Figure 1. In 2016, Republican presidential candidate Donald Trump became the center of the media’s horserace coverage. As the field winnowed from over twenty candidates down to three, the media incessantly compared everyone else in the field to Trump. (credit: Max Goldberg)

PUBLIC OPINION AND ELECTIONS

Elections are the events on which opinion polls have the greatest measured effect. Public opinion polls do more than show how we feel on issues or project who might win an election. The media use public opinion polls to decide which candidates are ahead of the others and therefore of interest to voters and worthy of interview. From the moment President Obama was inaugurated for his second term, speculation began about who would run in the 2016 presidential election. Within a year, potential candidates were being ranked and compared by a number of newspapers.^[3] The speculation included favorability polls on Hillary Clinton, which measured how positively voters felt about her as a candidate. The media deemed these polls important because they showed Clinton as the frontrunner for the Democrats in the next election.^[4]

During presidential primary season, we see examples of the bandwagon effect, in which the media pays more attention to candidates who poll well during the fall and the first few primaries. Bill Clinton was nicknamed the “Comeback Kid” in 1992, after he placed second in the New Hampshire primary despite accusations of adultery with Gennifer Flowers. The media’s attention on Clinton gave him the momentum to make it through the rest of the primary season, ultimately winning the Democratic nomination and the presidency.

Polling is also at the heart of horserace coverage, in which, just like an announcer at the racetrack, the media calls out every candidate’s move throughout the presidential campaign. Horserace coverage can be neutral, positive, or negative, depending upon what polls or facts are covered. During the 2012 presidential election, the Pew Research Center found that both Mitt Romney and President Obama received more negative than positive horserace coverage, with Romney’s growing more negative as he fell in the polls.^[5] Horserace coverage is often criticized for its lack of depth; the stories skip over the candidates’ issue positions, voting histories, and other facts that would help voters make an informed decision. Yet, horserace coverage is popular because the public is always interested in who will win, and it often makes up a third or more of news stories about the election.^[6] Exit polls, taken the day of the election, are the last election polls conducted by the media. Announced results of these surveys can deter voters from going to the polls if they believe the election has already been decided.

Public opinion polls also affect how much money candidates receive in campaign donations. Donors assume public opinion polls are accurate enough to determine who the top two to three primary candidates will be, and they give money to those who do well. Candidates who poll at the bottom will have a hard time collecting donations, increasing the odds that they will continue to do poorly. This was apparent in the run-up to the 2016 presidential election. Bernie Sanders, Hillary Clinton, and Martin O’Malley each campaigned in the hope of becoming the Democratic presidential nominee. In June 2015, 75 percent of Democrats likely to vote in their state primaries said they would vote for Clinton, while 15 percent of those polled said they would vote for Sanders. Only 2 percent said they would vote for O’Malley.^[9] During this same period, Clinton raised $47 million in campaign donations, Sanders raised $15 million, and O’Malley raised $2 million.^[10] By September 2015, 23 percent of likely Democratic voters said they would vote for Sanders,^[11] and his summer fundraising total increased accordingly.^[12]

Presidents running for reelection also must perform well in public opinion polls, and being in office may not provide an automatic advantage. Americans often think about both the future and the past when they decide which candidate to support.^[13] They have three years of past information about the sitting president, so they can better predict what will happen if the incumbent is reelected. That makes it difficult for the president to mislead the electorate. Voters also want a future that is prosperous. Not only should the economy look good, but citizens want to know they will do well in that economy.^[14] For this reason, daily public approval polls sometimes act as both a referendum of the president and a predictor of success.

PUBLIC OPINION AND GOVERNMENT

The relationship between public opinion polls and government action is murkier than that between polls and elections. Like the news media and campaign staffers, members of the three branches of government are aware of public opinion. But do politicians use public opinion polls to guide their decisions and actions?

The short answer is “sometimes.” The public is not perfectly informed about politics, so politicians realize public opinion may not always be the right choice. Yet many political studies, from the American Voter in the 1920s to the American Voter Revisited in the 2000s, have found that voters behave rationally despite having limited information. Individual citizens do not take the time to become fully informed about all aspects of politics, yet their collective behavior and the opinions they hold as a group make sense. They appear to be informed just enough, using preferences like their political ideology and party membership, to make decisions and hold politicians accountable during an election year.

