Zamina Aliyeva

The share of crimes against human life and health on average is up to 10% worldwide, and losses are estimated from 0.3 to 3% of GDP. This study examines the dependence of the rate of violence and crime against human life and health on the state policy elements in the example of transitioning and developing countries. The crime index, the share of people reporting crime, the rate of violence or vandalism in the area, and the number of intentional homicide offenses in the largest cities were used as parameters characterizing the rate of violence and crime against human life and health. All parameters were divided into institutional, social, and economic. The dependence between the indicators was studied using fixed-effects and random-effects models; a grouping of countries according to the nature of this dependence employed the iterative separation method of k-means and tree clustering. Based on the results, it is justified that institutional and economic (highest GDP and real minimum wages) components significantly influence the level of violence and crime against human life and health. For example, the average value of the crime index for the fourth cluster is 29.98 compared to 54.09 for the first cluster. At the same time, strengthening responsibility for committed crimes has a more negligible impact on the crime level than increasing the material well-being of the population and supporting its vulnerable segments.

State policy of prevention, detection, termination, disclosure, and investigation of crimes against a person in Azerbaijan should be based on other countries’ best practices and experience. The choice of countries to be followed by Azerbaijan should be very well-founded, given that the dynamics of crimes against a person depend significantly on many social and economic determinants: income inequality, the dominance of the rule of law in the country, the level of literacy and financial literacy of citizens, or racial diversity.
50 countries are clustered according to the similarity of trends regarding the dependence of crimes against persons on these socio-economic determinants. Clustering is based on data of 2021 from the World Bank, World Population Review, UNODC, and WGI (the selection of countries is due to the availability of comparable statistical information, the choice of year – to the availability of the most up-to-date data). Clustering was carried out using two methods (DBSCAN and K-Means) to ensure the adequacy of the calculations. Clustering is performed for 3 combinations: 1) by the entire set of crimes and their determinants; 2) by the specific type of crime and all types of determinants; 3) by the entire set of crimes and a specific socio-economic determinant. Albania, Jordan, Mongolia, Romania, and Serbia were most often in the same cluster with Azerbaijan. Therefore, the best experience and best practices of these countries can be used by the state regulatory bodies of Azerbaijan in developing state policy on preventing crimes against the person.