This paper examines the determinants of identity theft, focusing especially on the influence of internet diffusion. Results, based on panel data across the U.S. states, show that a 10% in increase households with internet access would increase identity theft by about 9%, ceteris paribus. Other noteworthy findings point to states with greater corrupt activity having greater identity theft but greater police employment not having a significant deterrent impact. Dynamic panel regressions results reveal the presence of inertia in identity thefts. Some implications for policy are discussed.