When all cars and light trucks have CHMSL, the lamps will prevent an estimated 92,000-137,000 police-reported crashes per year, and approximately 102,000 unreported crashes. CHMSL will reduce property damage and its associated societal costs by approximately $655,000,000 per year (in 1994 dollars) in reported and unreported crashes. The lamps will prevent 58,000-70,000 injuries per year. CHMSL add $12 to the price of a car or light truck and one pound to its weight. The annual cost of CHMSL to consumers is about $206,000,000, and it is a fraction of the benefits. Even though the effectiveness of CHMSL has declined substantially from its initial level, this still is and will continue to be a highly cost-effective safety device.

Crashes avoided

Police-reported crashes: NHTSA's most recent analysis of the economic cost of motor vehicle crashes estimates that 9,701,040 crashes were reported to the police during 1994 in the United States [2], p. 9. However, CHMSL only has a potential for effectiveness in those crashes where at least one car or light truck had rear-impact damage. CHMSL cannot be expected to have any effect on crashes where all the involved vehicles had only frontal, side, or rollover damage, or on crashes where the only vehicles with rear-impact damage were heavy trucks, motorcycles, etc. The percentage of police-reported crashes in which at least one car or light truck had rear-impact damage is estimated from 1995 data from the same eight States whose files were employed in the effectiveness analyses of Chapter 2, using the same definitions of "rear-impact damage" as in that chapter:

Thus, there were .3271 x 9,701,040 = 3,173,000 crashes in which at least one car or light truck had rear-impact damage. The long-term (1989-95 and onwards) effect of passenger car CHMSL, as estimated in Chapter 2, is to reduce these crashes by 4.33 percent. Let us assume, for the time being, a similar effectiveness in light trucks. As a result, the annual benefit is estimated to be

For a more conservative estimate of benefits, we can utilize the estimate of police-reported crashes generated by NHTSA's General Estimates System (GES) and reported in Traffic Safety Facts [34], p. 14. (NHTSA's economic cost study, however, presents various reasons why the direct GES number is a substantial underestimate of actual police-reported crashes [2], pp. 16-21.) According to GES, there were 6,492,000 police-reported crashes in the United States in 1994. In that case the estimated long-term crash reduction for CHMSL would be

Unreported crashes: NHTSA's economic cost study estimates that 7,173,091 crashes that occurred in the United States during 1994 were not reported to the police [2], p. 9. The average unreported crash tends to have lower severity than the average police-reported crash: e.g., crashes that do not result in enough damage to meet a State's reporting threshold would not be reported. However, the analyses of Chapter 3 did not show any substantial differences in CHMSL effectiveness as a function of crash severity (except for zero effectiveness in fatal crashes) Let us assume the same 4.33 percent reduction in unreported crashes as in police-reported crashes, and also that the proportion of rear impact is the same in reported and unreported crashes. In that case the estimated long-term benefit of CHMSL would be

Police-reported plus unreported crashes: The preceding calculations suggest a range of 92,000 to 137,000 police-reported and 102,000 unreported crashes avoided each year. That adds up to a range of 194,000 to 239,000 crashes avoided per year.

Property damage avoided

Police-reported plus unreported crashes: NHTSA's analysis of the economic cost of crashes estimates that the total cost of property damage resulting from all motor vehicle crashes in the United States during 1994 was $52,119,000,000 [2], p. 7. (The study does not apportion that sum between police-reported and unreported crashes.) The eight State files analyzed in Chapter 2 do not supply any data on the cost of property damage in rear impacts vs. other impacts, but the 1994 North Carolina file (analyzed in Chapter 4) has nearly complete records of the estimated total damage in each crash: the sum of the damages to the various vehicles and other property. In North Carolina, the average property damage in crashes where at least one car or light truck was rear-impacted was only 84.07 percent as high as the average property damage in all other crashes. Let us assume that the same cost ratio prevails in other States' reported crashes, and also in unreported crashes. In that case, even though rear impacts account for 32.71 of all crashes, they only account for

Since, in the long term, CHMSL reduces rear impacts by 4.33 percent, the annual value of property damage avoided is:

(When CHMSL succeeds in preventing a crash, there is a twofold benefit: the CHMSL-equipped vehicle escapes being struck in the rear and damaged, but so does the would-be striking vehicle.)

