MEAN VALUE OF OWNER-OCCUPIED HOUSING AND MEAN RENT WERE THE TWO DEPENDENT VARIABLES. EACH WAS ESTIMATED SEPARATELY IN THE BASIC EQUATION. USING FORWARD INCLUSION REGRESSION, ONE VARIABLE ENTERED THE EQUATION AT EACH STEP UNTIL NO VARIABLES WITH INSIGNIFICANT COEFFICIENTS REMAINED. RESULTS SHOWED THAT TWO TYPES OF CRIME (VANDALISM AND RESIDENTIAL BURGLARY) EXERTED A STRONG INFLUENCE ON THE DEMAND FOR OWNER-OCCUPIED RESIDENCES AND HENCE THE VALUE OF SUCH RESIDENCES IN MINNEAPOLIS. FOR EVERY 1 PERCENT OF AN OWNER-OCCUPIED UNIT FELL ABOUT $336. ACROSS MINNEAPOLIS CENCUS TRACTS, THE RESIDENTIAL BURGLARY RATE RANGED FROM 1 PERCENT TO 14 PERCENT OF RESIDENTIAL UNITS. SIMILARILY, EVERY VANDALISM INCIDENT PER 1,000 POPULATION WAS ASSOCIATED WITH A RESIDENCE VALUE DROP OF ABOUT $117. VANDALISM RATES VARIED FROM UNDER 1 PER 1,000 POPULATION TO ABOUT 30 PER 1,000. FACTOR ANALYSIS COMPARING PROPERTY CRIMES AND STREET CRIMES SHOWED THAT ONLY PROPERTY CRIME SIGNIFICANTLY AFFECTED THE VALUE OF OWNER-OCCUPIED HOUSING. FOR MEAN RENT, THE FACTOR REPRESENTING STREET CRIME HAD A STRONG DEPRESSING INFLUENCE. RESULTS SHOWED THAT CRIME IMPOSES WINDFALL LOSSES ON PROPERTY OWNERS. RESULTS ALSO SUGGESTED THAT A GIVEN CRIME CONTROL PROGRAM WILL MORE LIKELY GENERATE AN UNAMBIGUOUS NET SOCIAL BENEFIT ONLY IF IMPLEMENTED ON A CITYWIDE BASIS. NEVERTHELESS, ADDITIONAL RESEARCH IS NEEDED, ESPECIALLY ON SUCH ISSUES AS DISPLACEMENT EFFECTS OF CRIME CONTROLS AND WHETHER CERTAIN NEIGHBORHOODS ARE 'CRIMINOGENIC.' NOTES, A REFERENCE LIST, AND AN APPENDIX CONTAINING AN ESTIMATE OF LOST PROPERTY TAX REVENUE ARE INCLUDED. (CFW)