DeVry University Causes and Consequences of Polarization Paper

Question Description



By now you know that you will need at least three scholarly sources for your Course Paper. Let’s avoid procrastinating and make sure we are on target.

Please do the following for this assignment:

  • Identify the topic or theme (concerning political polarization) you are thinking about researching;
    • For some ideas check out Causes and Consequences of Polarization
  • Identify three scholarly sources on the topic of political polarization;
    • Provide the correct citation (APA or MLA) for each source.

You do not need to provide a commentary about each source. At this point, I just want you to locate and cite some sources.

Unformatted Attachment Preview

2 Causes and Consequences of Polarization* Michael Barber and Nolan McCarty Rarely these days does a news cycle pass without new stories of political dysfunction in Washington, DC. New reports of stalemates, fiscal cliffs, and failed grand bargains have begun to erode the public confidence in the ability of our representative institutions to govern effectively. In May 2013, only one American in six approved of the way Congress has handled its job.1 Sadly, that level of support was a major improvement from the previous summer, when wrangling over the usually routine matter of raising the debt ceiling drove congressional approval down to 10%. The most common diagnoses of Washington’s ailments center on the emergence of excessive partisanship and deep ideological divisions among political elites and officeholders. In short, “polarization” is to blame. Consequently, the reform-minded have taken up the mantle of reducing polarization or mitigating its effects. In recent years, proposals for electoral reform to change electoral districting, primary elections, and campaign finance have been presented as panaceas. Other reformers have focused on changing legislative procedures such as those related to the filibuster, appropriations, and confirmation process to limit the opportunities for polarization to undermine government. Although there has been intense public discussion about the causes of polarization, its consequences, and possible cures, social science research has only recently begun to help shape those discussions. The intent of this chapter is to provide a more evidence-based foundation for these debates. Preliminaries The academic study on partisanship and polarization is based on a combination of qualitative and quantitative research. Noteworthy qualitative accounts, which often combine historical research and participant observation, include Rohde (1991), Sinclair (2006), Hacker and Pierson (2006), and Mann and Ornstein (2012). The starting point for many quantitative studies of polarization is the robust observation of rising partisan differences in roll-call voting behavior in Congress. The bipartisan coalitions of the 1950s and 1960s have given way to the party-line voting of the twenty-first century. Although * This piece was shaped profoundly by discussions of the American Working Group of the APSA Task Force on Negotiating Agreement in Politics. This group includes Andrea Campbell, Thomas Edsall, Morris Fiorina, Geoffrey Layman, James Leach, Frances Lee, Thomas Mann, Michael Minta, Eric Schickler, and Sophia Wallace. We also thank Chase Foster for his assistance with the Working Group. 1 Task Force on Negotiating Agreement in Politics 19 these trends are apparent in simple descriptive statistics about partisan divisions on roll calls, political scientists have developed more refined measures of partisan voting differences. A variety of techniques uses data on roll-call voting to estimate the positions of individual legislators on a set of scales.2 The primary scale—the one that explains most of the variation in legislator voting—generally captures partisan conflict. At the individual-legislator level, positions on these scales reflect a mix of ideological positioning and constituency interest as well as party loyalty and discipline. Political scientists continue to debate the exact weights of these factors. Some scholars argue that the scores primarily capture ideological differences (e.g., Poole 2007), whereas others interpret them as measures of partisanship (e.g., Lee 2009). Without taking a position on this debate, we refer to the primary roll-voting scale as the “party-conflict dimension.”3 However, consistent with common usage, we may also label positions on the scale as liberal, moderate, or conservative. All of these techniques for estimating the party-conflict dimension produce similar findings with respect to polarization. Consequently, we focus on the DW-NOMINATE measures developed by McCarty, Poole, and Rosenthal (1997). Generally, these scores range from -1 to +1 and are scaled so that the highest scores are those of conservative Republicans and the lowest are those of liberal Democrats. Given the estimated positions of legislators on this scale, we can measure partisan polarization by computing the difference in means (or medians) across the political parties, where a larger gap indicates