336 What Is the Relationship between Citizens and Their Government? should actually be quite distrustful of government officials, and whether democracies require a minimum amount of trust to function. "Public Trust in Government, 1958-2010." Pew Research Center for the People and the Press, http://people-press.org/zo10/04/18/public-trust-in-government-1958-2010/. This interactive Web site contains survey data regarding the public's trust in government. You can graph changes in trust over time alongside other measures, such as consumer confidence. There is even a Government Satisfaction Quiz. Check it out and see how your level of satisfaction stacks up against the views of others. CHAPTER 12 mpact of Public Opinion on Policy ON THE EVE of the U.S. invasion of Iraq in March 2003, nearly 60 percent of the American public supported U.S. military intervention to force Saddam Hussein from power.1 In fact, according to polls conducted by the Pew Research Center, a majority of the public had favored sending troops into Iraq since November 2001 (see Figure 12-1). Polls conducted by other organizations showed "exceptionally consistent" results, with all demonstrating majority support for using U.S. troops to unseat Hussein.2 With public support on his side (as well as congressional approval), President George W. Bush ordered the military into Iraq on March 19, 2003. In a nationally televised speech that evening, Bush announced this action as follows: "My fellow citizens, at this hour, American and coalition forces are in the early stages of military operations to disarm Iraq, to free its people and to defend the world from grave danger."3 During the first eighteen months of the war, most Americans continued to express support for Bush's decision. Beginning in October 2004, however, less than 50 percent of the public felt that using military force in Iraq was the "right decision."4 As the months ticked by, more and more Americans concluded that military action in Iraq should not have happened. By January 2007, only 40 percent labeled the military intervention as the right decision, whereas 51 percent believed it was the wrong decision. Furthermore, over these years, gradually increasing percentages of the public supported bringing U.S. troops home from Iraq. Whereas only 32 percent of people expressed this view in September of 2003, nearly 50 percent did by the beginning of 2007.5 It was in this climate of decreased public support for the Iraq war that Bush proposed sending more than 20,000 additional troops to Iraq to serve alongside the approximately 140,000 servicemen and -women already there. He argued that these extra troops were necessary to combat sectarian violence between Shiite and Sunni Muslims in Iraq and therefore improve safety and security in the nation, particularly in Baghdad.6 Given the public's dissatisfaction with the war and increasing preference to return troops to the United States, it is not surprising that support for the proposed surge was not high. In fact, only 31 percent of the 338 What Is the Relationship between Citizens and Their Government? Impact of Public Opinion on Policy 339 I CT s 01 -t-t -a c ro c tt O _u q- .3 r > tu bß o n o _c c >1 „ Cut & o S C 2 J C _H -2 'S Mfc V C - 2 Ss C ^ G 8 'sp-S .SP E a 8*1 t, -3 & a c E ^5 • u a h o P =5 a o » ft ^ ä ° Oh'- .O *, J3 o CT1 O Oh r) E B M) Oh 9 P ät8 0 -0 U 1- -a 4-i « w , r *i ..... rt O C JÜ u O. x o _a 3 1 & Q.^ g «§ is a £ 3 b g J » * J ^ Uh s 5 P o 1 «'S) Q S 5 .S I E s JS ST a, 0^ i> ^a co qQ J H bp 2 #13 H PJ public favored this proposal in January 2007. And of those that opposed it, fully 69 percent believed that Congress should actively try to prevent Bush from implementing his plan, even by withholding funding for the troops if necessary.7 Despite this lack of public support, Bush did order more troops into Iraq in early 2007. Throughout 2007 and 2008, even more citizens began to support bringing U.S. military troops home from Iraq. In fact, a majority of the public held this view during most of those two years (refer to Figure 12-1). Also, only a minority of the public believed that military action in Iraq had been the right decision, a trend that continued through 2010. Clearly, public support for U.S. military involvement in Iraq was not high. In one of his first major foreign policy speeches as president, Barack Obama announced in February 2009 that U.S. combat forces would be gradually removed from Iraq. His goal, ultimately met, was to remove all such forces by August 31, 2010, while retaining up to 50,000 members of the military in Iraq after that date to support the Iraq government and its military.8 Drawing down troop levels in Iraq was favored by the public. In both April 2009 and August 2010, clear majorities (74 and 68 percent, respectively) of the public approved of "Obama's decision to end the combat role of U.S. troops and remove most but not all U.S. troops from Iraq by August 31, 2010."9 These three examples of policy decisions made by Bush and Obama present contrasting views of the relationship between public opinion and public policy. Bush's choice to engage in military action in Iraq and Obama's decision to remove combat troops from Iraq both coincided with the publics preferences. In