THE ENTRENCHING OF INCUMBENCY: REELECTIONS IN THE U.S. HOUSE OF REPRESENTATIVES, 1790-1994

Vol. 14 No. 3

Stephen C. Erickson

The "Intuitive Impulse"

On November 8, 1822 an article on term limits, or "Rotation in Office" as they were called, appeared in the Richmond Enquirer. The author argued that long tenure in public office leads to abuse of power. He declared that the principle of rotation of public offices was "long cherished" in America and was impressed upon the public mind "by a kind of intuitive impulse, unassailable to argument or authority."

The overwhelming popularity of the current term limits movement suggests that this sentiment remains alive and well in modern American political culture, despite very high incumbent reelection rates. As a general principle, Americans continue to believe in a centuries-old democratic tradition that power held over extended periods of time corrupts representative institutions. And today, many Americans seem convinced that their political institutions are thoroughly corrupt.

Nevertheless, of those who choose to run for reelection in the U.S. House of Representatives, over 90 percent are assured of victory, a fact it is easy to lose sight of given the historic Republican takeover of both houses of the U.S. Congress in the 1994 election. The Republican "revolution" was based on their ability to toss a small percentage of Democratic incumbents out of office, and to win open seats. In fact, the power of incumbency masked the true scale of the popular sentiment in favor of Republicans. The advantages of incumbency not only make elections unfair to challengers, but also minimize the impact of popular partisan swings on Congress.

This paper traces the path of reelection rates and congressional turnover historically, through the fertile valleys of citizen legislatures to the mountains of immovable incumbency. The focus is on the U.S. House of Representatives, since Senators were subject to indirect election up until after the turn of the 20th century. The paper is an attempt to provide historical context for the current state of American democracy, and in so doing explain the peaks and valleys of House reelection rates, as illustrated in Figure 1.[iu1] Indeed, Figure 1 is a kind of topographical map of incumbency to accompany the following description of the reelection landscape since the nation's founding.

The history of congressional turnover readily breaks into four parts: (1) the Republican Era of Social Deference and Public Service, 1787-1828; (2) the Era of Jacksonian Democracy, 1829-95; (3) the Era of Safe Seats, 1896-1957; and (4) the Era of Entrenched Incumbency 1958-present. Each of the four periods represents a distinct set of circumstances that shaped elections and reelections for the U.S. House of Representatives.

The Republican Era of Social Deference and Public Service

James Madison was immersed in political theory during the winter of 1786-87. He read voraciously prior to the Constitutional Convention that met in Philadelphia the following spring. Although the convention was billed as a means to improve the existing Articles of Confederation, Madison and some of his allies had far more ambitious plans for the gathering. They were going to construct a new political system from the ground up, that would usher in "a new order of the ages" (Novus Ordo Seclorum). Madison and his allies understood that the success of the great American republican experiment rested on their ability to properly structure the new government. His political consultants included the great philosophers of antiquity, generations of English republican theorists, and the leading figures of the European Enlightenment. Among the subjects Madison's reading encouraged him to consider, when he contemplated the proper design for a national legislature, was "rotation in offices."

The principle that new representatives should regularly be circulated through the political system was an integral part of republican political theory. Aristotle (Politics, IV, Sec. 2: 2091) wrote approvingly of how office rotation was practiced in ancient Greece, where citizens were retired from public office to live under the laws they had made, and replaced by a new group of citizens. Livy recorded that offices were also rotated in the Roman Republic. Sixteenth-century Italian humanist writers advocated republican political structures, including mandatory rotation of offices, that encouraged rulers to exercise virtue over self-interests. From the Italian humanists the Englishman James Harrington adopted many of his political ideas. Like James Madison much later, Harrington aspired to design and build a republic. The result was his The Republic of Oceana ([1656] 1977), a constitutional blueprint that Harrington tried, unsuccessfully, to sell to Oliver Cromwell during the English Civil War. For Harrington, rotation of offices was absolutely essential to the well-being of a republic. A legislator returning to the body of the people was analogous to a blood cell returning to the lungs to be enriched with fresh oxygen (Pocock 1975: 393-95). Generations of English republican thinkers adopted Harrington's ideas on rotation (Robbins 1968). John Trenchard and Thomas Gordon, who were especially influential in America, wrote on the subject in Cato's Letters ([1755] 1995: Nos. 43, 61; Bailyn 1967: 35-36) and warned that long tenure in office frequently led to aristocracy.

Madison was familiar with all of these political philosophers, and no doubt appreciated their arguments for mandatory rotation of offices. He was also well acquainted with the various laws of the American states. Not only had there been rotation under the Articles of Confederation, but many states incorporated term limits in their early constitutions. Virginia, Pennsylvania, Maryland, and Delaware all rotated their governors out of office. Virginia, New York, and Delaware limited the number of terms that Senators could serve. Most radically of all, Pennsylvania limited the terms of the members of its unicameral legislature. Other states regularly rotated sheriffs and coroners (Adams 1980: 308-9). In history, in theory, and in practice, Madison and his fellow constitutional framers had many models from which to borrow concepts for republican government, and many of these included term limits.

Behind the scenes Madison prepared the Virginia delegation for the Constitutional Convention to be held at Philadelphia in 1787. In letters to George Washington (16 April 1787) and Edmund Randolph (8 April 1787), Madison outlined what would become known as the Virginia Plan, the working draft of the Constitution itself. In the Virginia Plan, Madison included radical term limits for the House of Representatives, consisting of mandatory retirement after a single two-year term in office (Madison's "Notes on the Federal Convention" in his Papers).

Madison made a unique contribution in defense of term limits. He prophetically warned of the danger that factions, or special interests, presented to a republican form of government. He worried that legislators, absorbed by their desire for reelection, might serve their own narrow self-interest at the expense of the overall national interest. By mandating periodic retirement for legislators, and therefore making reelection impossible at times, Madison hoped to minimize the problem of special interest politics (Erickson 1993).