a greater level of polarization. Figure 2.1 presents the difference in party means on the party-conflict scale from 1879 through 2011. Figure 1: Polarization in Congress .6 .2 .4 Polarization .8 1 Average Difference on Party Conflict Scale 1878 1894 1910 1926 1942 Year Polarization in House 1958 1974 1990 2006 Polarization in Senate Figure 2.1: Average Distance between Positions across Parties. The y-axis shows the difference in mean positions between the two parties in both the House of Representatives and Senate from 1879 to 2011 using the DW-NOMINATE measures. Congress is more polarized than it has been in over 125 years. From the 1930s until the mid-1970s, these measures of polarization were quite low. Not only were differences between the typical Democratic and Republican legislators small, but there 2 See Poole and Rosenthal (1997); Groseclose, Levitt, and Snyder (1999); Clinton, Jackman, and Rivers (2004). 3 It is important, however, to distinguish these scores from party loyalty. Some members who have extreme positions on these scales are not always loyal partisans (e.g., “Tea Party” Republicans). 20 American Political Science Association also were significant numbers of conservative Democrats and liberal Republicans. Since the 1970s, however, there has been a steady and steep increase in the polarization of both the House and Senate. Other measures of party conflict confirm the trend of increasing polarization in the past 40 years.4 Although conventional wisdom often asserts that polarization resulted from the changing behavior of both parties (i.e., with Democrats moving to the left and Republicans to the right), the evidence shows that the behavioral changes are far from symmetric and are largely driven by changes in the positioning of the Republican Party.5 Figure 2.2 plots the average positions of the parties by region. In the past 40 years, the most discernible trend has been the marked movement of the Republican Party to the right (for qualitative evidence, see Hacker and Pierson 2006; Mann and Ornstein 2012). It is important to note that the changes in the Republican Party have affected both its Southern and non-Southern members. The movement of the Democratic Party to the left on economic issues in the past 50 years is confined to its Southern members—reflecting the increased influence of African American voters in the South. However, it is important that the implied asymmetry may pertain only to the issues (primarily economic) that dominate the congressional agenda. It may well be the case that on some social issues (e.g., gay marriage), polarization is the result of Democrats moving to the left. Figure 2.2: Mean Party-Conflict Score by Party and Region. The y-axis shows the mean position of each party by region. In this plot, the South is defined as AL, AR, FL, GA, KY, LA, MS, NC, OK, SC, TN, TX, and VA. There were no Southern Republican Senators between 1913 and 1960 and only two before that. 4 Although Figure 2.1 shows a steady movement by the average Republican, the Republican caucus in Congress has not become more homogeneous in the same time period. The standard deviation of Republican ideal points has remained around 0.15 since the 1950s. Democrats, conversely, have become much more homogeneous in the same period with the disappearance of conservative southern Democrats. 5 For a discussion of methodological issues underlying this claim, see Hare, McCarty, Poole, and Rosenthal 2012. Task Force on Negotiating Agreement in Politics 21 Another important aspect of the increase in party polarization is the pronounced reduction in the dimensionality of political conflict. Many issues that were once distinct from the partyconflict dimension have been absorbed into it. Poole and Rosenthal (1997) and McCarty, Poole, and Rosenthal (1997) both noted that congressional voting can be increasingly accounted for by a single dimension that distinguishes the parties. This situation directly contrasts with that of the mid-twentieth century, when the parties divided internally on a variety of issues primarily related to race and region. Figure 2.3 quantifies these changes, showing the percentage of individual roll-call vote decisions in the House that can be correctly classified by one- and twodimensional models. 6 The two-dimensional spatial model accounts for most individual voting decisions since the late nineteenth century. Classification success was highest at the turn of the twentieth century, exceeding 90 percent. However, the predictive success of the two-dimensional model fell during most of the twentieth century, only to rebound to the 90% level in recent years.7 Increasingly, most of the work is being done by the party-conflict dimension. In the period from 1940 to 1960, adding a second dimension to account for intraparty divisions on race and civil rights led to a substantial improvement to fit. A second dimension often explained an additional 3% to 6% of the voting decisions in the House. However, in recent years, the second dimension adds no additional explanatory value. In the 112th Congress, the second dimension explains only an additional 1,800 votes of the almost 600,000 cast by House members. Although polarization and the reduction in dimensionality tend to coincide, there is no necessary logical connection between the two trends. One possibility is that partisan polarization 6 When legislators cast a vote in the way that is predicted by their estimated position on the scales, we say their vote is “correctly classified.” Therefore, the figure simply plots the total number of correctly classified votes divided by the total number of votes in a given congressional session. Patterns for the Senate are similar. 