contrast, Bush's troop surge decision went against the wishes of a majority of Americans. Which is more common? Do policymakers generally enact policies that a majority of citizens support? Or is the discounting of public opinion more common? Under what circumstances will policy reflect the public's policy opinions? Does the type of policy matter? Are officials more or less likely to be responsive to public opinion when formulating foreign versus domestic policy? We address these questions in this chapter. For a minute, however, let's think about the first example a bit more carefully. Before the Iraq war began, Bush attempted to persuade the public and other policymakers (notably members of Congress) that the United States should invade Iraq and topple Saddam Hussein. He presented two primary reasons why he felt this action was necessary.10 First, he warned that Hussein was developing weapons of mass destruction and America's safety depended on stopping this effort. Second, he argued that Hussein must be removed from power because he was connected with al-Qaeda, the group responsible for the 9/11 attacks. Could his arguments have convinced the public to support his plan, which then became policy? If so, is this an example of Bush responding to public opinion or shaping opinion to support his own goals? These questions suggest that to truly understand the role of public opinion in policymaking, we must consider how officials 340 What Is the Relationship between Citizens and Their Government? monitor and use public opinion to further their own policy preferences, topics that we also take up in this chapter. First, however, we summarize democratic theorists' views on the role of the public in policymaking. Should Public Opinion Influence Policy? Regardless of whether public policy does respond to the wishes of the public, should'it? As you have no doubt guessed by now, democratic theories provide quite different answers to this question. Before discussing these differences, recall that all democratic theorists support popular sovereignty. This is the belief that power in a democratic society ultimately rests with the citizenry. Differences across the theories emerge over how the people should exercise their power and how capable the public is for democratic governance. Participatory democrats expect the influence of the public on policy to be quite substantial. Policymakers should, in their view, enact policy that coincides wirK the wishes of the majority. These theorists further believe that officials should debate policy openly, providing the public with meaningful and relevant information about policy options. Leaders should not attempt to manipulate or mislead the public. After listening to an information-rich policy debate, the public can form opinions about public policy, opinions that they communicate to the leaders. Because political equality is also important to participatory democrats, they expect that policymakers will respond to the wishes of the entire public, not only those citizens who are especially involved in politics or who have the financial means to express their opinions most loudly (such as by contributing to candidates for elective office). Pluralistsexpect that public policy will reflect publicj)£inion, but they prefer that citizens be somewhat removed from the policymaking process. For them, opinions are best expressed via organized groups.11 Whereas citizens are not knowledgeable enough about policy issues to express clear preferences to officials, plural-ists assume interest group representatives are. These groups lobby policymakers directly, trying to convince them to support policy that is in the interest of the groups members. Pluralists also argue that citizens are not attentive enough to politics to follow the goings-on of their elected officials but that interest group representatives do this and communicate details back to their members. Through the interest group link, however, only those people who are represented by groups are likely to have their preferences communicated to officials. This model thus privileges the opinions of those who are organized over those who do not have a group actively involved in lobbying government officials. Of the democratic theories we have been profiling in this book, elite democrats posit the smallest role for the public in policymaking. These theorists believe that the public should be involved in electing officials but should then leave the policy details up to the leaders. Elite democrats view citizens as disinterested in Impact of Public Opinion on Policy 341 following politics closely enough to make decisions about complex policy matters and as incapable of seeing beyond their own interest to make choices that are in the best interest of the nation. Thus, elite democrats prefer to leave policymaking to those with expertise—the leaders. Policymakers can attempt to educate the public about the best policy option, but at the end of the day, the officials should do as they see fit. This view was clearly expressed in a July 2007 opinion column written by the late David Broder, a former Washington Post journalist. After describing examples in which the "dangerously compliant congressional leadership" followed the wishes of