Extensions Winter 2015

A Journal of the Carl Albert Congressional Research and Studies Center

Winter 2015

The Geography of American Politics

The Carl Albert Congressional Research and Studies Center Established in 1979 by the Oklahoma Regents for Higher Education and the Board of Regents of the University of Oklahoma, the Carl Albert Congressional Research and Studies Center is a nonpartisan institution devoted to instruction and scholarship related to the United States Congress. The mission of the Center is defined broadly in terms of academic inquiry into the history, structure, process, personnel, and policies of the Congress, and the relationship between the Congress and other agencies and actors in the American political system. In the most general sense, the Center is concerned with the problems of modern representative democracy, as exemplified by the Congress.

In pursuit of this goal, the Carl Albert Center performs four principal functions. The first is the development of academic programs in congressional studies at both the graduate and undergraduate levels, which are sponsored in cooperation with the University of Oklahoma’s Department of Political Science. At the graduate level the Center offers a four-year, specialized fellowship program leading toward the doctoral degree. Each Fellow receives a fully financed program of study. At the undergraduate level the Center sponsors a research fellowship program designed to foster collaborative research between faculty and undergraduates.

Second, believing that professional research is the foundation upon which its academic programs rest, the Center promotes original research by faculty members and students into various aspects of politics and the Congress. The Center encourages

publication and provides its faculty and students with institutional and financial support to travel for research purposes and to present research findings at professional conferences.

The third function of the Center is the development of resource materials related to the Congress. The Center’s Congressional Archives, which are among the largest in the country, include the papers of more than fifty former members of Congress. Such prominent Oklahomans as Speaker Albert, Dewey F. Bartlett, Page Belcher, Mickey Edwards, Glenn English, Robert S. Kerr, Sr., Fred Harris, Steve Largent, Dave McCurdy, Mike Monroney, Tom Steed, Mike Synar, and J. C. Watts have donated their papers to the Center along with such distinguished nonOklahomans as Dick Armey, Helen Gahagan Douglas, and Carl Hatch.

Fourth, the Center actively strives to promote a wider understanding and appreciation of the Congress through various civic education programs. The Center sponsors conferences, speakers, television appearances, and the biennial Julian J. Rothbaum Distinguished Lecture in Representative Government.

The Center also publishes Extensions, a journal which focuses on issues related to the Congress.

Taken together, these diverse aspects of the Carl Albert Center constitute a unique resource for scholarship and research related to the United States Congress.

Winter 2015

extensions

A Journal of the Carl Albert Congressional Research and Studies Center

Contents

Director and Curator Cindy Simon Rosenthal Associate Director Michael H. Crespin

Editor’s Introduction

Regents’ Professor Ronald M. Peters, Jr.

The Geography of American Politics . . . . . . . . . . 2

Ronald M. Peters, Jr.

Managing Editor and Assistant to the Director LaDonna Sullivan

Special Orders

Assistant Director for N.E.W. Leadership Lauren Schueler

Where Are the Working Class Republicans and Is There Something the Matter with Them? . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 James G. Gimpel

Archivist Nathan Gerth Bailey Hoffner National Advisory Board David Albert Richard A. Baker David L. Boren Richard F. Fenno, Jr. Joseph S. Foote Jess Hay Joel Jankowsky Thomas J. Kenan Dave McCurdy Frank H. Mackaman Thomas E. Mann Chuck Neal Michael L. Reed Catherine E. Rudder James C. Wright, Jr. Hon. Tom Cole 4th District, Oklahoma ex officio

extensions

Winter 2015

Change and Continuity in Congressional Districts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Michael H. Crespin

Reapportionment, Regional Politics, and Partisan Gain . . . . . . . . . . . . . . . . . . . . . . . . . 18 Kenneth C. Martis J. Clark Archer Robert H. Watrel Fred M. Shelley Gerald R. Webster

For the Record

News from the Center . . . . . . . . . . . . . . . . . . . 22

LaDonna Sullivan

Winter 2015

Extensions is a copyrighted publication of the Carl Albert Congressional Research and Studies Center. It is published twice each year and distributed free of charge. To receive copies of Extensions, or to obtain permission to reprint, please contact Katherine McRae, at (405) 325-6372 or e-mail to mcrae@ou.edu.

Extensions may also be viewed on the Center’s web site at www. ou.edu/carlalbertcenter.

1

Editor’s Introduction

The Geography of American Politics Ronald M. Peters, Jr. James Madison’s most famous contribution to political science is his theory of the “extended sphere,” offered in the tenth Federalist and other writings.

Standing the “celebrated Montesquieu” on his head, Madison argued (contrary to Montesquieu) that republican government was most viable in a large, rather than a small, country. The size and diversity of America in the 1780s, and its prospects for future geographic expansion and population growth, boded well for the viability of its republican institutions. By “extending the sphere (and) taking in a greater number and variety of interests” the American republic, by the scope and diversity of its population, would counter the tendency of democracy toward majority tyranny.

Perhaps less well-known is Madison’s concern with the composition and distribution of the population. In a series of articles written for The National Gazette, the organ of the nascent Republican Party of that day, Madison ruminated on the forces that would likely shape the American polity.1 In “Population and Immigration” he offered a robust brief on the advantages of immigration to both Europe and America. The surplus labor of Europe would fill the empty spaces of America, creating new markets for European goods. In “A Republican Distribution of Citizens” he considered the impact of a growing industrial economy on the character of the American people and their capacity for republican citizenship. Echoing Jefferson’s preference for the virtues of agrarian life, Madison suggested that “In appreciating the regular branches of manufacturing and mechanical industry, their tendency must be compared with the principles laid down, and their merits graduated accordingly.

Whatever is least favorable to vigor of body, to the faculties of the mind, or to the virtues or the utilities of life, instead of being forced or fostered by public authority, ought to be seen with regret as long as occupations more friendly to human happiness, lie vacant.” Clearly Madison thought that the “extended republic of the United States” would be served well by the abundance of land on the western frontier available for settling and farming.

But what if the arable land were to become taken up as population advanced? Richard Matthews, in his book If Men Were Angels, argues that this development, which Madison thought inevitable, would lead to the rise of an urban underclass and a division of American society into economic classes, the propertied and the un-propertied, that would eventually swallow 2

the pluralism that the theory of the extended sphere anticipated.2 In the Constitutional Convention, Madison had predicted that, eventually, “a great majority of the people will not only be without landed, but any other sort of property.” What then?

Madison’s speculations and ruminations about the shape of the American polity were always set in the context of his mechanistic political science, which looked to institutional structure to combat, channel, or shape social forces. Recognizing that majority rule was always the fundamental principle of republican government, Madison acknowledged that public opinion must, in the end, prevail.

In his Gazette article “Public Opinion,” he put it this way: “Public opinion sets bounds to every government, and is the real sovereign in every free one.” The practical problem, then, was to design institutions to channel, and where necessary constrain (or at least delay), the influence of public opinion. These considerations lay at the root of Madison’s views on issues such as the structure of representation, federalism, the Electoral College, and the suffrage. He argued powerfully (although unsuccessfully) for proportional state representation in the U.S. Senate. That would have offered a Senate bonanza to large population states like Virginia, similar to the representational advantage offered by the 3/5ths compromise in the House. Although the Constitution did not specify qualifications for suffrage, Madison always thought that property should be protected in a scheme of representation.

And he was apparently satisfied with the Electoral College that, while skewed toward the small states, put a barrier between the presidency and the direct expression of popular will.

These institutional arrangements sought to resolve several tensions that Madison viewed as inherent in the republican enterprise: stability versus responsiveness; elites versus masses; propertied versus un-propertied; country versus city; majority rule versus minority rights. Madison, and other members of the elite, sought to structure the rules of American republicanism so as to manage, if not resolve, these tensions.

So for Madison, the art of constitutional construction was an exercise on the border of population, geography, and institutional design. This issue of Extensions offers three articles that are similarly situated. Each offers insight into the interface of geography and politics within the framework of the rules of the game. In “Where Are the Working Class Republicans and Is There Something the Matter with Them?” James Gimpel takes Winter 2015

extensions

up the celebrated analysis of Thomas Frank in Frank’s book What’s the Matter with Kansas?3 Frank had argued that the working class citizens of his home state were voting Republican contrary to their economic interests because of their preoccupation with social issues. Religious conviction was the main motivation of these working class Republicans, running against a tendency toward working-class solidarity that Frank seemed to assume should normally occur.

Frank’s analysis was largely anecdotal and focused, as his title suggested, on the state of Kansas. Gimpel seeks to extend the analysis beyond Kansas’s borders by considering working class Republicans nationwide. To do so, he identifies counties that voted more than 50 percent

importance of property to republican stability. Contrary to Madison’s fear, but in accordance with his theory, Americans’ economic identities are complex. Just as there are many economic interests in a pluralist society, there are also many aspects of economic security to be taken into account. And how people feel about their circumstances is more salient to their voting preferences than an external perspective might assume. On this accounting, there is nothing the matter with Kansas’s working class Republicans.

Frank Republicans are located across the country; in fact there are relatively fewer Frank Counties in Kansas than elsewhere. Gimpel locates most in rural areas across the nation, in both red and blue states. Collectively, he estimates that

elections due to their dispersal around the country, it is accentuated within the confines of the states, and may as well be reflected in a stable House Republican majority in Congress.

Gimpel suggests that the Frank Republicans are more economically secure than Frank supposes. But it appears that many working-class voters are trending conservative both here and abroad. As Thomas Edsall notes: “In France, a September survey in Le Figaro shows that supporters of Marine Le Pen’s right-wing National Front are, in comparison to supporters of other parties, less well educated, employed in manual and blue-collar jobs and live in rural areas. The parallel with white working-class voters in the United

“T

HE SURVEY DATA CONFIRM THAT RESPONDENTS OWNING THEIR HOMES AND CARRYING

LOWER DEBT BURDENS ARE MORE LIKELY TO IDENTIFY AS REPUBLICANS EVEN IF THEIR INCOME

”

IS BELOW $40,000 PER YEAR.

for both John McCain and Mitt Romney, yet whose median income was below the median income of the states in which they are located. He then analyzes three economic variables associated with income security: self-employment, home ownership, and levels of debt. He finds that in these “Frank Counties” self-employment and home ownership are relatively high and levels of debt relatively low. Gimpel then draws on survey data to reinforce these findings. The survey data confirm that respondents owning their homes and carrying lower debt burdens are more likely to identify as Republicans even if their income is below $40,000 per year. Gimpel thus infers that there are economic factors shaping the attitudes of these voters and not simply the religious convictions that Frank had stressed. Madison had emphasized the extensions

Winter 2015

they comprise as much as 5 percent of the national electorate, enough he notes, to make a difference in national elections.

