Review Essay 41 on The Increasingly United States: How and Why American Political Behavior Nationalized23 min read

The quote from baseball legend (and wordsmith) Yogi Berra throws caution at legendary Democratic House Speaker Tip O’Neill’s famous reminder to all politicians at election time. And for good reason, judging from the titles of the two new headlines above about the 2018 primary and general elections: politics are not local but national, with President Donald Trump front and center; Jonathan Martin can readily make this case. Regarding former President Barack Obama’s endorsements of congressional and state-level candidates, Alexi McCammond then goes on to say Why it matters: Obama is the left’s answer to President Trump’s continued presence in the primaries. Not only will Obama announce another round of endorsements before Nov. 6, but he also plans to campaign in several of these states throughout the fall.”

H-Diplo | ISSF Review Essay 41

Daniel J. Hopkins.  The Increasingly United States:  How and Why American Political Behavior Nationalized.  Chicago:  University of Chicago Press, 2018.  ISBN:  9780226530239 (hardcover, $105.00); 9780226530376 (paperback, $35.00).

Reviewed by Robert Y. Shapiro, Columbia University

Published 13 September 2018 |

Editor:  Diane Labrosse
Web and Production Editor:  George Fujii


“…All politics is local.”                    

—Tip O’Neill, former Speaker of the United States House of Representatives

“The future ain’t what it used to be.”

—Yogi Berra, baseball great

“At State Level Just One Issue Counts: Trump. All Politics Are Local? Not in GOP Races”

—Jonathan Marten (front page The New York Times, 31 July 2018)

“Meet the 81 candidates Obama is endorsing in 2018”

—Alexi McCammond (1 August 2018),

The quote from baseball legend (and wordsmith) Yogi Berra throws caution at legendary Democratic House Speaker Tip O’Neill’s famous reminder to all politicians at election time. And for good reason, judging from the titles of the two new headlines above about the 2018 primary and general elections: politics are not local but national, with President Donald Trump front and center; Jonathan Martin can readily make this case. Regarding former President Barack Obama’s endorsements of congressional and state-level candidates, Alexi McCammond then goes on to say Why it matters: Obama is the left’s answer to President Trump’s continued presence in the primaries. Not only will Obama announce another round of endorsements before Nov. 6, but he also plans to campaign in several of these states throughout the fall.”

Martin’s piece emphasizes Trump’s endorsements and efforts to promote in various states his preferred Republican candidates who have supported him and his policy efforts. Martin focuses on Trump’s upcoming appearance that night at a rally in Tampa, Florida, along with his preferred Republican candidate for governor, Representative Ron DeSantis, and on how Trump caused DeSantis’s primary opponent, Adam Putnam, to collapse in the polls. DeSantis enthusiastically welcomed Trump’s support, whereas other politicos and observers thought that his close ties to the strident and relatively unpopular president would weaken his chances in the general election. But Martin’s story extends beyond presidential support to how DeSantis’s campaign was outside the local orbit in other ways, in particular, as it appeared throughout the media in the context of national issues, not state and local matters that state candidates in the past had to pay close attention to. Martin observes that DeSantis’s rise “reflects the broader nationalization of conservative politics, in which a willingness to hurl rhetorical lightning bolts at the left, the news media, and the special counsel Robert S. Mueller can override local credentials, local endorsements and preparedness for a state-based job.”

In contrast, Putnam’s claims that the state needed a Floridian who would put Florida first had been falling on deaf ears, as well as his criticism of DeSantis for his “ubiquity on Fox News”—at “an out of state television studio.” DeSantis appeared “on Fox prime-time shows at least 41 times since Mr. Trump was inaugurated” to support and defend Trump. And Sean Hannity and other lesser Fox personalities were campaigning for him. This all helped DeSantis, whose own campaign polls found that 66 percent of Florida’s likely Republic voters were regular Fox News viewers. The national context, that would be the same for any state, dominated the election, whereas the Florida state government in Tallahassee and state issues that Putnam tried to emphasize were far from the minds of voters. Anecdotally, Martin cited a local Republican activist who, when asked about the most important issue facing Florida, answered that it’s “the illegals”; “it’s not immigration, it’s an invasion,” and the “migrants are ‘absolutely ruining Europe’.” Local politics this is not.

The purpose of this lengthy description is to ask: Is this nationalization of the 2018 election now typical or is it peculiar to the current era of Donald Trump and the election as a referendum on his presidency? Is this possible national “wave” election—hurting the incumbent party, like the elections of 1974, 1994, 2006, 2010, and 2014—the new norm, or is it just a deviation from the primacy of localized politics?