the public by not enacting policies that Broder thought would be good for the country, he concluded, "Politicians are wise to heed what people want. But they also have an obligation to weigh for themselves what the country needs. In today's Washington, the 'wants' of people count far more heavily than the nation's needs."12 Despite this view, elite democrats expect that policy might minimally correspond with public preferences. They see this outcome occurring via elections. Citizens elect leaders who they hope will follow their general policy preferences and then have the chance to remove these officials from office if they do not. Thus, elections can produce a connection between the opinions of voters and policy. The voters have control over who is elected, and officials have an incentive to keep the preferences of voters in mind or risk losing their office come Election Day. Finally, elite democrats do not hope for or expect that the preferences of all members of the public will be expressed through elections but, rather, only the wishes of those who are most attentive to and involved in politics. For elite democrats, it is perfectly fine, even preferred, if the opinions of other members of the public do not influence policy decisions. Judgments about politicians' responses to public opinion are not restricted to democratic theorists. Everyday conversations include negative descriptions of leaders who do not adopt the speaker's preferred behavior. Politicians who are perceived to follow the public's wishes too quickly and uncritically are said to pander, whereas those who make decisions contrary to public opinion axejiecrJ£d as shirkers or worse. Indeed, you might have strong preferences about whether officials should or should not enact policy that corresponds with the wishes of the public. Whether you do or not, we encourage you to ponder the normative democratic theories as we present findings from empirical research on public opinion and public policy. Is Public Opinion Related to Policy? As one scholar puts it, "No one believes that public opinion always determines public policy; few believe it never does."13 True enough, but sorting out how often public opinion is related to policy is not an easy task. Numerous decisions need to be made, including whether to examine public opinion for one or more issues 342 What Is the Relationship between Citizens and Their Government? and whether to focus on national or state policymaking. Thus, researchers have taken very different approaches to studying the relationship between opinion and policy. One method that has been used is a case study, which entails an in-depth analysis of one policy area (such as health-care policy) or one policy decision (the passage of a specific bill). Although many case studies of the opinion-policy relationship have been conducted,14 we instead focus our attention in this section on research that examines many different policy areas and decades at once. After all, if a case study of tax policy in the 2000s finds that opinion influenced policymaking, we cannot be certain that this relationship exists for other issues or for other time periods. In contrast, aggregate studies, which examine many issue domains and years, provide more conclusive evidence about the overall relationship between public opinion and public policy. Issue-Specific Opinion and Policy When public support toward an issue changes, does policy then change? This question was addressed by Benjamm Page and Robert Shapiro.15 They examined public opinion survey dataT)etweeni935 and 1979 and identified 357 cases in which policy preferences changed significantly over time. Page and Shapiro then studied national and state policies for each case at the time of the first public opinion measure and then again one year after the second measure to see if there had been any change in the policy. Their goal was to^s^s^opinion-poliqrcon-^ gruence. For example, imagine that the percentage of the public that wanted the ""government to spend more money to protect the environment increased from 48 percent in 1978 to 58 percent in 1983. To assess the relevant government policy, Page and Shapiro compared the level of actual spending on the environment in 1978 with that in 1984. Such an approach allowed them to determine whether public policy changed in the same direction as the opinion change (this is congruence), changed in the opposite direction, or did not change. Theirunderlying^ assumption is that if policy does respond to public opinion, this jhojLiid_be^sp_e-ciaily likely when public opinion changes.1 Across all their cases, Page and Shapiro found that when both opinion and policy changed, 66 percent of these changes were congruent. They further demonstrated that congruence was more likely the larger the opinion change (see Figure 12-2). When public support for a policy changed only 6 or 7 percentage points and when the relevant public policy changed, it did so in the same direction as the opinion change 53 percent of the time. Opinion-policy congruence fairly steadily increased for larger opinion changes and reached 100 percent when the opinion change was 30 percentage points or higher. A related approach to assessing the opinion-policy relationship is to examine