PurposeThe purpose of this paper is to complement the scant macroeconomic literature on the development outcomes of social media by examining the relationship between Facebook penetration and violent crime levels in a cross-section of 148 countries for the year 2012.Design/methodology/approachThe empirical evidence is based on ordinary least squares (OLS), Tobit and quantile regressions. In order to respond to policy concerns on the limited evidence on the consequences of social media in developing countries, the data set is disaggregated into regions and income levels. The decomposition by income levels included: low income, lower middle income, upper middle income and high income. The corresponding regions include: Europe and Central Asia, East Asia and the Pacific, Middle East and North Africa (MENA), Sub-Saharan Africa and Latin America.FindingsFrom OLS and Tobit regressions, there is a negative relationship between Facebook penetration and crime. However, quantile regressions reveal that the established negative relationship is noticeable exclusively in the 90th crime quantile. Further, when the data set is decomposed into regions and income levels, the negative relationship is evident in the MENA while a positive relationship is confirmed for Sub-Saharan Africa. Policy implications are discussed.Originality/valueStudies on the development outcomes of social media are sparse because of a lack of reliable macroeconomic data on social media. This study primarily complemented three existing studies that have leveraged on a newly available data set on Facebook.