Daniel Hopkins’s book provides the most definitive answer to date. In short, it puts the kibosh on Tip O’Neill’s assertion that “All politics are local.” While all politics may have been local before, they are clearly a lot less local than they used to be.

The big question about the 2018 election is how much of a national wave will it be. For Hopkins this is a not just another midterm election blip but rather part of a long-term trend of nationalizing forces. His book masterfully presents a wide range of evidence about changes in partisan politics that are front and center today and about changes in American engagement politics at the national versus state and local level, with state politics as we used to know it, falling by the wayside and playing out much more so than the past as a consequence of national political forces. President Trump and other leaders in both political parties endorsing congressional and gubernatorial candidates on the basis of ideology and national issues reflects the polarization of the Democratic and Republican parties that began forty years ago.[1] We have also seen the reverse development, in the recent case of far left congressional candidate Alexandria Ocasio-Cortez’s surprise defeat of the more moderate Democratic incumbent Joseph Crowley in a New York City district. In instances like this, the success of more extreme local candidates (which can get more national media attention than a more moderate candidate’s win) might push their parties further to the left of right.

Hopkins marshals a broad and rich range of data and other evidence to make his case. These include conventional public opinion survey data and election exit polls; data from survey experiments and responses to open-ended questions; aggregate statistics at different levels; state-of-the-art computerized analysis of text to compare and study the content of news media coverage and the policy issue agenda at the national and state level; as well as his synthesis of the latest relevant research. (With regard to these extensive data, it is frustrating that the extensive accompanying appendix is not available in the book but rather online.) Hopkins nicely lays out the important parts of his claims: that changes in national partisan politics along ideological and identity lines have penetrated and come to increasingly dominate state politics as well; that the cross-pressures that state and local issues used to provide and that made state parties different from the national parties (that came together only every four years to nominate a president) have diminished; and that citizens engagement with national politics and issues has come at the expense of attention to and participation in state and local politics. Lower levels of attentiveness and knowledge about local politics, and abysmal voter turnout, especially in primary elections, has been the result of both changes in party politics and the nationalization of the news media – especially the new national media provided by cable networks and the internet, which have competed with local television news. Meantime, newspapers that provided local news – with multiple daily editions in the forgotten past — have fallen dramatically by the wayside.

Hopkins provides an excellent discussion of the consequences of nationalization. There are obvious implications for democracy and representation—with respect to distinguishing state and local governments from the national government in order to hold them accountable for the policies and issues for which they are responsible. In this and other contexts, the idea of states as “laboratories of democracy” (230) goes out the window. Another consequence, and perhaps one that is more important today, is that nationalization feeds back into the partisan polarization that has caused it. State and local politics no longer provide a moderating force, with legislators cooperating through log-rolling and supporting earmarks for localities, and the like. The book would have benefitted from a further discussion of what led to the polarization or highly conflictual—and now highly emotional—‘partisan sorting’ that has occurred since the late 1970s on nearly all major issues (from New-Deal era Democratic and Republican parties that were divided only on economic and other ‘big government’ issues, and that had moderating forces—respectively, southern Democrats who were conservative on racial and labor issues, and liberal Republicans on issues of racial and other rights), with the parties now more evenly matched than they were when the Democrats controlled Congress for most of the post-World War II period. This aspect of current partisan politics would not have looked different if another Republican or Hillary Clinton (or another Democrat) were president.[2] It would still look like there are two Americas today—a liberal and conservative one[3]—along with political gridlock in government.

Hopkins nicely emphasizes both the upside and other downsides of the nationalization politics. When it expands the ‘scope of conflict,’ it can make issues, such as civil rights in the 1950s, more salient and force them on to the political agenda for remedy.[4] But this greater focus on national issue draws attention away from important and less partisan issues ‘on the ground’—such as roads, schools, and so forth—in state and local politics that affect people’s daily lives and toward which elected leaders should be responding and held accountable to the preferences of their district constituents or statewide electorates.[5] As Hopkins observes, nationalization undercuts the representation of local “unique communities of interest,” (6) when legislators ask “Is my party for or against this bill?” instead of asking, “how will this particular bill affect my district?” (7). This undercuts the purpose of a federal system and its institutions that acknowledge the importance of local interests and issues, which these institutions are better positioned to deal with than the national government. Moreover, in Hopkins’s review of other research, when individuals vote based on partisanship as related to one’s social group and identity, they are acting expressively and not instrumentally in terms of how state and local government actions benefit them, their families, or others in their community.