public preference for change rather than.actual opinion, change,. That iS) instead of finding instances when public opinion did change over time as Page and Shapiro Impact of Public Opinion on Policy 343 Figure 12-2 Congruence between Opinion Change and Policy Change, by Size of Opinion Change 100 « 80- 60 B 40 20 6 to 7 8 to 9 10 to 14 15 to 19 20 to 29 Size of opinion change, in percentage points 30+ Source: Data from Table 3 of Benjamin I. Page and Robert Y. Shapiro, "Effects of Public Opinion on Policy," American Political Science Review 77 (1983), 180. did, we could see whether the public desires that a specific government_policy be changed and then examine whether government policy does in fact change in response. This was the approach favored by Alan MonroeTPerusing the results from many public opinion polls, Monroe identified many examples of questions asking about policy change. He did this for two time periods: 1960-1976 and 1980-1993.17 For each measure of public preferences, he then examined whether national policy changed or stayed the same To assess consistency (Monroe's term for congruence), Monroe compared public preferences for change with actual policymaking for each policy issue in his analysis. When a majority of the public favored policy change and change did occur in the direction of the public's wishes, opinion-policy consistency was present. Similarly, consistency also occurred when the majority opinion favored the status quo (no change) and policy did not change. Overall, Monroe concluded that public preferences and policy were consistent 64 percent of the time for the first time period and 55 percent for the second. As you can see in Table 12-1, however, the preferences of the public were more in line with public policy when the public favored the status quo.18 Between i960 and 1976, for instance, when the public preferred that a specific policy not change, this policy stayed the samekT 76 percent of the cases Monroe analyzed. When the public wanted change, they rg"6t it only S9~percent of thetime. And during the second time period, when the public preferred that a policy be changed, this, change occurred only 45 percent of the time, yet public preferences for the status quo were met for 70 percent of these 344 What Is the Relationship between Citizens and Their Government: Table 12-1 Consistency between Public Opinion and Public Policy Consistency for 1960-1976 Overall consistency: 64% (222 cases) Public policy Public preference Status quo Change Status quo Change 76% 24 41% 59 Consistency for 1980-1993 Overall consistency: 55% (566 cases) Public policy Public preference Status quo Change Status quo Change 70% 30 55% 45 Sources: Data tor top halt from laDie 1 or ruan u. iviumuc, v^t-m^Ltn^ ^y»^.» * Preferences and National Policy Decisions," American Politics Quarterly 7 (1979). 9- Data for bottom half from Table 1 of Alan D. Monroe, "Public Opinion and Public Policy, 1980-1993," Public Opinion Quarterly 62 (1998), 13. cases. This status quo bias is perhaps not surprising. The U.S. political system with its three branches of government that share power and check each other's power was set up to make policy alterations occur slowly, if at all. Opinion Trends and Policy Other studies of the opinion-policy relationship have used broader measures of public opinion. This approach assumes that when formulating policies, policymakers focus on general opinion trends (such as liberal or conservative swings) in public opinion rather than opinion toward specific policy issues.19 One broad measure of public opinion used in opinion-policy studies is labeled policy mood or policy sentiment.20 The measure is obtained by aggregating across opinion toward dozens of specific issues. It captures whether the public feels that the gov-ernmentjjn general^ is doing too much or not doing enough. In other words, it measures "global preferences for a larger, more active federal government as opposed to a smaller, more passive one across the sphere of all domestic policy controversies."21 Examining variation in the domestic policy mood over time is interesting in its own right. Americans' global preferences were quite liberal in the 1960s and Impact of Public Opinion on Policy 345 Figure 12-3 Correspondence between Public Mood and Policy -*— Public mood Public policy Source: Figure 2.2 of Robert S. Erikson, Michael B. MacKuen, and James A. Stimson, "Public Opinion and Policy: Causal Flow in a Macro System Model," in Navigating Public Opinion: Polls, Policy, and the Future of American Democracy, ed. Jeff Manza, Fay Lomax Cook, and Benjamin I. Page (Oxford: Oxford University Press, 2002), 43. then turned more conservative until 1980 (refer to the solid line in Figure 12-3). A more liberal policy mood returned and stayed until the early 1990s. The first half of the 1990s witnessed a trend toward conservatism. For our purposes now, however, we want to know whether changes in the public's policy sentiment were accompanied by changes in public policy This did happen for domestic policy issues, according to research conducted by Robert Erikson, Michael MacKuen, and James Stimson.22 These authors assessed the direction of policymaking by determining the ideological direction of key laws passed