This blog post examines how voters’ ideological leanings influence the strategic manipulation decisions of those in power, analyzing under what conditions democracy begins its path toward regression.
The erosion of democratic values by elected politicians using legal procedures constitutes a crisis for democracy as severe as a sudden, violent coup. In modern political processes, those in power possess the capacity to weaken democracy’s foundations through various institutional means. This ‘manipulation’ is particularly insidious and deadly because it formally maintains the appearance of legality. The manipulation referred to here is not overtly illegal acts, but rather actions that are not clearly unlawful—such as amending regulations on the right to vote or eligibility for office, intervening in public opinion through media regulation, or increasing partisan influence over state institutions—that unfairly skew the political process to enhance the likelihood of staying in power. Let us examine two models that describe the process of democratic regression resulting from the combination of such manipulation by those in power and the responses of voters.
First, the Svolik model posits that voters decide whom to support based on utility inversely proportional to the distance between a candidate’s policy ideology and their own. Crucially, when judging incumbents, voters factor in the extent of manipulation, calculating a utility reduction proportional to the level of manipulation. In other words, it assumes voters possess an inherent preference not only for policy ideologies but for democratic values themselves. For example, consider a scenario where a right-wing incumbent engages in manipulation in a given country. For a moderate right-wing voter, the utility loss from the erosion of democracy may outweigh the potential ideological dissatisfaction that could arise if a left-wing challenger were to take power. In this case, the moderate right-wing voter is highly unlikely to support the incumbent. Conversely, extreme right-wing voters, viewing a left-wing challenger’s victory as the worst-case scenario, are far more likely to vote for the incumbent even at the cost of democratic erosion.
Meanwhile, an incumbent who chooses manipulation to secure re-election faces a situation where the effect of gaining votes through manipulation and the effect of losing votes due to the erosion of democracy caused by manipulation act simultaneously. This loss of votes primarily occurs among centrist or center-right voter groups. Ultimately, the incumbent calculates the gains and losses of manipulation and determines the extent of manipulation at the level that secures the maximum votes in the election. Recent studies have reconfirmed that this strategic calculation is frequently observed in actual phases of democratic regression.
Meanwhile, the Loureiro and Szebrowski model posits that voters cast ballots based on a candidate’s overall ‘attractiveness,’ which encompasses policy positions and competence. Even if citizens do not inherently value democracy itself within this model, they still tend to prefer being governed by more appealing politicians. Consequently, citizens prioritize the capacity to choose their rulers—that is, the capacity for democracy. This implies that value is placed on the democratic capacity of the society itself. During an election, voters face a trade-off between the utility derived from the attractiveness of the currently selected candidate—that is, the quality of the government formed through the election—and the utility derived from future democratic capacity—that is, the ability of citizens to replace the ruling party through elections whenever a more attractive challenger emerges.
When a more appealing challenger emerges, citizens desire a change in leadership. Consequently, leaders with authoritarian tendencies are inclined to engage in manipulation during their tenure. This increases the likelihood of maintaining power despite citizen opposition—the so-called incumbent premium. This model starts with a premium of zero, sets up a situation where the game between the leader and the electorate repeats, and concludes that democracy is threatened in both cases where the incumbent’s appeal is very high or very low.
When the incumbent’s appeal is high, citizens are highly satisfied and judge challengers are less likely to be more appealing. At this point, the incumbent feels little burden to engage in manipulation, a phenomenon termed ‘regression in support’. Conversely, when the incumbent’s appeal is low, if the incumbent wins several elections by luck, causing the incumbent premium to exceed a certain threshold, citizens become concerned about prolonged rule. Consequently, they desire regime change in the upcoming election, even if the challenger’s appeal is somewhat low. Anticipating this, the incumbent resorts to all possible manipulation to further increase the premium, termed ‘regression amid opposition’. This analysis reflects a trend widely observed in recent cases of democratic backsliding and has established itself as a significant explanatory framework in academia.
In both types of regression, authoritarian-inclined incumbents advance in the direction of regression to the point where they risk being replaced by means such as coups or popular uprisings, ultimately plunging the foundations of democracy into a severe crisis. Such theoretical models demonstrate that democracy does not collapse abruptly but can be gradually eroded through strategic interactions between rulers and voters. They continue to provide crucial insights for analyzing the democratic backsliding observed in numerous countries today.
