This paper studies the relationship between unemployment and property crime rates for the United States using aggregate data from 1993 to 2022. Given the mixed results of previous studies, we construct a new unemployment rate: the difference between U-6 and U-3 unemployment rates. The U-6 unemployment measure includes marginally employed and underemployed workers as unemployed, differing from the U-3 unemployment measure which includes these individuals as employed. As such, our measure isolates the number of discouraged, underemployed and marginally employed individuals. We argue that this group is the appropriate population for whom the marginal benefit of property crimes may exceed marginal costs, relative to unemployed individuals according to the U-3 unemployment rate definition. The results of our double-difference error correction model confirm our hypothesis. We find a positive relationship between unemployment and total property crime rates, theft rates and burglary rates in the short and long run. Policy implications are discussed. JEL Code: C33, J69, Z10