by Congress (and not vetoed by the president) between 1954 and 1996. That is, for each two-year session of Congress, this measure shows whether laws enacted tended to be conservative or liberal. When comparing the ideology of policy with the ideology of the public, Erikson et al. found that when the public mood changed public policy generally followed in the same direction (see the dotted line in Figure 12-3). In other words, a liberal shift in the public mood was followed by a liberal shift in policy; likewise, conservative opinion shifts were accompanied by policy moving in a conservative direction. As Erikson et al. describe this relation-sKip7"Tublic sentiment shifts. Political actors sense the shift. And then they alter their policy behavior."2' More recently, Stuart Soroka and Christopher Wlezien used a similar dynamic approach to assess policy responsiveness to public opinion.24 Much like Erikson 346 What Is the Relationship between Citizens and Their Government? and his co-authors, Soroka and Wlezien examined public opinion and policy output over time (specifically, in 1973-2004). Their policy focus was, however, narrower than Erikson et al.'s. Rather than a wide range of policy opinions, Soroka and Wlezien explored public opinion only toward government spending in specific policy domains (such as defense, social welfare, health, education, and foreign aid). They compared trends over time in public preferences for spending to trends over time in actual spending by the national government. Their conclusion? Policy responds to public views. When the public prefers more spending in a domain, actual spending often does increase. When the public holds the opposite preference, spending is likely to decrease. The relationship between public opinion and policymaking in the U.S. states has also been examined. In one study, researchers used people's responses to poll questions asking their ideological identification to determine the average ideology for eachjstate.15 For the years of their study (1976-1988), the states with the most conservative citizens were Utah, Idaho, and Oklahoma. Massachusetts, Rhode Island, and New York were the most liberal. Each state's policy ideology was also assessed by examining the liberalness or conseryativeness of state policies in specific areas (such as education, criminal justice, and taxes). When the ideology of the public was compared with that of a state's policy, a clear pattern emerged: policy was more liberal in states where the citizens were more liberal; when more of a state's citizens identified as conservative, the state's policies were more conservative.26 More recent work on the states reaches a similar conclusion.27 In this study, researchers determined state public opinion for 2000 and 2004 in two policy domains: economic policy (for example, taxes, minimum wage, and education spending) and social policy (for example, gun control, abortion, and school prayer). When comparing opinion with state policy outputs, the authors found that public preferences in a state corresponded with policy in that state for both policy domains. Variation in Congruence Despite the dissimilarities in research designs and goals, all these scholars conclude that public opinion and public policy tend to be related.28 They and others have also explored whether congruence is higher for certain policies or people. It is. First, issue salience matters. The more that the public is especially concerned about an issue or the more that an issue receives national attention, such as in the news media, the more likely it is that public opinion about the policy will closely match actual policy.29 For his 1980-1993 analysis, for example, Monroe found that public preferencesjforchange and policy outcomes were 6q percent consistent for issues that the public felt were the most important for the nation.30 Consistency dropped to 60 percent for issues that were ranked the second to fifth most important and to 56 percent for issues that were ranked lower than this. Among issues Impact of Public Opinion on Policy 347 that were not considered important by the public, opinion and policy were consistent only 46 percent of the time. Public attentiveness is also relevant. Congruence is higher for citizens who follow politics most attentively than for citizens who pay lesser attention.31 Members of the public who care deeply about a policy area as well as the attentive public are likely to know how their elected officials stand on issues and to take this into consideration when they vote.32 Elected officials also know this, and if they wish to remain in office, they pay attention to the wishes of their aware and engaged constituents. Officials are also vigilant about especially salient issues. Such issues can provoke those citizens who are normally only marginally attentive to politics to tune in and become informed about the topics.33 And if these citizens do not like their representatives' positions on salient issues, they can vote them out of office. Finally, an emerging body of work suggests that the more income that a person has, the more likely that public policy corresponds to his or her prefer-ences, for both national3** and state public policies.35 This conclusion is particularly