Madison may have worked with fellow Virginian George Mason when drafting the Virginia Plan. Mason was an ardent believer in rotation of offices. He had included term limits in the widely imitated Virginia Declaration of Rights, which suggests that like other specifically enumerated rights, mandatory term limits were considered an essential guarantee against tyrannical government. Mason was disappointed by the exclusion of term limits from the Constitution, along with other provisions that would later be incorporated as the Bill of Rights. Likewise, upon reading the Constitution for the first time, Thomas Jefferson declared his hope that some of its shortcomings might be remedied. "One thing I dislike, and greatly dislike," Jefferson wrote to Madison (20 December 1787), "is the abandonment in every instance of the necessity of rotation in office." The absence of any assurance that there would be rotation of office was also a sticking point for many Americans who would become known as Anti-Federalists.

The provision for term limits was dropped, without discussion, from the Virginia Plan. Madison would again briefly return to the subject during the course of the debate over the Constitution at Philadelphia. Yet, it appears that among the framers, Madison's and Mason's view on the necessity of mandated rotation was decidedly in the minority. Term limits were also not included in the Bill of Rights, where Mason might have incorporated them. It would be a mistake, however, to assume that the framers favored professional politicians. The historical context in which the framers designed their republic helps explain the exclusion of term limits.

The 13 states emerged from the Revolution victorious, but hobbled with enormous financial difficulties. They faced a mountain of public and private debt, as well as a severe economic depression. Specie drained out of America into the coffers of large merchants in England, individual states were insolvent, foreign loans needed to be paid off, and internal trade barriers between states discouraged commerce. Western Massachusetts erupted in outright rebellion. The government created by the Articles of Confederation was hopelessly too weak to adequately address the young republic's formidable problems.

To the gentlemen who wrote the Constitution, politics had also taken a dangerous turn after the Revolution. They saw demagogues appealing to the passions and narrow self-interests of the masses. Men who had formerly been of low social stature in their communities were now exalted to high public office, while representatives of the most elite families shunned public service along with the new brand of relatively democratic politics.

No radical development better illustrated the fears of elite republican culture than the adoption of the Pennsylvania Constitution of 1776, which included a unicameral legislature, radical suffrage, and mandatory rotation in office. The new Pennsylvania Constitution swept the old Quaker elite from power, and replaced them with radical evangelical Scots-Irish. Arguably, Pennsylvania's legislative term limits provision was less radical in practice than it was in spirit. It openly distrusted those elite characters who would normally be elected to serve in government, implying that they would not necessarily retire of their own volition. Thus rotation was forced through the power of law (Wood 1969).

In sum, the authors of the Constitution failed to include a provision for mandatory rotation in office because they perceived that the "best" men were no longer serving in government. New unsophisticated and ignorant politicians had, in many cases, come along to take the place of America's leading patriot families. By creating large congressional districts, the framers hoped that men would win office based not on their political connections or demagoguery, but on sterling public reputations known far and wide. Simply put, they hoped that under a new system, men like themselves would be elected. Given the public's apparent predisposition to throw gentlemen out of office, term limits seemed at least unnecessary, and at most counterproductive.

While in retrospect the framers sound elitist--indeed, they were elitists--their attitude was a product of the time and place in which they lived. The world of 18th-century America, even within most republican circles, was strictly hierarchical. The best families asserted their social positions in every way possible. They lived in the nicest homes, wore elegant clothing, sat in the front pews of churches, and served in public office. Tipping their hats in deference to their community's "natural leaders," common folk generally went along with this hierarchical system. To be sure, significant social mobility existed in America, but only successful men who carried themselves as gentlemen could assume leadership in the community.

The framers hoped to encourage the nation's leading men to serve in government. Naturally, the prominent men were wealthy. Wealth meant financial independence. A gentleman of means, it was believed, would be more likely to serve the needs of his community and was less tempted by the corrupting influences of money and power associated with high public office. Furthermore, it was quite natural that only those with the leisure time to read broadly could govern justly. There was no congressional staff, no think-tanks, and no bureaucracy to guide elected representatives. There were few books around by today's standards, and those that existed were expensive items. Colleges and universities stressed theology. Practical training for running a government had to come on the job, without the help of public policy experts. Given the complexity of the economic and political problems in post-revolutionary America and the diversity of local conditions throughout the 13 states, the new nation seemed to need experienced legislators.

However, the framers did not favor professional politicians, whose motivations could not be disinterested. Looking back on the long struggles of English Whigs against the Stuart monarchies, American republicans understood the danger that sprang from a government class, self-serving, and distant from the people. The generation that made the American revolution, both Federalist and Anti-Federalist alike, scorned those who made their living from government. The ideal republican was embodied in Cincinnatus, a Roman farmer who put down his plow to take up arms in defense of the community. When the crisis ended, he returned to his plow. The motivation driving Cincinnatus was public virtue; not money or power, but a reputation as a virtuous citizen, was the permissible reward under a republican system of government. George Washington's behavior toward public office--his outward reluctance to hold power, climaxing in his refusal to serve a third term as President--exemplified the ideal of Cincinnatus.

The political values of 18th-century republicanism faded slowly and continued to dominate electoral behavior well into the 19th century. A high turnover rate in Congress, with about one-third of its members replaced each election cycle, was precisely the rate James Harrington and his many American political heirs had called for in mandatory rotation laws. Madison did not fight very hard for term limits, probably because he knew that the sentiment at the Philadelphia Convention was against them, and the specter of entrenched incumbency was then only a theoretical menace. In the early years of the republic, Madison's hopeful speculation proved true, that a "few" members of the legislative branch would be "frequently reelected" (Madison's Notes, 21 June 1787), but "new members would always form a large Proportion" (Federalist No. 53). Concerns that Congress would become a separate, entrenched class seemed unwarranted. It appeared that no term limits law was necessary.

