Testing the law of crime concentration in China: Do transportation networks matter?

While the Law of Crime Concentration (LCC) has been widely validated in Western contexts, less is known about its applicability and structural determinants in other settings. This study examines whether the LCC holds across Chinese cities and whether variation in transportation networks help explain differences in spatial crime concentration. A national dataset of 584,047 court-validated robbery, snatching, and theft cases from 2014 to 2019 was analyzed. Incidents were geocoded and aggregated to 1 km × 1 km grid cells across 333 prefectural-level units and 4 municipalities. Crime concentration was measured using generalized Gini coefficients and percentile indicators (P25/P50). Two-way fixedeffects models and lagged specifications were employed to assess both contemporaneous and delayed effects of transportation networks. The findings confirm a highly skewed and stable pattern of crime concentration, consistent with the LCC. This regularity persists across crime types, spatial resolutions, time periods, city hierarchies, alternative measurement strategies, and specific geographic locations. Denser and better-connected rail networks are significantly associated with lower levels of crime concentration in the contemporaneous year, while these effects weaken and become insignificant over longer lags. The pattern suggests that opportunity-diffusion effects operate more rapidly, whereas potential social-disorganization processes may take longer to materialize. This study provides the first macro-level evidence on spatial crime concentration and its temporal dynamics in China. The findings highlight the significant role of transportation networks in shaping the spatial distribution of crime, underscoring the importance of considering where crime occurs, not just whether it occurs, when designing crime prevention strategies. • The law of crime concentration holds across Chinese cities at the national scale. • Crime concentration remains stable across types, scales, time, city hierarchies, and specific geographic locations. • Gini coefficients are more robust than P25/P50 in measuring cross-city crime concentration. • Denser and better-connected rail networks are associated with lower crime concentration in the short term. • Opportunity-diffusion effects are immediate, while potential social-disorganization effects may take longer to emerge.