How much difference they might make in presidential elections depends, of course, on where they are located. The Frank Counties in eastern Washington, Georgia, or southern Illinois are of less consequence than those in swing states like Virginia, Colorado, and Pennsylvania.

The Electoral College matters.

But perhaps we should not be looking to presidential elections to find the main effect of Frank Republicans. In Kansas, after all, their main impact is at the state level, as witness policy developments over the past four years. Across the nation, Republicans have surged to legislative and executive majorities not seen in aggregate in over a century. If the impact of Frank Republicans is limited in presidential

States who have moved into Republican columns is clear.”4 Here, the suggestion seems to be that many downscale Republicans are worried, rather than self-confident and secure, about their economic circumstances. The question for them of course is, what policies will best secure their economic future? Frank assumes that working class consciousness ought to point in the direction of progressive Democratic policies. But it appears that working class consciousness either does not exist or it now points in a different direction for many working class voters.

James Gimpel’s concern is with the social and economic causes of partisan identification. Michael Crespin’s focus is on the consequences of redistricting.

Decennial redistricting upsets established representational relationships and can 3

lead to shifts in partisan control. How significant are these effects and what are the consequences? To answer these questions Crespin draws upon Geographic Information System (GIS) technology to create a measure of “percentage new,” i.e., the percentage of new voters in districts created off of the 2010 census and put into effect in the 2012 congressional elections. Drawing on this measure, he determines that states in which redistricting commissions or

“O

done so under Democratic control, and now Ohio, under Republican majorities, has proposed a bipartisan reform plan that will empower the minority party. The end result should be greater competition for state legislative seats. Importantly, however, the Ohio reform plan does not address congressional districts.5 Crespin’s analysis addresses representation and congressional redistricting. The decennial census also instigates reapportionment of House

clear, at least with respect to presidential elections. The Electoral College gains that the Republicans have made in many solidly red states have been offset by an increasingly competitive Florida, and the solidly Democratic behemoth California.

On net, the Republican advantage from two generations of population shift is as little as six Electoral College votes.

Not too long ago the Republican Party was regarded as having an advantage in the Electoral College derby due to

N NET, THE REPUBLICAN ADVANTAGE FROM TWO GENERATIONS OF POPULATION SHIFT IS AS

”

LITTLE AS SIX ELECTORAL COLLEGE VOTES.

the courts drew the new boundary lines experienced greater instability (a higher percentage new) than those drawn by state legislatures. Relating this finding to outcomes, he discovers that a higher percentage new is associated with both the election of new members and turnover in party control.

Assessing these findings from a representational point of view, we see an interesting example of Madison’s concern for stability and responsiveness.

Extra legislative redistricting upsets the gerrymandered apple cart, leading to more new constituents per district, more new members, and more turnover in party control. This appears advantageous from the perspective of creating competition for seats and chamber control. But at the same time, taking redistricting away from the legislature upsets established constituency relationships and may even invite more out-of-district money into congressional races. By contrast, legislative redistricting purchases stability and constituency continuity at the expense of genuine electoral competition, at least in general elections.

There appears some prospect that states may move in the direction of less partisan redistricting mechanisms. California has 4

seats among the states. The shifting of House seats over time has evident implications both for partisan control of the House and for the Electoral College balance in presidential elections. In “Reapportionment, Regional Politics, and Partisan Gain,” co-authors Ken Martis, Clark Archer, Robert Watrel, Fred Shelly, and Gerald Webster draw on their recent political cartography to examine shifts in apportionment in historical context.

The general story their data reveal is hardly surprising: a long-term shift from the Snow Belt (the northern tier of states, especially in the Northeast and Midwest) to the Sun Belt (the southern and southwestern states). This trend, which traces to the 1970 census, continued as a result of the 2010 census and consequent reapportionment.

What are the consequences of this shift in population and House seats to the Sun Belt? Given the transformation of the South from Democratic to Republican control, and the historical roots of the Goldwater wing of the Republican Party in the Southwest, one might assume that the Electoral College advantage has tipped decisively in favor of the Republicans. Martis et. al. suggest that the picture is not quite so

the solidly Republican character of the great “L” stretching through the Great Plains from the Canadian border to the Rio Grande, and then east to the Atlantic Ocean across the deep South.

The great “L” secured the presidency for Ronald Reagan and George H.W. Bush. Bill Clinton was twice elected in part because he was able to penetrate the “L” in Arkansas, Tennessee, and (in 1996) Florida. Population shifts into the “L” states should have solidified the Republican advantage, but clearly have not. And now we often hear that the Democrats have a step up in presidential elections due to the number of electoral votes counted in the blue states.

In fact, the balance of power seems likely to reside in large states like Florida and California. Historically, Republicans were particularly attentive to California, with good results. From 1948 until 1984 the Republicans placed a Californian on their national ticket eight times. They carried California in seven of these eight elections, and won the presidency in six of the eight. The Democrats have never placed a Californian on their ticket, and since 1988, neither have the Republicans.

Still, George H.W. Bush carried both California and the election in 1988.

Winter 2015

extensions

Since then, the Democrats have carried California in every election. George W. Bush lost California in 2000 and 2004, but was able to win the presidency because he carried Florida in both of those elections. The grimmest harbinger for the Republicans today is the fact that Florida seems increasingly amenable to the Democrats in presidential elections.6 Viewed politically, this is what President Obama’s immigration and Cuban initiatives are all about.

But what about the House of Representatives? Visiting the University of Oklahoma in March 2005, former House Republican Majority Leader Dick Armey predicted that the GOP would maintain

seats are in play is a function of state population, affected by natural population growth, in-migration, and out-migration.

How the House districts are designed is a function of the choices that elites (commissioners, judges, or legislators) make. The effect of reapportionment on presidential elections is due to the dynamics of civil society in relationship to that curious structural mechanism the Founders designed, the Electoral College.

It may affect how presidents govern, but has more to do with how they are elected.

Redistricting, by contrast, has more direct representational effects. But try as they might, political elites, be they Founders or redistricting authorities, cannot

Notes 1. Madison’s National Gazette articles cited here as well as his Convention remarks can be found on the Constitution Society’s web site at http://www. constitution.org/jm/jm.htm.

2. Richard K. Matthews, If Men Were Angels (Lawrence, Kansas: University of Kansas Press, 1995).

3. Thomas Frank, What’s the Matter with Kansas? (New York: Henry Holt, 2004).

4. Thomas B. Edsall, “The Rise of Welfare Chauvinism,” New York Times, December 17, 2014, accessed at: http:// www.nytimes.com/2014/12/17/opinion/ the-rise-of-welfare-chauvinism.html?hp&

“B

UT TRY AS THEY MIGHT, POLITICAL ELITES, BE THEY FOUNDERS OR REDISTRICTING

AUTHORITIES, CANNOT DETERMINE THE CHOICES THAT VOTERS MAKE. THESE ARE SHAPED BY THE

”

ECONOMIC AND POLITICAL FORCES AT PLAY IN CIVIL SOCIETY.

its House majority for perhaps forty years due in large part to the demographic trends that Martis et. al. describe.7 Yet in 2006 and 2008 the Democrats swept to initial and then substantial House majorities. The backlash that occurred in 2010 enabled the Republicans to regain the House majority and to control redistricting in many states, including some Democratic and swing states.

The likely result now appears to be a Republican House majority extending to or past the 2020 census.

We are now able to bring the reapportionment analysis of Ken Martis and his co-authors into relationship with Michael Crespin’s analysis of congressional redistricting and James Gimpel’s diagnosis of voter preference.

The effects of reapportionment are shaped by the nature of population shifts from state to state. The effects of redistricting are shaped by the mechanisms by which it is undertaken. How many House

extensions

Winter 2015

determine the choices that voters make.

These are shaped by the economic and political forces at play in civil society, as James Gimpel argues, and may have more to do with how voters feel than about their objective circumstances. Kansas’s voters have choices.

These are matters on the interface of institutional design and the shape of civil society. As Madison understood, political theory is a matter of structuring institutions in contemplation of the dynamics of civil society. Civil society is constituted by the citizens’ character and concerns, arising from life choices that are shaped by social and economic forces. Madison saw politics in terms of population, space, and time. He could not have imagined the powerful tools available to modern researchers for describing and analyzing the geography of American politics. He would be envious.

action=click&pgtype=Homepage&modu le=c-column-top-span-region&region=ccolumn-top-span-region&WT.nav=ccolumn-top-span-region&_r=0, December 17, 2014.

5. Trip Gabriel, “In a Break From Partisan Rancor, Ohio Moves to Make Elections More Competitive,” New York Times, December 21, 2014, A25.

6. If we count the 2000 election as essentially a tie, we see that in the 1988 through 2012 elections, the Republicans and Democrats have each carried Florida three times, with one tie. However, the Democrats have carried Florida in the last two presidential elections. It is not surprising that there are now two plausible Republican nominees from Florida, Jeb Bush and Marco Rubio.

7. The Carl Albert Center proudly serves as the repository for Congressman Armey’s papers.

5

Where are the Working Class Republicans and Is There Something the Matter with Them?

James G. Gimpel

University of Maryland

I

n his widely read 2004 book, What’s the Matter with Kansas?, the Democratic pundit, Thomas Frank, raises the question of why so many Kansans vote contrary to their economic interests for Republican Party candidates. Frank’s “Kansans” are not just people who live in Kansas, of course. Kansas is a metaphor for legions of white working-class voters inside and outside the state who earn modest incomes and can often be described as financially marginal but who reliably vote for Republican candidates in presidential, state, and local elections. To be sure, there is some ambiguity about just who Frank classifies as white and working-class.

Since they are “working” or at least have worked, apparently he does not mean the very poor, but some combination of the working poor and the lower middle class.

The precise size of this population is unknown and where, exactly, they live is largely unspecified.

Among this group, Republican voting is considered a bad political habit because GOP policies are adverse to the material interests of such voters, favoring government policies that benefit the rich first and foremost, while shunting aside the needs of those of low-and-moderate income. Frank goes on to conclude that figurative Kansas has been duped. Voters there are fooled into supporting Republicans on the basis of cultural and social themes, e.g., family values, only to have Republican insiders ignore these concerns once in power, even as they govern in a contrary manner on economic issues, pursuing an agenda that encompasses tax cuts, deregulation, free trade, and corporate subsidies.

This is not a new critique of working-class white voters, or of the Republican Party. Journalists and scholars have puzzled over working-class support for Republicans at least since the 1980 election of Ronald Reagan. Whatever its size, this group has been around for a while. One question to be addressed here is whether there is a distinctive geography associated with these working-class Republicans or “Frank voters.” Beyond Kansas, where might they be found?

