Working Paper: NBER ID: w13992
Authors: William Easterly; Shanker Satyanath; Daniel Berger
Abstract: Do superpower interventions to install and prop up political leaders in other countries subsequently result in more or less democracy, and does this effect vary depending on whether the intervening superpower is democratic or authoritarian? While democracy may be expected to decline contemporaneously with superpower interference, the effect on democracy after a few years is far from obvious. The absence of reliable information on covert interventions has hitherto served as an obstacle to seriously addressing these questions. The recent declassification of Cold War CIA and KGB documents now makes it possible to systematically address these questions in the Cold War context. We thus develop a new panel dataset of superpower interventions during the Cold War. We find that superpower interventions are followed by significant declines in democracy, and that the substantive effects are large. Perhaps surprisingly, once endogeneity is addressed, US and Soviet interventions have equally detrimental effects on the subsequent level of democracy; both decrease democracy by about 33%. Our findings thus suggest that one should not expect significant differences in the adverse institutional consequences of superpower interventions based on whether the intervening superpower is a democracy or a dictatorship.
Keywords: superpower interventions; democracy; Cold War; CIA; KGB
JEL Codes: O1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Superpower interventions (F51) | Declines in democracy (O17) |
| US interventions (F51) | Declines in democracy (O17) |
| Soviet interventions (P33) | Declines in democracy (O17) |
| Distance from Moscow (R53) | US interventions (F35) |
| US security perimeter (F52) | Soviet interventions (P33) |