This dissertation uses theoretical and empirical tools to answer applied questions of design with an emphasis on issues relating to safety and crime. The first essay incorporates safety in implementation theory and studies when and how safe mechanisms can be designed to obtain socially desirable outcomes. I provide general conditions under which a social choice rule can be implemented using safe mechanisms. The second essay is an empirical study of how criminals respond to changing profitability of crime, a question that informs the policy debate on the most effective crime fighting methods. I find that the price elasticity of theft is about 1 in the short term and increases to about 1.2 over a seven-month horizon, suggesting that policies that directly affect crime profitability, such as policies that shut down black markets or those that reduce demand for illegal goods, can be relatively effective. The third essay shows that any standard implementation problem can be formulated as a question about the existence of a graph that solves a graph coloring problem, establishing a connection between implementation theory and graph theory. More generally, an implementation problem can be viewed as a constraint satisfaction problem, and I propose an algorithm to design simple mechanisms to solve arbitrary implementation problems. / text