Injuries avoided

Police-reported crashes: NHTSA's economic cost analysis estimates that 4,089,930 people were injured, nonfatally, in police-reported crashes in the United States during 1994 [2], p. 9. CHMSL only has a potential for effectiveness in those crashes where at least one car or light truck had rear-impact damage; however, if CHMSL successfully prevents such a crash it will not only save the occupants of the rear-impacted vehicle from injury, but also the occupants of the other vehicles involved in the crash. The proportion of crash injury victims who were involved in crashes in which at least one car or light truck had rear-impact damage is estimated from 1995 data from the eight States:

Thus, there were .3353 x 4,089,931 = 1,371,000 people injured in reported crashes in which at least one car or light truck had rear-impact damage. The long-term (1989-95 and onwards) reduction of injury crashes by passenger car CHMSL, as estimated in Chapter 3, is 4.01 percent. Let us assume a similar effectiveness in light trucks. The annual benefit is estimated to be

For a more conservative result, we can use the estimated number of people injured in police-reported crashes, as generated by NHTSA's General Estimates System (GES) and reported in Traffic Safety Facts [34], p. 15. (According to NHTSA's economic cost study, the direct GES number is a substantial underestimate [2], pp. 16-21.) Based on GES, 3,215,000 people were injured in police-reported crashes in the United States in 1994. In that case the estimated long-term injury reduction for CHMSL would be

Unreported crashes: NHTSA's economic cost study estimates that 1,126,000 people were injured in unreported motor vehicle crashes during 1994 [2], p. 9. Let us assume the same 4.01 percent injury-crash reduction for CHMSL in unreported crashes as in police-reported crashes, and also that the proportion of injuries occurring in rear impacts is the same in reported and unreported crashes. In that case the estimated long-term benefit of CHMSL would be

Police-reported plus unreported crashes: The preceding calculations suggest a range of 43,000 to 55,000 fewer people injured per year in police-reported crashes and 15,000 in unreported crashes. That adds up to a range of 58,000 to 70,000 injuries avoided per year.

Cost analysis

During 1985-88, a NHTSA contractor estimated the purchase price increase and weight added to passenger cars by CHMSL, based on detailed inspection and disassembly of the lamps in a representative set of 30 make-models [4], [26]. NHTSA's 1989 evaluation of CHMSL took a sales-weighted average of those estimates [19], pp. 46-49. On average, the lamps added $9.05 (in 1987 dollars) to the purchase price and 0.95 pounds to the weight of a car. The added weight resulted in a small penalty of added fuel consumption over the life of the car, amounting to a net present value of 95 cents. One additional consumer cost was identified: CHMSL bulbs may burn out and require replacement. The net present value of that replacement cost was estimated at 48 cents. In 1987 dollars, the lifetime consumer cost per car, equal to the sum of the purchase price increase, the fuel penalty and bulb replacement, was $9.05 + .95 + .48 = $10.48 per car. Each of those costs can be inflated to 1994 dollars, based on price indices supplied by the Bureau of Labor Statistics [3]. The index for consumer goods, such as lamps and replacement bulbs, rose from 113.6 in 1987 to 148.2 in 1994. The average price of a gallon of fuel rose from 98 cents in July 1987 to $1.199 in July 1994. Thus, the lifetime consumer cost of CHMSL, in 1994 dollars, is:

NHTSA's Final Regulatory Impact Analysis for light truck CHMSL predicted that CHMSL could be provided for the overwhelming majority of pickup trucks, vans and sport utility vehicles at the same price as for passenger cars. However, in certain multistage vehicles, produced at the rate of 300,000 per year, there might be complications that could increase the cost by 50 percent [14], pp. 28-32.

Almost exactly 15,000,000 passenger cars and light trucks were sold annually in the United States during 1994-96 [1]. We can assume that the lifetime cost of CHMSL was $13.60 in 14,700,000 of those vehicles, and $20.40 in the 300,000 multistage trucks. Thus, the total annual cost of CHMSL is

That is just a fraction of the estimated property damage reduction of $655,000,000. Since the damage reduction alone pays for the lamps several times over and, in addition, the lamps prevent many nonfatal injuries, it is obvious that CHMSL, despite the substantial decline in their effectiveness from 1987 values, are still highly cost effective.