This book's core purpose is to develop new indicators of electoral fraud. The authors argue that existing methods of measuring such fraud are insufficient and should be supplemented by a new approach. This approach uses official electoral data disaggregated to the lowest possible administrative units and assesses the likelihood that the patterns observed in these data are natural or contrived.

Three specific indicators are used. First, the authors look at the frequency distribution of turnout across electoral districts. If the data are genuine, we would typically expect this distribution to be normal. Deviations from normality—such as a substantial tail of districts reporting very high turnouts—give cause for suspicion. Second, the authors look at how each candidate's share of the total electorate correlates with turnout. If voting figures are reported accurately, we would expect a candidate's share of the electorate to increase with turnout in rough proportion to his or her overall vote share: if turnout increases by 100 votes, we would expect a candidate who scores 60% of the vote overall to pick up around 60 of these votes. If, by contrast, turnout is inflated by the stuffing of ballot boxes, the proportion of additional votes captured by the favored candidate will be far higher. Third, the authors use econometric methods to estimate the flow of votes among parties or candidates from one election to the next. Again, we have cause for concern if a candidate picks up many votes from an improbable source.

These indicators usefully add to our ability to identify and gauge electoral fraud. The cases that the authors explore—primarily recent elections in Russia and Ukraine, but briefly also the United States—illustrate this value well. The Russian cases do not tell us anything that we did not know already, but demonstrate that the indicators do pick up fraud in cases where we know it to have existed, particularly in the presidential elections of 2004 and the Duma elections of 2007. Similarly, the Ukrainian cases capture known fraud, too: The manipulation of the second round of the presidential election of 2004, which triggered the “Orange Revolution,” is clearly visible. The Ukrainian cases also identify possible instances of fraud that have received less attention. Evidence points toward fraud on behalf of the pro-Western Yushchenko as well as the pro-Russian Yanukovich in 2004. The patterns observed suggest continuing, though narrowly targeted, fraud in the parliamentary elections of 2007. Thus, the methods used do more than merely tell us what we already knew anyway; they also prompt us to investigate particular localities and issues in greater depth.

The authors fully acknowledge that their approach has limitations. It cannot simply be applied across cases without considerable case-specific knowledge: “[T]here is no black box—no index or magical configuration of numbers—into which one inputs the data and out of which emerges a grading of an election's legitimacy” (p. 271). This is because the natural distribution of data can vary widely between cases: There is no ideal pattern to which data from actual elections can be compared. For example, while a non-normal distribution of turnout might signal electoral fraud, it could also be entirely natural; there could be two types of district (say, urban and rural), which tend, respectively, toward low and high turnout. Investigation of these possibilities, as the authors repeatedly argue, requires considerable context-specific understanding. Because of this, they offer their approach as a supplement—not an alternative—to ground-level observation of election procedures: “[O]ur methods can at best augment the studies of those with substantive knowledge and first-hand experience” (p. 270).

In addition, the authors emphasize that their methods can detect only large-scale fraud, using the example of the United States as illustration. These methods cannot tell us whether there has been fraud in recent U.S. presidential elections since any such fraud would have been too limited or too subtle to be picked up (e.g., p. 237). This greatly narrows the scope for the methods to tell us anything new; large-scale fraud is likely to be evident by other means.

This relates closely to a further constraint that the authors hint at in the final pages. The method identifies several patterns in election data that should arouse suspicions. But does this not mean that future fraudsters who wish to remain undetected will simply avoid these patterns? They will, for example, manipulate election returns while maintaining a broadly normal distribution of reported turnout. Certainly, such manipulations require far more central coordination than the ad hoc and often locally motivated tampering observed particularly in Russia. But this simply adds to the impression that these methods will capture only some types of fraud. They will capture locally driven and incompetent fraud, not fraud that is sophisticated and managed from the center.

These various limitations raise significant doubts about the extent to which the methods developed here are likely to generate new knowledge. Still, as I have said, the empirical analyses of Russia and Ukraine show that these methods can have real value. Used in conjunction with other sources of evidence, such as the reports of journalists and external observers, they may corroborate suspicions or point us toward issues that need further attention. In an area where powerful empirical evidence is inevitably difficult to come by, the addition of further methods, even if flawed, is to be welcomed.

Two presentational issues need also to be raised. First, the book sometimes lapses, particularly in the first two chapters, toward the polemical. The authors laud objective analysis (e.g., p. 18), but they do not always practice it. They are notably scathing about external election monitors (p. 14), even though they later argue that we need such monitors (see pp. 226–27). They cast irrelevant aspersions on the New York Times (p. 8). They say of the 2008 Russian presidential contest that “calling it an election denigrates the meaning of the word” (p. 6), but they make no attempt to verify this assertion through their own methods. Given the importance of judgment in applying the methods that the authors propose, such intimations of prior commitment to particular conclusions are unfortunate.

Second, the authors barely engage with the existing literature on their subject. The bibliography has just 43 entries, 19 of which are self-references. In consequence, the case that existing measures of electoral fraud are inadequate is never properly made, the possibility of building on the work of others is not explored, and the opportunity to understand how the processes studied here relate to the broader phenomenon of electoral manipulation is not taken.

In short, The Forensics of Election Fraud is a valuable but also a limited and frustrating book. It develops useful methods, but the utility of these methods is tightly bounded. It makes a significant contribution to a rapidly growing literature, but the authors' understanding of that contribution is not made explicit.