In Print: Incentives for Policy Pandering By Brandice Canes-Wrone March 2006

The assumptions of the Conditional Pandering Theory produce a relatively intricate set of policy incentives . . . In the theory, the president and electorate either agree about the appropriate course of action or have different beliefs about which policy should be selected. When their beliefs conform, the president’s incentives are straightforward. His desire for reelection and a positive historical legacy induce him to choose the popular policy, which he expects will produce a good outcome. It is when the president thinks a popular policy will ultimately fail that his incentives can be more complex. In this circumstance, he faces motivations to pursue the option he believes will succeed, but he can also have motivations to pander to public opinion.

The incentives for pandering are related to the electoral motivations. When the president is not running for reelection (i.e., when he is in his second term), his desire for a positive historical legacy induces him to support the policy he believes to be in citizens’ interests even if it is currently unpopular. Of course, in reality American presidents who are serving a second term may behave as if they are running for reelection. For example, a chief executive could want his vice-president or another fellow partisan to succeed him . . .

Although the incentives for pandering derive from the president’s electoral motivations, voters in the theory paradoxically want to elect a candidate who is unlikely to pander to them. Their interests are best served by having a leader who will choose the optimal policy, regardless of whether it is popular. The president’s incentive to pander to public opinion derives from the electorate’s uncertainty about whether the president or the challenger is most likely to choose good policies. Voters only have an initial estimate of each candidate’s level of competence and update their estimate of the president’s competence immediately after his policy decision. At that point, they increase their assessment of him if he chooses the policy they believe to be in their interests, regardless of the accuracy of the belief.

The president thus faces two types of tradeoffs when his information implies citizens are misinformed about the optimal course of action. The first tradeoff concerns the distinction between short- and long-run public opinion. Immediately after the president enacts a popular policy, voters think more highly of him. However, as soon as the outcome of the policy becomes known, their evaluations depend entirely on whether it succeeded or failed. The president’s concern about short-run public opinion thus encourages him to pursue the policy he believes will produce a good outcome. The second tradeoff involves the president’s desire for a strong historical legacy. Under certain conditions, this desire may propel the president to choose an unpopular policy that he believes will produce a good outcome. Under other conditions, however, the president would prefer to choose the popular policy in order to get reelected for another term, during which he could establish a favorable legacy.

Precisely when will the president’s optimal strategy be to exercise policy leadership as opposed to pandering? [These incentives are] a function of his popularity relative to that of his challenger. This relative popularity can be divided into three main categories: highly popular, unpopular, and marginally popular. The first category refers to situations in which the incumbent’s relative popularity is high enough that voters would reelect him even if he enacted a policy that lacked public support and they did not learn the consequences of the policy. The unpopular category encompasses the inverse set of circumstances. In that case, the president’s relative popularity is low enough that he would not be reelected if he supported a popular policy that did not produce an outcome prior to the election. Marginally popular presidents are the ones whose relative popularity is between these two thresholds. In this category, if voters did not learn before the election whether the president’s choice was a good one, they would reelect him if and only if he had chosen the option they initially supported.

. . . One might suppose that the electorally motivated presidents always have the incentive to increase their popularity by pandering to public opinion. However, the Conditional Pandering Theory predicts that a highly popular president will not pander. Such an incumbent will lose an impending election only if voters realize beforehand that he has made a bad policy choice. Therefore, he has the incentive to support policies he believes will succeed. By doing so, he minimizes the probability that a policy failure occurs before voters cast their ballots . . .

Given highly popular presidents’ incentives to avoid policy pandering, one might expect unpopular presidents to be strongly motivated to engage in this behavior. After all, an unpopular incumbent cannot retain his position unless he improves voters’ perception of him, and pandering would increase his perception right after the policy decision is announced. An unpopular president’s optimal strategy, however, is to exercise policy leadership. These presidents fall into two subgroups. In the first, the president is so unpopular that regardless of his policy decisions, he will be removed from office. Given this lack of electoral motivations, the incumbent’s desire for a strong historical legacy induces him to pursue the program he believes will succeed. The second subgroup concerns presidents who, while unpopular, may still retain office if they can prove their policy competence to voters. For these presidents, the best way to demonstrate such competence is to endorse an option that succeeds before the election. An unpopular incumbent’s electoral incentive is therefore to choose the option recommended by his policy information and hope voters learn his judgment was sound before casting their ballots.

Consider the case of an unpopular president who is deciding whether to deploy U.S. troops to fight anti-U.S. rebels in another country. The president believes the rebels could pose a danger to U.S. interests and has information suggesting the operation could be accomplished without a loss of troops. The public, while it does not support the rebels, believes they do not pose any danger and would prefer not to send troops because of fears of casualties. If the president wages the mission and it is accomplished without casualties, voters’ evaluations of the president’s competence will rise dramatically. If, on the other hand, the president follows public opinion, his popularity will increase slightly until voters learn whether the decision was a good one, but the increase is not sufficient to win him reelection. Moreover, if the president panders to public opinion and voters learn before the election that the lack of military engagement has harmed American interests, then the pandering actually decreases his prospects for reelection.

The Conditional Pandering Theory thus finds that highly popular and unpopular presidents do not engage in policy pandering. The only presidents who do so are marginally popular ones, and even they do not always pander. For example, given a sufficiently high probability that voters will learn before the election whether a policy choice succeeded, even marginally popular presidents want to exercise policy leadership. In this circumstance, voters will likely base their electoral decisions on the policy outcome, not the choice in and of itself. Presidents therefore want to endorse the option most likely to produce a good outcome.

It is only when voters are unlikely to learn before the election whether a policy choice succeeded that a marginally popular president will pander. Under these circumstances, the increase in approval the incumbent receives from promoting a popular policy makes him likely to win reelection. The behavior still augments the possibility of a policy failure before the election and, in expectation, detracts from his historical legacy. However, because voters are unlikely to learn about the policy failure before they cast their ballots, the boost in popularity overwhelms these drawbacks. The president wants to pander now and hopefully create a favorable historical legacy in the next term.

Excerpted from Who Leads Whom? Presidents, Policy, and Public Opinion, by Brandice Canes-Wrone, published in the United States by the University of Chicago Press. © 2006 by the University of Chicago. All rights reserved.