James G. Gimpel (Ph.D., University of Chicago) is a professor at the University of Maryland, where he teaches social scientists how to use GIS tools in their research. His email address is jgimpel@umd.edu.

Locating the Frank Counties Since there are no surveys that sample densely enough to capture localized geographic patterns of class and party identification for the entire nation, one might begin by examining U.S. counties. Specifically, Figure 1 displays the location of counties that gave at least 50 percent of their votes to the Republican candidate in both the 2008 and 2012 presidential elections, but are also situated well below their state’s median income.1 The 2010 census indicates that these places are small in size, with an average voting age population of about 26,000.

6

Winter 2015

extensions

As the darkly shaded terrain in Figure 1 suggests, they are predominantly rural, but constitute entire regions in some states, for example much of Eastern Washington and the plains of Eastern Colorado as well as all of Southwest Virginia and Southeast Illinois. Central Minnesota and Southern Missouri fit the profile, as do rural counties in Georgia, New York, and Pennsylvania. In total, an estimated 16 million people reside in these 613 counties, about 5 percent of the total population, and they cast about 4.8 percent of the popular vote in the 2012 presidential election. Conspicuously, not even a majority of Kansas counties qualify. Only about 134 of these Frank counties are situated in the Agricultural Midwest region in which Kansas is grouped. New England contains only one Frank county, but they are also abundant in the rural reaches of the South and West (see Figure 1). The ellipses drawn on Figure 1 provide a summary of the geographic distribution of working class Republicans (in red) compared with all voters (in blue) weighting the counties by the numbers of voters in each, something impossible to grasp simply looking at the shaded counties alone. After all, shaded maps are always better at capturing land area than they are population. What is evident is that the geography of working class Republicans weighted for their electoral contribution is tilted slightly

major segment of the electorate either nationally or in most states. Approaches using survey research would likely find little evidence of their existence.

Moreover, in a typically sized survey, there might not be enough of these voters to make a striking difference to statistical estimates. Perhaps this is why researchers taking the survey approach have dismissed the Frank thesis as weakly supported, or even wrong. Of course

“I

N THIS MANNER, OWNER OCCUPATION OF PROPERTY,

LIKE SELF-EMPLOYMENT, FRAGMENTS THE WORKING CLASS, OBSTRUCTING THE COHESION NECESSARY FOR THE SEEDS OF

”

RADICALISM TO GERMINATE.

more southeastern and northwestern than the population of voters in the nation as a whole (Figure 1).

Even if one were to credit all of this population to a coalition group labeled “low-to-moderate income Republican voters,” surely a fallacy since county populations are not homogeneous, this does not appear at first glance to be a extensions

Winter 2015

locating and counting Frank counties is not the same as counting Frank voters.

Working class Republicans could arguably be found just about anywhere, and a sufficiently dense survey sample might reveal them to be living in Manhattan, and in Hawaii, and in Key West. I will turn to individual level data on working class Republicanism shortly.

For now, suffice it to say that any bloc of voters that could sum to as much as 5 percent of the voting population is of consequence. When parsing out other voting groups by, say, age group, party, occupation, religion, region, and/ or a few other characteristics, 5 percent amounts to a bloc of substantial size. For instance, the entirety of New England’s vote amounted to 5.4 percent of the total vote cast in 2012. Win just a handful of groups this size, and you win the election.

The next question is the one Frank asks: what explains the propensity of these locations to cast Republican votes? Some might point to Evangelical Christian religious adherence, and that might help to account for socially conservative supporters in some places, for instance, in the South. However, the South has an abundance of white Evangelicals and Fundamentalists, compared to other regions, but they are not particularly concentrated among the working class, instead showing a presence at all levels of income. Other parts of the country can also boast large proportions of Evangelical Christian adherents, but they do not stand out as centers of lower income Republican strength.

7

Homeownership and Self-Employment as Explanations In fact, one does not require the discovery of a particular religious confession to find Republican support at low and moderate levels of income.

An alternative explanation goes to the local business and home economies of these locations. Specifically high rates of property ownership and selfemployment, coupled with low rates of indebtedness, have been an obstacle to the development of a working-class consciousness. Evangelical Protestant religious traditions are not necessarily linked to these patterns of economic organization, in spite of Max Weber’s theory that Protestantism and capitalism went hand in hand through a shared emphasis on individual responsibility.

In contemporary decades, there are too many individuals expressing Christian beliefs who are not themselves involved in small business enterprises to regard these as united forces. At best they are complementary, but not fused.

The argument that owning property or a small business, however modest, is an obstruction to the development of class consciousness (quite without religion) is not new. Being self-employed offers workers a greater measure of freedom, flexibility, and self-determination, quite the opposite of hierarchical workplaces where rules and constraints are imposed by superiors. Self-employment is certainly no guarantor of higher income or success in business pursuits. But it has been associated with greater job satisfaction and happiness in numerous social science studies. The self-employed, therefore, do not share with similarly paid wage workers on the shop floor the same sense of hierarchical and tense relations in the workplace that can make lower income workers feel vulnerable and powerless, resulting in calls for working class solidarity and political mobilization.

A similar logic applies to property ownership. Ownership of private property and, specifically, owner occupation of a residence, signifies not just the possession of a significant asset, but also a degree of empowerment that tenants do not possess. After all, ownership 8

offers a measure of control as well as the potential for gains from an investment.

Property need not be of great value to be a source of enablement and confidence.

Presumably, many lower income property owners are vested in land and dwellings that have little market value and are not productive – the stereotypical single-wide mobile home comes to mind. But simply owning something is sufficient to give a citizen a stake in the community, spur involvement, and thereby replace the potential for unrest with social stability.

In this manner, owner occupation of property, like self-employment, fragments the working class, obstructing the cohesion necessary for the seeds of radicalism to germinate.

Finally, there is the element not only of owner-occupied residences, but also low levels of indebtedness that hinder working class cohesion. Debt is a known stressor, averse to mental and physical health, self-esteem, and marital stability.

Mortgage debt, in particular, is associated with housing insecurity and the risk of foreclosure, an unpleasant event that hundreds-of-thousands of Americans have recently experienced. Secure mortgage loans tend to assume the stability of

employment over a long period of time, something that in contemporary times fewer low and moderate income citizens can count on, placing debtors in a highly exposed position. This is why those who own their homes outright, or owe less, typically possess a measure of autonomy over their lives and express a level of satisfaction that the debt-ridden do not.

In summary, self-employment, home ownership, and low debt levels do not add up to affluence. But they are commonly associated with fewer grievances, less of a sense of vulnerability and classoppression, and reduced support for a left-leaning political agenda. This is why we find a statistically close relationship between self-employment, homeownership, and the locations of Frank counties as defined by their Republican loyalty and modest income.

The data displayed in Figures 2 and 3 help to make the connection. Figure 2 shows the geographic distribution of homeowners who have no mortgage debt.

The darker shades indicate locations that are relatively debt free. Perhaps what is most noteworthy is the relative scarcity of such locations on the West Coast (California, Oregon, Washington) and in Winter 2015

extensions

the Southeast (Virginia, North Carolina, South Carolina, Georgia, Florida). Places dotting the middle of the country, running from Texas and Louisiana to North Dakota and Minnesota, contain the smallest shares of indebted homeowners. Low debt homeownership is also associated with small towns and elderly populations.

Figure 3 maps the percentage of selfemployed workers by county. There is an evident north central and northwest concentration of counties in the top quintile (darker shades). There are a few other locations on the map, but nothing compared to the agriculture and retail trade-driven entrepreneurship of the Upper Plains and Mountain states. Selfemployment is notably lowest in the most populous and urban locations, and most robust in the smallest towns.

To be sure, inspection reveals that the three maps do not perfectly match up. There are locations where selfemployment is high, but home ownership is middling or low. But where both are high, there is some tendency to find a Republican voting county containing low and moderate income voters of the kind Thomas Frank describes.

Evidence from Statistical Comparisons A simple statistical analysis clarifies the relationship that is only suggested by viewing the maps. The comparison in Table 1 exhibits characteristics of “correct” voting Democratic counties of low income, with the “incorrect” voting Republican counties meeting the same low income criteria. Comparing the mean values for the two groups reveals some striking differences on a variety of measures (see Table 1).

First, it is of considerable moment that there are far more low income Republican counties (N=613) than there are Democratic counties (N=239), a reminder that places and individuals can be entirely different. It is no secret that most low income people live in densely populated cities that register higher median incomes because of their tendency to also attract the affluent.

At the county level of observation, however, lower income locations supportive of Barack Obama are racially and ethnically diverse mid-sized cities in the South and West, not the nation’s largest cities.

If Frank’s analysis is correct, the Evangelical Christian population, in number of adherents per 1000, should exhibit a higher mean level in the working class Republican counties than in the Democratic ones. And indeed the Republican counties show both higher Evangelical adherence and less than half the rate of reported Catholic adherence (see Table 1).

Table 1. Difference in Means Between Low Income Democratic and Low Income Republican Counties Mean Democratic

Mean Republican

6.2

9.0

% Manufacturing

10.7

11.9

% Retail Trade

11.1

11.6

5.8

5.2

% Government

22.8

19.2

% On Social Security

35.0

38.8

% Agriculture

% Prof/Sci

% Self-Employed

7.5

10.0

% Renter Occupied

33.2

25.2

% Owner Occupied

66.8

74.8

% Owner Occupied No Mortgage

53.9

60.2

Home Cost as % of Income

37.3

32.1

% Age 65 Up

14.8

18.0

% White

59.0

86.5

% Age 18 Up

66.935

19.524

Total Population

89,008

25,632

Median Age

38.1

42.1

In Metro Area

16.7

5.7

Gini Index

46.3

43.8

Catholic (per 1000)

190.9

87.7

Evangelical (per 1000)

164.4

275.5

N

239

613

All differences statistically significant at .05 level.

Sources: 2012 American Community Survey, 5-Year Estimates for population, occupation, income, and housing. 2010 Glenmary Research Institute Survey of Congregations for religion information.

extensions

Winter 2015

9

The economic differences between the Democratic and Republican locations are also large, in spite of their similarly low incomes. Expressed as percentages, there are more homeowners and more homeowners without mortgages in the Republican counties. Self-employment is also higher, and it is not just agricultural employment that is captured here, although that is also a more important economic activity in the Republican counties than in the Democratic ones.