7 The high rates of classification success that we observe do not result simply because most votes in Congress are lopsided votes, where members say “Hurrah.” On the contrary, Congress continues to have mostly divisive votes, with average winning majorities between 60% and 70%. 22 American Political Science Association might occur simultaneously across any number of distinct dimensions. For example, parties could polarize on distinct economic and social dimensions. However, this would imply varying intraparty disagreements on the different dimensions. To the contrary, the evidence points to similar intraparty cleavages on almost all issues. For example, the most anti-tax Republican legislators are generally the most pro-life, pro-gun, and anti-marriage equality. Similarly, the Democrats most likely to support a minimum-wage hike are those most supportive of abortion rights and gay marriage. Using the terminology of Converse (1964), issue constraint at the congressional level has expanded dramatically. A second logical alternative is that polarization might coincide with the displacement of the primary dimension of partisan conflict by another issue dimension, consistent with the theory of realignments put forward by Schattschneider (1960), Burnham (1970), Sundquist (1983), and others. Such a situation also seems inconsistent with the data on roll-call voting. As McCarty, Poole, and Rosenthal (2006) documented, the partisan division on economic issues has remained the primary dimension of conflict, and other issues—such as social, cultural, and religious issues—have been absorbed into it. Although there is a broad scholarly consensus that Congress is more polarized than any time in the recent past, there is considerably less agreement on the causes of such polarization. Numerous arguments have been offered to explain the observed increase in polarization, and these causes can be divided into two broad categories: (1) explanations based on changes to the external environment of Congress, and (2) those based on changes to the internal environment. The external explanations provide arguments about how shifts in the social, economic, and electoral environments have altered the electoral incentives for elected officials to pursue moderation or bipartisanship. The internal explanations focus on how the formal and informal institutions of Congress have evolved in ways that exacerbate partisan conflict (or generate the appearance of such an increase). Although we think it is productive to divide the literature along external-internal lines, it is important to note that explanations are not mutually exclusive. Indeed, many of the internal explanations presume a shift in the external environment that stimulates revisions of legislature rules, procedures, and strategies. In the following sections, we review the current literature on each of these suggested causes and evaluate the evidence for and against each argument. External Explanations A Polarized Electorate Perhaps the simplest explanation for an increasingly polarized Congress is one grounded in the relationship between members of Congress and their constituents. If voters are polarized, reelection-motivated legislators would be induced to represent the political ideologies of their constituents, resulting in a polarized Congress. Evidence of voter-induced polarization is elusive, however. Empirical support for the voter-polarization story requires evidence for two specific trends. First, it requires that voters be increasingly attached to political parties on an ideological basis. Liberal voters should increasingly support the Democratic Party and conservative voters should increasingly support the Republican Party. This process has been labeled partisan sorting. Second, Task Force on Negotiating Agreement in Politics 23 the hypothesis requires that voters must be increasingly polarized in their policy preferences or ideological identification. Extreme views must be more common so that the distribution of voter preferences becomes more bimodal. There is considerable evidence for the first trend—voters have become better sorted ideologically into the party system. Layman and Carsey (2002) and Levendusky (2009) found that over time, voters have increasingly held political views that consistently align with the parties’ policy positions. Using data from the National Election Study, Layman and Carsey (2002) found evidence for a pattern of conflict extension, in which differences in the policy preferences of partisans have grown in