Aided by a vigorous press, an energetic civil society, and the determination of whichever party comprises the “loyal opposition” at each level of the system, today’s American is free to choose who holds political power and has at least as much influence on the workings of government as citizens in any other democracy. The process through which citizens compete for political office, and more generally for influence over government actions and the substance of public policy, is relatively transparent and accessible to those who choose to take part.
Yet, as the world famously learned in November and December 2000, the United States has a remarkably decentralized and diverse set of electoral arrangements in which a closely decided contest can lead to political firestorms and diminished public confidence in the integrity of the process. Controversy continues about how the country can compile an accurate registry of eligible voters, ascertain that these persons each vote only once in elections when they desire to participate, and be confident in the announced results. Several major efforts have been undertaken to find solutions consistent with the American federal system. A bipartisan commission led by former President Jimmy Carter and former Secretary of State James Baker produced one of the more compelling proposals as part of a comprehensive set of recommendations in September 2005.1 Yet efforts to improve electoral administration are often stymied by partisans who insist that the main issue is either voter fraud or voter disenfranchisement. Dismayingly, because the most consistent demographic correlation in voting habits is that African Americans tend disproportionately to vote for Democrats, partisan scuffling for advantage in this regard often veers into racial stereotyping and race-baiting that complicates resolution.
Other aspects of the national political process are problematic or controversial. In late 2007, the rules of the game are still being established for the presidential contest that is already well under way, as conflict simmers within the political parties and among the states about the calendar of primary election dates.2 As in much of American society, the proper boundaries between private and public responsibility are difficult to discern and often disputed. (Are the primaries an internal matter for privately organized political parties? Or does the public interest in the large sums of money spent on elections at every level of government, and the interest in fair elections that extends well beyond party members, mean that these are matters of public policy?) The role of money in politics—who gets to spend how much on which aspects of the electoral and governing processes—is hotly debated. The power of incumbency is much studied and analyzed: while some argue that the institutionalization of constituency service by legislators confers an unfair electoral advantage on those already in office, others see the same dynamic as proof that a democratic system makes officials responsive to citizen demands.
The predominance of two political parties that have successfully established barriers to entry for independent candidates and third parties, draw most electoral boundaries for the protection of incumbents on both sides, and control the electoral machinery itself, is also noteworthy. Opinion surveys suggest popular disquiet with the existing system, though voters also tend not to support alternatives when they appear. Moreover, major third-party bids for the presidency, even when relatively well-funded and successful in attracting voters, have generally been highly personalized epiphenomena rather than the products of lasting political organizations. In the current presidential election season, a new effort called Unity08 has brought together prominent campaigners from both the Republican and Democratic establishments who seek to launch a bipartisan ticket, avoiding both the formation of a permanent third party and a focus on any single personality. Using secure internet voting, the project’s backers intend to craft a centrist alternative to what they see as polarizing and unconstructive partisanship.3
More than two centuries after the adoption of the Constitution, the American political system is, in many critical aspects and depending on one’s perspective, either still a work in progress or in need of serious repair. Some points of contention date back several decades (low voter turnout) or even to the adoption of the Constitution (the Electoral College), while others have crystallized in the modern era (campaign finance regulation). No American would say the United States is a perfect democracy, or even that it administers elections as well it should. All three branches of government, every one of the 50 states, and more than 3,100 counties each play a distinct role in defining and implementing the political process in our federal system. This means not only that ours is a broadly inclusive, locally based democracy, but that many, many people and entities have collaborated in the construction of an electoral framework that has lately dismayed America’s admirers, and brought glee to critics of the United States.
Nevertheless, and notwithstanding the flaws discussed below, the American political system basically works. New actors constantly enter the arena, while incumbents and majority parties are turned from office at each election. Public opinion matters, and those citizens who choose to mobilize can affect outcomes very directly. The country is almost continually engaged in debate about how to improve the system.
