NHTSA Report Number DOT HS 809 835December 2004

Cost Per Life Saved By The Federal Motor Vehicle Safety Standards

Charles J. Kahane, Ph.D.

pdf graphic The complete report is available here in pdf format.

Abstract

The National Highway Traffic Safety Administration (NHTSA) began to evaluate its Federal Motor Vehicle Safety Standards (FMVSS) in 1975. By December 2004, NHTSA had evaluated the life-saving benefits as well as the consumer cost for a substantial "core" group of safety technologies for passenger cars and LTVs (pickup trucks, sport utility vehicles and vans).

In 2002, these technologies added an estimated $11,353,000,000 (in 2002 Dollars) to the cost of new cars and LTVs of that model year. They saved an estimated 20,851 lives in the cars and LTVs on the road during that calendar year. That amounts to $544,482 per life saved in 2002.

These technologies added a total of $189,842,000,000 to the consumer cost of new cars and LTVs over model years 1968-2002. They saved 252,989 lives in model year 1968 and later vehicles during calendar years 1968-2002. That amounts to $750,782 (in 2002 Dollars) per life saved in 1968-2002.


Two reports summarize the benefits and costs of the Federal Motor Vehicle Safety Standards (FMVSS).  The first report estimates that the FMVSS and other safety technologies saved 328,551 lives in calendar years 1960-2002, including 24,561 in 2002.[1]  The second report estimates that the FMVSS added $839 (in 2002 Dollars) to the cost of a car in model year 2001 and $711 to the cost of an LTV[2], smaller amounts in earlier years.[3]  The results of the two reports are not directly comparable, primarily because the first report includes benefits of several voluntary, pre-FMVSS and/or no-cost technologies omitted from the cost study, whereas the second report included the cost of a side-impact standard for passenger cars for which benefits have not yet been evaluated.

Nevertheless, there is a substantial core group of safety technologies attributable to the FMVSS for which both benefits and costs have been estimated, and the cost per life saved can be computed.  The computation is not based on a complex economic analysis, but on a direct comparison of consumer cost and lives saved.  The purpose is to demonstrate, at an order-of-magnitude level, the heuristic cost effectiveness of the FMVSS, not to compute an exact, societal benefit-cost ratio. 

In other words, this analysis does not include any:

For example, the cost of the 8 million new cars and 7.9 million new LTVs sold in model year 2002 included $11,353,000,000 for the vehicles to meet the core FMVSS (at $710 per car, excluding the $129 cost of the side impact standard, and $711 per LTV).  During calendar year 2002, the core FMVSS saved 20,851 lives, some in these model year 2002 vehicles, but most in earlier vehicles already on the road before 2002.  Thus, in 2002 the payment per life saved was $544,000.

The core group of technologies cost an estimated $544,000 (in 2002 Dollars) per life saved in 2002, and an average of $751,000 (in 2002 Dollars) per life saved over the longer time period, 1968-2002.  These numbers indicate the FMVSS are, on the whole, highly cost-effective, far below the $3,000,000 per life saved that the Department of Transportation considers acceptable.

Exclusions from the “core” group of safety technologies

The core group of technologies includes all those associated with regulations of greater-than-zero cost, whose benefits and costs have both been evaluated and included in the summary reports, but limited to the model years where benefits and costs are both known.  The core group is most easily defined by describing what is excluded.

The cost report begins with model year 1968, because the initial FMVSS went into effect on January 1, 1968.  Therefore, any benefits in vehicles before model year 1968 are excluded from the core group, even though some of the technologies were implemented before 1968 and their benefits estimated, back to calendar year 1960, in the lives-saved report.

