Our students have produced a wealth of descriptive statistics and data visualizations to help users navigate the Democratic Erosion Event Dataset. You can find more examples in the Texas A&M capstone reports from 2018 and 2019, which also include some suggested uses of the dataset.
All Events by Year
The current version of the dataset captures 1,763 unique democratic erosion-related events in 98 countries between 2000 and 2018. Precursors and symptoms of democratic erosion constitute 42% and 38% of the dataset, respectively, while acts of resistance comprise the remaining 20%.
The number of events jumps after 2012. This is due in part to the increasing number of countries exhibiting precursors and symptoms of democratic erosion in recent years. But it is also partly due to bias towards more recent news coverage in the country case studies from which the dataset is built.
While the dataset does not encompass all democratic erosion-related events in these countries, it is the most systematic and granular attempt so far to capture the dynamics of democratic erosion across countries and over time. A description accompanies each event, making the dataset a powerful tool for analyzing democratic erosion both quantitatively and qualitatively.
All Events by Type
Events are classified into 44 “types.” Of these, corruption, nonviolent protest, and media repression are the most prevalent. Media repression is the most common event in the dataset, recorded nearly 200 times. The fact that the dataset is compiled from news reports suggests a possible bias towards media-related democratic erosion, as it is in the interest of news sources to expose these events.
Media Repression Worldwide
Media repression occurs in all regions, but it is most prevalent in Russia and India, followed by Guyana, Ecuador, and Turkey. Qualitative descriptions of the events reveal that government officials in Guyana have launched several high-profile lawsuits against the press for investigations into corruption, and have pressured private news sources to conform to the state media.
Horizontal and Vertical Events by Region
Each event in the dataset is characterized as “horizontal” or “vertical.” Horizontal events refer to relationships between the branches of government, while vertical events refer to the relationship between the government and citizens and civil society organizations. Vertical erosion is far more common than horizontal erosion around the world. This may be due to the fact that horizontal erosion requires structural changes in government, while vertical erosion, such as media repression, can occur more spontaneously.
Latin America and South/East Europe display similarly high levels of vertical erosion, driven in particular by restrictions on the independence of the judiciary. Horizontal erosion events in Latin America tend to involve attempts by the president to extend term limits. In Eastern Europe, in contrast, horizontal erosion usually involves efforts by the legislature to undermine checks and balances and stifle opposition parties.
Nicaragua and Bolivia account for most of the horizontal erosion events in Latin America. Both of these countries have long been ruled by strongmen presidents. In contrast, in South/East Europe, horizontal erosion events are concentrated in countries with more recently elected executives, such as in Hungary and Macedonia.
Precursors, Symptoms and Resistance
The dataset also allows us to explore the relationship between precursors, symptoms, and acts of resistance against democratic erosion. In these plots, the bubbles represent the number of resistance events in each country, ranging from 0 to 14 (South Korea). The dashed line shows the correlation between the number of precursors and the number of symptoms. In the plot above, we test the correlation between precursors and symptoms in the same period. In the plot below, we instead test the correlation between symptoms in the present (after 2012) using precursors in the past (before 2012).
The correlation is weak in both cases, with an R2 of under 0.05. Taken together, these plots seem to suggest that the transition from precursor to symptom is not inevitable or automatic. Democracy may erode without much warning. Or, conversely, democracy may remain intact even in the face of threats.
Precursors (pre-2012), Symptoms (post-2012) and Resistance
