"'Polling has turned leaders into followers,' laments columnist Marueen Dowd of The New York Times. Well, that's news definitely not fit to print say two academics who have examined the polls and the legislative records of recent presidents to see just how responsive chief executives are to the polls. Their conclusion: not much. . . . In fact, their review and analyses found that public opinion polls on policy appear to have increasingly less, not more, influence on government policies."—Richard Morin, The Washington Post

Copyright

Preface

The way Congress handled the impeachment of President Bill Clinton revealed a lot about American politics. Commentators and the American public were visibly struck by the unyielding drive of congressional Republicans to remove Clinton from office in the face of clear public opposition. The Republicans' disregard for the preferences of the great majority of Americans contradicted perhaps the most widely accepted presumption about politics—that politicians slavishly follow public opinion.

There was little ambiguity about where Americans stood on Clinton's personal behavior and impeachment. The avalanche of opinion polls during 1998 and early 1999 showed that super-majorities of nearly two-thirds of Americans condemned the president's personal misdeeds, but about the same number approved his job performance, opposed his impeachment and removal from office, and favored a legislative censure as an appropriate alternative punishment.

Despite Americans' strong and unchanging opinions, congressional Republicans defied the public at almost every turn. Beginning in the fall of 1998, the Republican-led House of Representatives initiated impeachment proceedings; its Judiciary Committee reported impeachment articles; and it passed two articles of impeachment on the House floor. Neither the House nor the Senate allowed a vote on the option supported by the public—censure. For all the civility in the Senate trial of the president on the House-passed articles of impeachment, the Republicans' pursuit of Clinton was checked not by a sudden attentiveness to public opinion but rather by the constitutional requirement of a two-thirds vote and the bipartisan support that this demanded.

The impeachment spectacle reveals one of the most important developments in contemporary American politics—the widening gulf between politicians' policy decisions and the preferences of the American people toward specific issues. The impeachment of Clinton can be added to the long list of policies that failed to mirror public opinion: campaign finance reform, tobacco legislation, Clinton's proposals in his first budget for an energy levy and a high tax on Social Security benefits (despite his campaign promises to cut middle-class taxes), the North American Free Trade Agreement (at its outset), U.S. intervention in Bosnia, as well as House Republican proposals after the 1994 elections for a "revolution" in policies toward the environment, education, Medicare, and other issues.

The conventional wisdom that politicians habitually respond to public opinion when making major policy decisions is wrong.

Recent research, which we review in chapter 1, provides evidence that this list is not a quirk of recent political developments but part of a trend of declining responsiveness to the public's policy preferences. The conventional wisdom that politicians habitually respond to public opinion when making major policy decisions is wrong.

While the impeachment crisis illustrated unresponsiveness to public opinion, it was also atypical in important respects. One of its most unusual features was the attention that journalists and political observers devoted to the decision of elected officials to ignore public opinion. Politicians' disregard for public opinion on other policy decisions was never an issue.

On impeachment, however, the press devoted generous coverage to Republicans' open discussion of their unresponsiveness and their insistence that they would "not listen to polls at all" (Bill McCollum quoted in Rosenbaum [1998]). Senator Rick Santorum spoke for many Republican legislators when he defiantly announced at the end of the Senate trial that "I don't accept that . . . we should simply do what the polls say" (quoted in Bruni [1999]). Standing for "principle," "constitutional duty," and the "rule of law" against the onrush of ill-guided public opinion was the defense that Republicans offered. 1

The Republicans' unwavering decision to sail against public opinion and then to openly defend their undemocratic actions was just one aspect of the impeachment process that was atypical. Perhaps most surprising was that Republicans rebuffed public opinion on an extraordinarily salient issue that put at risk their ambitions to expand congressional majorities and capture the White House in the year 2000. Even after the disappointing results of the 1998 elections, when Republicans unexpectedly lost seats in the House and failed to widen their margin in the Senate, they persisted in pressing onward. What happened to politicians who were easily cowered by the threat of electoral retribution?

The Republicans' handling of impeachment fits into a larger pattern in contemporary American politics. This book examines the connections of politicians to public opinion and the impact of this relationship on the mass media and the public itself.

We use the cases of Clinton's health care reform campaign and Newt Gingrich's first Congress as House Speaker to make three points that are echoed in the impeachment episode. First, Republicans disregarded public opinion on impeachment because their political goals of attracting a majority of voters was offset by their policy goals of enacting legislation that politicians and their supporters favored. The ideological polarization of congressional Republicans and Democrats since the mid-1970s, the greater institutional independence of individual lawmakers, and other factors have raised the political benefits of pursuing policy goals that they and their party's activists desire. Responding to public opinion at the expense of policy goals entailed compromising their own philosophical convictions and risked alienating ideologically extreme party activists and other supporters who volunteer and contribute money to their primary and general election campaigns. Only the heat of an imminent presidential election and the elevated attention that average voters devote to it motivate contemporary politicians to respond to public opinion and absorb the costs of compromising their policy goals.