striking when we consider instances in which the preferences of high-income Americans differ from the opinions of either low- or middle-income Americans. In such situations, public policy is significantly likely to be related to the opinions of the high-income group but unrelated to the preferences of people with lower incomes. The primary instance when lower-income citizens are likely to see their preferences reflected in public policy is when they have the same opinion as those with more means.36 This finding led one scholar in this area to conclude that "public policy in the United States would look rather different if poor Americans had the influence over government policy that affluent Americans appear to enjoy."37 Finally, other work has focused not on actual public policies but, rather, on policy-related behavior. These studies also demonstrate income inequalities. Specifically, the preferences of the affluent are more likely to be related to the roll call votes of U.S. senators^orto have been reflected in policy statements made by President Ronald Reagan.39 These findings concern proponents of participatory democracy greatly. Not only do the research conclusions suggest that not all Americans' preferences are equally likely to be related to the policy decisions of leaders, but they also suggest that economic inequalities are one source of such differing levels of responsiveness. This provides further evidence to these theorists that inequalities among the public need to be minimized. Elite democrats, in contrast, would likely applaud the findings. These theorists prefer that, if the government's policies do match the preferences of the public, the opinions of the most involved and more aware citizens are followed. These citizens, after all, are most likely to have well-considered opinions on policy matters, according to elite democrats. 348 What Is the Relationship between Citizens and Their Government? Impact of Public Opinion on Policy 349 Public Opinion in Comparative Perspective Box 12-1 Comparing Opinion-Policy Congruence across Democracies Is the level of opinion-policy congruence in the United States typical among democratic nations? If not, why might levels of governmental policy responsiveness to citizens differ across nations? Unfortunately, we have only preliminary answers to these questions. Studies of opinion-policy congruence for other countries have been rare, with research directly comparing the United States and other nations rarer still. One of the first scholars to study this topic was Joel Brooks. He compared public opinion and policy in five nations across different time periods: the 1960s and 1970s for the United States, Canada, and Great Britain; the 1940s through 1980s for France; and the 1970s and 1980s for West Germany.1 Brooks's analyses demonstrated very similar, and fairly modest, levels of opinion-policy consistency for these nations. More recent scholarship, incorporating data from as recent as 2005, has uncovered greater discrepancies across nations, thus leading to attempts to explain why policy is more likely to reflect the wishes of citizens in some nations. Whether the country has a parliamentary or presidential system seems to matter. In the latter, power is separated between the executive and the legislature. Not only are both branches of government involved in policymaking, but the executive and members of the legislature are all directly elected. This seems to create incentives for these officials to be more responsive to public wishes when formulating policy. Results from two studies provide support for this explanation. In both, the United States was found to have higher levels of policy responsiveness than nations with parliamentary governments (specifically, Denmark, Great Britain, and Canada).2 Some features of public opinion matter also. In both parliamentary and presidential systems, policy is more likely to reflect public preferences when the popularity of the executive is low.3 This is probably because the president or prime minister is worried about remaining in power and wants to keep the public happy. In the domain of social welfare policy, government spending is more likely to reflect the public's wishes in nations where the citizens prefer high levels of social welfare spending.4 Put another way, when citizens send a signal to leaders that they want a high level of social welfare spending, they are more likely to get the policy they prefer than when the signal is for low spending levels. (continued) 1. Joel E. Brooks, "Democratic Frustration in the Anglo-American Polities: A Quantification of Inconsistency between Mass Public Opinion and Public Policy," Western Political Quarterly 38 (1985): 250-261; Joel E. Brooks, "The Opinion-Policy Nexus in France: Do Institutions and Ideology Make a Difference?" Journal of Politics 49 (1987): 465-480; Joel E. Brooks, "The Opinion-Policy Nexus in Germany," Public Opinion Quarterly 54 (1990): 508-529. 2. Sara Binzer Hobolt and Robert Klemmensen, "Government Responsiveness and Political Competition in Comparative Perspective," Comparative Political Studies 41 (2008): 309-337; Stuart N. Soroka and Christopher Wlezien, Degrees of Democracy: Politics, Public Opinion, and Policy (Cambridge, UK: Cambridge University Press, 2010), Chap. 7. 