In the 2006 midterm elections, owing to the corrupt practices of a handful of officeholders and rising opposition to the president’s policies in Iraq, the dominant Republican Party lost its majorities in both chambers of Congress. It remains to be seen, of course, whether the Congress now controlled by Democrats will address either the corruption issue or the war in Iraq in ways that satisfy the recently energized voters. But the election results demonstrated anew that, despite concerns about the impact of gerrymandering, the presumed fund-raising and other advantages of incumbency, and efforts at vote suppression in key areas, the American people can change who governs them when they choose to do so.
Gerrymandering. The practice of gerrymandering, perhaps the most visible means by which incumbents seek to engineer their reelection, is an American political tradition rooted in the system of single-member legislative districts. It dates back at least to 1812, when Governor Elbridge Gerry of Massachusetts, at the behest of Jeffersonian Democrats, approved a district that had been tortured for political purposes into the shape of a salamander. He was promptly excoriated in the press by opponents of the plan, who preserved the governor’s name for posterity by coining the term “gerrymander.” Two centuries later, the growing power of computing enables mapmakers to draw lines with great precision based not only on election statistics, but on data about voters’ history of political contributions, race, education, income level, and other factors that may correlate with voting behavior. Because gerrymandering is designed to increase the number of “wasted votes” among the electorate, as the majority interest packs its opponents’ voters into designated districts, the relative representation of particular groups, whether ideological, ethnic, or economic, can diverge from their actual share of the voting population.
Gerrymandering has become a central feature of the American political system, as state legislators fulfill their constitutional obligation to accommodate changes in the population by adjusting state and federal legislative districts after each decennial census. In the process, both major parties utilize their positions in the state legislatures and gubernatorial mansions to carve out personal and partisan advantage. Even in states where bipartisan commissions have been established to draw districts, they sometimes break down amid partisan standoffs, prompting the courts to step in and replace district maps that would have egregiously violated equal protection clauses of the Constitution and federal legislation.
The self-interested nature of this process, as it occurs in 49 states, is prima facie corrupt. That alternatives are possible is made clear by the practice in the state of Iowa, where redistricting is entrusted to the nonpartisan Legislative Services Bureau with specific instructions not to consider the location of incumbents or party interest. Indeed, in 2002, two of Iowa’s incumbent House members—Jim Leach (R) and Leonard Boswell (D)—had to move their place of residence in order to run in newly configured districts.4 Those looking for proof that gerrymandering achieves its aims typically point to the longevity in office of its practitioners and beneficiaries. Since the 1950s, elections to the U.S. House of Representatives in particular have been marked by an increasingly conspicuous lack of competitiveness. In 2002, only four incumbents in the 435-seat House lost to nonincumbent challengers (though others retired, and four were defeated by rival incumbents in newly reconfigured districts)—the fewest ever in the country’s history. In 2004, only five of the 404 incumbents seeking reelection were defeated at the polls, a reelection rate of 99 percent.5
Nevertheless, American scholars are divided on the impact of redistricting and gerrymandering on incumbency, with some asserting that there is, in fact, little hard evidence that it works as well as popularly believed.6 A major reason for the lack of consensus is the difficultly of separating the effect of redistricting from other incumbency advantages, discussed below, as well as from social and political trends at the national, regional, and local levels. Complicating the issue further—and apparently leading many scholars to support the null hypothesis when, in fact, a small but real effect may exist—is the observation that the effects of redistricting can vary widely depending on the type of gerrymander employed.