The high turnover rate in the late 18th and early 19th centuries, as Figure 1 shows, was dependent on regular voluntary retirement from public office. Certainly the spirit of Cincinnatus, which played an important role in the political culture, encouraged such behavior. However, public service in early America was in many ways a genuine hardship. Living costs and social expectations at the capital outweighed the financial compensation for serving in Congress. Most Congressmen probably left Washington, D.C. less wealthy than when they arrived. Poor transportation, consisting primarily of notoriously bad roads, made trips to the banks of the Potomac an arduous journey. Once in Washington, trips home for business or to visit family were necessarily few and far between. The incentives to remain in the District of Columbia were also underwhelming. Hardly a bustling metropolis, the capital consisted of only a few buildings scattered about in a physical and social wilderness. The activities of the federal government were often mundane, and arguably, the most important decisions were made at the state level throughout much of the 19th century. In fact, members of Congress would often return home to serve in state legislatures (Kernell 1977).

For all of these reasons, Congress experienced a high rate of retirement in the early republic. Turnover had much less to do with defeat at the polls, especially in the years immediately following the establishment of the Constitution. The egalitarianism of the Pennsylvania Constitution was premature. It was replaced by a more conservative document, made in the mold of the Massachusetts and federal Constitutions. Well into the 19th century, leading families continued to send representatives to serve in government, who were dutifully elected and reelected, when they chose to stay in office, by their fellow citizens occupying lower steps on the social ladder. Turnover remained high due to frequent voluntary retirements. Yet the American political system increasingly accommodated itself to more egalitarian political ideas and practices (MacNeil 1963). As the largest-scale democratic meetings ever assembled in America, the state conventions that ratified the Constitution were themselves radical departures in self-government created by the federalists (Wood 1969). Indeed, the new social and political order planted at the time of the Revolution took root slowly, but its growth was unmistakable. To trace the declining rate of those who stood for reelection and won between 1800 and 1850 in the U.S. House, is to mark the decline of a system of social and political deference, and the rise of values commonly associated with Jacksonian democracy.

The Era of Jacksonian Democracy, 1829-95

The language and ideas of the American Revolution created an ideology more politically and socially radical than many of the makers of the new republic intended. The notion that "all men were created equal," for example, could be taken far beyond the basic principle that all men merely had equal rights. Political equality, taken to an extreme, might mean that all men were equally capable of governing as well as being governed.

While Andrew Jackson considered himself a political heir to Jefferson, his attacks on "privilege" and the growing suspicion that no representative could legislate with disinterest made the whole idea of "public virtue" appear almost chimerical. The intensified skepticism over the ability of any officeholder to remain in power and uncorrupted for any significant length of time led to comprehensive applications of the idea of rotation of offices.

To be sure, members of the revolutionary generation, like Madison and John Adams, held no illusions about human nature. What changed was the notion that elite gentlemen were better equipped to legislate for the public good, and were naturally more disinterested. Americans became steadily more convinced that all power corrupted, and that wealthy planters and lawyers were as self-interested and corruptible as anyone else. Elite political leaders could not be counted on to legislate for the common good or represent the common man. The idea that gentlemen allegedly had a monopoly on reason was replaced by a widespread faith in common sense (Wiebe 1984, Wood 1993).

Andrew Jackson (as quoted in Richardson 1896: vol. 2, 448) said as much in his address to the 21st Congress,

There are perhaps few men who can for any great length of time enjoy office and power without being more or less under the influence of feelings unfavorable to a faithful discharge of their public duties. The duties of all public officers are, or, at least, admit of being made so plain and simple, that men of intelligence may readily qualify themselves for their performance; and I cannot but believe the more is lost by long continuance of men in office, than is generally gained by experience.

Jackson's deep distrust of the holders of government-granted power and privilege led to his well-known war on the Bank of the United States, and attacks on other government monopolies. Jackson's presidency also marked the beginning of the high point of office rotation, "a leading principle of the Republican Creed," as Old Hickory called it.

The number of House members who sought reelection and won after 1800 declined over the course of the next half century as the era of citizen government reached its height. Incumbents who chose to run again were 95 percent successful at the beginning of the century, when old hierarchical and deferential social structures remained in place. But in the election of 1842 the survival rate of incumbents bottomed out at 64 percent. That same year so many incumbents chose not to run for reelection that Congress experienced a 76-percent turnover. By the 1840s, the reelection rate for incumbents attempting to maintain office settled into an average range where 20 to 30 percent regularly went down to defeat. This attrition rate due to defeat continued almost up until 1900. Adding in voluntary retirements, the total turnover rate in Congress averaged more than 50 percent per election between 1840 and 1880.

Many of the founding fathers would have looked askance at the U.S. House of Representatives of the mid-19th century, dominated as it was, not by the nation's leading characters, but by quite ordinary men. The famous French traveler and aristocrat, Alexis de Tocqueville, while admiring much of what he saw in America, was disdainful of the lower house of the United States legislature. "On entering the United States House of Representatives in Washington," sniffed de Tocqueville ([1835] 1965: 103), "one is struck with the vulgar demeanor of that great assembly." He added that "the eye frequently does not discover a man of celebrity within its walls." The body contained "almost all obscure individuals," including "village lawyers, men in trade, or even persons belonging to the lower classes of society." The U.S. House of Representatives in the mid-1800s came as close as it ever would to a true citizen legislature. In the same report, de Tocqueville pointed out that the Senate, by contrast, contained gentlemen who compared favorably with their European counterparts. Madison and Jefferson would have at least appreciated the contrast between the two houses. Madison's goal of a lower house that was immediately reflective of the people was realized in the 19th century.

While the belief in social and political deference that existed during the age of the nation's founding faded, the republican principle of voluntary rotation did not. In fact, the custom became more prevalent as a result of the rise of party politics and the birth of "the spoils system" (McCormick 1986) under Jackson. The spoils system constituted a new form of rotation, where tenure in appointed office became limited by the duration of the authority of the appointee's political allies. When Jackson spoke of the disadvantages of allowing individuals to hold public office for extended periods of time, he meant all public offices, including administrative ones. Since all offices were public property, they were free to be distributed to the party faithful. According to the Jacksonians, such a rotation not only rewarded political allies, but also discouraged the corruption that stems from long tenure in office.