3. Hobolt and Klemmensen, "Government Responsiveness." 4. Clem Brooks and Jeff Manza, "Social Policy Responsiveness in Developed Democracies," American Sociological Review 71 (2006): 474-494. Do Politicians Follow or Lead the Public? You just learned that there is considerable evidence that public opinion is related to policy outputs for both the national and state governments in the United States. Does this mean that the views of the public influence the decisions of policymakers? We need to know more before we can answer this question. Just because we see a correlation berween opinion and policy, we cannot be sure that opinion caused the policy outcome. It is possible, after all, that the policy change caused the public's opinions to change. We can rule out this possibility for most of the studies we just discussed, however, because of the methods used by the researchers. For example, both Page and Shapiro and Erikson et al. measured public opinion at one time and policy at a later date. Thus, their results cannot be due to policy influencing opinion because public opinion was assessed before the policy changed.40 There isjmothe^possibility, however. Before public policy is changed, politi-cians unveil proposals and publicly debate their preferred policies. The news media often cover these developments. While politicians are debating the merits of various policy proposals, the public thus has the opportunity to learn about the pros and cons of each. And, because it can take many weeks, months, or even years before policy is changed, the opinions of the public can be influenced by the policymaking process. For example, as we suggest at the beginning of this chapter, Bush's appeals to the public to support his plan to invade Iraq probably influenced public opinion on this matter. Why do we care about this? It is important to sort out the precise nature of the opinion-policy connection for many reasons, not least of which are democratic theory implications. Democratic responsiveness refers to leaders enacting policy that the public wants. Such responsiveness assumes that governmental policy 35° What Is the Relationship between Citizens and Their Government? reflects the genuine opinions of the jpublic. This is what participatory democrats hope for. However, what if the public opinion that the officials seem to be responding to was actually largely created by these officials? If policymakers persuade the public to support the officials' preferred policy and then policymakers enact this policy, can we conclude that democratic responsiveness has occurred? It depends. If public support for the proposal is genuine, then politicians' influence on public opinion would not worry participatory democrats so much. What if however, the public has been manipulated?. Manipulation occurs when leaders use "false or misleading arguments or information to turn the public against its true interests (the preferences it would hold if information were accurate and complete)."41 Such a circumstance would undermine participatory democrats' goal of responsiveness. So, before we can conclude that participatory democrats are happy with the evidence showing that public opinion and public policy are related, we need to know whether leaders are truly responding to the public or are trying to shape, or even manipulate, citizens' opinions. Unlike participatory theorists, elite democrats actually prefer that public preferences be shaped by leaders. The leaders, after all, are the ones who are most aware of what is in the best interests of the nation and who have the expertise to put forth specific policy proposals. Policymakers should share their proposals and reasons for supporting them to the public, thus providing opinion leadership. According to this view, educating the public is an important goal for leaders. If policy reflects the wishes of the public at all, according to elite democrats, it is best if the public's preferences have been influenced by the wisdom and expertise of the leaders. To navigate among these competing theories, it is important to consider the context in which public policy opinions are formed. This entails moving beyond the aggregate studies of opinion and policy we have just reviewed and into the "whole big, messy realm of public opinion and policymaking in the United States."42 In particular, delving into the processes of policymaking and trying to determine the goals of politicians will help to address the debate between participatory and elite democrats. We do that in this section, first by examining how leaders learn about public opinion. Then, we discuss what politicians do with this public opinion information. Do they use it to inform their policy decisions? Or do they use it to try to direct public opinion? How Do Politicians Learn about Public Opinion? One way that politicians try to determine the opinions of the public is through opinion polls. This is especially the case with presidents. Before modern opinion polling techniques were developed in the 1930s, presidents used other means to learn of public preferences. These included reading letters that were sent to the White House, talking to citizens, reading newspapers, interpreting past election