Indeed, the Republicans' relentless pursuit of impeachment was largely driven by the priority that the domineering conservative wing of the party attached to their policy goal (removing Clinton) over their political goals (appealing to a majority of Americans). Moderate Republicans could not ignore the risk of opposing impeachment—it could lead to a challenge in the next primary election and diminished campaign contributions.

Politicians track public opinion not to make policy but rather to determine how to craft their public presentations.

Our second point is that politicians pursue a strategy of crafted talk to change public opinion in order to offset the potential political costs of not following the preferences of average voters. Politicians track public opinion not to make policy but rather to determine how to craft their public presentations and win public support for the policies they and their supporters favor. Politicians want the best of both worlds: to enact their preferred policies and to be reelected.

While politicians devote their resources to changing public opinion, their actual influence is a more complex story. Politicians themselves attempt to change public opinion not by directly persuading the public on the merits of their policy choices but by "priming" public opinion: they "stay on message" to highlight standards or considerations for the public to use in evaluating policy proposals. Republicans, for example, emphasized "big government" to prompt the public to think about its uneasiness about government. Politicians' efforts to sway the public are most likely to influence the perceptions, understandings, and evaluations of specific policy proposals such as Republican proposals in 1995 to significantly reduce spending on Medicare to fund a tax cut. But even here, politicians' messages promoting their policy proposals often provoke new or competing messages from their political opponents and the press that complicate or stymie their efforts to move public opinion. In addition, efforts to influence the public's evaluations of specific proposals are unlikely to affect people's values and fundamental preferences (such as those underlying support for Medicare, Social Security, and other well-established programs). We distinguish, then, between political leaders' attempts to alter the public's perceptions, evaluations, and choices concerning very specific proposals (which are susceptible but not certain to change) and Americans' values and long-term preferences (which tend to be stable and particularly resistant to short-term manipulation). Politicians' confidence in their ability to move public opinion by crafting their statements and actions boosts their willingness to discount majority opinion; but the reality is that efforts to change public opinion are difficult and are often most successful when deployed against major new policy proposals by the opposition, which has the more modest task of increasing the public's uncertainty and anxiety to avoid risk.

Politicians respond to public opinion, then, but in two quite different ways. In one, politicians assemble information on public opinion to design government policy. This is usually equated with "pandering," and this is most evident during the relatively short period when presidential elections are imminent. The use of public opinion research here, however, raises a troubling question: why has the derogatory term "pander" been pinned on politicians who respond to public opinion? The answer is revealing: the term is deliberately deployed by politicians, pundits, and other elites to belittle government responsiveness to public opinion and reflects a long-standing fear, uneasiness, and hostility among elites toward popular consent and influence over the affairs of government. 2

Politicians respond to public opinion in a second manner—they use research on public opinion to pinpoint the most alluring words, symbols, and arguments in an attempt to move public opinion to support their desired policies. Public opinion research is used by politicians to manipulate public opinion, that is, to move Americans to "hold opinions that they would not hold if aware of the best available information and analysis" (Zaller 1992, 313). Their objective is to simulate responsiveness. Their words and presentations are crafted to change public opinion and create the appearance of responsiveness as they pursue their desired policy goals. Intent on lowering the potential electoral costs of subordinating voters' preferences to their policy goals, politicians use polls and focus groups not to move their positions closer to the public's but just the opposite: to find the most effective means to move public opinion closer to their own desired policies.3

Political consultants as diverse as Republican pollster Frank Luntz and Clinton pollster Dick Morris readily confess that legislators and the White House "don't use a poll to reshape a program, but to reshape your argumentation for the program so that the public supports it" (quoted in Cannon 1998; Morris 1999). Indeed, Republicans' dogged pursuit of impeachment was premised on the assumption that poll-honed presentations would ultimately win public support for their actions. We suggest that this kind of overconfidence in the power of crafted talk to move public opinion explains the political overreaching and failure that was vividly displayed by Clinton's health reform effort during the 1993-94 period and the Republicans' campaign for their policy objectives beginning with their "Contract with America" during 1995-96. Crafted talk has been more effective in opposing rather than promoting policy initiatives partly because the news media represent and magnify disagreement but also because politicians' overconfidence in crafted talk has prompted them to promote policy goals that do not enjoy the support of most Americans or moderate legislators.

