The variables used in Peltzman's analysis were reviewed. II was concluded that some of them were arbitrarily chosen, that some were correlated. and that important factors were omitted. This ma) cause spurious and biased correlations. Peltzman's time series regression equations were reconstructed and found unstable. which makes them useless for predictions which are one basis for Peltzman's conclusions. The cross-sectional analyses were found to he unvalidated. Their results for important factors disagreed with those from the time series analysis. Peltzman's conclusions on the role of young drivers were compared with their actual accident involvement and found to disagree. To illustrate the misuse of trend models. a "model" for pedestrian deaths was constructed which ieads to conclusions contradicting Peltzman's. Sam Peltzman's claim that motor vehicle safety regulation has had no effect on the highway death toll, that it may have increased the share of this toll borne by pedestrians, and increased the total number of traffic accidents [Peltzman, 1975]i has attracted wide attention [Anonymous, 1975;Guzzardi, 1975; Irvin, 1975; U.P.I., 19751. I will demonstrate that his study provides no basis for his claim, even more, following Peltzman's approach I will develop an empirical "model" which shows that pedestrians too did benefit from the motor vehicle safety regulations (but I suggest that this model not be taken seriously). I will show that (i) Pehzman's analysis omits factors of known influence, (ii) he does not consider well known sources of bias and spurious relations, and that, therefore, the results of his regression analyses are meaningless, (iii) the results of his analyses disagree amongst each other and with independent information. 'Part of this study was supported by the Insurance Institute for Highway Safety. $This paper is based on a version of Peltzman's study which he kindly made available to me in 1974. 4Tentative empirical evidence to the contrary was found by The Center for the Environment and Man 119731: the frequency of singie car accidents for 1967 and later model year cars did not differ from or was lower than that for earlier model years.