Thank you--I think the Japan case is especially compelling.

Very interesting report! Thanks for reading.

Thanks for this comment Michael--I think you make a great point about risks from malevolent actors. In terms of the longermist economic growth aspect, I was thinking more along the lines of institutional quality in the 1600's explaining a lot of the more recent economic growth trajectories, with substantial consequences for global poverty.

Thank you so much for this comment. The evidence from a bunch of good papers seems to suggest that it's about incentives to make democratic reforms. In terms of whether EA could contribute enough money---the Carnegie and Marinov paper I cite finds small but still noticeable improvements in democracy in response to relatively insignificant increases in conditional aid from the EU (for example, going from 20 million dollars in aid to 25 million).

Thank you for this comment--I think one advantage of the Acemoglu et. al. (2019) paper published in the Journal of Political Economy (https://economics.mit.edu/files/16686) is that it accounts for the economic crises that generally precede transitions to democracy--democracies initially grow slowly as they emerge out of depressions, and then grow faster as they invest in capital, education and health. One reason existing studies had found mixed effects is that they hadn't properly accounted for these dynamics. I think there are also strong second-order effects on global growth of democracy because democracies impose lower tariffs on other countries--especially on other democracies--and democracies rarely if ever fight each other. Peace and trade are likely to enhance global growth--though these externalities may be difficult to properly measure.

I think this a really strong argument in favor of democracy promotion. Thank you for your comment!

Thank you so much for these thoughtful comments! A few responses:

1. While there are of course differences of opinion on this issue outside of the research community, the social science research literature universally considers Russia and Iran to be non-democratic (see for example, the Polity IV project or the recent Acemoglu et. al. 2019 Democracy dataset). These regimes might be considered "competitive authoritarian" regimes (see Way and Levitsky) or hybrid regimes/"anocracies"--the benefits of democracy stated in the article do not apply to these states. While liberal democracy is likely preferable for outcomes like democratic peace, other outcomes like higher spending on public goods are linked primarily to electoral democracy, rather than to liberal norms.

2. In terms of the democratic peace--it's true that not all scholars agree with the consensus about the democratic peace--though Mearsheimer is definitely an outlier to the extent of believing power politics to be the only thing that matters (i.e. he thinks Europe isn't at war because of US troops in Germany). Scholars that have traditionally emphasized power politics (like Robert Jervis) acknowledge that the current situation-- in which powerful countries in Europe/Japan/South Korea don't even contemplate war against one another--is unique historically and likely linked to democratic norms.

3. I agree on China--I think pro-democracy aid can be effective when a country is in transition, like a mixed regime (and when a country actually needs aid enough to be influenced). Conditional aid can boost the fragile institutions of new democracies and make democratic consolidation more likely ( a very important long term outcome).