Our argument flips the widespread image of politicians as "pandering" to public opinion on its head. Public opinion is not propelling policy decisions as it did in the past. Instead, politicians' own policy goals are increasingly driving major policy decisions and public opinion research, which is used to identify the language, symbols, and arguments to "win" public support for their policy objectives. Responsiveness to public opinion and manipulation of public opinion are not mutually exclusive: politicians manipulate public opinion by tracking public thinking to select the actions and words that resonate with the public.

Politicians' muted responsiveness to public opinion and crafting of their words and actions has a profound impact on the mass media and on public opinion itself.

Our third point is that politicians' muted responsiveness to public opinion and crafting of their words and actions has a profound impact on the mass media and on public opinion itself. In contrast to others who emphasize the nearly unlimited independence and power of the mass media, we argue that press coverage of national politics has been driven by the polarization of politicians and their reliance on crafting their words and deeds. The press focuses on political conflict and strategy because these are visible and genuine features of contemporary American politics. The combination of politicians' staged displays and the media's scrutiny of the motives behind them produced public distrust and fear of major government reform efforts. We do not treat policymaking, media coverage, and public opinion as parts that can be studied one at a time; rather, we study their dynamic configurations and processes of interdependence. Democratic governance and the process of public communications are inseparably linked.

We do not claim that polls, focus groups, and other indicators of public opinion play no important role in the policymaking process. Information about public opinion does play a role in the making of symbolic decisions (such as the location of presidential vacations), minor policy decisions (Clinton's proposal before the 1996 election for school uniforms), and some important policy decisions (raising the minimum wage in the summer of 1996). Our main point is that the influence of public opinion on government policy is less than it has been in the past and certainly less than commonly assumed by political pundits and some scholars. In addition, public opinion research in American politics does play a critical role in how politicians and other elites craft their actions and statements to elicit public support. Finally, politicians are not shy about brandishing polls that support their positions in order to justify and promote them further.

This book is motivated by the central premise of representative democracy: popular sovereignty and the notion of government responsiveness in which the public's policy preferences point government officials in specific directions. The Declaration of Independence was animated by a demand for "consent of the governed" and the promise of popular control has inspired a long and, at times, violent struggle for the right to vote by all Americans, the full and equal right to freedom of speech and assembly, and other essential rights.

This book revisits the fundamental premise of representative democracy (popular sovereignty and government responsiveness) and asks, Does the American government respond to the broad public or to the interests and values of narrowly constituted groups committed to advancing their private policy agendas? On one side lies democratic accountability; on the other a closed and insular government that is ill-suited to address the wishes or wants of most citizens. When politicians persistently disregard the public's policy preferences, popular sovereignty and representative democracy are threatened.

The responsiveness of national policymakers to what most Americans prefer has declined and remained low for almost two decades.

Can we rely on competitive elections to fend off muted responsiveness to centrist opinion? After all, congressional Democrats suffered stunning setbacks in the 1994 elections following Clinton's campaign for an unpopular health care reform plan and the Republicans' congressional majorities were reduced in the 1996 and 1998 elections after they pursued policies that defied strong public preferences. We argue that electoral punishment may not be enough to improve the public's influence on government: the responsiveness of national policymakers to what most Americans prefer has declined and remained low for almost two decades despite electoral setbacks to Democrats and Republicans. Politicians have worked hard to obscure their true positions and to distort the positions of their opponents, which makes it hard for the electorate to identify the policy positions of elected officials and to punish politicians for pursuing unpopular policies. In addition, most members of Congress today attach greater electoral importance to following the policy goals of party activists than responding to centrist opinion. The bottom line is that most politicians are keenly motivated and amply skilled at evading electoral accountability for long periods. Their success has impaired our system of accountability and sullied the quality of citizenship by eroding public trust and fuelling the news media's increasing focus on political conflict and strategy rather than on the substantive issues raised by government policy.

Our analysis should not be confused, however, with naive populism. We recognize that the sheer complexity and scope of government decisions require elite initiative, at times without public guidance. And, on occasion, elites may need to defy ill-informed and unreasoned public opinion in defense of larger considerations and, instead, rely upon the public's post hoc evaluations of their actions and their arguments justifying their actions. Franklin Roosevelt's arming of merchant marines prior to the United States' entry into the Second World War and Richard Nixon's opening to China represent such cases.

What we see today in contemporary American politics, however, far exceeds responsible leadership in a representative democracy. What concerns us are indications of declining responsiveness to public opinion and the growing list of policies on which politicians of both major political parties ignore public opinion and supply no explicit justification for it. The practice of American government is drifting from the norms of democratic responsiveness.

