Search

The widespread adoption of written constitutions is one of the most notable developments in institutional design in politics over the past 250 years. The American states offer a rich place to study constitutional innovation and imitation as being among the first political bodies to adopt constitutions and also given that they often replaced them, in both cases innovating and learning from one another. In this paper, we use quantitative text analysis to identify constitutional innovation and to investigate patterns of imitation. First, we find substantial textual borrowing between state constitutions. On average, 20 percent of a state’s constitutional language was borrowed directly from another state constitution. Second, states were more likely to borrow text from geographically proximate states, from temporally proximate state constitutions, and from states that shared similar partisan profiles. Finally, we offer a brief discussion of the most influential constitutions as an exploratory example for extending our approach of identifying textual innovation and imitation. These findings offer new contributions to both the study of constitutional design and institutional diffusion.

US federalism grants state legislators the authority to design many aspects of election administration, including ballot features that mediate how citizens understand and engage with the choices available to them when casting their votes. Seemingly innocuous features in the physical design of ballots, such as the option to cast a straight ticket with a single checkmark, can have significant aggregate effects. Drawing on theoretical insights from behavioral economics and extensive data on state ballot laws from 1888 to the present, as well as in-depth case studies, this book shows how strategic politicians use ballot design to influence voting and elections, drawing comparisons across different periods in American history with varying levels of partisanship and contention. Engstrom and Roberts demonstrate the sweeping impact of ballot design on voting, elections, and democratic representation.

Models of voting behavior typically specify that all voters employ identical criteria to evaluate candidates. We argue that moderate voters weigh candidates’ policy/ideological positions far less than non-moderate voters, and we report analyses of survey data from the 2010 Cooperative Congressional Election Study that substantiate these arguments. Across a wide range of models and measurement strategies, we find consistent evidence that liberal and conservative voters are substantially more responsive to candidate ideology than more centrist voters. Simply put, moderate voters appear qualitatively different from liberals and conservatives, a finding that has important implications for candidate strategies and for political representation.

Considerable debate exists over the impact of electoral institutions on turnout in U.S. national elections. To address this debate, I exploit the rich variation in electoral rules present throughout the nineteenth and early twentieth century. Using a newly constructed dataset of district-level turnout results for the U.S. House from 1840 to 1940, I find that electoral institutions and political competition jointly provided incentives, and by the turn-of-the-century disincentives, for political leaders to mobilize the electorate. The results demonstrate that changes in electoral institutions and varying levels of political competition help explain congressional turnout across districts and over time.

Distributive politics represents one of the most important and controversial aspects of legislative policymaking. In the U.S. Congress, controversies over distributive politics are most evident in the area of legislative earmarking. In this article, we employ a unique set of data matching earmarks to their legislative sponsors to assess the leading explanations of distributive politics. We find that members of the majority party do considerably better than equally situated members of the minority. Moreover, party leaders target earmarks to those holding pivotal agenda-setting positions and to electorally vulnerable members. These findings have direct implications for both the extensive political science literature on distributive politics and the practical politics of earmarking reform.

Most political observers agree that incumbent legislators have a considerable advantage over nonincumbents in modern congressional elections. Yet there is still disagreement over the exact source of this advantage and the explanation for its growth over time. To address this debate we utilize a unique set of historical elections data to test for the presence of an incumbency advantage in late-nineteenth-century House elections (1872–1900). We find a modest direct effect of incumbency and a substantial candidate quality effect. Moreover, the cartel-like control of ballot access by nineteenth century political parties created competition in races that the modern market-like system simply does not sustain. Our results suggest that candidate quality is a fundamental piece of the puzzle in understanding the historical development of the incumbency advantage in American politics.

Considerable debate exists over the impact of redistricting on the partisan composition of the U.S. Congress. I address this debate by turning to an era of congressional redistricting that has received little systematic attention—the politics of gerrymandering in the 19th century. Using statewide-, county-, and ward-level electoral data from 1870 to 1900, I show that when a single party controlled the districting process, they used districting to systematically engineer a favorable partisan bias. These partisan biases affected the partisan composition of state congressional delegations and at times even helped determine party control of the House of Representatives.

The effects of redistricting on candidate entry patterns in contemporary House races has received growing attention in the scholarly literature, yet virtually no consideration has been given to this question in the context of historical elections. This is unfortunate as the wider variation in congressional redistricting during the nineteenth century gives us increased leverage in terms of understanding strategic candidate behavior. Utilizing a new dataset of candidate quality for nineteenth-century House races, we examine whether candidates with prior electoral experience are more likely to run in districts that are altered during the redistricting process, and provide an account of how differences in the prevalence of redistricting may affect strategic entry decisions of politicians. Our results suggest that entry decisions and electoral outcomes are affected by redistricting in this era. Moreover, our analysis provides an opportunity to use history to test contemporary theories of congressional elections in a broader context. W hen the Texas state legislature convened in January 2003, Republicans enjoyed unified control of the state government for the first time in 130 years. Recognizing the unique political opportunity of the situation, the Texas legislature began debating a new redistricting plan drawn to shore up Republican dominance in the U.S. House of Representatives. In an effort to prevent the plan from taking shape, Democratic state legislators twice fled Texas during the legislative session to prevent the quorum necessary to allow Republicans to vote on the redistricting plan. Eventually, however, Republican state legislators succeeded in pushing their revised redistricting plan through a special session of the legislature during the fall. 1

Conventional wisdom suggests that incumbent politicians use the resources of office to create an electoral advantage. But Cox and Katz (2002) argue that at least part of this incumbency advantage in the United States House of Representatives can be attributed to the strategic entry and exit decisions of incumbents. We test this claim by taking advantage of the natural experiment provided by state legislative term limits in California. By comparing different types of open seats, we identify the strategic component of the incumbency advantage that exists above and beyond the resource-based advantage. The vote loss suffered by the incumbent party is smaller in term-limited seats than in voluntary open seats, indicating that incumbents do sometimes leave when their electoral prospects look dim. Further evidence of this strategic component is that quality challengers run more often in voluntary open seats, while quality incumbent-party replacements run disproportionately in term-limited seats.

The modern history of divided government in America suggests that the framers succeeded in creating a government unresponsive to popular passions. Yet in the nineteenth century the party winning the presidency almost always captured control of the House of Representatives. Why and how could nineteenth century national elections be so responsive that they resemble parliamentary outcomes? We identify electoral institutions present in the states that directly linked congressional elections to presidential coattails. Specifically, we estimate the impact of state ballot laws and the strategic design of congressional districts on presidential coattail voting from 1840 to 1940. We find that presidential elections, as mediated by state electoral laws, strongly account for unified party control of the House and the presidency throughout the nineteenth century. W orried that unified control of government would breed majority tyranny, the Constitution’s Framers created separate executive and legislative branches and provided each with the authority to check the excesses of the other. They also sought an electoral remedy by varying constituencies, term lengths, and modes of election across the branches and legislative chambers. The modern history of inconclusive national elections and divided party control of government suggests that they did their job well. From 1948 through 2000 only five of 13 presidential elections found the same party winning the presidency and both chambers of Congress. Equally revealing, only once (Democrats in 1952) did a governing party collapse and wholly surrender control of Congress and the presidency. Limiting comparison to control of the president and the House of Representatives, a pairing more likely to turn up unified outcomes, only seven of 13 presidential elections ended with the same party controlling both institutions.