It is difficult to do justice to Hopkins’s systematic data and evidence. In only a couple spots are they less convincing than they could be, mainly due to limitations in the available data, and in a few instances where the presentations of data and labeling in a few of the figures could be clearer. One caveat, that he acknowledges, is that his non-national data are largely state and not local data, but this speaks more than well enough to the nationalization of American politics. To start, in Chapter 3, regarding the nationalization of elections, Hopkins’s multilevel modeling of survey data shows quite clearly that the party identification of voters has never varied much by state since the 1950s once individuals’ demographic characteristics are controlled, and what variation there has been that is attributable to the state has decreased over time. Further, and more directly relevant to vote choices, while there has been variation over time in county-level voting for president and governor, the correlation since the 1970s has increased from less than the 0.2 to 0.8 or better depending on how the states are parsed by region and by presidential or midterm election years. Exit poll data since 1990 show an increasing correlation from 0.6 to more than 0.8 between vote choices for governor versus president. And the home state advantage of presidents declined after the mid-1970s. So state as a context has been less relevant in these respects.

Chapter 4 looks at knowledge of and engagement with politics and leaders below the national level. For one, knowledge of the identity of one’s state governor has declined after the 1970s. Further, survey respondents have more to say about the president than their governor. And when asked whether they care more about whether their party wins the presidency or the governorship, more than twice as many individuals say the presidency. Looking at Google Trends searches, it is also clear that people are much more interested in seeking information about the president than their state governor. The available data show that since 1990 House and Senate candidates have gotten much greater contributions for their campaigns from sources outside their states, so that local interests are not driving them. And in comparing contributions within state to Senate candidates versus candidates for governor, the former have come to raise more money by a factor of as much as five-fold.  The comparisons of changes in relative voter turnout in presidential elections to mayoral, gubernatorial, and Senatorial elections, taking into account the election cycle, show that voter participation in presidential elections relative to the others has increased substantially.

In Chapter 5, Hopkins examines the extent to which local context affects political attitudes and preferences. Comparing a number of issues, he finds that local contexts in certain high salience cases—geographic closeness to a terrorist attack on 9/11, and locations with high crime rates and high unemployment—do affect politically related attitudes. But here, interestingly, these are issues that are highly nationalized, so that the localized influence is an extension of the national debate.

Chapter 6 turns to explanations for nationalization in which Hopkins elaborates on the already clear party-centered account, which is epitomized by how Republican gains in the House and the Senate in 2010 and 2014 were accompanied by Republican domination of state houses and state legislatures. Hopkins presents data that show the increasing correlation between the partisan presidential and gubernatorial vote closely tracking a measure of partisan polarization in congressional roll call voting. Through related correlational and other analyses, including the possible effect of the expansion of national government authority and scope, Hopkins rules out alternative explanations for the nationalization that has occurred.

To go beyond these correlations and to get more closely at causality, the next chapters are more important ones in getting more directly at the nationalization of partisanship as reflected in state government and politics. Chapter 7 examines changes in state parties and the American public’s perceptions of them, with an eye to looking for the same ideological cleavages that exist nationally and any differences across state parties that remain beyond that. There is persuasive evidence provided by Boris Shor and Nolan McCarty that state legislative parties—that is, politics at the leadership level—have become polarized in ways that are consistent with the pattern found in national data.[6] To further his case, however, Hopkins is interested in whether this state partisan positioning has become similar across states and also whether voters have come to see no difference between state parties and the national ones. This speaks to the question of whether the United States continues to have different state parties or just two national parties with state branches. Reviewing the Shor and McCarty data, Hopkins finds increased partisan differences in the same direction for all the state parties after 1995, when the available data begin. However, the Republicans and Democratic legislators across states vary in their ideological ‘ideal points’ (e.g., Massachusetts Republican legislators are less conservative than those even in other northeast states). Figure 7.2, which shows these trends, at first glance is not so easy to read. Hopkins examines state parties further by comparing state party platforms which cover a much larger period beginning in 1918. Here his computerized text analysis focuses on the amount of attention the platforms devote to different issues that have increasingly become partisan.  What is striking is that the Democratic and Republican state party platforms for the full period have touched on the same issues but have since the 1970s included the same issues around which national partisan conflict has occurred. Moreover, this increase over time is associated with strikingly less variation across states in the inclusion of these issues, and partisan differences have increased only slightly in their focus on particular issues.