The cost report does not include any technologies introduced on a strictly voluntary basis, even if there is a tenuous subject-matter relationship with one of the FMVSS.  The following voluntary technologies, whose benefits are included in the lives-saved report, are excluded from the core group:

Two FMVSS are excluded because they incorporate technologies that were implemented by the manufacturers well before any regulatory process by NHTSA or other government agencies, and are for that reason considered “no cost” in the summary report on costs:

Two safety technologies are excluded because they did not result in a cost increase, and possibly even saved money.  In the spirit of the cost report, we will not attribute “credit” for the lives they saved to the associated FMVSS because, at zero or negative cost, these technologies would presumably have been implemented even in the absence of the FMVSS:

Two technologies are excluded because they do not add anything to the consumer cost of a new car or LTV:

In addition, we will limit benefits to those that accrue directly to car and LTV occupants, and not include the pedestrian, bicyclist and motorcyclist lives saved by dual master cylinders on cars and LTVs.

Finally, two recent technologies are excluded because they have not been fully evaluated:

The “core” group of safety technologies

Despite these exclusions, a large core group of technologies remains that can be associated with specific FMVSS, and for which both costs and benefits have been estimated starting in model year 1968 or later.  The following technologies significantly reduced fatalities while adding cost to new cars and/or LTVs:

In addition, the following technologies were evaluated for cost and benefit.  They added to the cost of a new vehicle, and although there was not a statistically significant reduction of fatalities, there were significant benefits of reducing injuries or avoiding crashes:

One technology added cost, but the statistical evaluation did not identify any significant benefit:

Finally, four technologies were found to add cost.  No evaluations of benefit have been performed, but none are planned because there is little hope of finding a statistically significant effect in the limited crash data that would be relevant to the analysis.  We will assume, conservatively, zero lives saved for these technologies:

In other words, the core group comprises all technologies included in the Executive Summary and summary tables of the cost report except those related to the dynamic test requirements of FMVSS 214 in passenger cars (whose benefits have not been evaluated).

Consumer cost vs. lives saved

Table 1 computes the annual and cumulative consumer cost of the core group of FMVSS for passenger cars, in model years 1968-2002, in 2002 Dollars.  The first column of Table 1 indicates new-car sales by model year.  It is based on Polk NVPP registrations, for that model year, as of July 1 of the next calendar year.[4]  New car sales have been in the 7-11 million range annually.

TABLE 1 – PASSENGER CARS: COST OF THE FMVSS, MODEL YEARS 1968-2002
(in 2002 Dollars)

 

 

FMVSS Cost per Car

 

