Is democracy actually in retreat? There’s some evidence to the contrary

One could fill a library with the books and reports regarding the ongoing “crisis” of democracy. But what if democracy is not in decline? A new dataset appears to show quite the opposite.

Catastrophes and crises, real and imagined, trigger avalanches of proposed solutions. The most recent offer is “Constitutional Localism,” the idea that a devolution of power to local communities will solve the frustration and anger of the American voter.

Leaving aside my immediate questions and concerns about this proposal, Americans should take a step back and ask if proposed solutions to a perceived crisis are based on a false premise. What if democracy is actually doing fine and it is something else that needs correcting?

The REIGN (Rulers, Elections, and Irregular Governance) dataset by Curtis Bell at One Earth Future (OEF) Research would seem to show a twenty-year global expansion of democracy, directly contradicting Freedom House’s finding of a twelve-year democratic crisis. Whose data is more believable?

To unpack that charged question, first understand that methodologies produce the outcomes they are designed to.

Freedom House’s methodology is thorough, transparent, and has undergone stringent review processes. But the data points are ultimately generated from the assessments of “a team of in-house and external analysts and expert advisers from the academic, think tank, and human rights communities.”

The REIGN dataset was built from a variety of other widely-used datasets (Polity, Archigos, LEAD, and Geddes Wright Frantz) that were also ultimately created from expert assessments. REIGN is also different from Freedom House in that it measures by the month rather than the year, drills into the diversity of autocratic regime types, and recognizes that some forms of government are provisional.

My point: I am skeptical of all data, especially those claiming to measure highly politicized conceptual constructs like “democracy” or “freedom.” Perhaps before setting about to change the political order to favor more of something, we should consider the very real possibility that our thinking is based on a false premise.

In other words, we all should be very careful about drawing sweeping conclusions from datasets like Freedom House, REIGN, and others, which can lead analysts to radically different conclusions that stem from perfectly valid methodological differences.

Note: Jay Ufelder shared the REIGN dataset on Twitter, bringing it my attention.