This study examines the efficiency of tools for fighting software piracy in the conditional distributions of software piracy. Our paper examines software piracy in 99 countries over the period 1994–2010, using contemporary and non-contemporary quantile regressions. The intuition for modelling distributions contingent on existing levels of software piracy is that the effectiveness of tools against piracy may consistently decrease or increase simultaneously with the increasing levels of software piracy. Hence, blanket policies against software piracy are unlikely to succeed unless they (...) are contingent on initial levels of software piracy and tailored differently across countries with low, medium and high levels of software piracy. Our findings indicate that GDP per capita, research and development expenditure, main intellectual property laws, multilateral treaties, bilateral treaties, World Intellectual Property Organisation treaties, money supply and respect for the rule of law have negative effects on software piracy. Equitably distributed wealth reduces software piracy, and the tendency not to indulge in software piracy because of equitably distributed wealth increases with the increasing software piracy levels. Hence, the negative degree of responsiveness of software piracy to changes in income levels is an increasing function of software piracy. Moreover, the relationships between policy instruments and software piracy display various patterns: U-shape, Kuznets-shape, S-shape and negative thresholds. A negative threshold represents negative estimates with the increasing negative magnitude throughout the conditional distributions of software piracy. We also discuss the policy implications of our study. (shrink)