Model Year

New Car Sales All Excluded Core

Cost of the Core FMVSS

1968

  8,742,568

$169.24

 $ 0.00

$169.24

   $1,480,000,000

1969

  9,106,883

 216.05

   0.00

 216.05

   $1,968,000,000

1970

  8,704,374

 236.02

   0.00

 236.02

   $2,054,000,000

1971

  8,730,816

 241.47

   0.00

 241.47

   $2,108,000,000

1972

  9,948,136

 268.24

   0.00

 268.24

   $2,668,000,000

1973

 11,036,182

 291.23

   0.00

 291.23

   $3,214,000,000

1974

  9,649,885

 301.97

   0.00

 301.97

   $2,914,000,000

1975

  7,611,961

 299.54

   0.00

 299.54

   $2,280,000,000

1976

  9,452,325

 312.58

   0.00

 312.58

   $2,955,000,000

1977

 10,267,394

 306.66

   0.00

 306.66

   $3,149,000,000

1978

 10,573,362

 302.85

   0.00

 302.85

   $3,202,000,000

1979

 10,277,491

 299.58

   0.00

 299.58

   $3,079,000,000

1980

  8,707,110

 298.26

   0.00

 298.26

   $2,597,000,000

1981

  8,127,671

 297.87

   0.00

 297.87

   $2,421,000,000

1982

  7,303,353

 297.25

   0.00

 297.25

   $2,171,000,000

1983

  7,657,122

 297.61

   0.00

 297.61

   $2,279,000,000

1984

 10,407,770

 297.66

   0.00

 297.66

   $3,098,000,000

1985

 10,531,723

 298.29

   0.00

 298.29

   $3,142,000,000

1986

 10,694,040

 299.50

   0.00

 299.50

   $3,203,000,000

1987

 10,380,058

 338.32

   0.00

 338.32

   $3,512,000,000

1988

 10,303,489

 380.42

   0.00

 380.42

   $3,920,000,000

1989

  9,728,781

 421.31

   0.00

 421.31

   $4,099,000,000

1990

  8,695,605

 596.71

   0.00

 596.71

   $5,189,000,000

1991

  8,100,316

 593.10

   0.00

 593.10

   $4,804,000,000

1992

  7,739,082

 607.59

   0.00

 607.59

   $4,702,000,000

1993

  8,201,002

 650.01

   0.00

 650.01

   $5,331,000,000

1994

  8,150,339

 752.09

  12.71

 739.38

   $6,026,000,000

1995

  9,178,951

 777.93

  31.94

 745.99

   $6,847,000,000

1996

  7,695,487

 782.84

  51.38

 731.46

   $5,629,000,000

1997

  8,049,242

 838.81

 128.94

 709.87

   $5,714,000,000

1998

  7,714,249

 839.18

 129.44

 709.74

   $5,475,000,000

1999

  8,177,224

 839.16

 129.41

 709.75

   $5,804,000,000

2000

  8,882,145

 839.29

 129.58

 709.71

   $6,304,000,000

2001

  8,092,939

 839.13

 129.35

 709.78

   $5,744,000,000

2002

  8,053,834

 839.13

 129.35

 709.78

   $5,716,000,000

 

314,672,909

 

 

 

 $134,798,000,000


The next column, the cost of all FMVSS per passenger car, is copied from the cost report.[5]  Excluding the cost of the dynamic requirement for FMVSS 214 that phased in during 1994-97, the cost of the core FMVSS increased from $169 in 1968 to almost $710 in 1997-2002.  The total consumer cost of the FMVSS in passenger cars increased from $1,480,000,000 in 1968 to $5,716,000,000 in model year 2002.  It amounted to $134,798,000,000 in model years 1968-2002.

Table 2 estimates the lives of passenger car occupants saved by the core group of FMVSS.  The first column of Table 2, copied from Table 2-4 of the lives-saved report, estimates that 232,255 lives were saved in calendar years 1960-2002 by all of the safety technologies considered in that report, in cars of any model year.  Two steps reduce those numbers to the benefits of just the core FMVSS.  The middle column shows lives saved by all of the technologies, but only in cars of model year 1968 and later.  The same model as in Part 2 of the lives-saved report, but with the data limited to model year 1968 and later, generates those estimates.[6]  In the early years, a substantial proportion of the lives saved are in pre-1968 cars, but by the 1980’s the first two columns are quite similar, because most of the pre-1968 cars have been retired.  The lives saved by the core group of FMVSS, shown in the last column, is obtained by deducting that model’s estimates for the voluntary, pre-FMVSS, no-cost and other safety technologies excluded from the core group.  In 2002, 11,851 of the total 14,175 lives saved are included in the core group of FMVSS; altogether, 176,614 of 232,255 lives saved are in the core group.

Tables 3 and 4 compute the costs and benefits of the core FMVSS in LTVs.  The first column of Table 3 indicates that new-LTV sales increased from 1.5 million in 1968 to nearly 8 million in 2002.[7]  All of the LTV FMVSS in the cost report are included in the core group; the cost of the core FMVSS per LTV is copied from the cost report.[8]  The consumer cost of the FMVSS in LTVs increased from $158,000,000 in 1968 to $5,637,000,000 in model year 2002, adding up to $55,044,000,000 in model years 1968-2002.

The first column of Table 4, copied from Table 2-4 of the lives-saved report, estimates that 94,117 lives were saved in LTVs during calendar years 1960-2002 by all of the safety technologies in all model years.  When the analysis is limited to LTVs of model years 1968+, the estimate drops to 93,296.  The core group of FMVSS saved 76,245 of these lives in 1968-2002.  In 2002 alone, 9,000 of the total 10,331 lives saved are included in the core group.


