Â鶹ÒùÔº

Voters value intentions over outcomes in reelection decisions, experiment suggests

Are you better off than you were before the last election?

For incumbent politicians seeking reelection, it's a consequential question. But voters—at least those from the same party—assess records with more nuance, new Cornell research finds. In experiments involving more than 3,000 participants, the researchers found voters also give significant weight to incumbents' intentions—potentially more than to policy outcomes they've experienced.

"We care not only about what politicians do, but what they try to do—seeing that effort to improve our everyday lives," said Talbot Andrews, assistant professor in the Department of Government in the College of Arts and Sciences. "But we are more forgiving of co-partisan incumbents for bad outcomes outside their control. When evaluating opposing party members, we care more if they've delivered benefits or not."

Andrews is a co-author of "The Road to Reelection is Paved with Good Intentions: Experiments on the Role of Outcomes and Intentions in Voting Behavior," Dec. 6 in the Journal of Politics, with Scott Bokemper, an Iowa-based independent scholar.

To investigate the role of intentions in electoral accountability, the researchers designed online games presenting a simplified election environment, with players acting as either incumbents or voters. Given different incentives, incumbents chose between "good" or "bad" policies awarding voters more or fewer tokens if enacted after they won reelection.

Voters knew the policy outcome, and some saw the incumbent's choice—their intention—along with the probability that it would be enacted. They then voted either for the incumbent, earning more if the "good" policy was enacted, or for a computer challenger providing a fixed payout.

The results showed voters responded to policy outcomes: They were 11 percentage points more likely to vote for incumbents producing a good outcome. But across outcomes, they were twice as likely to vote for incumbents with good intentions, if they had that information.

A second game removed voters' incentive to see policy enacted, asking them to choose solely based on outcomes and intentions. This time, voters were 12.4 percentage points more likely to support incumbents providing good outcomes, regardless of intentions—but nearly 50 percentage points more likely to vote for incumbents with good intentions.

Why? The researchers propose that when evaluating incumbents, voters use the same psychology observed in cooperative social interactions, in which we seek out partners who can work together and deliver benefits, understanding they may not always succeed in an inherently noisy environment.

"This is a fundamental way that humans evolved to cooperate with each other, and it influences how we hold our elected officials accountable," Andrews said. "From these experiments, we can say that intentions matter. The question then is how they matter in a more complex, real-world political environment."

A third game introduced partisan competition, assigning incumbents and voters to orange or purple groups. Results showed both rewarded good outcomes compared to bad ones. But while good intentions gave same-group incumbents a big boost—17 percentage points—they hardly helped counterparts from the other group.

"Up against partisan competition, the reward for good intentions really holds for members of your own political party," Andrews said. "If the incumbent is a member of the opposing political party, then we might only be willing to keep them if we're better off than before they were elected."

On the road to reelection, the research suggests, incumbents should claim credit for positive outcomes (as they do), but also signal their intentions through promises and highlighting even unsuccessful legislative efforts, to inform evaluations by voters from the same party.

"Voters are sophisticated enough to reward that kind of effort," Andrews said, "even if they might not personally feel the benefits of those policies."