The Democratic counties report higher levels of government employment, and more of them lie within or near major metropolitan areas. The Democratic voting counties of low income exhibit greater inequality internally, judging from the Gini index, as well as a lower median age, along with a smaller share of the population drawing on Social Security (see Table 2). On multiple measures, in fact, the Republican counties have a larger share of their population in older age brackets. Finally, the Republican locations of low and moderate means carry less debt, and spend less on housing as a percentage of income. Undoubtedly some of these variables are related to one another, but what the averages in the Republican column of Table 1 depict is an image of the Republican base that fits perfectly in all respects except for affluence. To find the stereotypical rich Republicans, you have to look elsewhere.

In sum, the Republican counties are ill-suited to the propagation of workingclass consciousness because they offer support not only to socially conservative values, but reinforcement for economically conservative ones. Democratic locations, for their part, predictably exhibit interests and values that prop up that party’s support, including a higher percentage employed by the government, and more tenants and debtors. The economic and racial disparities that show up in the Democratic counties likely lead to more frequent calls for redistributive and regulatory policy intervention than one would find in the Republican low income locations.

Individual Level Comparisons Given that a majority of working class Democrats and Republicans may not live 10

in or even near the locations described by the statistics in Table 1, there is a need to include an analysis of individual level data if it is available. It is not easy to find information about self-employment, homeownership, personal debt, and partisan affiliation all in the same survey.

But there is sufficient intersection of the necessary questions to permit a few crosstabulations from recent years.

Consistent with the reasoning above, we should expect to find that low-tomoderate income homeowners and the self-employed to be Republicans more so than those similarly situated but renting or employed by someone else. Similarly, those occupying lower income brackets

location, particularly with the cost of housing. But surveys usually report only the respondent’s income ranges, and not specific income amounts, as from a tax return. Given that recent American Community Survey estimates place mean income for individuals at or near $40,000, and median income for individuals at a little under $30,000, a reasonable but conservative cut-off for defining “lowto-moderate” income should be at one of those two levels and below. Surveys commonly ask about family income, or household income, rather than individual income, so a threshold of $40,000 and below under those circumstances is appropriately conservative.

Table 2. Party Identification of Respondents with Family Income Below $40,000 Republican

Independent

Democrat

Total

Homeownership Rent

714 23.3%

471 15.4%

1879 61.3%

3064 100%

Own

708 40.5%

133 7.6%

907 51.9%

1748 100%

Total

1422 29.6%

604 12.6%

2786 579%

4812 100%

Self Employed

348 39%

188 21.1%

357 40.0%

893 100%

Not Self Employed

779 28.7%

403 14.8%

1532 56.4%

2714 100%

Total

1127 31.2%

591 16.4%

1889 52.4%

3607 100%

Borrow More

410 31.9%

141 11.0%

736 57.2%

1287 100%

Neither Borrow nor Cut Back

295 35%

137 16.2%

412 48.8%

844 100%

Cut Back

1101 33.1%

561 16.9%

1662 50%

3324 100%

Total

1806 33.1%

839 15.4%

2810 51.5%

5455 100%

Employment

Response to Recession

Sources: Pew Research Center, May 2010 and July 2012 Middle Class Update. Party leaners are reclassified as partisans, not independents.

but not carrying a heavy burden of debt should also be more Republican than those who are indebted.

The income threshold below which a person should fall in order to be considered working class does vary by

Table 2 reports results from simple crosstabulations of party identification by homeownership, self-employment, and propensity to borrow for those earning less than $40,000 per year. With respect to home ownership, a 2012 Pew survey Winter 2015

extensions

revealed that, although a solid majority of voters in these lower income brackets are Democrats (57.9%), it is also clear that homeownership is associated with Republican Party identification among this group. Among low-to-moderate income homeowners, 41% are Republican, compared with only 23% for renters.

The relationship is similar for those who are younger than age 65 (not shown here), an indication that it is not that Republicans in lower income brackets are just older, and have finished paying off their mortgages. Homeownership is also more highly associated with Republican Party identification for those earning low incomes than it is for those earning above $40,000. Similar Pew surveys from June and July 2011, and from May 2010 show similar results.

(even at young ages), but they also report a slightly greater tendency to have cut back more rather than borrowed in response to the financial pressure of the recession, although this difference is admittedly small. A higher percentage of Democrats than Republicans in this income bracket also reported losing their homes to foreclosure during the recession, and a substantially higher percentage of Democrats admitted that it was acceptable to walk away from a mortgage (table not reported). In summary, attitudes toward personal finance and tolerance for debt exhibit substantial partisan differences, independent of income level.

To conclude, there is support for the idea that the adherents of the two parties differ in important economic characteristics separate from income,

in low-and-moderate income locations with fewer Evangelical Christians shows that working class consciousness is impeded by more than socially conservative views on abortion and gay marriage.

In the last few years, both political parties have placed a far greater campaign emphasis on economic issues than on fighting the culture wars. In spite of this emphasis, Republicans have done exceptionally well in two recent mid-term contests, lagging behind only in the two presidential contests. This does not mean that the viewpoints of religious conservatives should be ignored, or that the Republican Party can win back the presidency without them. But it does suggest that economic and socially conservative messages act independently

“R

EPUBLICAN COUNTIES ARE ILL-SUITED TO THE PROPAGATION OF WORKING-CLASS

CONSCIOUSNESS BECAUSE THEY OFFER SUPPORT NOT ONLY TO SOCIALLY CONSERVATIVE VALUES,

”

BUT REINFORCEMENT FOR ECONOMICALLY CONSERVATIVE ONES.

As for self-employment, a Pew survey on recovery from the recession in May 2010 asked questions not only about homeownership but source of employment, as well as debt. The lower income self-employed are about 10 percent more Republican than those working for someone else (see Table 2).

Democratic identification is 16 percent lower among the self-employed in this lower income bracket than among those who are employed by another person.

For older people among the self-employed, the association with Republican Party identification is even stronger, but controlling for age does not appreciably weaken the relationship for those under age 65.

In terms of indebtedness, for those of low-to-moderate income it does appear true that Republicans carry less debt than Democrats, as the aggregate data appeared to indicate. Not only do fewer Republicans owe money on their homes extensions

Winter 2015

per se. On the one hand, there are many more Democrats than Republicans earning less than $40,000 per year.

On the other hand, why are there any Republicans at all on these lower rungs of the economic ladder? Where is their working class solidarity? According to Thomas Frank, they are disloyal to their income group.

Surely income is not the only economic characteristic that matters to self-perception and partisan identification.

Republicans earning modest incomes also carry less debt, are more likely to report owning their own homes and to be working for themselves than similarly situated Democrats. These characteristics add up to a temper of independence and empowerment that produces fewer calls for government intervention and regulation on which contemporary Democratic Party platforms are founded.

Religious belief unquestionably matters but the high levels of Republican support

of one another in crosscutting class cleavages and forming the nation’s characteristically exceptional political party coalitions. Frank’s “Kansans” stand out even lacking a religious dimension to political party conflict.

Notes 1. As estimated by the 2012 5-year American Community Survey. Since income levels and living costs vary by region and state, county median income was subtracted from state median income for all U.S. counties. Alternative definitions can be specified but here differences of more than $12,000 below the state median (or approximately onehalf standard deviation below) were deemed large enough to designate a county as lower-to-middle income.

11

Change and Continuity in Congressional Districts Michael H. Crespin

University of Oklahoma

O

nce a decade nearly all of the congressional districts in the United States undergo some degree of change. In a few states, district boundaries are drastically altered in the wake of political gerrymandering to reflect changing partisan and demographic landscapes. In others states with relatively stable populations and a limited number of representatives, lines have hardly changed over the past fifty or so years since the start of the redistricting revolution that begin with the Baker v Carr decision handed down in 1962. Of course, re-shaping districts is not a modern phenomenon and has been going on since the Founding. Although some question the claim,1 others assert that Patrick Henry manipulated district boundaries in Virginia prior to the 1789 elections in an attempt to clear a path for James Monroe to serve in the First Congress and keep James Madison from holding office. While the effort was unsuccessful, it took place over twenty years before the term “gerrymander” was coined after the Massachusetts state senate elections in 1812.

A great deal of research focuses on the political aspect of redistricting, while scholars have placed less emphasis on the simple fact that redistricting represents the shuffling of constituents into different districts. Even when the new voters are ideologically similar to existing ones, switching districts means old representational ties are broken and fresh ones must be forged. In this article, I will explore some of the ways changing districts have influenced representation and elections. First, I will discuss what the extant literature tells us about changes in constituents. Then I describe one way to measure change and continuity in congressional districts before presenting some original data and results related to the 2012 election outcomes.

Representation and Elections When large numbers of new constituents are introduced to a district, research shows there are both electoral and representational consequences. Theoretically, we would expect population changes to induce instability into the system. Members of Congress dedicate a great deal of time working in their districts to establish a positive identity, build trust with voters, and increase name recognition. They accomplish this through years of casework, working with the local media, and good old-fashioned hand shaking and baby kissing. Members also spend hours every week on the phones and attend local events to fundraise for future campaigns. All of this and more goes into building an incumbency advantage over any potential challengers. When districts change, most of this work goes for naught as new constituents have not witnessed any of the efforts of their new representative. On the flip side, political activists and donors who are drawn into different districts may find they have little to no pull with their new representative. In essence, the relationships built over the years are torn apart, and 12

Michael H. Crespin is associate director of the Carl Albert Congressional Research and Studies Center at The University of Oklahoma and an associate professor of political science. His research focuses on legislative politics, congressional elections, and political geography. Some of his work has appeared in American Journal of Political Science, Journal of Politics, Legislative Studies Quarterly, Political Analysis, and State Politics and Policy Quarterly. His email address is crespin.edu.

Winter 2015

extensions

“W

ITH THE ADVENT OF GEOGRAPHIC INFORMATION

SYSTEM (GIS) TECHNOLOGY, IT IS NOW POSSIBLE TO CREATE A PRECISE MEASURE OF DISTRICT CHANGE FOR ALL DISTRICTS, AT LEAST WHEN ELECTRONIC VERSIONS OF DISTRICT

”

BOUNDARIES AND BASE LAYERS EXIST.

political actors have to create them all over again.

Any changes can manifest themselves in the voting behavior of representatives.

For instance, members of Congress may have a difficult time adapting their voting patterns or representational style to the new district. While some researchers show that representatives become more liberal or conservative along with their districts2, others find that members do not always immediately adapt when there are big changes due to the uncertainty of the new district characteristics.3 It might take some time for representatives to learn just what the district wants on every issue. In the meantime, representation may be suboptimal.

Adjusting districts can also have unintended consequences on fundraising efforts. Recent research shows that in elections after a redistricting, incumbents decrease the proportion of money they raise from inside of the district. This out-of-district fundraising is especially pronounced as more and more new constituents enter the district. While the effect grows smaller as time passes, it never returns to pre-redistricting levels for the average member and reemerges after the next round of district adjustments.