An incumbent-protecting gerrymander creates safer seats for current members on both sides of the aisle, so it decreases the number of competitive, marginal districts contested by incumbents in the election cycle immediately following redistricting. In partisan gerrymandering, however, the party in power will often attempt to concentrate opposition supporters into as few districts as possible (and sometimes dilute its own advantage in secure areas) in order to increase its overall majority. Depending on how far this process is taken, it can have the effect of increasing the number of marginal seats. Moreover, redistricting plans are frequently hybrids combining some aspects of each kind of gerrymander. Many argue that, in the aggregate, the decrease in electoral competition created by incumbency-protection gerrymandering is balanced by the increase in competition resulting from partisan gerrymandering.7
Calculating incumbency advantage. Before addressing the other tools available to incumbents, it is worthwhile to note the difficulty of producing a precise calculation of their overall electoral edge. As noted above, recent House reelection figures are fairly impressive. Over the last five election cycles, incumbents have been returned to office an average of 97 percent of the time. Even in the Democrats’ sweeping victory of 2006, House members running for reelection managed to win 94.5 percent of their races. Similarly, an examination of the proportion of congressional elections over the last 50 years that have been decided by less than 20 percentage points provides a panoramic view of the rising level of stagnation in congressional elections. From 1960 to 1968, House incumbents won by at least a 20-point margin, 64 percent of the time.8 Then, during the 1970s, they won by that margin 73 percent of the time. From 1980 to 1988 the trend continued, with incumbents winning handily in 79 percent of all districts in which they ran for reelection. Between 1990 and 1998, the tide temporarily reversed, as only 71 percent of incumbents won reelection by more than 20 percent.9 But the elections since the turn of the millennium have been the most stagnant of any decade since Reconstruction. Over the last four election cycles, 83 percent of House incumbents won by more than 20 points. Again, in the tumultuous 2006 elections, 79 percent of those seeking reelection were returned via landslide margins of 20 points or more.
Nonetheless, incumbency reelection rates are an incomplete measure of incumbency advantage. They assume that incumbents and challengers are essentially equal, as if selected at random, when they are not. Incumbents have already won an election and are thus, ipso facto, better-than-average campaigners. Furthermore, challengers who choose—or are chosen—to run in districts that are deemed safe for the incumbent party or lawmaker are typically not as competitive as the average candidate. And incumbents who perceive that they are unlikely to be reelected, due to personal problems or broader political trends, often retire strategically. Thus a discussion focused mainly on reelection rates tends to overstate the incumbent’s entrenchment.
Among the first scholars to address this problem and present a more nuanced estimate of incumbency advantage was Robert Erikson, who in 1971 measured its magnitude as the difference between a candidate’s success in an initial electoral win and that of subsequent victories. His measure also controls for redistricting, partisan electoral tides, and the effect of running against an incumbent—for members who did not enter Congress through an open seat. Erikson found that, on average, incumbency in the 1960s was worth about two percentage points on election day.10 Recent studies, using essentially the same methodology, have shown that incumbency advantage has grown drastically since 1960 and place its current value at between 8 and 11 percentage points.11
Professionalization and earmarking. Some analysts attribute the longevity of lawmakers in office to the growing professionalization of the nation’s legislatures during the 20th century.12 On the one hand, professionalization increases legislative effectiveness, leads to higher productivity, expands membership diversity, and brings in savvier, more adept representatives.13 On the other hand, over the last 60 years, career-minded legislators have developed tools and provided themselves with resources that both raise the barrier for entry and serve them well in reelection bids.
