No, Democracy Has Not Been Discarded

Freedom House released the latest iteration of its annual Freedom in the World report yesterday (PDF) with the not-so-subtle subtitle “Discarding Democracy: Return to the Iron Fist.” The report starts like this (emphasis in original):

In a year marked by an explosion of terrorist violence, autocrats’ use of more brutal tactics, and Russia’s invasion and annexation of a neighboring country’s territory, the state of freedom in 2014 worsened significantly in nearly every part of the world.

For the ninth consecutive year, Freedom in the World, Freedom House’s annual report on the condition of global political rights and civil liberties, showed an overall decline. Indeed, acceptance of democracy as the world’s dominant form of government—and of an international system built on democratic ideals—is under greater threat than at any point in the last 25 years.

Even after such a long period of mounting pressure on democracy, developments in 2014 were exceptionally grim. The report’s findings show that nearly twice as many countries suffered declines as registered gains, 61 to 33, with the number of gains hitting its lowest point since the nine-year erosion began.

Once again, I’m going to respond to the report’s release by arguing that things aren’t nearly as bad as Freedom House describes them, even according to their own data. As I see it, the nine-year trend of “mounting pressure on democracy” isn’t really much of a thing, and it certainly hasn’t brought the return of the “iron fist.”

Freedom House measures freedom on two dimensions: political rights and civil liberties. Both are measured on a seven-point scale, with lower numbers indicating more freedom. (You can read more about the methodology here.) I like to use heat maps to visualize change over time in the global distribution of countries on these scales. In these heat maps, the quantities being visualized are the proportion of countries worldwide landing in each cell of the 7 x 7 grid defined by juxtaposing those two scales. The darker the color, the higher the proportion.

Let’s start with one for 1972, the first year Freedom House observes and just before the start of the so-called third wave of democratic transitions, to establish a baseline. At this point, the landscape has two distinct peaks, in the most– and least-free corners, and authoritarian regimes actually outnumber democracies.

Okay, now let’s jump ahead to 1986, the eve of what some observers describe as the fourth wave of democratic transitions that swept the Warsaw Pact countries and sub-Saharan Africa. At this point, the world doesn’t look much different. The “least free” peak (lower left) has spread a bit, but there are still an awful lot of countries on that side of the midline, and the “most free” peak (upper right) is basically unchanged.

A decade later, though, things look pretty different. By 1995, the “most free” peak has gotten taller and broader, the “least free” peak has eroded further, and there’s now a third peak of sorts in the middle, centered on 4/4.

Jump ahead another 10 years, to 2005, and the landscape has tilted decisively toward democracy. The “least free” peak is no more, the bump in the middle has shifted up and right, and the “most free” peak now dominates the landscape.

That last image comes not long before the run of nine years of consecutive declines in freedom described in the 2015 report. From the report’s subtitle and narrative, you might think the landscape had clearly shifted—maybe not all the way back to the one we saw in the 1970s or even the 1980s, but perhaps to something more like the mid-1990s, when we still had a clear authoritarian peak and a lot of countries seemed to be sliding between the two poles. Well, here’s the actual image:

That landscape looks a lot more like 2005 than 1995, and it looks nothing like the 1970s or 1980s. The liberal democratic peak still dominates, and the authoritarian peak is still gone. The clumps at 3/3 and 6/5 seem to have slumped a little in the wrong direction, but there are not significant new accretions in any spot. In a better world, we would have seen continued migration toward the “most free” corner over the past decade, but the absence of further improvement is hardly the kind of rollback that phrases like “discarding democracy” and “return to the iron fist” seem to imply.

Freedom House’s topline message is also belied by the trend over time in its count of electoral democracies—that is, countries that hold mostly free and fair elections for the offices that actually make policy. By Freedom House’s own count, the number of electoral democracies around the world actually increased by three in 2014 to an all-time high of 125, or more than two-thirds of all countries. Here’s the online version of their chart of that trend (from this page):

Again, I’m having a hard time seeing democracy being “discarded” in that plot.

So how can both of these things be true? How can the number of electoral democracies grow over a period when annual declines in freedom scores outnumber annual gains?

The answer is that those declines are often occurring in countries that are already governed by authoritarian regimes, and they are often small in size. Meanwhile, some countries are still making jumps from autocracy to democracy that are usually larger in scale than the incremental declines and thus mostly offset the losses in the global tally. So, while those declines are surely bad for the citizens suffering through them, they rarely move countries from one side of the ledger to the other, and they have only a modest effect on the overall level of “freedom” in the system.

This year’s update on the Middle East shows what I mean. In its report, Freedom House identifies only one country in that region that made significant gains in freedom in 2014—Tunisia—against seven that saw declines: Bahrain, Egypt, Iraq, Lebanon, Libya, Syria, and Yemen. All seven of those decliners were already on the authoritarian side of the ledger going into 2014, however, and only four of the declines were large enough to move a country’s rating on one or both of the relevant indices. Meanwhile, Tunisia jumped up two points on political rights in one year, and since 2011 its combined score (political rights + civil liberties) has gone up eight points, from 12 to 4. We see similar patterns in the declines in Eurasia, where nearly all countries already clustered around the “least free” pole, and sub-Saharan Africa, where only one country moved down into Freedom House’s Not Free category (Uganda) and two returned to the set of electoral democracies after holding reasonably fair and competitive elections (Guinea-Bissau and Madgascar).

In short, I continue to believe that Freedom House’s presentation of trends over time in political rights and civil liberties is much gloomier than the world its own data portray. Part of me feels like a jerk for saying so, because I recognize that Freedom House’s messaging is meant to be advocacy, not science, and I support the goals that advocacy is meant to achieve. As a social scientist, though, I also think it’s important that our analyses and decisions be informed by as accurate a sketch of the world as we can draw, so I will keep mumbling into this particular gale.

PS. If you want Freedom House’s data in .csv format, I’ve posted a version of them—including the 2014 updates, which I entered by hand this morning—on my Google Drive, here.

PPS. If you’re curious where I think these trends might be headed in the next 10 years, see this recent post.

PPPS. The day after I ran this post, I published another in which I tried to think of better ways to measure what Freedom House purports to describe in its annual reports. You can read it here.

Posted by Jay Ulfelder on January 29, 2015

https://dartthrowingchimp.wordpress.com/2015/01/29/no-democracy-has-not-been-discarded/