Bolivia’s Electoral Fraud Reckoning
While the preponderance of evidence indicates that Bolivia's October 2019 presidential election was marked by rampant fraud, a recent study by two researchers suggests otherwise. But the study's assumptions are questionable, and, even if correct, the authors' results do not show absence of fraud.
WASHINGTON, DC/EDINBURG, TEXAS – Bolivia finds itself at a critical juncture. After its long-serving president, Evo Morales, resigned and fled the country last November in the wake of a presidential election marred by alleged irregularities, a transitional government scrambled to fill the power vacuum, stopped violent clashes, and swiftly called for new elections.
Although Bolivia could have descended further into chaos and violence, the warring political parties somewhat miraculously reached a fragile détente and agreed to hold an electoral do-over in May. But the controversy surrounding the October 2019 presidential election recently reignited after researchers John Curiel and Jack R. Williams (C&W) claimed in The Washington Post to have “found no reason to suspect fraud.”
Given its narrow scope, the C&W study cannot, in fact, dispel doubts about the election’s fraud. Nonetheless, the researchers’ widely publicized blanket statement detonated in Bolivia like a cluster bomb, rekindling antagonisms and encouraging agitators. It is therefore imperative to scrutinize the study’s analysis and conclusions.
For starters, the C&W study lacks recent historical context. Morales’s ascent to the presidency in 2006 was indisputably democratic, and represented a victory for Bolivians who had been politically and economically marginalized throughout the country’s history. But as Morales tightened his grip on power, he started undermining democratic institutions.
Flouting term limits and using a questionable interpretation of the constitution, Morales won a third term in 2014. Emboldened, he called a referendum in 2016 to allow him to run for a fourth term. Although he lost, Morales, defiant, called for a repeal of term limits and the Constitutional Court caved in, thus invalidating the plebiscite and allowing him to run again. Only the most credulous believe that his party did not try to prevent a repeat of the referendum result. Indeed, the fact that Morales was even on the ballot at all last October indicated that fraud was afoot.
In addition, the preponderance of evidence that has since emerged points to rampant electoral fraud. An audit by the Organization of American States (OAS) revealed major irregularities and manipulation, including falsification of poll officials’ signatures, altered tally sheets and databases, and a broken chain of custody. Most damning of all, the transmission of voting data was redirected to two unauthorized hidden servers. European Union election monitors also noted worrisome irregularities. But despite all these red flags, C&W focused instead on the vote-reporting blackout on election night – a single event that had little to do with structural fraud.
Project Syndicate is conducting a short reader survey. As a valued reader, your feedback is greatly appreciated.
When Bolivia’s preliminary vote-counting system – called TREP – reported that Morales’s lead was insufficient to avoid a run-off, the Electoral Court shut down all further reporting. Yet, when reporting resumed a day later, the margin had grown large enough for Morales to declare victory.
C&W purport to show that there was no fraudulent intervention during the blackout – that the votes reported after it were similar enough to those reported beforehand to make fraud unlikely. But even if their analysis were convincing, it would not counter or discredit the ample evidence of fraud in other aspects of the electoral process.
The one claim for which C&W offer empirical support is that there is no evidence of a change in voting trends after the TREP blackout. However, we question their findings, for three reasons.
First, their results are suggestive, not conclusive. Both the OAS’s audit and the organization’s later response to C&W highlight an observable trend change after 84% of the TREP tally, and a remarkable break in trend at 95% of the tally.
Although C&W reject these observations, alternative methods of analysis corroborate the OAS’s findings. In a recent working paper with the University of Oklahoma’s Gary Hoover, one of us tested the null hypothesis of no electoral fraud during the blackout, and validated the results with pseudo-outcomes using a placebo analysis. Unlike C&W, we allowed for heterogeneous voting preferences across different regions of Bolivia without making assumptions regarding patterns among precincts. And we found statistically significant evidence of fraud.
Consistent with our results and the OAS report, Rómulo Chumacero of Universidad de Chile found that the TREP blackout produced an inexplicable margin in favor of Morales. C&W’s study therefore should be regarded skeptically in view of the robust evidence at variance with their results.
Second, lack of evidence of trend-change after the TREP blackout is not evidence of an absence of fraud. If fraud was already baked into the 84% of votes reported before the interruption, then C&W’s results suggesting no shift in trend when reporting resumed are incapable of shedding light on the existence of fraud before the blackout.
Third, C&W estimate voting behavior in a critical subset of small precincts not reported before the blackout (accounting for 2.2% of total ballots) by sampling precincts of similar size included in the 84% batch. If electoral fraud is easier to engineer in small precincts, C&W are unwittingly extrapolating artificially wide margins to Morales’s advantage.
C&W have produced a clever but inconclusive analysis that fails to address both relevant history and evidence of widespread fraud. Above all, they have stretched their conclusions beyond credibility. Given Bolivia’s delicate political situation, researchers participating in this debate should proceed with great caution.