This result has serious normative consequences if members find themselves beholden to donors who do not live in the district. When out-of-district donors are moved into recipient districts’, they are more likely to make a contribution than if they remained in the old district.4 Knowing this, mapmakers frequently make a conscious effort to herd donors into some districts and out of others.5 Antoine Yoshinaka and Chad Murphy have a series of related papers on the electoral consequences of district change extensions

Winter 2015

and argue mapmakers are strategic actors who try to manipulate the redistricting process to achieve the desired outcomes.

In their research they label redistricting plans as partisan, bi-partisan, or neutral and explore the extent to which plans foster population instability in some districts more than others. They find that mapmakers target cross-partisan incumbents and try to limit the degree of district continuity after a redistricting6, yet keep their own districts relatively intact.7 Furthermore, incumbents are more likely to find themselves in a competitive general election when district continuity decreases, although Yoshinaka and Murphy fail to find evidence that members will retire when faced with large changes in their districts.8 Measuring District Change There are many ways to measure how much a district has changed from year to year. Early research relied on an indicator variable to denote if a district changed at all, while other sources might classify redistricting as major or minor. Some scholars examined if a district changed politically after a redistricting by looking

at the presidential vote in the old and new district. By only looking at political change and ignoring other measures, we may miss out on important nuances, especially those related to representation.

After all, it is easy to imagine shifting boundaries to create a district that does not change its partisanship, but is filled with a new set of voters that previously were located in other districts.

The replacement residents may look ideologically similar, but it can take time for representatives to learn their personal needs and general issue demands. To this end, I am more interested in determining how many new constituents a member of Congress might represent after a redistricting than changes in partisanship.

In some of my previous research, I worked with co-authors and inspected paper copies of district maps and county population data to determine if a district was continuous or new from year to year using 50 percent change as a threshold.9 We used this method to analyze districts from 1962 through the 2002 redistricting.

Unfortunately, using county maps as a base layer made it difficult to see how much geographically small districts in urban counties varied before and after redistricting since there were frequently several districts in a county.

With the advent of Geographic Information System (GIS) technology, it is now possible to create a precise measure of district change for all districts, at least when electronic versions of district boundaries and base layers exist. To calculate this statistic, a researcher needs to spatially intersect the old and new district boundaries and then overlay the

“A

S THE PERCENTAGE OF NEW CONSTITUENTS INCREASES,

SEATS ARE MORE LIKELY TO BE REPRESENTED BY A NEW MEMBER. HIGHER LEVELS OF PERCENT NEW ALSO MEAN A DISTRICT HAS A GREATER CHANCE OF EXPERIENCING A

”

CHANGE IN PARTISAN CONTROL.

13

resulting shape files on top of a mapping of census tracts. The tracts contain a relatively small number of people and are usually the building blocks for legislative districts. The “output� then identifies both continuous and new tracts to a district and it is fairly easy to calculate what percentage of the constituents are new (or continuous) to a district.10 As the percentage of new increases, I expect to see a greater influence on election outcomes and other related behavior.

Change for the 2012 Elections How much did districts change in response to the 2012 redistricting cycle?

To explore this question, I used the techniques described above to create a measure, percent new, that conveys the percentage of new constituents residing in each district for elections to the 113th Congress.11 Figure 1 displays the measure in a few different ways.

First, the map depicts the percent new variable for each district. This number ranges from zero for twenty districts 14

all the way to 74 percent new for the sixth district of Illinois represented

by Republican Peter Roskam. In the average member’s district, just over a quarter of the constituents were new at 27.4 percent. The lower bound is characterized by the seven states that have only one representative (Alaska, Delaware, Montana, North Dakota, South Dakota, Vermont, and Wyoming) and thirteen other districts that saw their population grow between 2002 and 2010 and subsequently lost portions to other districts. The upper bound is arguably a conservative estimate, since eight states gained new seats in 2012 and had to create entirely new sets of district boundaries. However, since these new districts are carved out of existing ones, I do not classify them as 100 percent new.

If we dig a little deeper into the data and examine the map, some interesting patterns emerge. For example, all of the Iowa districts are in the top percent new category, and overall it is the state with the most change with an average score

of 48.3 percent. This should not come as a surprise to some since Iowa uses a unique system where mapmakers draw plans using only population statistics and cannot use political data to come up with new boundaries. Although the final plan does receive a vote in the legislature, the end result is more change and less continuity. California also experimented with a redistricting commission to draw districts for the 2012 election and the final outcome was a great deal of change.

The average member of the California delegation represented a district that was 42.6 percent new. The few states that had to turn to the courts to draw new districts also saw lots of change with an average of just under 32 percent new constituents. On average, non-legislative plans were 31.5 percent new compared to legislative-drawn districts that were only 24.8 percent new. A t-test tells us the difference is statistically significant. This suggests if states want to shake up their districts, taking the process out of the hands of the legislature is one way to get Winter 2015

extensions

there. If they want to maintain the status quo, following normal legislative practices is recommended.

Districts in Georgia, Ohio, and Illinois also experienced a lot of change. This is likely the effort of unified governments trying to lock in their advantage over the course of the next decade in current and future battleground states. In other states like Oklahoma, very little change is necessary to maintain the status quo. Finally, states experiencing little population growth such as Maine, New Hampshire, and Idaho did not undergo large movements in district boundaries.

A second way to see the amount and variation of change and continuity is through a frequency distribution of the percent new in each district. These are also provided in Figure 1 separated by political party of the incumbent. The clumping near zero for the Republicans shows us that members of the GOP represent a greater number of districts with little change compared to the Democrats. In fact, the difference was 3:1 with fifteen no-change districts for the Republicans. Many of the Democratic districts underwent more adjustments and are clustered in the 20-40 percent new range. On average, Republican districts were 25.9 percent new while Democratic districts were 29.2 percent new. The extensions

Winter 2015

difference between the two parties is statistically significant, meaning it was unlikely to occur by chance. Although Republican districts did see less change on average, they also represented the few districts that exhibit the most change.

The lower levels of change in Republicanheld districts are likely a reflection of their control over the process across the country. The Republicans had complete (or veto-proof) jurisdiction in more states and districts than the Democrats.

Now that we know some districts saw a great deal of change while others saw none, we can examine if the amount of new constituents had any influence on the outcomes of the elections themselves.

To explore this possibility, I looked in two places. First, I wanted to test if percent new corresponded with the district being represented by a freshman in the 113th Congress. This can happen if the incumbent retires, loses in the primary or general election, or in new districts. A simple theory suggests that incumbents might retire because all the new voters are of a different party, and this reduces the chances the incumbents can hold on to their current seat. They might not want to put forth the effort to earn the votes of the new residents. Gary Jacobson and Sam Kernell12 would call this a strategic retirement. If incumbents do decide to

run, they might lose because they face a quality challenger, the new district is a bad fit ideologically, or the new voters do not know the sitting member and the incumbency advantage is reduced.

The second place I looked was to see if districts that saw a change in party control from Democrat to Republican or vice-versa experienced more population change. Although the same party controls most seats from congress to congress with incumbents returning or being replaced by new members of the same party, they do occasionally change hands. Since most districts are originally drawn to benefit one party or another, it seems to reason that a large shock might be enough to force a change.

To test these hypotheses, I turned to a statistical technique called logistical regression. This method shows us if higher levels of percent new correlate in meaningful ways with a dichotomous outcome of interest. The method also tells us more as we can determine the predicted probability of either event happening over the available range of the percent new variable. That is, if a district is 50 percent new, we can predict the probability that a freshman will represent the district or if it will change between the two parties.

15

After estimating the models, I found support for both hypotheses. As the percentage of new constituents increases, seats are more likely to be represented by a new member. Higher levels of percent new also mean a district has a greater chance of experiencing a change in partisan control. After estimating the models, I created two figures that show how different levels of change influence the events under study. In both figures, the dashed line is the probability of the event happening and the shaded portion is a 95 percent confidence interval. The confidence intervals illustrates the degree of precision in the probability and is usually wider when there are a limited number of observations. This is why they are wider at the upper bound of percent new and smaller near the average, where there are lots of observations.

Figure 2 shows the probability of a

3. Michael Crespin, “Serving Two Masters: Redistricting and Voting in the U.S. House of Representatives,” Political Research Quarterly 63(4): 850-859.

4. Michael H. Crespin and Barry Edwards, “Redistricting and Individual Contributions to Congressional Candidates,” a working paper.

5. Justin H. Kirkland, “WalletBased Redistricting Evidence for the Concentration of Wealth in Majority Party Districts,” State Politics and Policy Quarterly 13(1): 49-69.

6. Chad Murphy and Antoine Yoshinaka, “Are Mapmakers Able to Target and Protect Congressional Incumbents? The Institutional Dynamics of Electoral Competition,” American Politics Research 37(6): 955-982.

7. Antoine Yoshinaka and Chad Murphy, “Partisan Gerrymandering and Population Instability: Completing the Redistricting Puzzle,” Political Geography 28: 451-462.

8. Antoine Yoshinaka and Chad Murphy, “The VERALL, WHAT DOES CHANGE DO TO THE POLITICAL ENVIRONMENT? Paradox of Redistricting: How Partisan Mapmakers FIRST AND FOREMOST, IT INTRODUCES INSTABILITY TO THE SYSTEM. Foster Competition but Disrupt Representation,” Political Research Quarterly 64: 435-447. freshman representing a district over This result certainly helped them hold 9. Jamie Carson, Michael Crespin, the range of percent new. At the lowest the House and may make it more difficult Charles Finocchiaro, and David Rohde, amounts, the probability is rather small, for Democrats to regain control of the “Redistricting and Party Polarization between 13 and 20 percent. As percent lower chamber. Absent a political wave, in the U.S. House of Representatives,” new increases, the chances that a winning back the House will be unlikely American Politics Research 35(6): 878-904. freshman will serve in the next Congress unless the Democrats are able to institute 10. This method is described more fully grows, topping off at 32 percent. This some big transformations to Republican in Michael Crespin, “Using Geographic is substantively large since the average districts.

Information Systems to Measure District reelection rate for a member of the House Change, 2000-2002,” Political Analysis hovers around 90 percent.

Notes 13(3): 253-260.