Acceptance of the idea of office rotation became so dominant during the 19th century that a set of unwritten rules evolved. In district political conventions, party leaders took turns holding elected office. In 1824 the New York Statesman (as quoted in Struble 1979: 659) reported that "perhaps most" of the turnover in the state's congressional delegation was the result of "an arrangement, by which it is stipulated that, after a given time, one aspirant for office is to succeed another."

Often counties took turns sending representatives to Washington, in a geographic rotation. In 1869, the Christian Examiner (as quoted in Struble 1979: 660) noted the "constant temptation, in a district made of an aggregation of counties or towns, to pass the office from town to town, or county to another each claiming in its turn the honor of furnishing the member."

One of the most well-known examples of the practice of rotation involved Abraham Lincoln. In describing Lincoln's 7th Illinois Congressional District, his secretaries, John Nicholay and John Hay, reported (as quoted in Struble 1979: 659-60),

The Sangammon district was one which the Whigs of Illinois had apparently the best prospect of carrying, and it was full of able and ambitious men, who were nominated successively for the only place which gave them the opportunity of playing part in the national theater at Washington.

Lincoln, by agreement, served only one term in Congress. When the subject of running for reelection came up, Lincoln wrote (as quoted in Struble 1979: 660), "to enter myself as a competitor of another, or to authorize anyone so to enter me, is what my word and honor forbid." The man who succeeded Lincoln, Steven Logan, was the same man who had nominated Lincoln two years earlier.

The strong party system of the 19th century contributed to rotation of offices by encouraging compromises between politicians and different geographic regions within congressional districts. A candidate's fate was closely tied to the fate of his political party and the presidential candidate that headed the ticket. Therefore a candidate would not campaign directly for himself, but for his party, on which his own political success depended. Major campaign contributions went through party channels to the candidates. Party control through money and closed nominating conventions helped party bosses to enforce rotation of offices. Rotation itself contributed to party discipline and coherence by allowing various members to play a role on the national stage in the U.S. Congress.

Yet the party system and the rotation of offices that often went along with it underwent dramatic changes at the end of the century. Such changes were ultimately the products of the tumultuous transformations occurring in the American economy and society toward the end of the 19th century.

The Era of Safe Seats, 1896-1957

The transformation of the agrarian republic into the industrial nation-state radically shook American social and political structures, including the electoral system. Industrialization, urbanization, sectionalism, the rise of a new ideology of professionalism, and large-scale immigration, combined to undermine the old practice of rotation of offices and made elections less competitive. Marginal congressional districts--districts rather evenly balanced between Republicans and Democrats--began to disappear, creating uncompetitive seats in both the South and the North. The era of safe seats was caused first, by a polarization of economic interests, and second, by a systematic disenfranchisement of minority groups during the progressive era.

Distinct sectional interests, given bold definition by northern industrialization as well as by southern slavery and its aftermath, dominated American politics both before and after the Civil War. The issue of the tariff in particular cleaved the nation into sections, with the industrial North and Midwest favoring protectionism, and the South and far-West opposed. Agricultural regions also favored a silver standard to ease debt, which was opposed by Eastern financial interests. The Populist political revolt of the 1890s underscored the sectionalist nature of the economic conflict. The Populists drew their strength from agricultural areas in the South and West. The election of 1896, when the Democrats nominated the Populist William Jennings Bryan as their standard-bearer, is often characterized as one of the great realigning elections in U.S. history (Bensal 1984, Kleppner 1987). While the effect of that particular election may be overstated, there is little doubt that the last decade of the 19th century, and first decade of the 20th century, saw the solidification of Republican strength in the North and a Democratic monopoly in the South. Clearly much of this polarization was based upon economic sectional differences.

With depressions setting in like plagues, huge numbers of non-Protestant immigrants pouring into overcrowded cities, labor unrest, urban corruption, and unhealthful living and working conditions, the scale of changes taking place was alarmingly unprecedented. One historian has aptly named the response to this upheaval "a search for order" (Wiebe 1967). Many progressive reformers believed that the chaos of the time required professional management. J.P. Morgan, the great financier, argued that government should be run like a business.

Historian Robert Struble contends that an ideology of professionalization replaced the belief in office rotation and slowed turnover in the U.S. House at the turn of the century. Certainly a waning of the practice of rotation accompanied the trend toward professionalization. However, Americans did not necessarily abandon the ideal of citizen government in favor of a professional class of lawmakers. The disappearance of voluntary rotation had other causes as well.

The radical transformation of political parties at the end of the century, brought about by the progressive belief that old party practices and structures were fundamentally corrupt, was a major reason for the change. The introduction of open primaries and the overall weakening power of party bosses removed policing forces that had helped enforce rotation of offices. Party leaders could no longer limit an officeholder's tenure, because the power to do so increasingly rested with the electorate, and, as we shall see, the electorate was increasingly configured in such a way as to discourage competition at the polls. With the dismemberment of old party structures, a candidate's electoral fate was, in part, cut free from the success or failure of the party as a whole.

Yet the decline in voluntary rotation began before both the rise of the ideology of professional management and well before the beginning of party decline. The growth of cities provides another explanation for the long-term decline of voluntary rotation. The process of intense urbanization began in many areas, especially in the Northeast, before midcentury. After 1910, rural America could no longer marshal a majority in the federal census. Not surprisingly, politically powerful urban centers had little enthusiasm or incentive to share power with sparsely populated agricultural communities. Naturally, as some congressional districts became heavily weighted in favor of urban concentrations, the imbalance discouraged geographic rotation of offices.

Native-born white Protestant rural America responded with alarm to the growth of ethnic urban centers. That agricultural America might lose its place of dominance had profound implications for a nativist culture still infused with a Jeffersonian agrarian ideal. Nativist anxieties about the effects of a growing ethnic urban influence on American democracy prompted the old Protestant classes to entrench themselves, politically, against newcomers who, it was believed, were generally unfamiliar with the habits of democracy. The nativists' first line of defense was to protect their interests through gerrymandering.

