Our models are significant (global p-value [Log-Rank] < 0,10) and the Concordance indexes oscillate from 0.59 to 0.68, which are the habitual results when Cox Proportional Hazard Ratios satisfactorily fit. Additionally, each model passed the linktest diagnostic for misspecification.

Figure 2 shows that de jure emigrant enfranchisement positively correlates to democracy. As expected, higher levels of inclusion and contestation, along with higher indicators in terms of rule of law and liberal principles, result in higher odds of enacting external voting rights. We find liberal democracies are around two times more likely to enact voting rights for nonresident citizens (HR 2.1; p-value < 0.05), as compared to closed autocracies. While electoral democracy is insignificant in the model, electoral autocracies are also positively correlated with de jure emigrant enfranchisement, more than twice as likely as compared to closed autocracies (HR 1.9; p-value < 0.05). This is unsurprising since most closed autocracies such as North Korea and Saudi Arabia do not hold elections. A few exceptions appear, such as Algeria under the command of Houari Boumédiène (1965–1978), which passed a modest external voting rights provision in the 1976 Constitution (Brand, 2010).

Figure 3 complements our previous results by showing the timing of de facto emigrant enfranchisement. On this occasion, all types of political regimes are significant. Similar to our first model in which democracy was conducive to de jure emigrant enfranchisement, Figure 3 shows that liberal democracies are about five times more likely to implement an external voting provision (HR 5.1; p-value < 0.001) and around four times more likely if the country is classified as an electoral democracy (HR 3.7; p-value < 0.001). Although probabilities are reduced if the country is an electoral autocracy (HR 3.2; p-value < 0.01), it still represents a robust positive correlation to de facto emigrant enfranchisement. As in Model 1, closed autocracy is the reference category. Overall, most countries classified as autocracies display a negative outcome or ‘0’, when referring to implementing external voting rights, regardless of whether they had previously passed a provision to enfranchise their nonresident citizens. Certain countries, such as Angola and Nicaragua, promulgated provisions to organize external voting decades ago, but have not implemented corresponding legislation (Ellis et al., 2007; Palop-García and Pedroza, 2019).

Four Cox Proportional Hazard Models ensured our results remain the same when including another measurement of democracy. Table 1 demonstrates that the probabilities of positive correlation between democracy and emigrant enfranchisement remain high. Adding two dummy variables as controls, we corroborate that a democratic regime is a strong predictor of both enacting and implementing external voting rights. Surprisingly, belonging to both the EU and OECD does not have a significant impact on the odds for origin states to undertake de jure and de facto emigrant enfranchisement. So, how and why do some democracies choose not to pass and/or implement external voting rights, while some autocracies do?”

In the original article, there was a mistake in Figure 2 as published. Hazard ratio data were displayed incorrectly. The corrected Figure 2 appears below.

In the original article, there was a mistake in Figure 3 as published. Hazard ratio data were displayed incorrectly. The corrected Figure 3 appears below.

In the original article, there was a mistake in Table 1 as published. Hazard ratio data were displayed incorrectly. The corrected Table 1 appears below.

The authors apologize for this error and state that this does not change the scientific conclusions of the article in any way. The original article has been updated.