TABLE 2 – PASSENGER CARS: LIVES SAVED BY SAFETY TECHNOLOGIES, CALENDAR YEARS 1960-2002

Calendar Year

All

All (MY 68+)

Core FMVSS

1960

      92

        0

         0

1961

      95

        0

         0

1962

     110

        0

         0

1963

     131

        0

         0

1964

     168

        0

         0

1965

     211

        0

         0

1966

     288

        0

         0

1967

     444

        0

         0

1968

     719

      252

       169

1969

   1,042

      586

       395

1970

   1,279

      867

       585

1971

   1,574

    1,180

       793

1972

   1,984

    1,591

     1,073

1973

   2,282

    1,921

     1,303

1974

   2,220

    1,958

     1,356

1975

   2,723

    2,492

     1,786

1976

   2,852

    2,659

     1,880

1977

   3,190

    3,015

     2,065

1978

   3,501

    3,360

     2,285

1979

   3,657

    3,546

     2,337

1980

   3,848

    3,764

     2,445

1981

   3,758

    3,697

     2,374

1982

   3,394

    3,337

     2,173

1983

   3,534

    3,494

     2,286

1984

   3,943

    3,909

     2,634

1985

   5,196

    5,160

     3,767

1986

   6,827

    6,786

     5,167

1987

   7,783

    7,751

     5,975

1988

   8,733

    8,702

     6,759

1989

   8,677

    8,652

     6,763

1990

   8,684

    8,666

     6,789

1991

   8,943

    8,926

     7,074

1992

   9,007

    8,994

     7,146

1993

   9,916

    9,903

     7,982

1994

  10,626

   10,612

     8,578

1995

  11,115

   11,108

     8,957

1996

  12,076

   12,068

     9,785

1997

  12,146

   12,138

     9,866

1998

  12,325

   12,319

    10,128

1999

  12,401

   12,393

    10,140

2000

  13,052

   13,044

    10,689

2001

  13,532

   13,526

    11,258

2002

  14,175

   14,168

    11,851

 

 232,255

226,543

   176,614




TABLE 3 – LTVs: COST OF THE FMVSS, MODEL YEARS 1968-2002(in 2002 Dollars)

Calendar Year

New LTV Sales

Core (All) FMVSS Cost Per LTV

Cost Of The Core FMVSS

1968

  1,483,642

$106.58

      $158,000,000

1969

  1,765,168

 106.64

      $188,000,000

1970

  1,629,097

 115.95

      $189,000,000

1971

  1,674,141

 116.24

      $195,000,000

1972

  2,238,132

 142.26

      $318,000,000

1973

  2,654,790

 154.61

      $410,000,000

1974

  2,573,930

 160.79

      $414,000,000

1975

  2,012,141

 160.87

      $324,000,000

1976

  2,688,814

 162.27

      $436,000,000

1977

  3,170,915

 173.14

      $549,000,000

1978

  3,463,941

 189.14

      $655,000,000

1979

  3,701,137

 191.06

      $707,000,000

1980

  2,167,721

 195.96

      $425,000,000

1981

  1,861,330

 196.63

      $366,000,000

1982

  1,996,118

 206.92

      $413,000,000

1983

  2,339,221

 212.12

      $496,000,000

1984

  3,557,322

 214.28

      $762,000,000

1985

  3,943,446

 216.33

      $853,000,000

1986

  4,488,215

 216.81

      $973,000,000

1987

  4,162,249

 233.52

      $972,000,000

1988

  4,598,016

 232.58

     $1,069,000,000

1989

  4,525,434

 232.42

     $1,052,000,000

1990

  3,869,770

 236.15

      $914,000,000

1991

  3,977,269

 250.26

      $995,000,000

1992

  4,077,077

 294.29

     $1,200,000,000

1993

  4,861,491

 308.55

     $1,500,000,000

1994

  5,677,239

 389.49

     $2,211,000,000

1995

  5,934,273

 547.71

     $3,250,000,000

1996

  5,407,762

 612.52

     $3,312,000,000

1997

  6,125,244

 658.58

     $4,034,000,000

1998

  6,331,463

 705.27

     $4,465,000,000

1999

  7,189,531

 708.22

     $5,092,000,000

2000

  7,410,544

 709.39

     $5,257,000,000

2001

  7,389,764

 710.86

     $5,253,000,000

2002

  7,929,365

 710.86

     $5,637,000,000

 

