Understanding Victory: Why Political Institutions Matter

In our book, Democracies at War, we asked the question: Why do democracies tend to win the wars they fight? We confirmed this pattern, first noted by David Lake in his "Powerful Pacifists" article, using statistical tests and numerous historical cases. ¹ Notably, this phenomenon confounds the traditional realpolitik fear that democratic liberalism is a luxury that states may be unable to afford. Our basic answer to the question is that democracies tend to win because they put themselves in a position to do so. The constraints that flow from democratic political structures lead the executives of liberal democracies to hesitate before starting wars, particularly wars where victory on the battlefield appears to be less than clear-cut.

Democracies' willingness to start wars only against relatively weaker states says nothing about the actual military efficiency or capacity of democratic states. Rather, it says that when they do start a fight, they are more likely to pick on relatively weaker target states. We also find, however, that in addition to this "selection effects" explanation of democratic success, democratic armies enjoy a small advantage on the battlefield.

Michael Desch, a prominent realist scholar, reviews these claims in his article "Democracy and Victory: Why Regime Type Hardly Matters." ² His assertion that regime type is irrelevant to the probability of military victory is consistent with the broader realist agenda, which argues that domestic politics matters little in the formation of foreign policy or the interactions between states. Desch makes a valuable contribution in advancing the debate over this question. There are many points about which Desch and we agree. Democracies do not win wars because of economic superiority, greater public support during wartime, or having more or better allies. "Realist" factors such as military strategy, leadership, and industrial capabilities directly account in part for the war outcomes we observe. Regarding the irrelevance of regime type, however, we fail to find Desch's claims convincing because his analysis contains serious errors in both logic and method. In the next section, we review our central argument and explain how best to test it empirically. In the following section, we take up and rebut Desch's critiques of our primary data set. We then address some of Desch's other critiques.

Our Theory and Our Tests

How have democracies managed to compile their impressive wartime record? Generally, there are three possible answers. First, they might simply pick on weaker states, thereby arriving on the battlefield with more of the material determinants of victory. We call this argument the "selection effects" explanation. Because democratic political institutions make elected leaders more vulnerable to losing office, their foreign policies are different from those of leaders of other kinds of states. Specifically, they tend to represent more closely the values and preferences of the people than do the foreign policies of more authoritarian states. States where the elites must generate the contemporaneous consent of the people tend to avoid starting wars in which they have only a moderate or low chance of victory, so that when they go to war, they win.

Second, democracies might be materially more powerful. Their armed forces might fight with greater efficiency or better strategies, they might be able to extract relatively more resources from their societies, or they might choose allies that are relatively more reliable or powerful. The third possibility is that both of these explanations are partly correct.

The theories are disarmingly simple. Testing them is also quite straightforward. To test our theories, we examined all interstate wars that occurred between 1815 and 1985. We also looked closely at a number of specific cases to be sure that we were not spuriously attributing causal power to the correlations we uncovered. To test the selection effects theory, we looked in particular at the record of all war initiators, as the theory implies that democratic war initiators, being more risk averse, should be more likely than autocratic war initiators to win...