Of course, gerrymandering was nothing new. But by the end of the 19th century, state legislatures, when revising political districts, increasingly turned away from drawing lines based on population. Instead they invoked a "federal principle," which justified overrepresentation for rural communities, which in turn tended to be white and Protestant. In defense of the federal principle, Elihu Root, arguing at the New York Constitutional Convention of 1894, pleaded that his state follow the lead of other states with "great cities" within their borders. He advocated putting small agricultural communities on an equal footing with urban centers. "Otherwise," declared Root (as quoted in Dixon 1968: 83-84), "we can never have a truly representative and truly republican government."

In the last decade of the 19th century, the federal principle created congressional districts in various states that were 40 percent larger than the state's smallest district (Dixon 1968: 91). The consequences of such redistricting isolated immigrants, who tended to be Democrats, in burgeoning urban districts. Hence, in the North, the federal principle encouraged proportionally fewer Democratic urban districts and more rural Republican ones. Likewise, the principle was one of several tools employed to disenfranchise blacks in the South, which meant undermining the Republican vote below the Mason-Dixon line. Disparities between the populations of urban and rural districts increased during the course of the 20th century until the Supreme Court intervened in 1964. Following the landmark Supreme Court case Baker v. Carr, which found unequal representation in state legislatures unconstitutional, Wesberry v. Sanders put an end to the "federal" apportionment system for U.S. House seats. The case involved congressional districts in Georgia, where the state's smallest district contained 272,154 rural inhabitants, compared with the 823,860 inhabitants of Georgia's largest district in Atlanta (Dixon 1968; Congressional Quarterly 1985: 694-95).

Redistricting was just one way Jim Crow established safe Democratic seats in the South. Northern efforts to reconstruct the South into a biracial democracy ended in 1877 when federal troops withdrew from the old confederacy. Yet Republicans won seats in the South through 1900. Sometimes as many as five or six Republicans represented North Carolina or Tennessee. However, in 1893, when the Democrats controlled the federal government, it formally abandoned supervision of federal elections in the South. Turning the South's black citizens into scapegoats, Southern populists and "Bourbon Democrats" firmly united under the banner of Jim Crow in 1896. In the years that followed, Southern blacks, who were predominantly Republicans, were systematically disenfranchised through the imposition of poll taxes, property qualifications, and literacy tests. With the exception of East Tennessee, the South was solidly Democratic by 1910, and remained so through the middle half of the 20th century (Woodward 1955, Grantham 1988).

One of the ironies of the period was that as blacks were disenfranchised, women obtained the vote. However, even this expansion of democracy undermined marginal districts by swelling Republican majorities in the North.

Women's suffrage was a long time coming. It grew out of abolitionism, and for most of the movement's 70-year history, based its appeal on the notion that all people were created equal and worthy of full citizenship rights. Yet as historian Aileen Kraditor discovered, women's suffrage only succeeded when, at the end of the century, suffragists began appealing to Protestant nativism. They won over support for their cause by arguing, first, that there were more native-born women than foreign-born men and women combined. Second, they advocated literacy tests, which would further exclude the foreign born from politics and further entrench the literate, native-born Protestant class (Kraditor 1965).

Literacy tests could be defended everywhere as a progressive reform to ensure an educated citizenry. However, in the North, like the South, literacy tests furthered the interests of native-born elites. Nine Northern states imposed such tests. As black Republicans were disenfranchised in the South, literacy tests and women's suffrage undermined Democrats in the North.[f1]

The era of safe seats also saw the birth and expansion of a number of practices associated with modern government, including the creation of legislative staff. There is, however, little evidence to suggest that such practices had much of an effect on congressional turnover during the first half of the 20th century. But the practice of voluntary rotation took a near fatal blow. As a result of the destruction of party power and the effects of urbanization, incentives that encouraged voluntary rotation disappeared. Sectionalist economic interests furthered the creation of safe seats, as did progressive era reforms that entrenched white Protestant Republicans in the North and Democrats in the South.

In 1910 dissident Republican members of Congress joined their Democratic peers in a successful revolt against Republican Speaker Joseph Cannon. In so doing they undermined the power of the Speaker of the House and created a congressional system based on seniority. It is noteworthy that Republicans broke with their party in order to create a system that furthered their own individual political interests. Longevity in office, not party loyalty, became the new route to power. Amid the chaos of the period, the new direction in American politics may have appeared uncertain. However, from a historical distance, it is clear that structural developments, many of which were still in their infancy, were beginning to systematically erode American democracy.

The Era of Entrenched Incumbency, 1958-Present

While the era of safe seats undermined marginal districts, part of an incumbent's safety arose from a natural, as well as an artificial, distribution of regional interests. Incumbents were safe because more often than not, they represented their constituent's interests, even if those constituencies were subject to some manipulation. Furthermore, in times of crisis, during the Great Depression, for example, more than one-third of the U.S. Congress might fail to be reelected. Yet by 1960, the power of incumbency itself, as distinct from a constituency manipulated to be homogeneous, began to climb. Burgeoning congressional staffs, massive use of the franking privilege, growing incumbent campaign war chests, mass media, and expanding federal power combined to provide incumbent members of Congress with advantages that made them almost immune to defeat.

Ground was broken for incumbent entrenchment early in the century. A few astute observers recognized the potential for abuse in what was then a new congressional system. De Alva Stanwood Alexander, a scholar of Congress writing in 1915, foresaw the dangers of new congressional resources, especially staff. "A member's clerk," declared Alexander ([1915] 1970: 152), "becomes the head of a campaign bureau, with the hope that franked favors may induce grateful constituents to remember the sender on primary and election days." Alexander may have fretted excessively over the impact of a single clerk and the franking privilege. Yet clearly politicians understood from very early on how to use public office to secure reelection. If Alexander's concerns seemed overly anxious in 1915, by 1975 they were prophetic.

