Most attempts at democratic government around the world have ended with coups or rebellions that have returned the countries in question to autocratic rule. These breakdowns of democracy have often happened within a decade of the elections that marked the start of democratic government, and poorer countries are especially susceptible to them. Yet, in spite of those broad patterns, some attempts at democracy in poorer countries have succeeded, and some of those successes have occurred in countries identified by current theories of democratization as unlikely candidates.
Why do some democratic regimes survive in apparently inauspicious conditions while others do not? In 2010, I did a statistical analysis for the Political Instability Task Force to try to shed some light on this question. The “key findings” from that research, summarized below, confirmed some elements of the conventional wisdom but also included some surprises (to me, anyway). You can find a permalink to the full report on that project here, on the Social Science Research Network.
- Strong legislatures make democracies in ‘developing’ countries more resilient. By contrast, proportional representation and federalism have no discernable impact on democratic resilience, even in culturally heterogeneous societies.
- Other things being equal, democracies are more likely to survive when they fully respect their citizens’ rights of assembly and association and avoid using security forces and the judiciary to political ends. This relationship is not just true by definition; apparently, violations of these rights contribute independently to the risk of democratic breakdown, probably by fostering a political climate in which the incentives for political parties and the military to try to usurp power are strengthened.
- Surprisingly, government censorship of the media (within a range in which a country can still be considered democratic) has little effect on the risk of democratic breakdown.
- Stronger state protections for women’s rights—especially economic ones, such as the right to own property and the right to receive equal wages for equal work—are associated with democratic resilience.
- Democracies in ‘developing’ countries are more likely to survive when economic growth is producing real improvements in their citizens’ quality of life and are more likely to fail when quality of life is declining. Year-to-year variation in the delivery of social services such as education and health care does not seem to have any immediate and direct effect on democratic resilience, however, and neither does foreign aid.
- Membership in the Commonwealth and the World Trade Organization (WTO) also seems to bolster prospects for the survival of democracy in non-OECD countries. At least in part, these effects appear to stem from the vigorous democracy-promoting efforts of the former and the institutional changes demanded by the latter. By contrast, a stronger commitment on paper to protecting citizens’ civil and political rights—as evidenced by accession to the First Optional Protocol of the U.N.’s International Covenant on Civil and Political Rights—does not appear to translate into better prospects for the survival of democratic government, nor does membership in la Organization Internaionale de la Francophie or the Organization of the Islamic Conference.