Reynaerts and Vanschoonbeek (2016) propose a semi-parametric procedure to estimate the economic
impact of secession, finding empirical evidence that declaring independence significantly lowered per capita GDP in newly formed states. To demonstrate that these findings appear to hold irrespective of the estimation procedure employed, this addendum formulates a parametric
approach to estimate the independence dividend. Our preferred parametric specifications comprise
a dynamic, quasi-myopic model of per capita GDP dynamics that controls for country and year fixed effects, the rich dynamics of GDP, finite anticipation effects and a vector of alternative growth determinants. The results indicate that declaring independence reduces per capita GDP by around
15-20% in the long run. These results are qualitatively confirmed when we use non-regional secession waves to instrument for local incentives to secede.