Much has been written about the more recent lack of congressional turnover since political scientists like Erikson (1971), Mayhew (1974a, [dc18]1974b), and Fiorina (1977) began noticing the trend in the early 1970s.[f2] There is general agreement that money from political action committees (PACs), institutional perks, and federal pork have all contributed to the preservation of incumbency. Many scholars have rightly pointed to the increasing role and power of the federal government to explain the transformed nature of congressional priorities. Acting as ombudsmen, members channel resources to their districts, and perform constituent services in order to help secure continual reelections. However, the entrenchment of incumbency appears to have begun at a rather static time in the expansion of federal authority and power. Precisely when and how incumbency began its entrenchment remains a point of some contention, but the broad outlines of the process are clear.

Political scholars have devised one helpful approach to distinguish whether it is the power of incumbency that reelects officeholders, or the safety of their seats. In other words, does a politician continue to get reelected because he or she represents like-minded people, or does the politician use the power of public office to gain advantage at election time? John Alvord and David Brady (1989) address this question by charting two measures. First, they take the margin of an incumbent's victory in his or her last reelection and subtract the margin of victory of the winner of the following open-seat election, provided that the open-seat election is won by a member of the same party. They also calculate the difference between the winner's margin in an open-seat race and his or her first reelection. Since the open-seat margin is almost always smaller, these measures are called the "freshman slump" and the "sophomore surge," respectively. Alvord and Brady average out these differences for each election cycle beginning in 1846. In this way they separate the effect of incumbency from the natural safety of the district. As illustrated in Figure 2, the results are rather dramatic. While there has always been an advantage to being an incumbent--usually it provided two or three extra percentage points at the polls--at about 1958 that advantage took off to levels of seven and eight points, and has recently soared even higher. Clearly something began happening in the late 1950s that significantly increased the advantage of incumbency.

Alexander's early observation in 1915 on the potential role of congressional staff points to the essential element that has institutionalized the advantages of incumbency. The ability of professional staff to run what are essentially full-time reelection campaigns is the necessary element in the rise of all of the subsequent advantages of office holding. Year-round fundraising, "favors" for constituents, management of direct mail, and the promotion of constituent-oriented legislation are all made possible by staffing. They are also all unavailable to challengers.

The Legislative Reorganization Act of 1946 was the backhoe that began the entrenching of incumbency by initiating an unprecedented expansion of congressional staff. Congress justified the explosive expansion of its personnel as a counter to the growth of the Executive Branch, and to address an increasing congressional workload. Yet no matter how much bigger congressional staff became, it somehow never caught up to its "workload," because new tasks that promoted reelection could always be found. House personal staff has grown enormously since the Legislative Reorganization Act. According to Norman Ornstein (1994: 126-27) the House employed 1,440 personal staff in 1947, 2,441 in 1957, 4,055 in 1967, 6,942 in 1977, and peaked at 7,584 in 1987. As a function of image making based on constituent services, staff has increasingly been moved closer to the voters. While in the late 1950s many congressmen had virtually no district staff, by 1992 almost one-third of all personal staff worked outside of Washington (Ornstein 1994: 131). And to confirm Alexander's warning, use of the franking privilege as a means of image making has increased from 45.6 million pieces of mail sent out in 1954, to regularly over 500 million pieces today (Ornstein 1994: 163).

Large staffs also run year-round campaign fundraising organizations, which they have done with increasing effectiveness. Although money has always been important in politics, it was never more so than in an age of modern media and incumbent image-building. Testifying before the Senate Subcommittee on Privileges and Elections in 1955, FCC Commissioner Freida Hennock noted,

A major change has taken place in the process of elections since television has been the most effective means of campaigning. ... The cost of reaching the public over television and radio at election time are so great that there is grave danger that the outcome of our elections will depend upon the war chests of the candidate and his party, and not upon the intrinsic worth of their qualifications.

Unfortunately, it is impossible to track campaign finances before 1974, when complete disclosure became mandatory. However the entrenching of incumbency seems to have coincided with the rise of television. Media scholar Robert Lichter (Lichter et al. 1986: 7) marks 1958 as the year "the age of television began," when the number of televisions approximately equalled the number of American homes. Perhaps not coincidentally, 1958 is also the precise year that the power of incumbency took off, according to Alvord and Brady's analysis. By 1960, the year of the first televised presidential debate, between John Kennedy and Richard Nixon, television was clearly having a powerful political effect. And if television is the engine to reelection, money is the fuel. With a full-time fundraising staff, incumbents have long had an advantage when it comes to building campaign war chests (Jacobson 1980, Malbin 1984). Television both decreased the unit cost of reaching voters, and provided the political process with a medium that was revolutionary in terms of its capacity to create public images. It maximized the impact of campaign funds by making possible, like never before, a personal appeal to voters.

Incumbents have dramatically expanded their capability to raise campaign funds, while challengers have clearly not been able to keep pace. Recently, PACs have provided decisive financial advantages almost exclusively to incumbents. In a circular arrangement, PACs today give to incumbents because incumbents win. Incumbents win largely because PACs give to incumbents. Party affiliation or ideology are practically irrelevant to the vast majority of PACs, who represent various economic interests. In 1976 House incumbents, on average, outspent challengers by $79,398 to $50,795. By 1992, the gap had widened to where incumbents spent $594,729 as compared with $167,891 for challengers (Ornstein 1994: 75). To illustrate incumbent advantage another way, the average amount of money spent on the congressional franking privilege during a single House term is about as much as the average challenger spends on an entire campaign.

Although the story of continued entrenched incumbency has been buried under news of the 1994 Republican "revolution," the recent election was, in fact, a conservative affair in terms of congressional turnover. The overall reelection rate in 1994 is consistent with four other elections in the era of entrenched incumbency. The U.S. House experienced a turnover of 20.8 percent in 1964, 21.1 percent in 1974, 18.6 percent in 1982, 25.3 percent in 1992, and now in 1994, 20 percent. Likewise, 91.3 percent of all incumbents who ran in 1994 defeated their challengers, a figure consistent with other elections in the era of entrenched incumbency.