When it comes to switches in partisan 1. Thomas Rogers Hunter, “The First 11. To create the data used in the control, the results shown in Figure 3 Gerrymander? Patrick Henry, James analyses of the 2012 election, I used are not quite so striking. When districts Madison, James Monroe, and Virginia’s the MABLE/Geocorr12: Geographic do not experience much change, the 1788 Congressional Districting,” Early Correspondence Engine at http://mcdc. chances that the other party will pick American Studies: An Interdisciplinary missouri.edu/websas/geocorr12.html, up a seat are pretty slim, starting at 4 Journal 9.3 (2011): 781-820. accessed 10/15/2014. percent. At the upper bound of percent 2. Robert G. Boatright, “Static 12. Gary C. Jacobson and Samuel new, the probability maxes out at 13 ambition in a changing world: Legislators Kernell, Strategy and Choice in percent. For the average district, there is preparations for, and responses to, Congressional Elections (New Haven: Yale only a 6 percent chance of a party pickredistricting,” State Politics and Policy University Press, 1981). up. So, while the results are statistically 4: 436-454.

“O

16

significant, the impact is substantively rather small.

Overall, what does change do to the political environment? First and foremost, it introduces instability to the system. Previous research and the results presented here show higher levels of change are associated with increased chances of turnover and lower vote levels for incumbents. If an incumbent is able to survive, he or she may have trouble adjusting to the new voters and this could spell trouble in subsequent elections.

The variation in districts change does not seem to be random either, but rather is related to who controls the process. In 2012, there was more disruption when commissions or courts got involved. Although the evidence is only speculative, Republicans were able to draw more total districts and the seats they held experienced less overall change.

”

Winter 2015

extensions

DO FACTS MATTER?

Information and Misinformation in American Politics ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★

JENNIFER L. HOCHSCHILD AND KATHERINE LEVINE EINSTEIN A democracy falters when most of its citizens are uninformed or misinformed, when misinformation affects political decisions and actions, or when political actors foment misinformation—the state of affairs the United States faces today, as this timely book makes painfully clear. In Do Facts Matter?

Jennifer L. Hochschild and Katherine Levine Einstein start with Thomas Jefferson’s ideal citizen, who knows and uses correct information to make policy or political choices. What, then, the authors ask, are the consequences if citizens are informed but do not act on their knowledge? More serious, what if they do act, but on incorrect information? $29.95 HARDCOVER · 248 PAGES ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★

VOLUME 13 IN THE JULIAN J. ROTHBAUM DISTINGUISHED LECTURE SERIES

Analyzing the use, nonuse, and misuse of facts in various cases—such as the call to impeach Bill Clinton, the response to global warming, Clarence Thomas’s appointment to the Supreme Court, the case for invading

Iraq, beliefs about Barack Obama’s birthplace and religion, and the Affordable Care Act—Hochschild and Einstein argue persuasively that errors of commission (that is, acting on falsehoods) are even more troublesome than errors of omission. While citizens’ inability or unwillingness to use the facts they know in their political decision making may be frustrating, their acquisition and use of incorrect “knowledge” pose a far greater threat to a democratic political system.

Do Facts Matter? looks beyond individual citizens to the role that political elites play in informing, misinforming, and encouraging or discouraging the use of accurate or mistaken information or beliefs. Hochschild and Einstein show that if a well-informed electorate remains a crucial component of a successful democracy, the deliberate concealment of political facts poses its greatest threat.

THE UNIVERSITY OF OKLAHOMA IS AN EQUAL OPPORTUNITY INSTITUTION. WWW.OU.EDU/EOO

2015 Extensions Ad.indd 1

extensions

1/26/15 1:04 PM

Winter 2015

17

Reapportionment, Regional Politics, and Partisan Gain Kenneth C. Martis, J. Clark Archer, Robert H. Watrel, Fred M. Shelley and Gerald R. Webster

G

eographers and demographers have been analyzing U.S. regional population change for many decades. From the perspective of politics and governance, understanding these population changes over time is very important because seats in the House of Representatives are reapportioned every decade in accordance with the U.S. Constitution. Representation in the House, in turn, affects the distribution of votes in the Electoral College and thus the impact of regional population change affects the presidency as well as Congress. As political geographers we have studied the possible impacts of this population change on elections, issues in Congress and the nation, and if a particular political party has gained or lost in this process. This article is adapted from our recently published co-edited book, Atlas of the 2012 Elections, which examines both the short-term and long-term state and regional gains and losses in the House of Representatives and Electoral College to see how the Democrats and Republicans have fared.1 The Regional Geography of Reapportionment Several seats in the House of Representatives shifted among states between the 2002 election (after the 2000 census), and the 2012 election (after the 2010 census), as shown clearly in Map 1. Seats are moving from the North and Northeast to the South, Southwest, and West. In fact, this general movement has been occurring for several decades. One of our previous books, The Historical Atlas of State Power in Congress, 1790-1990, analyzes long-term reapportionment change by calculating and mapping the state and regional ups and downs of the reallocation of seats since the first census in 1790.2 Our atlas divides American history into four geographical population/ reapportionment eras: Original States and New States (1790-1850); Free States and Slave States (1790-1860); Rural and Urban Places (1870-1930); and Sunbelt and Snowbelt (1970-1990 [and, as mentioned above, this era has continued to 2010]). Each of these four eras had a profound effect upon the regional balance of power in the United States, the issues and legislation brought before Congress, and the election of the president.

The Sunbelt is the term given by demographers and geographers to the combined population in-migration/immigration and strong economic growth along the southern tier of America, namely the South and West regions. The Snowbelt, sometimes called the Frostbelt or the Rust Belt, consists of states with out-migration and/or relatively slow population and economic growth along the northern tier.

Map 1 (see pg. 19) shows that nine of the ten states that lost seats in the 2010 census were in the North or Northeast. The only exception to this pattern is Louisiana, which experienced substantial out-migration following Hurricane Katrina in 2005. New York and Ohio were the biggest losers with two lost seats each. Eight states gained in electoral power: four in the South, with Texas (+4) and Florida (+2) the biggest winners, and four in the West. The 2000 census regional reapportionment numbers are very similar, with New York (-2) and Pennsylvania (-2) the biggest losers among many other 18

Kenneth C. Martis is Professor Emeritus of Geography, West Virginia University. His email address is ken.martis@mail.wvu.edu.

J. Clark Archer is a political, urban, and population geographer at University of Nebraska – Lincoln.

His work combines political science, economics, demographics, cartography, and Geographic Information Systems.

His email address is jarcher1@unl.edu.

Winter 2015

extensions

Robert H. Watrel is an associate professor of geography at South Dakota State University.

His research focuses on human geography, political geography, and cartography. Email address: Robert.Watrel@sdstate.edu.

Map 1

Fred M. Shelley is a professor and chair of the department of geography, University of Oklahoma. His research interests include political geography, resource and environmental policy, geography and the world economy, and electoral geography. His email address is fshelley@ou.edu.

Gerald R. Webster is a professor and chair of the department of geography at University of Wyoming. His research interests center on political geography, electoral geography, secessionist groups, and voting rights. His email address is gwebste1@uwyo.edu.

extensions

Winter 2015

northern states, and four Sunbelt states (Georgia, Florida, Texas, and Arizona) each gaining two additional seats among many other southern and western states.

In 2000 California gained one additional seat, making it the largest House of Representatives state delegation in American history with fifty-three members and fifty-five total electoral votes.

We began our long-term analysis with the 1952-1960 reapportionment cycle to take account of the addition of Alaska and Hawaii, to catch the tail end of the northern urbanization/industrialization period, and to encompass all of the next five cycles, which cover the present Sunbelt/Snowbelt era. The combined congressional losses of the­­­­ Northern industrial and Midwestern and Great Plains agricultural states over the last half century are staggering, and the historical trend noteworthy. The last northern states to gain in reapportionment were in 1960: Ohio and Michigan each added one seat because of the post-WWII auto boom and related manufacturing; New Jersey gained one, mostly from intra-regional suburbanization from New York City and Philadelphia; and Maryland gained one, reaping the benefit of the expansion of the federal government in Washington, D.C. and resulting suburbanization.

Since the 1952-1960 electoral cycle, New York has lost an incredible sixteen seats, Pennsylvania twelve, and Ohio and Illinois each seven.

The combined congressional/Electoral College losses of the Snowbelt states, of course, are equal to the combined gains of the Sunbelt states. The three growth poles of the Sunbelt, California (+23), Florida (+19) and Texas (+14), are the biggest gainers. Some of the biggest percentage gains are in the previously sparsely settled West. Arizona has gained seven seats, going from a delegation of just two in the Eighty-Seventh Congress (elected in 1960) to nine elected in 2012. Nevada has gone from one House member as recently as 1980 to four in the 2012 election.

Map 2 illustrates the geographical pattern of changes in apportionment to the U.S. House of Representatives, and thus the Electoral College, for the entire 1952-2012 period. During this time frame, eighty-seven seats, that is, 20 percent of the House, were transferred between the states. Only seven states did not have a net change in their House delegation during this period, including four smallpopulation states that kept the mandatory one representative: Alaska, Delaware, Vermont, and Wyoming. Forty-three states had their House delegations change: seventeen states gained representatives and twenty-six lost representatives.

As Map 2 shows, the regional pattern of reapportionment change since the 19521960 election cycle is striking. The states that lost representation are in one large area encompassing the Northeast, North Central, and much of the inland portion 19

Map 2

of the United States. In reality this area incorporates at least four major regional demographic trends. The Northeast and Great Lakes states have gone through deindustrialization; the inland Midwest and Great Plains states have experienced declining farm population; four southern states have gone through large AfricanAmerican out-migration and slow growth; and two Appalachian states have undergone mining mechanization and a general decline in extractive industry employment.

The states that gained in reapportionment are in two geographical sections. One comprises the Southeast coastal states, anchored and dominated

by Florida. That section has benefitted by such variables as individuals seeking a warmer climate, retirement trends, reasonable house prices, favorable business climate, and many other Sunbelt-related factors. Map 1 illustrates that, looking at the long term, the South is not one homogenous region of growth – more a region of sunspots rather than a continuous Sunbelt. The other growth section is the Southwest and Pacific Coast.

In this large region of an expanding economy and spectacular landscapes, numerous variables attract in-migration and immigration: the oil and gas industry and business advantages of Texas, the development of high-tech industry, the climate and environment of California, the affordability of Arizona and Nevada.

All in all, the three growth pole states of 20

Florida, Texas, and California comprise fifty-six of the eighty-seven seats, or 65 percent of all the seats gained.

The Political Geography of Reapportionment In the last several decades, there has been a clear reallocation of regional power in the United States. But how has this reallocation translated into partisan power? Has it benefited Republicans or Democrats? At first glance, one would assume the declining Democratic North and expanding Republican South would have made the Republican Party the clear winner in the reapportionment process.

