Why do armed groups recruit large numbers of children as fighters, often coercively? The international community has tried to curb these crimes by shaming and punishing leaders who commit them—in short, making the crimes costlier. Are these policies effective and sufficient? The answer lies in more attention to the strategic interaction between rebel leaders and recruits. We adapt theories of industrial organization to rebellious groups and show how, being less able fighters, children are attractive recruits if and only if they are easier to intimidate, indoctrinate, and misinform than adults. This ease of manipulation interacts with the costliness of war crimes to influence rebel leaders' incentives to coerce children into war. We use a case study and a novel survey of former child recruits in Uganda to illustrate this argument and provide hard evidence not only that children are more easily manipulated in war, but also how—something often asserted but never demonstrated. Our theory, as well as a new “cross-rebel” data set, also support the idea that costliness matters: foreign governments, international organizations, diasporas, and local populations can discourage child recruitment by withholding resources or punishing offenders (or, conversely, encourage these crimes by failing to act). But punishing war crimes has limitations, and can only take us so far. Children's reintegration opportunities must be at least as great as adults' (something that demobilization programs sometimes fail to do). Also, indoctrination and misinformation can be directly influenced. We observe grassroots innovations in Uganda that could be models for the prevention and curbing of child soldiering and counterinsurgency generally.

Is it possible to detect manipulation by looking only at electoral returns? Drawing on work in psychology, we exploit individuals' biases in generating numbers to highlight suspicious digit patterns in reported vote counts. First, we show that fair election procedures produce returns where last digits occur with equal frequency, but laboratory experiments indicate that individuals tend to favor some numerals over others, even when subjects have incentives to properly randomize. Second, individuals underestimate the likelihood of digit repetition in sequences of random integers, so we should observe relatively few instances of repeated numbers in manipulated vote tallies. Third, laboratory experiments demonstrate a preference for pairs of adjacent digits, which suggests that such pairs should be abundant on fraudulent return sheets. Fourth, subjects avoid pairs of distant numerals, so those should appear with lower frequency on tainted returns. We test for deviations in digit patterns using data from Sweden's 2002 parliamentary elections, Senegal's 2000 and 2007 presidential elections, and previously unavailable results from Nigeria's 2003 presidential election. In line with observers' expectations, we find substantial evidence that manipulation occurred in Nigeria as well as in Senegal in 2007.

How do episodes of intergroup violence affect political opinions toward outgroup members? Recent studies offer divergent answers. Some suggest violence deepens antagonism and reduces support for compromise, while others contend it encourages moderation and concessions to prevent further conflict. We argue that violence can fuel both hostility toward the outgroup and acceptance of outgroup objectives and provide evidence from a unique survey of 1,380 respondents implemented by the authors in greater Khartoum in Sudan in 2010 and 2011. We find that Northerners who experienced rioting by Southerners in Khartoum in 2005 were more likely to support Southern independence but less likely to support citizenship for Southerners remaining in the North. In combination, these results suggest that political violence hardens negative intergroup attitudes and makes individuals willing to concede separation to avoid living alongside outgroup members.

International mediation of violent conflicts is commonplace in today's world, and so is academic research on its features and effectiveness. But research that speaks to both the initiation and implementation of mediation remains relatively rare. This article outlines a theoretical and empirical argument that contributes to filling this gap and suggests a counterintuitive selection effect: potential mediators that are likely to resolve a dispute are unlikely to select into mediation. The argument hinges on the claim that mediation by biased third parties is relatively ineffective, and I provide qualitative evidence to suggest that this claim is plausible.

United Nations policy forbids its peacekeepers and other personnel from engaging in transactional sex (the exchange of money, favors, or gifts for sex), but we find the behavior to be very common in our survey of Liberian women. Using satellite imagery and GPS locators, we randomly selected 1,381 households and randomly sampled 475 women between the ages of eighteen and thirty. Using an iPod in private to preserve the anonymity of their responses, these women answered sensitive questions about their sexual histories. More than half of them had engaged in transactional sex, a large majority of them (more than 75 percent) with UN personnel. We estimate that each additional battalion of UN peacekeepers caused a significant increase in a woman's probability of engaging in her first transactional sex. Our findings raise the concern that the private actions of UN personnel in the field may set back the UN's broader gender-equality and economic development goals, and raise broader questions about compliance with international norms.