Overall, the collective public opinion of a country changes over time, even if party membership or ideology does not change dramatically. As James Stimson’s prominent study found, the public’s mood, or collective opinion, can become more or less liberal from decade to decade. While the initial study on public mood revealed that the economy has a profound effect on American opinion,^[15] further studies have gone beyond to determine whether public opinion, and its relative liberalness, in turn affect politicians and institutions. This idea does not argue that opinion never affects policy directly, rather that collective opinion also affects the politician’s decisions on policy.^[16]

Individually, of course, politicians cannot predict what will happen in the future or who will oppose them in the next few elections. They can look to see where the public is in agreement as a body. If public mood changes, the politicians may change positions to match the public mood. The more savvy politicians look carefully to recognize when shifts occur. When the public is more or less liberal, the politicians may make slight adjustments to their behavior to match. Politicians who frequently seek to win office, like House members, will pay attention to the long- and short-term changes in opinion. By doing this, they will be less likely to lose on Election Day.^[17] Presidents and justices, on the other hand, present a more complex picture.

Public opinion of the president is different from public opinion of Congress. Congress is an institution of 535 members, and opinion polls look at both the institution and its individual members. The president is both a person and the head of an institution. The media pays close attention to any president’s actions, and the public is generally well informed and aware of the office and its current occupant. Perhaps this is why public opinion has an inconsistent effect on presidents’ decisions. As early as Franklin D. Roosevelt’s administration in the 1930s, presidents have regularly polled the public, and since Richard Nixon’s term (1969–1974), they have admitted to using polling as part of the decision-making process.

Presidential responsiveness to public opinion has been measured in a number of ways, each of which tells us something about the effect of opinion. One study examined whether presidents responded to public opinion by determining how often they wrote amicus briefs and asked the court to affirm or reverse cases. It found that the public’s liberal (or non-liberal) mood had an effect, causing presidents to pursue and file briefs in different cases.^[18] But another author found that the public’s level of liberalness is ignored when conservative presidents, such as Ronald Reagan or George W. Bush, are elected and try to lead. In one example, our five most recent presidents’ moods varied from liberal to non-liberal, while public sentiment stayed consistently liberal.^[19] While the public supported liberal approaches to policy, presidential action varied from liberal to non-liberal.

Overall, it appears that presidents try to move public opinion towards personal positions rather than moving themselves towards the public’s opinion.^[20] If presidents have enough public support, they use their level of public approval indirectly as a way to get their agenda passed. Immediately following Inauguration Day, for example, the president enjoys the highest level of public support for implementing campaign promises. This is especially true if the president has a mandate, which is more than half the popular vote. Barack Obama’s recent 2008 victory was a mandate with 52.9 percent of the popular vote and 67.8 percent of the Electoral College vote.^[21] In contrast, President Donald Trump’s victory over Democratic nominee Hillary Clinton was a closer contest. While Clinton outdistanced him by 2.9 million votes nationally, after narrowly winning several states, Trump won a comfortable majority in the Electoral College.

When presidents have high levels of public approval, they are likely to act quickly and try to accomplish personal policy goals. They can use their position and power to focus media attention on an issue. This is sometimes referred to as the bully pulpit approach. The term “bully pulpit” was coined by President Theodore Roosevelt, who believed the presidency commanded the attention of the media and could be used to appeal directly to the people. Roosevelt used his position to convince voters to pressure Congress to pass laws.

Increasing partisanship has made it more difficult for presidents to use their power to get their own preferred issues through Congress, however, especially when the president’s party is in the minority in Congress.^[22] For this reason, modern presidents may find more success in using their popularity to increase media and social media attention on an issue. Even if the president is not the reason for congressional action, he or she can cause the attention that leads to change.^[23]

Presidents may also use their popularity to ask the people to act. In October 2015, following a shooting at Umpqua Community College in Oregon, President Obama gave a short speech from the West Wing of the White House. After offering his condolences and prayers to the community, he remarked that prayers and condolences were no longer enough, and he called on citizens to push Congress for a change in gun control laws. President Obama had proposed gun control reform following the 2012 shooting at Sandy Hook Elementary in Connecticut, but it did not pass Congress. This time, the president asked citizens to use gun control as a voting issue and push for reform via the ballot box.

In some instances, presidents may appear to directly consider public opinion before acting or making decisions. In 2013, President Obama announced that he was considering a military strike on Syria in reaction to the Syrian government’s illegal use of sarin gas on its own citizens. Despite agreeing that this chemical attack on the Damascan suburbs was a war crime, the public was against U.S. involvement. Forty-eight percent of respondents said they opposed airstrikes, and only 29 percent were in favor. Democrats were especially opposed to military intervention.^[24] President Obama changed his mind and ultimately allowed Russian president Vladimir Putin to negotiate Syria’s surrender of its chemical weapons.

However, further examples show that presidents do not consistently listen to public opinion. After taking office in 2009, President Obama did not order the closing of Guantanamo Bay prison, even though his proposal to do so had garnered support during the 2008 election. President Bush, despite growing public disapproval for the war in Iraq, did not end military support in Iraq after 2006. And President Bill Clinton, whose White House pollsters were infamous for polling on everything, sometimes ignored the public if circumstances warranted.^[25] In 1995, despite public opposition, Clinton guaranteed loans for the Mexican government to help the country out of financial insolvency. He followed this decision with many speeches to help the American public understand the importance of stabilizing Mexico’s economy. Individual examples like these make it difficult to persuasively identify the direct effects of public opinion on the presidency.