A number of commentators after the last two censuses have suggested this is the case. However, when examining the longterm Sunbelt/Snowbelt era historical data, the trend is somewhat more complex. To explore this further, we studied the 2012 elections using the reapportionment of seats in the Sunbelt/Snowbelt era. The gains and losses are calculated based upon the state House delegation size at the time of the 1952-1960 reapportionment cycle versus the House size at the 2012 election.

We divided the states into three categories based upon the consensus 2012 pre-election analysis: safe Republican (24), safe Democrat (18), and swing (8).3 The eight swing states are New Hampshire, Virginia, Ohio, Wisconsin, Iowa, Colorado, Florida, and Nevada.

When examining the reapportionment trends of the 2012 swing states, the results

are mixed. New Hampshire remained the same during this period; Virginia (+1), Colorado (+3), Florida (+19) and Nevada (+3) gained, but Ohio (-7), Wisconsin (-2), and Iowa (-4) lost. However, Florida’s gain is so large, the sum of the swing state total shows a net thirteen-seat gain, but, of course, President Obama eventually won all the swing states in 2012.

In 2012 eighteen states were considered safe for Democrats in the presidential election (plus the District of Columbia, which was given three electoral votes by the XXIII Amendment, providing safe electors to the Democrats since the 1964 presidential election). The safe Democratic states were California (+23), Connecticut (-1), Delaware (0), Hawaii (+1), Illinois (-7), Maine (-1), Maryland (+1), Massachusetts (-5), Michigan (-4), Minnesota (-1), New Jersey (-2), New Mexico (+1), New York (-16), Oregon (+1), Pennsylvania (-12), Rhode Island (0), Vermont (0), and Washington (+3). In the safe Democratic states, in 2012 at least, thirty seats were gained over the last half century (mostly in California), but forty-nine were lost, for a net loss of nineteen House seats and presidential electors. Since the political geography of the Electoral College in recent years favors the Democrats in the North, Northeast, and Pacific Coast states, the Democrats are on the losing trend in the North, but on the gaining trend in the Pacific Coast.

The twenty-four states considered safe for Republicans in the 2012 pre-election analysis were Alabama (-2), Alaska (+1), Arizona (+7), Arkansas (-2), Georgia (+4), Idaho (0), Indiana (-2), Kansas (-2), Kentucky (-2), Louisiana (-2), Mississippi (-2), Missouri (-3), Montana (-1), Nebraska (-1), North Carolina (+1), North Dakota (-1), Oklahoma (-1), South Carolina (+1), South Dakota (-1), Tennessee (0), Texas (+14), Utah (+2), West Virginia (-3), and Wyoming (0). At the 2012 elections the safe Republican states gained thirty seats, but also lost twenty-four, for a net gain of only six House seats and electoral votes since 1952-1960. These gains are mostly in Texas and Arizona, with the growing southern state of Georgia also contributing. Republican losses occurred in a number of Appalachian, Great Plains, and several slow-growing southern states.

The long-term data at this juncture shows that not all Republican-leaning areas are growing; in fact, seventeen of the twentyfour currently solid Republican states Winter 2015

extensions

have either lost or recorded no gain in the Sunbelt/Snowbelt era.

The long-term Sunbelt and Snowbelt era demographic trend shows the Democratic Party losing presidential electors and potential House members in Democrat-friendly states. Nevertheless, the Republicans have only gained slightly.

This seeming contradiction is accounted for by the swing states, primarily Florida.

In fact, changes in all three categories are driven by the Sunbelt’s major growth poles. The three large growth pole states mentioned above are split among California (+23) safe Democrat, Texas (+14) safe Republican, and Florida (+19) swing. The increase in the swing state Florida electors is one of the reasons it has played a critical, and sometimes decisive, role in recent presidential elections.

Table 1 compares the top ten states in House representation and presidential electors from the beginning of our study period to the most recent 2010 census and the 2012 election. It illustrates the astonishing growth of the influence of Florida on the American political scene. Florida is not even listed as one of the top ten states in the 1952-1960 reapportionment cycle. Shocking as it may seem, as recently as the 1960 election, states like Alabama and Minnesota had more representatives and electoral votes (9/11) than Florida (8/10).

five seats/electors. Although Republicans have made great strides in controlling the Deep South, Map 2 shows the longterm gains in the southeast coastal states have actually been quite small and offset with losses in the Great Plains and other places. Although the Democrats have recently dominated the North and Northeast, this region is declining in seats/ electors, but still has many states with a large population base. In addition, the Democrats control the sizable California vote (the fifty-five electors are 10.2 percent of the entire Electoral College and 20.4 percent of the 270 needed to secure the presidency) and other Pacific Coast states. Furthermore, Democrats have also made significant inroads in some growing western places like New Mexico, Colorado, and Nevada through immigration and in-migration from Democrat-friendly areas.

A number of political and demographic trends could alter the above alignment of safe or swing states for both parties.

Over time, whole regions have realigned.

The South, previously solid Democratic, is now mostly Republican; New England, previously staunch Republican, now leans Democratic. In addition, individual states can go through a quick and complete realignment. For example, West Virginia was one of the most loyal Democratic states in the last half of the twentieth

Table 1. Top Ten States in House Representation and Presidential Electors 1950 Census and 1952-1960 Election Cycle

1. New York 43/45 2. California 30/32* 3. Pennsylvania 30/32* 4. Illinois 25/27 5. Ohio 23/25 6. Texas 22/24 7. Michigan 18/20 8. Massachusetts 14/16* 9. New Jersey 14/16* 10. North Carolina 12/14

2010 Census and 2012-2020 Election Cycle

1. California 53/55 2. Texas 36/38 3. Florida 27/29* 4. New York 27/29* 5. Illinois 18/20* 6. Pennsylvania 18/20* 7. Ohio 16/18 8. Georgia 14/16* 9. Michigan 14/16* 10. North Carolina 13/15

*tied

In the 2012 presidential election, President Obama won twenty-six states and the District of Columbia with 332 electoral votes. Those twenty-six states have a net loss of five seats since 19521960. Obama’s Republican opponent, Governor Mitt Romney, won twenty-four states with a net reapportionment gain of extensions

Winter 2015

century, but now is solid Republican, at least on the presidential level. Also, the Hispanic vote is growing in such places as Arizona and North Carolina, which may even be categorized as swing rather than safe Republican states by the time of the 2016 or 2020 presidential elections. In fact, some observers initially placed North

Carolina as a swing state in 2012, since President Obama carried it in 2008. Our study is a snapshot of apportionment and long-term partisan gains or losses with full knowledge that possible future swings and realignments could alter present trends.

Notes 1. J. Clark Archer, Robert H. Watrel, Fiona M. Davidson, Erin H. Fouberg, Kenneth C. Martis, Richard L. Morrill, Fred M. Shelley, and Gerald R. Webster, eds., Atlas of the 2012 Elections (Lanham, Maryland: Rowman & Littlefield, 2014).

2. Kenneth C. Martis and Gregory A. Elmes, The Historical Atlas of State Power in Congress: 1790-1990 (Washington, DC: Congressional Quarterly Press, 1993).

3. The determination of safe Democrat, safe Republican, and swing states was based on a consensus of numerous sources. Each source has its own methodology and access to a wide variety of polling data. See http://blog. constitutioncenter.org/2012/05/definingthe-swing-states-for-2012/ and http:// thecaucus.blogs.nytimes.com/2012/05/07/ explaining-the-timess-battleground-stateratings/. Presidential polling became more meaningful and prevalent once Governor Mitt Romney secured the Republican nomination, and, of course, polling was conducted intensively up to Election Day. During the polling process, various sites had upwards of sixteen states in the toss-up category. In a survey of twelve media organizations, our eight swing states were most mentioned. See http://blog. lib.umn.edu/cspg/smartpolitics/2012/08/ will_the_real_battleground_sta.php. In addition to our eight swing states, eight others were included; most widespread among these were Wisconsin and North Carolina. However, as the election neared, Wisconsin was determined by most to be leaning or safe Democrat, and North Carolina leaning or safe Republican.

See http://www.cnn.com/election/2012/ ecalculator#?battleground, http://www. forbes.com/sites/thestreet/2012/11/06/ battleground-states-to-watch-for-thiselection/, http://www.politico.com/2012election/swing-state/, http://www. washingtonpost.com/blogs/the-fix/post/ the-9-swing-states-of-2012/2012/04/16/ gIQABuXaLT_blog.html, and http:// abcnews.go.com/blogs/politics/2012/10/abcnews-moves-pennsylvania-minnesota-fromsafe-to-lean-obama/. All these websites were accessed in October 2014.

21

for the record

News from the Center LaDonna Sullivan Managing Editor

Professors Cindy Simon Rosenthal and Ronald M. Peters, Jr. teamed up during the recent fall semester to initiate an archives-based undergraduate course focused on Congress and the Constitution. Graduate Research Assistant Heather Walser worked with them in the Carl Albert Center Congressional Archives to identify appropriate documents for the class projects. Students formed groups to explore the intricacies of presidential succession involving Speaker Albert during the Watergate crisis and the institutional struggle between Congress and President Nixon over the Budget and Impoundment Act. The project will move into a digital humanities phase to bring original documents to a larger audience.

Rosenthal and Peters reported on this digital humanities project when they spoke in The People, the Congress, and the Constitution Symposium on November 3, 2014. The symposium was sponsored by OU’s Institute for the American Constitutional Heritage and hosted by IACH Director Kyle Harper. Other speakers at the symposium included Jeremy Bailey, University of Houston; Amanda Demmer, University of New Hampshire; Rebecca Zietlow, University of Toledo College of Law; and Gordon S. Wood, Brown University. Carl Albert Center Associate Director Mike Crespin also participated in the symposium.

Alumnae Carl Albert Graduate Fellowship Alumna Jocelyn Evans, professor and chair of the political science department at University of West Florida, was promoted on January 15, 2015 to be the first associate dean in the university’s newly formed College of Arts, Social Sciences and Humanities. Her specific portfolio of duties will include: facilitation of academic processes, supervision of 22

curriculum improvements, collaboration with college units to advance strategic objectives, promotion of community engagement initiatives for the college, preparation of college reports, and active service on committees promoting the college’s mission.

Archives Archivist Robert Lay left the Carl Albert Center in September 2014 and headed south to Denton, Texas where he serves as Special Collections Librarian at University of North Texas.

After conducting a national search for a new archivist, the Carl Albert Center welcomed Nathan Gerth from the University of Notre Dame, where he earned his Ph.D. in history. While completing his doctorate, Nathan organized and described a wide range of manuscript collections for Notre Dame’s rare books and special collections department. He also holds a master’s degree from the University of Kansas, where he studied foreign policy issues related to the former Soviet Union. In the Carl Albert Center Congressional Archives, Nathan coordinates the processing and curation of the Center’s holdings, supervises the archive’s social media presence, and assists with classroom and patron outreach.

