9% 7 55 5 5.9 5 4.4% 6-88 68.45.9 Acura MDX 4dr.% 97-6 86.56.45 Acura RDX 4dr 5.5% 59-65 755.889 Acura ZDX 4dr.59 8% 84.499.7567 5.66 < Audi Q5 QUATTRO 4dr.4 4.5% 4.86.999 5.5 < BMW X 4dr.57 6.% 66 79. 6.99 < BMW X5 4dr.4 6.9% 95.56.77.64 < BMW X6 4dr.646 8.% 4.596.6896 94.55 < Cadillac SRX 4dr 8 5.5% 88.47.6. < Infiniti EX5 4dr..% 485 7.7 6.4 8 Infiniti FX5 4dr.5 8.4% 9.74.67 4.87 < Infiniti FX5 4dr.7 7.%.78 86.5479 7.4 7 Land Rover LR 4dr.46 7.9% 754 985.99.67 Lexus RX 5 4dr.87 46.% 7.78.47 98.48 < Lincoln MKT 4dr.4 8.% 56..47 8.6 < Lincoln MKX 4dr.7.7% 4 46.778.45 Mercedes-Benz GLK Class 4dr.65.% 9 9.4 7.6 < Mercedes-Benz M Class 4dr.6 5.% 97.655.88 56.79 < Saab 9-4X 4dr -.7-8.6%.76.7 47..6 Saab 9-7X 4dr 87 8.6%. -.74.95.4.578 Volvo XC9 4dr 677 7.% 9-89.44. 8 Volvo XC6 4dr State Alabama.554 6.8% 487 6.57.9 4 Arizona.9.5% 98.87 7.5 6 HLDI Bulletin Vol, No. : April 5 8

Parameter Appendix: Illustrative regression results - collision frequency for Volvo XC6 age Degrees of freedom Estimate Effect Standard error Wald 95% confidence limits Chi-square P-value Arkansas.775 9.4% 79 46 4 5.9 49 California.678.7% 9 5 96.5 < Colorado 776 8.% 79.59 4.9 47 Connecticut..7% 45 46.697 8.78 Delaware 94.% 8 -.79.867..74 District of Columbia.79 45.% 68.4.576 < Florida -.5.8% -.667-87 5. < Georgia -854-8.% -.5-6 6.67 98 Hawaii.84.4% 67-5.4.59.7 Idaho -.488.8%.44 -.4.8 6 Illinois 869 9.% 4 96.4.95 Indiana.96.8% 58.9 6.5 7 Iowa 45 4.% 8 -.47.997.8.596 Kansas 6 6.% 6-564.786 4 84 Kentucky -48-4.% 67 -.647 77.5.478 Louisiana.69 7.5% 4 78.46 4.57 Maine.78 9.5%.4-79.457.84.749 Maryland 77 7.6% 5 8.5 5.8 58 Massachusetts.96 5.8% 47.479. < Michigan.48 54.% 9.7.49 95.7 < Minnesota -4-4.% 47 -.5 54.8.7 Mississippi -7-7.% 87 -.55 889.79.756 Missouri.% 47-8.4 6.85 Montana -.4565-6.7%.54 -.898-47 4. 48 Nebraska -.7 5.6% 867 -.4 -.85 499 Nevada 895 9.4% 559 -.99.57.9 New Hampshire.65.4% 74.5.45.8 New Jersey 96.% 56.4 8.68 < New Mexico -5-5.% 89 -.78..4.555 New York.488 6.% 94.7.869 58.65 < North Carolina -.85-4.8% 97 -.69-7 5.6 < North Dakota 8 4.7%.976 -.665.68.5.68 Ohio -596-5.8% -.4 5.7 74 Oklahoma 8.9% 554-7.467.48.49 Oregon 45 4.6% 564-655.557.64.444 Pennsylvania.5 4.9% 4.754.696 85.68 < Rhode Island 4.6% 78 796.6 9.4 South Carolina -.66.8% 54 -.4-6.55 5 South Dakota.59 9.6%.69-6.585.5.7 Tennessee.67 4.6% 4 54.9.57 Utah 59 5.% 749-958.977.46.4964 Vermont.76 4.%.5-47.48.59.7 Virginia.44 5.5% 7 97.974 7.99 < Washington.64 7.8% 7 98.6 9.44 < HLDI Bulletin Vol, No. : April 5 9

Parameter Registered vehicle density Appendix: Illustrative regression results - collision frequency for Volvo XC6 age Degrees of freedom Estimate Effect Standard error Wald 95% confidence limits Chi-square P-value West Virginia 9 9.8% 99-849.75 5 49 Wisconsin 866 9.% 58-49.88.8 944 Wyoming.8.7%.77 -.88.4754.5.4688 Alaska.4 7.5%.6 7.4658 4.59 Texas Unknown -.55-4.9%.485.6.75.65.984 <5 -.78-4.% -.96 -.69 79. < 5 99 -.9 -.% 47 -.874 -.96 9.65 < 49 -.6 -% 78 -.7-4 76.64 < 5 499 -.984 8.% 6 - -.669 5.4 < 5 999-987 -9.4% 9 -.4-74 58.9 <,+ Deductible range to -65-6.% 9-997 -5.8 to 5.76.% 8.465.968 447.94 < 5+ -.74-7.% -.4 -.96 68.5 < 5 to 5 Rated driver age group 5 9 9 9.7% 4 7.77 4.5 4.55.4% 4 69.87 5.95 < 5 9.5.% 694.66 4.4 < 9 684 7.% 8 45 954 4.7 < 5 59 -..6% 6 -.88-854 67.8 < 6 64-778 -7.5% 7 -.6-44.4 < 65 69 5.5% 86-57.67.45 7 74 769 8.% 5 8..67 6 75+ 87.% 9.68.55 8.8 < Unknown -48-4.% 6-9 5.44 66 4 49 Rated driver gender Male -54-5.% 7-75 - 5.5 < Rated driver marital status Unknown -.4-9.9% 74 -.579-54 4.44 Female Single.9 % 6.657.5 9.9 < Unknown.545 6.7% 68. 7. < Married Risk Nonstandard.75 8.6% 4.46.985 4.8 < Standard The Highway Loss Data Institute is a nonprofit public service organization that gathers, processes, and publishes insurance data on the human and economic losses associated with owning and operating motor vehicles. DW4 SK&KW Highway Loss Data Institute 5 N. Glebe Road, Suite 7 Arlington, VA + 7 47 6 iihs-hldi.org COPYRIGHTED DOCUMENT, DISTRIBUTION RESTRICTED 5 by the Highway Loss Data Institute. All rights reserved. Distribution of this report is restricted. No part of this publication may be reproduced, or stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the copyright owner. Possession of this publication does not confer the right to print, reprint, publish, copy, sell, file, or use this material in any manner without the written permission of the copyright owner. Permission is hereby granted to companies that are supporters of the Highway Loss Data Institute to reprint, copy, or otherwise use this material for their own business purposes, provided that the copyright notice is clearly visible on the material.