A popular perception of congressional dynamics juxtaposes the major political parties in fierce battle with one another, locked in a zero-sum game with the majority attempting to push their partisan agenda through, while the minority works diligently to stymie such efforts. Scholars of congressional organization, though, offer a very different picture: a more decentralized legislature with relatively weak parties, where members seek seats on committees that have jurisdiction over the portion of the budget they believe is most important to the localities they represent.14 Members then engage in a positive-sum game: they use their committee positions to appropriate funds for projects that benefit a particular interest or visibly bring federal resources into their district or state, and then trade votes in order to secure funding for their pet projects.15
By catering in this manner to diverse—and sometimes deserving—special interests, members create powerful allies and potential campaign contributors, which provide them with financial and organizational advantages in subsequent elections. Given the electoral benefits of this distributive brand of politics, it is little surprise that pork-barrel projects, in the form of “earmarks” attached to appropriations bills, are both quite common and on the rise. According to the Congressional Research Service, in 1994 members of Congress allocated $27 billion on 4,202 pet projects; in 2005, they set aside $52.1 billion for 16,050 such projects.16
The 2006 election season was roiled by unfolding news about the outsized influence that lobbyist Jack Abramoff seemed to have on the work of Congress and the practice of earmarking. (Abramoff, Representative Robert Ney of Ohio, and at least 10 others have since been convicted of criminal charges in this context.17) In an effort to increase the transparency of these spending decisions and decrease the influence of hired lobbyists on Washington lawmakers, the Congress elected that year passed an ethics reform bill entitled the Honest Leadership and Open Government Act of 2007. The new law prohibits members of Congress and their staffs from accepting gifts, free meals, and travel, and extends the time a former member must wait before becoming a lobbyist. In addition, the bill requires the disclosure of, among other things, the legislative author of all earmarks for federal projects in appropriations bills, and an accounting of any earmarks contained in bills and conference reports before they are voted on. Critics of the legislation point out that while it does restrict the use of earmarks, it also creates a number of loopholes, for instance by increasing—from one or two to 40—the number of senators it takes to prevent provisions not passed by the House or the Senate from being inserted into legislation. That would make it easier, not harder, to enact special interest measures without public scrutiny.18
While earmarks have been popularly described as a means for lawmakers to divert taxpayer dollars into wasteful pork-barrel projects of limited public utility, some elected officials have spoken up in favor of the practice. Democratic Representative Rahm Emanuel of Illinois, for instance, noted that while “some members of Congress, on both sides of the aisle, eschew earmarks…most members believe it is their prerogative and their duty to channel federal resources to important public purposes.”19
Nearly every president in the past century has sought to obtain the authority of a line-item veto as a tool for controlling supposedly wasteful spending added to bills by lawmakers. Most governors have similar authority over state spending. In a curious triangulation, Democratic President Bill Clinton collaborated with congressional Republicans to enact the Line Item Veto Act of 1996, and Clinton used the new power 82 times before it was declared unconstitutional in a 6-to-3 ruling by the Supreme Court in 1998.20
Constituency service and district ties. Constituency service is another means by which members of Congress can improve their image within their districts and create an electoral advantage over challengers. Although helping constituents to obtain veterans’ benefits, Social Security checks, and civil service pensions, or to navigate some other facet of the federal bureaucracy, is a legitimate and important function performed by representatives, it is also a significant electoral advantage, as it allows members to stand apart from their party, and from Congress as a whole, at election time.21 In contrast to Canada or the United Kingdom, where party organizations play a stronger role in politics, a large and growing proportion of U.S. representatives’ electoral support is attributed to their personal qualities and local activities, making it less susceptible to changing partisan tides and negative views of Congress.22 Not surprisingly, opinion polls consistently find that voters rate their own representatives much higher than they rate Congress.23 To more effectively reap the electoral benefits of casework, members maintain increasingly well-staffed offices in their districts. Some may wonder whether there are more efficient, systematic fixes to the bureaucratic problems these offices seek to address for constituents, but the current system in which elected officials’ serve as brokers for federal benefits clearly is seen to be worth continuing.
In addition, federal legislators have allocated to themselves resources that enable them to maintain visibility in their districts throughout the year. From 1962 to 1997, for instance, members gradually raised the number of annual trips they could take to their districts at public expense elevenfold, from 3 to 33.24 Of course, it must be noted on behalf of these perquisites that frequent visits also increase legislators’ ability to understand their constituents’ concerns and advocate better for them in Washington.
