Backlash Hypotheses Sample Clauses

Backlash Hypotheses. ‌ A simple regression of protests on being under an IMF program in a given month yields the result, showing that a xxxx estimate indicates that being under an IMF program will result in 12 more protests and 5 more repressive events per month on average versus those months not under a program. This result is also statistically significant, pointing in the direction that the critics of globalization theorize. The second model has country and month level fixed effects, and the result remains statistically significant; albeit the estimate decreases from 12 to 2.5 protests and from 5 to 1.6 repressive events per month. I run these models to show the results without the use of matching and in order to illustrate the extent to which the estimates for protests and repression may be biased due to the lack of a valid counterfactual case. The models in Table 1 essentially produce the average increase in the number of protests or repressive events in those country months under IMF programs, versus all of those country months in which states were not under IMF programs. Though the models take into account relative covariates – the ones discussed in the previous sections – they do not use those covariates to inform the estimates writ large. That is, we are not necessarily making an appropriate comparison between groups and our estimates may be biased as a result. Dependent variable: Protests Repression