(2) host state’s international security threat increases the chance of withdrawal. In order to test the hypotheses, I use quantitative and qualitative research methods. An event history analysis is used to analyze the entire universe of cases controlling for potentially confounding factors. Qualitative examination of the Philippines case is conducted to provide additional confirmation to the statistical analysis and shed light on underlying causal mechanisms.The results of the analysis offer considerable support for the hypotheses. Host state’s high-level domestic political unrest and international security threat increase the risk of foreign nuclear withdrawals. The findings show some support for alternative explanations. Patron states are less likely to withdraw nuclear weapons from host states with which they share a common rival. Nuclear weapons are more likely to be withdrawn during Mutual Assured Destruction and post-Cold War periods. The results of this study carry important implications for our understanding of foreign nuclear withdrawals and nuclear force structures in general. At the same time, in the aspects of realpolitik, this study can help understand the situational conditions when nuclear weapons will be withdrawn.