The Evolution of Political Regimes, Freedom House Version

A year and a half ago, I posted animated heat maps that used Polity data to look at the evolution of national political regimes at the global level over the past two centuries (here and here). Polity hasn’t posted new data for 2013 yet, but Freedom House (sort of) has, so I thought I’d apply the same template to Freedom House’s measures of political rights and civil liberties and see what stories emerged.

The result is shown below. Here are a few things to keep in mind when watching it:

• The cells in each frame represent annual proportions of all national political regimes worldwide. The darker the gray, the larger the share of the world’s regimes that year.
• Freedom House’s historical depth is much shallower than Polity’s—coverage begins in 1972 instead of 1800—so we’re missing most of the story the Polity version told about the advent and spread of contemporary democracy in the 19th and 20th centuries. Oh, well.
• The order of the Freedom House indices is counter-intuitive. One is most liberal (“freest”) and 7 is least. So in these plots, the upper right-hand corner is where you’d find most of Europe and North America today, and the lower left-hand corner is where you’ll find what Freedom House calls “the worst of the worst.”
• One year (1981) is missing because Freedom House made some changes to its process around that time that meant they effectively skipped a year.
• For details on what the two measures are meant to represent and how they are produced, see Freedom House’s Methodology Fact Sheet.

Now here are a few things that occur to me when watching it.

• The core trend is clear and unsurprising. Over the past four decades, national political regimes around the world have trended more liberal (see this post for more on that). We can see that here in the fading of the cells in the lower left and the flow of that color toward the upper right.
• You have to look a little harder for it, but I think I can see the slippage that Freedom House emphasizes in its recent reports, too. Compared with the 1970s, 1980s, and even 1990s, the distributions of the past several years still look quite liberal, but it’s also evident that national political regimes aren’t marching inexorably into that upper right-hand corner. Whether that’s just the random part of a process that remains fundamentally unchanged or the start of a sustained slide from a historical peak, we’ll just have to wait and see. (My money’s on the former.)
• These plots also show just how tightly coupled these two indices are. Most of the cells far from the heavily populated diagonal never register a single case. This visual pattern reinforces the idea that these two indices aren’t really measuring independent aspects of governance. Instead, they look more like two expressions of a common underlying process. (For deep thoughts on these measurement issues, see Munck and Verkuilen 2002 and Coppedge et al. 2011 [gated, sorry].)

You can find the R script used to produce this .gif on GitHub (here) and the data set used by that script on Google Drive (here). Freedom House hasn’t yet released the 2013 data in tabular format, so I typed those up myself and then merged the results with a table created from last year’s spreadsheet.

Posted by Jay Ulfelder on February 13, 2014

https://dartthrowingchimp.wordpress.com/2014/02/13/the-evolution-of-political-regimes-freedom-house-version/