U.S. Republican presidential nominee Donald Trump holds a rally in Delaware, Ohio, on Thursday. (Jonathan Ernst/Reuters)

Talk of rigged elections may be new to American politics, but it is relatively common in many countries. In some, sham elections result in leaders receiving 99 percent of the vote. In others, results are fabricated so that the incumbent gets just enough votes to win. The damage done by rigged elections extends beyond the immediate electoral outcome, however; they also undermine public confidence in political institutions and processes.

But might rigged elections also undermine public confidence in democracy itself? The answer, as we report in a study of Algerian elections published in Comparative Political Studies, is that this is indeed the case. A belief that elections are not free and fair lowers agreement with the proposition that democracy, whatever its flaws, is the best political system.

Algeria may not seem like a natural case with which to examine the effect of rigged elections. It is not a democracy, and the military plays a dominant role in politic affairs. Yet, after a brutal civil war between the regime and Islamist opposition, there appeared to be a political opening during the run-up to the 2004 presidential election. The military did not endorse the incumbent’s bid for reelection, something it had done in the past, and stated that it would accept the election of any candidate. Nine candidates then announced their intention to run. The front-runners were Abdelaziz Bouteflika, the incumbent, and Ahmed Taleb-Ibrahimi, a moderate Islamist.

Hopes for a more open and honest political environment were soon crushed. The constitutional court declared invalid many of the signatures Ibrahimi had collected to have his name placed on the ballot, a determination that seemed unconvincing and politically motivated because Ibrahimi had received 1.2 million votes in the 1999 presidential election.

A less popular opponent of Bouteflika was permitted to run even though he ended up receiving far fewer votes than the number of valid signatures required to be on the ballot. This indicated that the scrutiny applied to Ibrahimi was not applied to other candidates and reinforced the belief of his supporters that his disqualification was politically motivated.

We consider the Algeria case to examine the link between perceptions of fairness in elections and attitudes toward democracy. We use data from three nationally representative public opinion surveys conducted in Algeria over time by the Arab Barometer, by the World Values Survey and as part of a project funded by the National Science Foundation. We employ multiple regression analysis, with appropriate controls, and then compare the results across the three surveys, conducted, respectively, in 2002, 2004 and 2006. We find that the belief that an election is rigged leads to a loss of support for democracy.

In 2002, before the presidential election, supporters of political Islam, who were Ibrahimi’s core constituency, were as likely as those who did not support political Islam to express support for democracy. By contrast, in 2004, a few months after the election, our analysis found that supporters of political Islam were significantly less likely to support democracy compared with other Algerians.

The 2006 Arab Barometer survey contained a question about the fairness of the last election and thus makes possible a more direct test of the link between judgments about the election and judgments about democracy. First, we reran the model employed in earlier surveys and again found that support for political Islam is inversely related to support for democracy.

We also ran a second model that included an evaluation of the fairness of the 2004 election, and, with this variable included in the analysis, the inverse relationship between support for political Islam and support for democracy disappeared. The finding that Islamists are just as likely as other Algerians to support democracy when electoral evaluation is controlled suggests, as we had expected, that support for political Islam was a proxy for judgments about the 2004 election in the 2004 survey.

Why does this matter? Democratic consolidation — the process whereby democracy becomes entrenched — depends on all key political groups believing that democracy is not only the best system but also “the only game in town.” Conversely, a loss of faith in democracy among a significant segment of the population weakens the foundation for democracy.

In countries like Algeria, holding competitive elections may give citizens experience with democratic procedures and be a step on the path toward democracy, but if these elections are perceived as unfair, they may not help to foster but rather serve to hinder the emergence of the citizen orientations that democracy requires.

Although the political systems of Algeria and the United States have few similarities, the broader lesson of our research is that a belief that elections are rigged has the potential to seriously undermine the public’s commitment to democracy. Moreover, any such effect might not fade away quickly. More than two years after the election in our study, those who felt the election was unfair continued to be more skeptical of democracy than their peers. Our study shows that a belief that elections are rigged — whether based on real flaws or on politically motivated fabrications — has important and detrimental consequences for the functioning of democracy.

Mark Tessler is the Samuel J. Eldersveld Collegiate Professor of Political Science at the University of Michigan. Michael Robbins is the director of the Arab Barometer.