THIS PAPER SUMMARIZES THE RESULTS OF A NUMBER OF STUDIES OF THE EFFECTIVENESS OF CONTROL OF DRUNKEN DRIVING IN NORWAY AND SWEDEN. ECONOMETRIC TECHNIQUES ARE USED FOR THESE EVALUATIONS WITHIN A SIMULTANEOUS SYSTEMS FRAMEWORK THAT TAKES ACCOUNT OF VARIATIONS IN LAW ENFORCEMENT EFFORT, LEVELS OF ALCOHOL COMSUMPTION, AND ENVIRONMENTAL EFFECTS THAT INCLUDE INDICES OF DISTANCE DRIVEN, VEHICLE MIX AND/OR TRAFFIC DENSITY, AND ROAD QUALITY. EFFECTIVENESS IS MEASURED IN TERMS OF THE IMPACT ON FATAL AND SERIOUS INJURY ACCIDENTS. BOTH CROSS-SECTION AND TIME SERIES ANALYSES HAVE BEEN CONDUCTED FOR PERIODS IN WHICH LEGAL STATUTES HAVE REMAINED RELATIVELY UNCHANGED. ACROSS ALL OF THE DATA SETS INVOLVED THERE IS A SUBSTANTIAL UNANIMITY SUPPORTING THE HYPOTHESIS THAT INCREASES IN ALCOHOL CONSUMPTION LEVELS ARE ASSOCIATED WITH HIGHER ACCIDENT LEVELS AND INCREASES IN LAW ENFORCEMENT EFFORT LEADING TO A GREATER PROBABILITY OF SANCTIONS WILL RESULT IN LOWER LEVELS OF FATAL AND SERIOUS INJURY ACCIDENTS THAN WOULD OTHERWISE PREVAIL.(Author/publisher).
Abstract