The Carl Albert Center Visiting Scholars Program provided funding recently for research conducted in the archives by Senior Lecturer Scott Walker, Department of Political Science and International Relations, University of Canterbury. His research is on the evolution of human rights attitudes in the United States.

Archivist Nathan Gerth welcomed a class of undergraduate students to the archives from The American West History Capstone Course, gave them a tour of the archives, and provided orientation to prepare them for research projects they will work on later this semester. The class is taught

In addition to his work at the Carl Albert Center, Archivist Nathan Gerth volunteers as a historian and board member for the Rock of Ages Lighthouse Preservation Society. Through his work with the society he hopes to help restore and preserve one of the most remote and powerful lighthouses on the Great Lakes.

by Professor Kathleen A. Brosna, the Paul and Doris Eaton Travis Chair of Modern American History.

Archivists Nathan Gerth and Bailey Hoffner spoke to Library and Information Studies graduate students in the class Archival Concepts and Traditions, taught by Associate Professor Kelvin White.

Awards The Oklahoma Political Science Association presented its 2014 Saundra Mace Service Award to LaDonna Sullivan. The award was established to honor support staff persons for excellence in job performance, attitude, and commitment in support of the political science discipline throughout the state. LaDonna is retiring after 27 years of service to the Carl Albert Center, and her service will be recognized in the next issue of Extensions.

Winter 2015

extensions

issues that the state of Oklahoma is facing. Starting July 1, the exhibit can be explored online or at one of the 18 host sites around the state.

Publications Associate Director Mike Crespin recently published three articles:

• “Procedural Signaling, Party Loyalty, and Traceability in the U.S. House of Representatives,” with Jamie Carson and Anthony Madonna, Political Research Quarterly (2014) 67(4): 729–742; and

Civic Engagement Fellows accept their award, for registering the most out-of-state students to vote, from State Regents for Higher Education Chancellor Glen D. Johnson (fourth from right) and Chair Michael C. Turpen (third from right). Also pictured from left to right: OU Director of Government Relations Scott Mason, Victoria Bautista, Jon Torres, OU Vice President for External Relations and Planning Danny Hilliard, Lauren Schueler, and Madison Hobson.

Civic Engagement Fellows The Carl Albert Center Civic Engagement Fellows for the 2014-2015 school year are Victoria Bautista, Madison Hobson, and Jon Torres.

They kicked off the year competing in the Oklahoma State Regents for Higher Education’s annual voter registration contest. The fellows partnered with other student organizations on campus to host tables in the union and residence halls where students could fill out in-state or out-of-state voter registration forms. These efforts helped the University of Oklahoma win the award for getting the most out-of-state registrations.

At the end of September all three fellows attended a conference at Harvard University, where they met with students from other universities that are members of the National Campaign for Political and Civic Engagement, a consortium of schools dedicated to improving their communities through bipartisan politics and public service.

The Civic Engagement Fellows also hosted a Politics and Pizza session for OU students in October. Guest speaker Akash Patel, founder of the Aspiring Americans Initiative, talked about the services his non-profit foundation provides to help immigrant students in Oklahoma learn about their educational opportunities, legal rights, and resources. Lauren Schueler, extensions

Winter 2015

who also serves as assistant director for N.E.W. Leadership, is directing the center’s civic engagement program this year.

Grants Associate Director Mike Crespin received funding from the College of Arts and Sciences to bring a scholar to campus to speak to students during a weeklong celebration of the remarkable variety of people and programs that make up the college. On February 26, Professor Frances Lee of the University of Maryland discussed why Congress cannot get along and break through partisan gridlock. Professor Lee also conducted research from the Albert, Armey, and Edwards papers during her visit.

Archivist Bailey Hoffner and Graduate Assistant Arielle Hughes are creating a traveling exhibit on the history of water policy in Oklahoma, entitled WATER: Congressional Representation to Protect a Precious Resource. The project is made possible by a grant from the Oklahoma Humanities Council and the National Endowment for the Humanities with matching funds from the Southern Climate Impacts Planning Program (SCIPP) and the Oklahoma Mesonet.

The water exhibit will provide a narrative of historical events, highlight pivotal congressional figures through time, and explain current water-related

• “The Political Geography of Distributive Politics,” with Austin Clemens and Charles Finocchiaro, Legislative Studies Quarterly (2015) 40(1): 111–136;

• “Re-evaluating the Effects of Redistricting on Electoral Competition, 1972-2012,” with Jamie Carson and Ryan Williamson, State Politics and Policy Quarterly (2014) 14(2): 162–174.

In Memoriam Betty Woodhouse French 1924–2014

Betty French passed away on September 29, just two weeks before her 90th birthday. Those who had the pleasure of working with her at the Carl Albert Center will always remember Betty as a gracious, intelligent woman with meticulous attention to detail. She began her employment at the University of Oklahoma in 1981 in the Western History Collections and then moved to the Carl Albert Center Congressional Archives the following year. She worked here with Archivist John Caldwell until her retirement in 1992. Initially, Betty performed clerical tasks for the archives, but her keen interest and aptitude led to on-the-job training that enabled her to process manuscript materials alongside the archivists. She worked primarily on the Carl Albert, Sidney Clarke, Elmer Thomas, Dick T. Morgan, and W. H. Murray collections. Betty and her late husband David, a long-serving Professor of English, were devoted to the University of Oklahoma. They are missed.

23

THE CARL ALBERT GRADUATE FELLOWSHIP PROGRAM ~ A Commitment to the Study of Representative Government ~ Each Carl Albert Fellow pursues a rigorous and individualized program of study while working closely with faculty. The fellowship is a four-year program leading to the acquisition of the Ph.D. degree in cooperation with the Department of Political Science at The University of Oklahoma.

Carl Albert Fellows focus their program of study on fundamental issues in representative government. The central focus is in the field of American government and includes institutions, processes, and public policy. In addition, Fellows pursue two additional fields of study selected from among comparative politics, international relations, methods, political theory, public administration, or public policy.

The fellowship program values both instructional development and research productivity. Carl Albert Fellows are expected to develop original research leading to professional conference presentation and publication. The Center’s resources enable Fellows to pursue field research where appropriate to the dissertation research design.

Carl Albert Fellows Tyler Hughes (far left) and Victoria Rickard met with Professor Tom Patterson while he was on campus to deliver the 2013 Julian J. Rothbaum Distinguished Lecture in Representative Government. Professor Patterson is Bradlee Professor of Government and the Press, Kennedy School of Government, Harvard University.

The fellowship package includes four years of financial support in teaching or research appointments, full tuition and fees, funded research and conference travel, summer support, participation and course work at the Inter-university Consortium for Political and Social Research, and dissertation research funds.

Carl Albert Fellows are introduced to nationally known political leaders and scholars through special guest lectures and seminars. Visitors have included former Ambassador James R. Jones, former U.S. Senator George McGovern, and former Congressmen Dick Armey and Mickey Edwards as well as distinguished scholars James E. Campbell, Morris Fiorina, Jennifer Hochschild, Tom Patterson, Jack Rakove, and Steven S. Smith.

Carl Albert Fellows access a rich and diverse selection of other resources at The University of Oklahoma: • Carl Albert Center Archives http://www.ou.edu/special/albertctr/archives • Public Opinion Learning Laboratory (P.O.L.L.) http://www.ou.edu/oupoll • Political Commercial Archives http://www.ou.edu/pccenter • Center for Applied Social Research http://casr.ou.edu

CARL ALBERT GRADUATE FELLOWSHIP Application Deadline: February 1 of each year.

Apply Online http://www.ou.edu/carlalbertcenter/student/grad-fellow.html.

24

Winter 2015

extensions

extensions

Winter 2015

25

Non-Profit Organization U.S. Postage

The Carl Albert Congressional Research and Studies Center 630 Parrington Oval, Room 101 Norman, Oklahoma 73019-4031 (405) 325-6372 http://www.ou.edu/carlalbertcenter

PAID University of Oklahoma

Visiting Scholars Program The Carl Albert Congressional Research and Studies Center at the University of Oklahoma seeks applicants for its Visiting Scholars Program, which provides financial assistance to researchers working at the Center’s archives. Awards of $500-$1000 are normally granted as reimbursement for travel and lodging.

The Center’s holdings include the papers of many former members of Congress, such as Speaker Carl Albert, Robert S. Kerr, and Fred Harris of Oklahoma, Helen Gahagan Douglas and Jeffery Cohelan of California, and Neil Gallagher of New Jersey. Besides the history of Congress, congressional leadership, national and Oklahoma politics, and election campaigns, the collections also document government policy affecting agriculture, Native Americans, energy, foreign affairs, the environment, and the economy.

Topics that can be studied include the Great Depression, flood control, soil conservation, and tribal affairs.

At least one collection provides insight on women in American politics. Most materials date from the 1920s to the 1990s, although there is one nineteenth-century collection. The Center’s collections are described on the World Wide Web at http://www.ou.edu/carlalbertcenter and in the publication titled A Guide to the Carl Albert Center Congressional Archives (Norman, Okla.: The Carl Albert Center, 1995) by Judy Day, et al., available at many U. S. academic libraries. Additional information can be obtained from the Center.

The Visiting Scholars Program is open to any applicant. Emphasis is given to those pursuing postdoctoral research in history, political science, and other fields. Graduate students involved in research for publication, thesis, or dissertation are encouraged to apply. Professional writers and researchers are also invited to apply. The Center evaluates each research proposal based upon its merits, and funding for a variety of topics is expected.

No standardized form is needed for application. Instead, a series of documents should be sent to the Center, including: (1) a description of the research proposal in fewer than 1000 words; (2) a personal vita; (3) an explanation of how the Center’s resources will assist the researcher; (4) a budget proposal; and (5) a letter of reference from an established scholar in the discipline attesting to the significance of the research. Applications are accepted at any time.

For more information, please contact: Archivist, Carl Albert Center, 630 Parrington Oval, Room 101, University of Oklahoma, Norman, OK 73019. Telephone: (405) 325-5835. FAX: (405) 325-6419. E-mail: cacarchives@ou.edu The University of Oklahoma is an Equal Opportunity Institution. www.ou.edu

Extensions is a copyrighted publication of the Carl Albert Congressional Research and Studies Center. It is distributed free of charge twice a year. All Rights Reserved.

Extensions and the Carl Albert Center symbol are trademarks of the Carl Albert Center. Copyright Carl Albert Center, The University of Oklahoma, 1985.

Statements contained herein do not necessarily reflect the views of the Carl Albert Center or the regents of The University of Oklahoma.