Cross-country analysis of spambots

Abstract

Spam is a vector for cybercrime and commonly legally prohibited. Why do certain national jurisdictions produce a higher percentage of spam than others despite its prohibition? Why do some countries have a higher percentage of systems acting as spambots compared to other countries? We begin to answer there questions by conducting a crosscountry empirical analysis of economic factors that correlate with the prevalence of spam and associated botnets. The economic factors under consideration are grounded in traditional theories of crime offline, as well as prior research in security economics. We found that more than 50% of spam can be attributed to having originated from merely seven countries, indicating that deterrence through policy is both feasible and economically rational. As expected, higher Internet adoption is correlated with higher percentage of spam from a country. Counterintuitively, Internet adoption is also positively correlated with the percentage of infected machines.