138,875,712

 

$55,044,000,000




TABLE 4 – LTVs: LIVES SAVED BY SAFETY TECHNOLOGIES, CALENDAR YEARS 1960-2002

Lives Saved By Safety Technologies

Calendar Year

All

All (MY 68+)

Core FMVSS

1960

    22

       0

       0

1961

    22

       0

       0

1962

    25

       0

       0

1963

    28

       0

       0

1964

    34

       0

       0

1965

    38

       0

       0

1966

    46

       0

       0

1967

    55

       0

       0

1968

    80

      22

      16

1969

   114

      60

      43

1970

   136

      92

      63

1971

   161

     124

      81

1972

   194

     161

     100

1973

   238

     208

     128

1974

   244

     222

     138

1975

   272

     251

     162

1976

   324

     306

     188

1977

   409

     395

     238

1978

   462

     448

     260

1979

   561

     549

     304

1980

   608

     596

     300

1981

   614

     602

     318

1982

   585

     576

     304

1983

   641

     635

     353

1984

   815

     804

     483

1985

 1,118

   1,107

     749

1986

 1,620

   1,606

   1,174

1987

 2,115

   2,105

   1,576

1988

 2,457

   2,439

   1,865

1989

 2,741

   2,733

   2,132

1990

 2,957

   2,948

   2,317

1991

 3,188

   3,180

   2,530

1992

 3,417

   3,408

   2,763

1993

 3,818

   3,813

   3,139

1994

 4,469

   4,462

   3,698

1995

 4,942

   4,931

   4,096

1996

 5,679

   5,673

   4,754

1997

 6,356

   6,350

   5,362

1998

 6,998

   6,988

   5,938

1999

 7,486

   7,480

   6,365

2000

 8,682

   8,676

   7,479

2001

 9,016

   9,012

   7,830

2002

10,331

  10,330

   9,000

 

94,117

93,296

76,245

Finally, Table 5 compares the consumer cost and lives saved by the core group of FMVSS for cars and LTVs, combined.  Adding the right columns of Tables 1 and 3, the first column of Table 5 shows the core FMVSS added a total of $189,842,000,000 to the consumer cost of new cars and LTVs in model years 1968-2002.  They saved 252,989 lives in model year 1968+ cars during calendar years 1968-2002 (sum of 176,614 from Table 2 and 76,245 from Table 4).  That averages out to a cost of $750,782 per life saved in 1968-2002.

In 2002 alone, the core FMVSS added $11,353,000,000 to the cost of new cars and LTVs of that model year.  They saved 20,851 lives in the cars and LTVs on the road during that calendar year.  They cost $544,482 per life saved.  The cost per life saved generally decreased from year to year during 1968-2002, primarily because belt use increased: the more people buckle up, the more benefit the public will obtain from equipment they have already paid for whether they use it or not.


TABLE 5 – CONSUMER COST AND LIVES SAVED BY THE CORE FMVSS, CARS + LTVs, 1968-2002 (in 2002 Dollars)

Year

Cost Of The Core FMVSS

Lives Saved

 

 

 

1968

   $1,638,000,000

      185

 

 

 

1969

   $2,156,000,000

      438

 

 

 

1970

   $2,243,000,000

      648

 

 

 

1971

   $2,303,000,000

      874

 

 

 

1972

   $2,986,000,000

    1,173

 

 

 

1973

   $3,624,000,000

    1,431

 

 

 

1974

   $3,328,000,000

    1,494

 

 

 

1975

   $2,604,000,000

    1,948

 

 

 

1976

   $3,391,000,000

    2,068

 

 

 

1977

   $3,698,000,000

    2,303

 

 

 

1978

   $3,857,000,000

    2,545

 

 

 

1979

   $3,786,000,000

    2,641

 

 

 

1980

   $3,022,000,000

    2,745

 