Hopkins then provides evidence that the public sees little distinction between the state and national parties. This is reflected in low levels of split ticket-voting but also in similar distributions in national, state, and local party identification based on a survey designed to measure these different identifications. This one-time survey cannot, however, provide evidence for this being a change that has occurred since the 1970s. These similarities in perceptions of the state and national parties is reflected also in data from open-ended responses concerning how people evaluate these parties (again, we do not know if or how this differs from the past). In Chapter 7, Hopkins provides a neat analysis of exit poll data that show that once voters’ individual level partisanship and demographic characteristics are controlled, the positioning of the state parties, as state level factors, adds no further explanation for voters’ gubernatorial choices.

Chapter 8 reviews research that shows that state-based identities are not much related to political attitudes and behavior beyond their overlap with other social and partisan identities (as measured by individual-level characteristics). Hopkins finds further evidence in the text of Google Books collections that American identity was increasingly expressed more that state identity in the last part of the twentieth century. He finds this as well in survey data in which respondents are asked about “how important various things are to your sense of who you are?” (178). This is reflected further in a clever survey experiment which shows that individuals take it more as a personal insult “when someone criticizes my country” compared to “my state” or “my city or town” (182-183). Further, based on responses to an open-ended question asking what respondents are most proud of about the United States and about their state, there are many more references to national political factors and values regarding the United States, compared to the state (there also many more responses overall about the nation than about the state).

The last empirical chapter focuses on the media and the overall declining audiences for state and local news, and the present-day dominance of national news. Tracking this is critically important in that it speaks to the existence of a national community that is the target audience for the changes that have occurred in partisan politics and the nationalization of American political behavior. But in the back-drop there was evidence for a certain primacy of national political leaders to begin with. Hopkins’s computerized text analysis of a sample of daily political articles from the Los Angeles Times and Chicago Tribune from the 1930s to 1989, compared the visibility of news coverage of presidents versus governors and mayors. This analysis revealed that, historically, there was more coverage of the national chief executives than local ones, with state governors coming in third. For the periods since 1989, Hopkins’s analysis of 51 of the 67 largest newspapers (from Newsbank), found a local-to-national coverage ratio of 1 to 2, and less than 1 to 5 for state to national (with more state coverage in newspapers found in state capitals.) In terms of change over time, interestingly, there is has been a trend since 1985 of slightly increased local and state (relative to national) politics newspaper coverage, perhaps, as Hopkins suggests, due to increased competition from media outlets in the coverage of national politics. The evidence from local television new, based on available transcripts since 2006 (also from Newsbank), also reveals much greater local/national news coverage (just over a 4 to 5 ratio) compared to newspapers, and greater state/national coverage (just over 1 to 2 ratio) compared to newspapers – along with a similar increase over time to what was found for local papers.

While the availability of political news at all levels has been stable in local newspapers and television, Hopkins presents data that confirms that people rely on these local outlets for news about local and state politics. The big change, of course, is the decline in the audiences for local newspapers and local television news (Figure 9.5 shows this but is very hard to decipher [212]), compared to other sources, including cable news and the internet, which focus much more heavily on national news. The other driving forces for the changing news audience is generational change, with new/younger generations preferring national news outlets over “spatially bounded media sources” (213). What follows from this and reflects the overall audience trend, as Hopkins shows with a simple survey experiment, is that when individuals were given a choice of a headline on a President, Governor, or Mayor who “apologizes for a remark” (215), they were would devote their attention to the President. With regard to how news sources affect political knowledge and engagement, Hopkins shows persuasively from survey data that individuals who know more about state governors are those who rely on local TV news and newspapers, and those who live in a media market in which the state capital is located (so that local news outlets are more likely to cover what is going on in state government). This is shown strikingly further when respondents are asked to name members of Congress, in order to see which primary news sources are associated with them naming in-state versus out-of-state House members and Senators. As expected, people who used local/state news sources tended to name more in-state members of Congress, and those who reported using national news media were more likely to name out-of-state members. Last, and most notably, the media exposure that people get to state politics has increasingly affected voter turnout in gubernatorial elections. Analyzing county level turnout since 1928, Hopkins’s data and multivariate analysis show that living in state capital media market has had an increasing effect on county level turnout for governor versus president starting in the 1970s through 1990; and further on point, living in an out-of-state media market has led to an increasing negative effect on this gubernatorial to presidential turnout ratio during the same period. But after 1990 these effects lessoned, which is consistent with the effect of increasingly nationalized news media coverage and with the increasingly correlated partisan vote for governor and president.