economic as well as social and racial domains. Their results, updated through 2004, are presented in Figure 2.4. The trends presented in Figure 2.4 are consistent with the finding that fewer voters today than in the past hold a mix of Democratic and Republican positions. As the parties become more coherent in their policy positions, voters sort themselves accordingly. This may well account for the finding of Bartels (2000) that partisan identification is a better predictor of voting behavior. Also, because the terms “Republican” and “Democrat” now represent increasingly distinct clusters of policy positions, citizens who identify with one party expect the other party’s identifiers to hold dramatically different political views. Consequently, party identifiers report that they dislike one another more than they did a generation ago (Shaw 2012) and state that they would be less likely to feel “comfortable” with their child marrying someone who identifies with the opposite party than was the case in the 1960s (Iyengar, Sood, and Lelkes 2012). 24 American Political Science Association Fiorina (2013) argues that the patterns described herein reflect party sorting and not polarization in voters’ policy positions. A lively debate has emerged about the mechanisms underlying the better sorting of voters into parties. Sorting may improve for two distinct reasons. First, voters may shift their allegiance to the party that takes their policy position. Alternatively, voters may adjust their policy views to match those of the party with which they identify. Levandusky (2009) found evidence for both mechanisms but determined that position switching is more common than party switching. Carsey and Layman (2006) also found that party switching does occur, but that it is limited to those voters who have a salient position on one issue and are aware of the partisan differences surrounding it. However, Lenz (2012) finds little evidence favoring the party-switching mechanism. Ultimately, however, both processes are facilitated by greater polarization of partisan elites, suggesting that the trends in Figure 2.4 may be the consequence of elite polarization rather than the cause. 8 Whereas few scholars doubt that substantial voter sorting has occurred, the evidence for voter-policy polarization is less clear. The emerging consensus is that most voters have been and remain overwhelmingly moderate in their policy positions (Fiorina, Abrams, and Pope 2005; Ansolabehere, Rodden, and Snyder 2006; Fiorina and Abrams 2008; Levendusky, Pope, and Jackman 2008; Bafumi and Herron 2010). In studies that produce estimates of voter-issue positions that are comparable to legislator positions, representatives were found to take positions that are considerably more extreme than those of their constituents (Clinton 2006; Bafumi and Herron 2010). 8 McCarty, Poole, and Rosenthal (2006) and Gelman (2009) also found that voters have become better sorted into parties by income over time. The question of whether partisan voters are more sorted by geography is controversial (see Bishop 2009; Klinkner 2004). Task Force on Negotiating Agreement in Politics 25 Figure 2.5 illustrates the main finding of Bafumi and Herron (2010). In the 109th Congress, almost every Senator was more extreme than the median voter of his or her state. The ideological distance between representative and constituent may well have increased, but some distance seems to have existed since the introduction of our earliest measurements. As early as 1960, McClosky and his colleagues found that delegates to the party conventions took positions that were more extreme than those of the voters identifying with each party.9 Recently, Abramowitz (2010) found a more bimodal distribution of preferences among those voters most likely to participate in politics compared to the average party identifier, with further polarization still among party activists and donors.10 The phenomenon of the more-and-more active being more-and-more extreme probably results in part from self-selection, with those having intense feelings being more willing to spend time and money on politics, and in part from the dynamic of group polarization (Sunstein 2002), in which people who talk with one another in relatively homogeneous groups end up taking more extreme positions than the party’s median members. Regarding moderate voters, some have chosen middle-of-the-road positions for substantive policy reasons. Others, however, are uninformed, unengaged, or apathetic, checking off the middle position on surveys due to lack of an opinion. Although the lack of evidence of voter polarization casts doubt on the simple link between voter and elite polarization, a dynamic version may hold more promise. As voters sort in response to elite polarization, the incentives for parties to take positions that appeal to supporters of the other party will diminish. This leads to greater partisan polarization and greater incentives for voters to sort. Although this mechanism is not ruled out by existing evidence, it has not yet been subjected to formal t …
Purchase answer to see full attachment