This book is intended for a wide audience. Chapters 1 and 2 situate our claims about the responsiveness of politicians to public opinion in the context of debates among academic scholars. Two long-standing explanations for the motivations of politicians predict contradictory behavior. One account expects competition for the median voter to motivate parties and politicians to respond to public opinion; the other predicts that politicians will engage in "strategic shirking" to pursue policy goals favored by themselves and their partisan and interest group supporters. We argue that changes in political and institutional conditions since the 1970s have elevated the importance attached to policy goals above that of majority opinion; only the threat of imminent elections produces a temporary rise in responsiveness to public opinion. The shift of politicians toward pursuing their policy goals and crafting their public stances to lower its potential political damage has contributed to changes in press coverage of politics and public opinion.

Chapters 3 through 8 examine in-depth policy debates during the 1990s. Readers uninterested in the contending claims of scholars in the first two chapters may wish to skip ahead to these sections.

Chapters 3 and 4 examine the debates that occurred within the Clinton administration and Congress over health care reform in 1993 and 1994. Here we argue that policymakers were driven by ideology, personal preferences, and political calculations about the demands of their allies. Contrary to popular belief, public opinion did not drive their policy decisions. Rather, the White House and party leaders in Congress tracked public opinion in order to carefully craft their preferred policy options in order to win (rather than follow) public opinion.

Chapters 5 and 6 examine press coverage of health care issues since the mid-1970s. Over the course of nearly two decades, the volume and content of press coverage reflected policy debates and the growing polarization among authoritative government officials in Washington. Although the media represented these genuine changes in health policy debates, they also magnified the political conflict.

Chapter 7 examines the public's reaction to the health care reform debate and, specifically, the roughly 20 percentage point decline in public support for the Clinton plan between September 1993 and the following summer. The decline resulted from polarized policy debates and the press's coverage of them, which produced a shift in public opinion from collective considerations to uncertainty and fears about personal self-interest.

Public opinion is equated by many contemporary policymakers, journalists, and scholars with the adding up of the preferences of disparate individuals and with its measurement by public opinion surveys. (For the historic evolution of the concept, see Gunn 1995.) This concept of public opinion also obscures the communal process that shapes it; its measurement by surveys obscures some of the complexity and thoughtfulness of public thinking (Blumer 1948; Herbst 1993). For example, pollsters' probes of whether respondents "favor or oppose" Clinton's health plan corresponded with the terrain of the political battlefield (they rarely asked about support for the unpassable single-payer plan), but these survey questions reduced Americans' complex cluster of attitudes to a thumbs-up or thumbs-down choice. Despite these limitations, survey results provide a means to investigate the mass public's reasoning, complex motivations (from collective considerations to more narrow self-interest), and multiple and competing tendencies.

Chapter 8 shows that the dynamics of Clinton's health reform crusade also defined Gingrich's drive to spark a revolution in American social policy. What we found for 199394 continued in the quite different political context of 199596.

Chapter 9 turns to the larger normative issues of whether elected officials should respond to public opinion or pursue their own preferences and judgments. We suggest that the decline in responsiveness and the rise of opinion manipulation and fierce partisan conflict have reduced the effectiveness of the governing process and the public's confidence in it. Rather than accepting the false choice between responsiveness and independent leadership, we propose that each has an important place in a process in which politicians generally respond to strong, sustained public preferences. We conclude in chapter 10 by proposing several changes aimed at raising the costs of discounting public opinion and making sustained responsiveness more politically attractive.

Footnotes

1. Hyde 1999; Barr 1999; Rogan quoted in Gugliotta and Eilperin 1999; Rosenbaum 1998.
2. The Oxford English Dictionary (1989) mostly clearly traces the term "pander" to the 16th century when it was used in two quite negative ways: as a description of a pimp and as a characterization of an individual who "ministers to the baser passions or evil designs of others." While the reference to pimps has remained consistent, the second meaning has been expanded. In Thesaurus of English Words and Phrases by Peter Roget (1860, 1909, 1994), the synonyms for pander widen from "pimp" and "indulge" in its 1860 edition to "flatter" in the 1909 edition and "toady to, truckle to, cater to" in the 1994 edition. The 1973 edition of Webster's New Collegiate Dictionary connects it with "someone who caters to or exploits the weaknesses of others."
   The evolution of pandering as a disparaging description of politicians who respond to public opinion is not entirely clear but its use was common by the early 1970s. Politicians used the term to undercut rivals who sided with public opinion and to fortify their unpopular positions. On 27 March 1971, for instance, the New York Times reported that Robert Dole (Senator and Chairman of the Republican National Committee) lashed out at Democratic politicians who had sided with the growing opposition to the Vietnam War for "pander[ing] to the war-weariness [of Americans.]"
3. We treat the terms "polls" and "survey" as synonyms, recognizing that the quality of research associated with them can vary.