Is there anything missing in this book that would have strengthened its overarching argument? It has plenty of relevant data and statistical analysis to make its case. What is missing is a narrative on certain national issues that play out in state politics that are a visible part of the nationalization of partisan politics. “Obamacare” is mentioned on page 1 but not in the context of state-level debates about the expansion of Medicare and the establishment of state health insurance exchanges, all of which were substantially decided in the states along partisan lines (with exceptions that may have warranted attention). Neither abortion politics, gay rights/gay marriage, immigration, nor gun control issues are singled out for discussion; again these have played heavily along strong partisan lines in the states. Discussing these issues and perhaps other specific ones would have made the book’s discussion a bit fuller and richer.

Two concluding observations: One, readers should not be left with the conclusion that no local politics remain. The book perhaps should have said more about this. For one, when one party dominates in a state, intraparty politics is important and candidates vying for party nomination have to pay attention to local political issues. To be sure, nationalized intraparty politics—with candidates staking out positions as moderates or more extreme liberals and conservatives—is currently playing out bigtime within the Democratic and Republican parties, as in the Ocasio-Cortez and Crowley race cited earlier. But local issues—ones like political corruption, service delivery, state and local economies and economic development, taxes, crime and criminal justices, government budgets, pensions, and so forth—cannot be ignored; just ask politicos, for example, in New York, New Jersey, and Illinois. But from Hopkins’s substantial and persuasive evidence and discussion, politics is unequivocally a lot less local than it used to be.

Second, Hopkins might somewhere have noted that that there were two tensions relevant to his account in the 1950 recommendations to the American Political Science Association in “Toward a More Responsible Government: A Report on the Committee on Political Parties.”[7] The first tension bears on the partisan conflict and gridlock that we have today, in that the Report recommended that the national parties move in the direction of giving voters clearer choices in their policy agendas. The parties today have clearly moved far in that direction. Second, and front and center for this book, the Report also recommended that the parties should become more tightly organized, disciplined, and nationally oriented (emphasis added). Indeed.


Robert Y. Shapiro is the Wallace S. Sayre Professor of Government in the Department of Political Science and Professor of International and Public Affairs at Columbia University, and Interim President of the Academy of Political Science. His most recent book is Selling Fear: Counterterrorism, the Media, and Public Opinion, with Brigittte L. Nacos and Yaeli Bloch-Elkon.

©2018 The Authors | Creative Commons Attribution-NonCommercial-NoDerivs 3.0 United States License



[1] Nolan McCarty, Nolan, Keith T. Poole, and Howard Rosenthal, Polarized America: The Dance of Ideology and Unequal Riches (Cambridge: MIT Press, 2006).

[2] On this rising partisan conflict, see Alan I. Abramowitz, The Great Alignment: Race, Party Transformation, and the Rise of Donald Trump (New Haven: Yale University Press, 2018); Joseph Bafumi and Robert Y. Shapiro, “A New Partisan Voter,” The Journal of Politics 71 (January 2009): 1-24; Morris P. Fiorina, with Samuel J. Abrams and Jeremy C. Pope, Culture Wars? The Myth of Polarized America, 3rd ed. (New York: Pearson Longman, 2011); Matthew Levendusky, Matthew, The Partisan Sort: How Liberals Became Democrats and Conservatives Republicans. (Chicago: University of Chicago Press, 2009); and McCarty, Poole, and Rosenthal, The Dance of Ideology (Cambridge: MIT Press, 2006).

[3] Ronald Brownstein, “There are absolutely two Americas. Sometimes in the same state.” 20 July 2018,

[4] See E.E. Schattcschneider, The Semi-Sovereign People: A Realist’s View of Democracy in America (New York: Holt, 1960).

[5] To say nothing of issues that are already under the radar screen on which business and special interests exert decisive influence at the state and local level. See Grant McConnell, Private Power and American Democracy (New York: Random House, 1966).

[6] Boris Shor and Nolan McCarty, “The Ideological Mapping of American Legislatures,” American Political Science Review 105:3 (2011): 530-551.

[7] American Political Science Association, “Toward a More Responsible Government: A Report on the Committee on Political Parties,” American Political Science Review Supplement 44 (September 1950), Part 2; Mark Wickham-Jones, “This 1950 political science report keeps popping up in the news. Here’s the story behind it,” 24 July 2018,