 

 

1981

   $2,787,000,000

    2,692

 

 

 

1982

   $2,584,000,000

    2,477

 

 

 

1983

   $2,775,000,000

    2,639

 

 

 

1984

   $3,860,000,000

    3,117

 

 

 

1985

   $3,995,000,000

    4,516

 

 

 

1986

   $4,176,000,000

    6,341

 

 

 

1987

   $4,484,000,000

    7,551

 

 

 

1988

   $4,989,000,000

    8,624

 

 

 

1989

   $5,151,000,000

    8,895

 

 

 

1990

   $6,103,000,000

    9,106

 

 

 

1991

   $5,799,000,000

    9,604

 

 

 

1992

   $5,902,000,000

    9,909

 

 

 

1993

   $6,831,000,000

   11,121

 

 

 

1994

   $8,237,000,000

   12,276

 

 

 

1995

  $10,097,000,000

   13,053

 

 

 

1996

   $8,941,000,000

   14,539

 

 

 

1997

   $9,748,000,000

   15,228

 

 

 

1998

   $9,940,000,000

   16,066

 

 

 

1999

  $10,896,000,000

   16,505

 

 

 

2000

  $11,561,000,000

   18,168

 

 

 

2001

  $10,997,000,000

   19,088

 

 

 

2002

  $11,353,000,000

   20,851

$544,482 per life saved in 2002

 

 

 

$189,842,000,000

252,859

$750,782 per life saved in 1968-2002

 

 



[1] Kahane, C.J., Lives Saved by the Federal Motor Vehicle Safety Standards and Other Vehicle Safety Technologies, 1960-2002 – Passenger Cars and Light Trucks, NHTSA Technical Report No. DOT HS 809 833, Washington, 2004.

[2] Light truck or van, including pickup trucks, sport utility vehicles (SUVs), minivans and full-size vans.

[3] Tarbet, M.J., Cost and Weight Added by the Federal Motor Vehicle Safety Standards for Model Years 1968-2001 in Passenger Cars and Light Trucks, NHTSA Technical Report No. DOT HS 809 834, Washington, 2004.

[4] For example, MY 2002 registrations as of July 1, 2003; by then nearly all MY 2002 cars have been sold and few retired.  NHTSA only has Polk files back to calendar year 1975.  The numbers for MY 1968-73 are from MVMA Motor Vehicle Facts & Figures ’77, Motor Vehicle Manufacturers Association, Detroit, 1977, p. 34 (registrations in the calendar year following the model year).  They are also derived from Polk data, but NHTSA’s numbers for 1974 and 1975 are only 97.9% of MVMA’s numbers; thus, the MVMA numbers for 1968-73 have also been multiplied by .979.

[5] The cost for 2002 is assumed to be the same as in 2001 (the last year in the cost report), since there was little or no change in the specific FMVSS considered in the cost report.

[6] The 252 lives saved in 1968 include 12 lives saved during CY 1967 in MY 1968 cars sold before January 1, 1968.  These 12 were added to the 240 lives actually saved in calendar year 1968 for more direct comparability with the cost estimate (cost of the FMVSS for MY 1968 cars).

[7] Estimates for MY 1974-2002 are based on Polk NVPP registrations, for that model year, as of July 1 of the next calendar year.  The numbers for MY 1968-73 are from MVMA Motor Vehicle Facts & Figures ’77, p. 35.  The MVMA numbers include heavy trucks as well as LTVs; they are multiplied by .938 to make them comparable to subsequent NVPP numbers for LTVs only.  Furthermore, the NVPP file for LTV registrations on July 1, 1977 is not available to NHTSA.  We used the registrations of MY 1976 LTVs on July 1, 1978, multiplied by 1.0027 (the inverse of the average attrition of MY 1974 and 1977 LTVs from the first to the second calendar year after the model year).

[8] The cost for 2002 is assumed to be the same as in 2001, since there was little or no change in the specific FMVSS considered in the cost report.



pdf graphic The complete report is available here in pdf format.

Return to Regulatory Evaluation Page