This paper offers a cross-country analysis of street-level corruption. In line with earlier studies on general corruption (where data did not allow a distinction between street-level and high-level corruption, e.g. Treisman 2000 and La Porta et al . 1999), nine hypotheses based on general economic theories of law enforcement are tested using data from the International Crime Victims Survey (ICVS) held in forty countries (N = 27,456). The analyses employ multilevel logistic regression models. The results show that citizens living in countries with (a) high levels of economic development, (b) high levels of economic freedom, (c) long exposure to democracy, (d) a non-federal structure, (e) Protestant traditions, and (f) a British legal culture, experience less corruption. A relation between corruption levels and economic development, economic freedom, long democratic traditions, and a British legal culture, however, is found only when no distinction is made between industrialized and developing countries. As for individual characteristics, a higher risk of being victimized by corruption is correlated with a lower age and higher levels of education and income.