FORMULATIONS PORTRAYING THE ECONOMICS OF THEFT AND PROTECTION CHOICES ARE DEVELOPED. VARIABLES IN THE FORMULA FOR THE ECONOMICS OF THEFT INCLUDE A SET OF ALTERNATIVE TARGETS, EXPENDITURE OF RESOURCES (TIME, ENERGY, AND OTHER INPUTS), NUMBER OF THIEVES, COMPETITION OF THIEVES, NUMBER OF THEFTS, AND TARGET PROTECTION AND WEALTH VARIABLES WHICH REPRESENT A TRANSFER FUNCTION. GIVEN THESE VARIABLES, A FORMULA IS DERIVED TO COMPUTE GAINS FROM THE THEFT. IMPLICIT IN THE FORMULATION IS TOTAL GAIN FOR EACH THIEF. ASSUMING MAXIMIZATION OF WEALTH BY THE INDIVIDUAL, PROPERTY PROTECTION VARIABLES CAN THEN BE USED TO PREDICT DISTRIBUTION OF THEFT ACTIVITY ACROSS TARGETS. SIMILARLY, A FORMULA BASED ON A COST-BENEFIT ANALYSIS OF SECURITY DECISIONS AND LOSS PARAMETERS IS DERIVED WHICH DESCRIBES THE LEVEL OF SECURITY THAT RESULTS IN WEALTH MAXIMIZATION FOR EACH TARGET OWNER WHEN THEFT OCCURS UNDER COMPETITIVE CONDITIONS. THIS OPTIMAL AMOUNT OF SECURITY GENERALLY LEAVES A RESIDUAL NUMBER OF THEFTS WHICH WILL OCCUR BECAUSE ON THE AVERAGE THEY WILL EARN A COMPETITIVE NET RETURN. USING THE ECONOMETRIC MODEL IT CAN BE SHOWN HOW THEFT AND PROTECTIVE FORCES SIMULTANEOUSLY DETERMINE THE LEVEL OF THEFT AND THE AVERAGE RETURN OF THEFT ACTIVITIES, AS WELL AS HOW THE EQUILIBRIUM WOULD RESPOND TO SOME CHANGE IN EXTERNAL FACTORS. THE EFFECTS OF THE FOLLOWING CHANGES ARE ANALYZED: (1) INCREASE IN THE COST OF ENGAGING IN THEFT SUCH AS WOULD FOLLOW INCREASES IN ALTERNATIVE WAGES OR AN INCREASE IN CRIMINAL PUNISHMENTS; (2) INCREASE IN THE COST OF ENGAGING IN THEFT AGAINST SELECTED TARGETS; (3) INCREASE IN THE PRICE OF SECURITY DEVICES; AND (4) INCREASE IN THE STOCK OF WEALTH BEING PROTECTED BY EACH TARGET OWNER. TO TEST THE IMPLICATIONS OF THE MODEL, A CROSS-SECTIONAL ANALYSIS OF BANK ROBBERY DATA FROM 1966 THROUGH 1971 IS PERFORMED. IN GENERAL, FINDINGS ARE SUPPORTIVE OF THE MODEL'S HYPOTHESIZED RELATIONSHIPS. WHETHER A MORE SOPHISTICATED ECONOMETRIC FORMULATION WOULD YIELD GREATER EXPLANATORY POWER, AND WHETHER THE THEORY IS CONSISTANT WITH A WIDER RANGE OF DATA THAN USED IN THE ANALYSIS PRESENTED WILL REQUIRE FURTHER TESTING. RESULTS OF A TIME-SERIES ANALYSIS OF THE BANK ROBBERY DATA ARE PROVIDED IN AN APPENDIX. TABULAR DATA BUT NO REFERENCES ARE INCLUDED. (JAP)