In a recent New Republic piece entitled “The Great Democracy Meltdown: Why the World Is Becoming Less Free,” Joshua Kurlantzick responds to the optimism spurred by the so-called Arab Spring by trying to set the record straight about recent trends in the global spread of democracy:
The Arab Spring is something of a smokescreen for what is taking place in the world as a whole. Around the globe, it is democratic meltdowns, not democratic revolutions, that are now the norm. (And even countries like Egypt and Tunisia, while certainly freer today than they were a year ago, are hardly guaranteed to replace their autocrats with real democracies.) In its most recent annual survey, the monitoring group Freedom House found that global freedom plummeted for the fifth year in a row, the longest continuous decline in nearly 40 years. It pointed out that most authoritarian nations had become even more repressive, that the decline in freedom was most pronounced among the “middle ground” of nations—countries that have begun democratizing but are not solid and stable democracies—and that the number of electoral democracies currently stands at its lowest point since 1995.
The idea that political rights and civil liberties are slipping around the world is not Kurlantzick’s invention. In 2008, democratization scholar-cum-advocate Larry Diamond published a paper in the influential journal Foreign Affairs called “The Democratic Rollback” in which he argued that the “democratic wave” triggered by the end of the Cold War “has been slowed by a powerful authoritarian undertow, and the world has slipped into a democratic recession.” Diamond’s paper echoed themes found in Freedom House’s annual “Freedom in the World” reports, which for the past few years have sounded alarm bells about threats to gains in “freedom” that occurred in the early 1990s.
I’ve got to say, though: I’m just not seeing it. When Diamond talks about a “democratic rollback,” I imagine a noticeable expansion followed by a sharp reversal. When Kulantzick writes that global freedom has “plummeted” for five years in a row, it implies a dramatic decline in political rights and civil liberties around the world. Plots of statistics summarizing the Freedom House data show neither.
Let’s start with a plot of global-average scores on Freedom House’s two indices, political rights and civil liberties. These indices range from 1 to 7, with 1 indicating the freest and 7 the least free. To make the plot more intuitive, I have inverted those scores here so 7 is freest and 1 is least free, but I have kept the original scaling.
I look at that plot, and I see a period of major gains in the early 1990s; a period of slower gains in the late 1990s and early 2000s; and something like a plateau to minor slippage since the mid-2000s. We can see the reported five-year decline in civil liberties referenced in Freedom House’s press release and Kurlantzick’s story, but the decline is slight–only about one-tenth of a point on a six-point scale, or a couple of percentage points–and the global average remains well above all other observed values except the previous few.
In addition to its scalar measures of political rights and civil liberties, Freedom House also makes a yes/no call on whether or not each country’s national political regime qualifies as an electoral democracy, which boils down to holding competitive and mostly free and fair elections. The following plot shows the percentage of states worldwide that qualified for this designation from 1989 (when Freedom House began tracking it) through year-end 2010.
Here, the downward trend noted by Freedom House (and amplified by Kurlantzick) is more evident. Since peaking in 2006-2007 at 64%, this measure has fallen steadily, to 59% at the end of 2010. (The counts on which those percentages are based are 123 of 193 countries in 2006-2007 and 115 of 194 countries in 2010.) Nevertheless, the only major trend clearly evident in that plot is the shift from the Cold War era to the post-Cold War era. In historical perspective, the slippage of the past few years looks potentially worrisome, but it is not yet large and is hardly dramatic. The end result is a percentage that’s largely unchanged since the mid-1990s, hovering around 60% for the past 15 years. It’s hard for me to think of such subtle shifts as a “rollback” or “recession,” let alone a series of “plummets.”
Perhaps the global summary statistics are masking more dramatic declines in electoral democracy in specific parts of the world. To see if that’s the case, the following chart compares proportions of electoral democracies in each region in 2005 and 2010, the start and end points of the five-year trend called out in Freedom House’s latest report. Here we can see that the largest drop occurred in sub-Saharan Africa, which suffered a net loss of 6 democracies (from 21/45 to 15/45) during that five-year period. Smaller but still noticeable drops occurred East Asia and Latin America, with net losses of 2 democracies in each (from 12/21 to 10/21 and from 21/23 to 19/23, respectively). The Near East and South Asia saw a net loss of just 1 democracy (from 4/25 to 3/25), and Europe and the former Soviet Union actually had a net gain of 1 democracy (from 36/46 to 38/49).
So, nearly all regions have suffered a reversal in the past five years in the long-term trend toward democratic government, and some of those reversals have been larger than others. What’s striking to me about those regional variations, though, is how consistent they are with historical patterns in democratic breakdown. Other things being equal, attempts at democracy are much more likely to fail in poorer countries than in richer ones, and they usually fail in their second, third, or even fourth election cycles–that is, between four and 20 years after they start. In other words, the slippage we’ve seen in the past several years is happening where and when we would have expected it to happen, given that so many of those democracies were “born” in a wave of transitions that occurred in the early 1990s. If those reversals were to continue until they had reversed most or all of the post-Cold War gains, then we should be both surprised and alarmed. In the meantime, while we can and should care about each reversal for its own sake, we should also be careful to keep short-term shifts in proper perspective so we can avoid making historical mountains out of line-chart molehills.
If the scope and scale of democratic backsliding in recent years are less dramatic than the “democratic recession” meme suggests, how did that meme get started? I haven’t tried to trace the idea in press coverage, but my hunch is that Freedom House deliberately amplified the small changes in the data to dramatize real concerns they had about what they saw as a loss of momentum in the democratization process in many countries in the 2000s. It is also worth noting that, as a nonprofit group, Freedom House relies on grants and donations for its funding, more than half of which comes from the U.S. Government. As such, Freedom House as an organization arguably stands to benefit when law-makers and other donors believe that democracy and freedom are profoundly threatened.
I’m sympathetic to Freedom House’s goals, and I don’t begrudge their attempt to “leverage” their analysis to support their advocacy and action work. What I do begrudge are press accounts and other analyses that take those advocacy-driven statements as scientific facts. Data aren’t inherently objective, and advocacy and rigorous analysis don’t naturally complement each other. Freedom House is first and foremost an advocate and implementer of democracy-supporting programs, so it would be nice if we didn’t lean so hard on their own analysis of their data (or even their data, period) when trying to assess global trends in governance.
UPDATE: See this follow-up post for links to some other views on the subject and my response to one of them.