After 37 years in power, a coup has finally toppled Zimbabwe’s longtime leader, Robert Mugabe. Many Zimbabweans and foreign observers are celebrating Mugabe’s downfall as an opportunity for economic revival and political reform. So, how likely is it that we’ll see significant democratization in Zimbabwe in the next few years?

We usually try to answer questions like this one by digging into the details of the case of interest. How organized and cohesive are pro-democracy forces right now? Do they have any allies inside the government or security forces? Will ordinary people turn out to demand change, and how will they respond to arrests and violence?

The details of the case at hand are what psychologist Daniel Kahneman calls the “inside view.” As Philip Tetlock and Dan Gardner describe in their book Superforecasting, “It’s natural to be drawn to the inside view. It’s usually concrete and filled with engaging detail we can use to craft a story about what’s going on.”

This detail-oriented, narrative-building approach contrasts with the “outside” view. From this farther vantage point, we try to identify the broader class of events or situations to which our case belongs. If we can identify a broader reference class, we can summarize how things unfold and turn out in them, ideally with tables and numbers.

Most of us eschew the outside view, or we only turn to it after we’ve reached a preliminary conclusion based on the case’s details, as a way to check our work. This is usually a mistake. As Tetlock and Gardner explain, our conclusions rarely deviate much from our initial estimates, and the inside view is an unreliable path to that initial estimate. When we start with the outside view and then use case knowledge to fine-tune that first guess, we tend to get more accurate results.

Most of us eschew the outside view, or we only turn to it after we’ve reached a preliminary conclusion based on the case’s details, as a way to check our work. This is usually a mistake.

Koto's Analogy Engine offers a rigorous but efficient way to take the outside view on situations like the one in Zimbabwe right now. We can use it to quickly identify historical cases where coups have occurred under structural conditions like Zimbabwe’s, and then to see how various conditions—including a country’s degree of democracy—changed (or not) over the ensuing several years.

The exported chart below summarizes the results. The Analogy Engine identified a half dozen cases in the past 25 years where successful coups occurred amid conditions fairly similar to contemporary Zimbabwe’s, as measured by their level of economic development (low), degree of democracy (also low), level of corruption (high), and how long the ousted leader had held office (a long time). Among that handful of similar cases, two (Togo 2005, Mauritania 2005) experienced significant democratic gains over the ensuing five years; two (Guinea 2008, Egypt 2011) saw little net change over that period; and two (Rwanda 1994, Guinea-Bissau 1999) saw substantial declines. In short, the average outcome among cases in this set was approximately no democratization.


We can also use the similar cases identified by the Analogy Engine to look for clues about forces that might push Zimbabwe up the y-axis in that chart of post-coup trajectories. Unfortunately, the qualitative evidence is not much more encouraging. Togo in 2005 is the case most similar to Zimbabwe now, according to our similarity-scoring algorithm, and it’s also the one that saw the most democratization in the five years following the coup. In Togo, sustained social unrest kept pressure on the post-coup government, but key foreign donors’ decisions to condition their aid on political reform probably had a stronger effect. Mauritania was the other major gainer among the five most-similar cases, and its democratization arguably followed from concerns about the loss of Western aid as well.

That causal chain from conditional aid to democratization is unlikely to recur in Zimbabwe. In contrast to Togo and Mauritania in the 2000s, contemporary Zimbabwe depends relatively little on aid from donors who might press for a democratic transition. Instead, Zimbabwe has much stronger financial ties to China and Russia, neither of which is likely to lean on Harare to reform.

In short, the outside view tells us that democratic gains are possible in Zimbabwe, but we shouldn’t expect them, and a deeper slide into dictatorship is about as likely. Exploring Zimbabwe’s particulars might lead us to nudge that estimate upward or downward, but a large body of evidence tells us that this pessimistic view is a smart place to start.

Jay Ulfelder

Political science Ph.D. (Stanford 1997), research director for the US government-funded Political Instability Task Force (2001-2011), Good Judgment Project superforcaster.