The cycle of violence in the Second Intifada: Causality in nonlinear vector autoregressive models

We contest Jaeger and Paserman's claim (Jaeger and Paserman, 2008. The cycle of violence? An empirical analysis of fatalities in the Palestinian–Israeli conflict. American Economic Review 98(4): 1591–1604) that Palestinians did not react to Israeli aggression during Intifada 2. We address the differences between the two sides in terms of the timing and intensity of violence, estimate nonlinear vector autoregression models that are suitable when the linear vector autoregression innovations are not normally distributed, identify causal effects rather than Granger causality using the principle of weak exogeneity, and introduce the “kill-ratio” as a concept for testing hypotheses about the cycle of violence. The Israelis killed 1.28 Palestinians for every killed Israeli, whereas the Palestinians killed only 0.09 Israelis for every killed Palestinian.