If the advantage of incumbency is at least 7 percent, as Alvord and Brady suggest, then the power of incumbency disguised the true scale of popular sentiment in favor of the Republicans over the Democrats in 1994. If each challenger had gained 7 percent at the expense of the incumbent in the last election, then the Republicans would have won between 32 and 41 additional seats. Indeed, the "gridlock" experienced in recent times appears to be a function of Congress's lack of responsiveness to changing moods in the electorate, especially as compared with the relative responsiveness of the Executive Branch. Entrenched incumbency shortens presidential coattails, and retards the force of popular sentiment on Congress.[f3]

The Reforming Impulse

Although the advantages of incumbency have not gone unnoticed, most reforms proposed fall far short of addressing the problem. Not unexpectedly, Congress itself has not seriously considered any reform that would affect the system in a meaningful way. A host of proposals addressing the subject of campaign finance are regularly kicked around by public interest groups. They include various degrees of PAC reform, public campaign financing, and channeling campaign contributions through political parties. There have also been proposals to further restrict franked mail beyond the minimal reforms that Congress has already enacted. In addition, Freshman Republicans recently tried to implement modest cuts in House personal staff, but were thwarted by the leadership. So far, the 104th Congress has restricted cuts in personnel to the staff of abolished committees. Not only have most meaningful reforms failed to gather any significant degree of political momentum, but most only address a single aspect of incumbent power. Term limits is not only the sole reform with popular political clout behind it, but it also provides a comprehensive solution by guaranteeing turnover, ensuring periodic competitive open-seat races, and undermining cravings for perpetual reelections. By comparison, all other proposals, taken individually, only tinker with a system that requires structural reform.

Members of Congress have sensed the political power of the term-limit debate. Republicans have exploited the popularity of the issue by incorporating term limits in their "Contract with America." However, a majority of House Republicans who claim to support term limits have also tried to water down the reform by allowing six elections for members of the House. By contrast, only 3 of the 22 states that passed term-limit laws would have allowed such generous tenure for their federal representatives. Even the watered-down version failed to garner enough votes to send a constitutional amendment on to the states.

Permitting as many as five reelections for members of the House would both keep its members susceptible to the narrow and short-term interest pressures that fuel reelections, and also insulate that body from popular mandates. Only strict term limits will tend to make the U.S. House more immediately reflective of the American people and discourage faction, or special interests, as Madison intended.

Indeed, Madison had proposed a one-term limit for members of the U.S. House when he designed the Virginia Plan, as if he understood that open-seat elections most accurately reflect the immediate will of the people on their government (Erickson 1995). Madison and the other founders knew that special interests, especially economic interests, would always try to manipulate the political system, and that officeholders might use their positions to secure their own reelections by serving special interests. The challenge was to create a political structure that discouraged such manipulation. For all of their vision, the founders could not have foreseen the rise of an industrial democracy, or the regulatory state and technological innovations that have accompanied it.

In fact, rotation of offices operated naturally during the first 100 years of the republic's history. During the progressive era, however, reformers lamented the growth of corrupt urban centers, and were stunned to discover the extent to which big business influenced government (McCormick 1981). They responded by trying to protect the old rural America they had known and idealized. In so doing they gave government new authority to manage the frightening new world in which they found themselves, but the power awarded to government was also necessarily placed in the hands of legislators. No new limits were placed on the people's representatives to correspond with the radically expanded role of government that arose out of the progressive era, the New Deal, and the Great Society.

Since about 1958, professional politicians have learned how to use the power of incumbency to perpetuate their tenure in office. In response, millions of American citizens, motivated by the "intuitive impulse" that assumes political careerism is inheritently dangerous, passed term-limit laws in every state where the issue came up for a vote. This historic popular reform effort was, however, brought to a screeching halt in the spring of 1995. Ruling against the constitutionality of state-mandated term limits in U.S. Term Limits v. Thornton, the Supreme Court invalidated the state laws in a decision with possibly profound consequences. In its wisdom, the High Court has left the fate of term limits legislation largely in the hands of those who benefit most from the status quo, professional politicians in the U.S. Congress. Given the unlikelihood of a Constitutional Convention, the vast majority of Americans who perceive the need for term limits must now rely on members of Congress to legislate against their own immediate self-interests, a very rare thing in politics. Whether or not Congress finally passes on to the states a constitutional amendment for meaningful term limits will prove to be a crucial test of the ultimate health of American democracy.

References

Adams, W.P. (1980) The First American Constitutions. Chapel Hill: University of North Carolina Press.

Alexander, de A.S. (1970) History and Procedure of the House of Representatives. New York: Ben Franklin.

Alford, J.R., and Brady, D.W. (1991) "Personal and Partisan Advantage in U.S. Congressional History, 1846-1986." In Silbey, J.H. (ed.) The United States Congress: The Electoral Connection, 1789-1989, vol. 1, 23-40. Brooklyn, N.Y.: Carlson Publishing.

Aristotle (1984) Politics. In The Complete Works of Aristotle, Book IV, Sec. 2. Edited by Jonathan Banes. Princeton: Princeton University Press.

Bailyn, B. (1967) Ideological Origins of the American Revolution. Cambridge, Mass.: Belknap Press.

Bensal, R.F. (1984) Sectionalism in American Political Development, 1880-1980. Madison: University of Wisconsin Press.

Bernstein, R.A. (1989) Elections, Representation, and Congressional Voting Behavior: The Myth of Constituency Control., N.J.: Prentice Hall.

Brady, D.W., and Grofman, B. (1991) "Sectional Differences in Partisan Bias and Electoral Responsiveness in U.S. House Elections, 1850-1980." British Journal of Political Science 21(2): 247-56.

Cain, B.E.; Ferejohn, J.A.; and Fiorina, M.P. (1987) The Personal Vote: Constituency Service and Electoral Independence. Cambridge, Mass.: Harvard University Press.