While presidents have at most only two terms to serve and work, members of Congress can serve as long as the public returns them to office. We might think that for this reason public opinion is important to representatives and senators, and that their behavior, such as their votes on domestic programs or funding, will change to match the expectation of the public. In a more liberal time, the public may expect to see more social programs. In a non-liberal time, the public mood may favor austerity, or decreased government spending on programs. Failure to recognize shifts in public opinion may lead to a politician’s losing the next election.^[26]

House of Representatives members, with a two-year term, have a more difficult time recovering from decisions that anger local voters. And because most representatives continually fundraise, unpopular decisions can hurt their campaign donations. For these reasons, it seems representatives should be susceptible to polling pressure. Yet one study, by James Stimson, found that the public mood does not directly affect elections, and shifts in public opinion do not predict whether a House member will win or lose. These elections are affected by the president on the ticket, presidential popularity (or lack thereof) during a midterm election, and the perks of incumbency, such as name recognition and media coverage. In fact, a later study confirmed that the incumbency effect is highly predictive of a win, and public opinion is not.^[27] In spite of this, we still see policy shifts in Congress, often matching the policy preferences of the public. When the shifts happen within the House, they are measured by the way members vote. The study’s authors hypothesize that House members alter their votes to match the public mood, perhaps in an effort to strengthen their electoral chances.^[28]

The Senate is quite different from the House. Senators do not enjoy the same benefits of incumbency, and they win reelection at lower rates than House members. Yet, they do have one advantage over their colleagues in the House: Senators hold six-year terms, which gives them time to engage in fence-mending to repair the damage from unpopular decisions. In the Senate, Stimson’s study confirmed that opinion affects a senator’s chances at reelection, even though it did not affect House members. Specifically, the study shows that when public opinion shifts, fewer senators win reelection. Thus, when the public as a whole becomes more or less liberal, new senators are elected. Rather than the senators shifting their policy preferences and voting differently, it is the new senators who change the policy direction of the Senate.^[29]

Beyond voter polls, congressional representatives are also very interested in polls that reveal the wishes of interest groups and businesses. If AARP, one of the largest and most active groups of voters in the United States, is unhappy with a bill, members of the relevant congressional committees will take that response into consideration. If the pharmaceutical or oil industry is unhappy with a new patent or tax policy, its members’ opinions will have some effect on representatives’ decisions, since these industries contribute heavily to election campaigns.

There is some disagreement about whether the Supreme Court follows public opinion or shapes it. The lifetime tenure the justices enjoy was designed to remove everyday politics from their decisions, protect them from swings in political partisanship, and allow them to choose whether and when to listen to public opinion. More often than not, the public is unaware of the Supreme Court’s decisions and opinions. When the justices accept controversial cases, the media tune in and ask questions, raising public awareness and affecting opinion. But do the justices pay attention to the polls when they make decisions?

Studies that look at the connection between the Supreme Court and public opinion are contradictory. Early on, it was believed that justices were like other citizens: individuals with attitudes and beliefs who would be affected by political shifts.^[30] Later studies argued that Supreme Court justices rule in ways that maintain support for the institution. Instead of looking at the short term and making decisions day to day, justices are strategic in their planning and make decisions for the long term.^[31]

Other studies have revealed a more complex relationship between public opinion and judicial decisions, largely due to the difficulty of measuring where the effect can be seen. Some studies look at the number of reversals taken by the Supreme Court, which are decisions with which the Court overturns the decision of a lower court. In one study, the authors found that public opinion slightly affects cases accepted by the justices.^[32] In a study looking at how often the justices voted liberally on a decision, a stronger effect of public opinion was revealed.^[33]

Whether the case or court is currently in the news may also matter. A study found that if the majority of Americans agree on a policy or issue before the court, the court’s decision is likely to agree with public opinion.^[34] A second study determined that public opinion is more likely to affect ignored cases than heavily reported ones.^[35] In these situations, the court was also more likely to rule with the majority opinion than against it. For example, in Town of Greece v. Galloway (2014), a majority of the justices decided that ceremonial prayer before a town meeting was not a violation of the Establishment Clause.^[36] The fact that 78 percent of U.S. adults recently said religion is fairly to very important to their lives^[37] and 61 percent supported prayer in school^[38] may explain why public support for the Supreme Court did not fall after this decision.^[39]

Overall, however, it is clear that public opinion has a less powerful effect on the courts than on the other branches and on politicians.^[40] Perhaps this is due to the lack of elections or justices’ lifetime tenure, or perhaps we have not determined the best way to measure the effects of public opinion on the Court.