Congressional Quarterly Guide to U.S. Elections. (1985) 2nd ed. Washington, D.C.: Congressional Quarterly.

De Tocqueville, A. ([1835] 1965) Democracy in America (2 vols.) In Great Books of the Western World. Edited by M.J. Adler. Chicago: Encyclopaedia Britannica.

Dixon, R.G. (1968) Democratic Representation: Reapportionment in Law and Politics. New York: Oxford University Press.

Erickson, S.C. (1993) "A Bulwark Against Faction: James Madison's Case for Term Limits." Policy Review (Winter): 76-78.

Erickson, S.C. (1995) "James Madison, a Term Limits Radical: The Father of the Constitution's Solution to Congressional Gridlock." U.S Term Limits Outlook Series, Washington D.C.

Erikson, R.S. (1971) "The Advantages of Incumbency in Congressional Elections." Polity 3: 395-405.

The Federalist Papers ([1787-88] 1976) Edited by L. DeKoster. Grand Rapids, Mich.: William B. Eerdmans.

Fiorina, M.P. (1977) Congress, Keystone of the Washington Establishment. New Haven, Conn.: Yale University Press.

Grantham, D.W. (1988) The Life and Death of the Solid South. Lexington: University of Kentucky Press.

Harrington, J. ([1656] 1977) The Commonwealth of Oceana and The Prerogative of Popular Government. In The Political Works of James Harrington. Edited by J.G.A. Pocock. Cambridge: Cambridge University Press.

Hennock, F.B. (1955) Testimony before the Subcommittee on Privileges and Elections Regarding the Federal Elections Act of 1955. 84th Cong., 1st sess.

Jacobson, G. (1980) Money in Congressional Elections. New Haven, Conn.: Yale University Press.

Kernell, S. "Toward Understanding Nineteenth Century Congressional Careers." American Journal of Political Science 21: 669-93.

Kleppner, P. (1987) Continuity and Change in Electoral Politics, 1893-1928. New York: Greenwood Press.

Kraditor, A. (1965) The Ideas of the Women's Suffrage Movement, 1890-1920. New York: Columbia University Press.

Lichter, S.R.; Rothman, S.; Lichter, L. (1986) The Media Elite: America's New Power Brokers. Bethesda, Md.: Abler and Abler.

Madison, J. (1962) The Papers of James Madison. Edited by W.T. Hutchinson, W.M.E. Rachal, and R.A. Rutland. Chicago: University of Chicago Press.

Malbin, M. (1980) Unelected Representatives. New York: Basic Books.

Mayhew, D.R. (1974a) "Congressional Elections: The Case of the Vanishing Marginals." Polity 6: 295-317.

Mayhew, D.R. (1974b) Congress: the Electoral Connection. New Haven, Conn.: Yale University Press.

McCormick, R.L. (1981) "The Discovery that Business Corrupts Politics: A Reappraisal of the Origins of Progressivism." American Historical Review 86: 247-74.

McCormick, R.L. (1986) The Party Period and Public Policy: American Politics from the Age of Jackson to the Progressive Era. New York: Oxford University Press.

MacNeil, N. (1963) Forge of Democracy: The House of Representatives. New York: David McKay.

Ornstein, N.J.; Mann, T.E.; and Malbinn, M. (eds.) (1994) Vital Statistics on Congress, 1993-1994. Washington, D.C.: Congressional Quarterly.

Pocock, J.G.A. (1975) The Machiavellian Moment: Florentine Political Thought and the Atlantic Republican Tradition. Princeton, N.J.: Princeton University Press.

Richardson, J.D. (ed.) (1896) A Compilation of the Messages of the Presidents, 10 vols. Washington D.C.: U.S. Government Printing Office.

Richmond Enquirer, 8 November 1822.

Robbins, C. (1968) The Eighteenth Century Commonwealthman. New York: Athenaeum.

Struble, R., Jr. (1980) "House Turnover and the Principle of Rotation." Political Science Quarterly 94 (Winter): 649-67.

Trenchard, J., and Gordon, T. ([1755] 1995) Cato's Letters. Edited by R. Hamowy. Indianapolis, Ind.: Liberty Classics. (The essays were first published in London from 1720 to 1723.)

Wiebe, R. (1967) The Search for Order, 1877-1920. New York: Hill and Wang.

Wiebe, R. (1984) The Opening of American Society: From the Adoption of the Constitution to the Eve of Disunion. New York: Knopf.

Wood, G. (1969) The Creation of the American Republic, 1776-1787. Chapel Hill: University of North Carolina Press.

Wood, G. (1993) The Radicalism of the American Revolution. New York: Vintage Books.

Woodward, V. (1955) The Strange Career of Jim Crow. New York: Oxford University Press.

The author is a former Resident Scholar at U.S. Term Limits and a doctoral candidate in Early American History at the College of William and Mary.

Notes

[f1]A statistical analysis by David W. Brady and Bernard Grofman (1991), who compare district swing ratios over time, supports the notion that safe seats were multiplied around the turn of the century and that progressive reforms were largely responsible. While Brady and Grofman note the effects of Jim Crow in the creation of safe Southern seats, their data also reveals the proliferation of safe seats in the North after the passage of women's suffrage.

[f2]For a collection on much of the best scholarship on the subject, see The United States Congress: The Electoral Connection, 1788-1989 (1991). See also the other volumes in the series for a thorough review of periodical literature on the U.S. Congress.

[f3]If House elections do not accurately reflect the mood of the electorate, then members of Congress may not necessarily represent the beliefs of their constituents either. For more on this, see Robert A. Bernstein (1989).

The Cato Journal is published in the spring/summer, fall, and winter by the Cato Institute, 1000 Massachusetts Ave., NW, Washington, D.C. 20001-5403. The Views expressed by the authors of the articles are their own and are not attributable to the editor, editorial board, or the Cato Institute. Printed copies of the Cato Journal may be ordered by calling 1-800-767-1241. Back issues are also available on the Cato Institute Web site: http://www.cato.org. Email comments or suggestions to subscriptions@cato.org.