Included among each representative’s personal communications expenditure is the franking privilege, which allows members of Congress to use the U.S. Postal Service for all “official business,” at no charge to them personally or to their campaign organizations. Although the privilege has been around in one form or another since 1789, in 1968 the standards for what constitutes official business were relaxed considerably, and representatives began blanketing their districts with literature that was likely to assist them in their reelection efforts. Yet members’ use of the mail is on the wane in today’s Congress, as the use of more advanced technologies increases sharply. Incumbents, for example, now cultivate huge taxpayer-funded databases with detailed information about their constituents’ interests and interactions with the federal bureaucracy, which they can use to send out issue-specific e-mails to well-targeted groups of voters within their districts.25
Term limits. Concern about the permanent hold that some politicians, especially legislators, seem to have on their jobs has given rise to a civic movement in favor of legal limits on the number of terms they may serve. Some jurisdictions have always required the regular rotation of officeholders, and 36 of the 50 states have limits on the ability of governors to be reelected, usually restricting them to one or two terms. At the presidential level, a two-term tradition launched by George Washington, who declined to stand for a third term, was broken only by Franklin D. Roosevelt, who was elected four times. The presidential two-term limit became mandatory under a 1951 constitutional amendment. In the 1980s and early 1990s, a private organization called U.S. Term Limits Inc. persuaded a large number of candidates for state and federal office to pledge that they would only serve three terms in a lower house or two terms in a senate, and the Republican Party’s “Contract with America,” the platform for their successful takeover of Congress in the 1994 elections, included term-limit provisions. In 23 states, measures were enacted to limit federal legislators’ terms in office, though in 1995 the Supreme Court ruled that states could not impose limits on the terms of federal lawmakers.26 That same year, the Republican leadership in the House brought a constitutional amendment on the subject to the full chamber. It secured a majority, but not the two-thirds required to pass.
In 2007, political action was under way in California, Pennsylvania, and Missouri to trim or eliminate existing term limits on state legislators. However, U.S. Term Limits Inc. remains active in supporting initiatives, filing related lawsuits, and publicizing the issue. Opponents of term limits argue that they are an unnecessary and unreasonable restriction on the rights of voters and candidates alike, insisting that voters should be able to decide for themselves whether to back an incumbent or one of the alternatives on the ballot.
Enhanced Fairness in Redistricting and Reapportionment
One man, one vote. Notwithstanding the real and imagined advantages that decision makers seek to extract in the course of redrawing district boundaries, the American system has seen a clear improvement in many dimensions of fairness over several decades, often due to the intervention of the courts. Since the Supreme Court handed down Baker v. Carr in 1962, the level of both partisan bias and malapportionment within the electoral framework has decreased markedly, at both the state and national level. Prior to 1964, when, in Reynolds v. Sims, the Supreme Court applied its “one man, one vote” principle nationwide,27 legislative districts in many states were seriously malapportioned, and a strong bias favoring rural over urban areas prevailed. In the 1962 Alabama state legislature, for example, a majority of state senators represented just 25 percent of all Alabamans, and the rural county of Bullock (population 13,500) had two House seats while Mobile (population 314,000) had just one.28 Similarly, based on the 1960 census, the most populous district in Vermont’s General Assembly had 33,000 residents, while the least populated district contained only 238.29 Other egregious examples include Connecticut, where the population ratio between the most and least populated districts was 424 to 1; California, where the ratio in the state Senate was 428 to 1; and Utah, with a 196 to 1 ratio in the state House. In the U.S. House of Representatives, the level of malapportionment was much less extreme, but ratios of 2 to 1 were not uncommon.
The redistricting revolution that followed Baker v. Carr and other decisions in the early 1960s shifted the balance of power from rural to urban and suburban voters, and greatly reduced malapportionment at both the state and federal level.30 Subsequent court decisions outlined more specific standards and have worked to further reduce malapportionment. By the conclusion of the 2002 round of redistricting, the deviation between the most and least populous U.S. House district of any state was less than two-thirds of one percent. In short, Supreme Court decisions and the redistricting efforts of the 1960s, 1970s, and 1980s have nearly eradicated malapportionment and, as a result, greatly increased representative equality in the United States. A recent study by leading political scientists concludes that population equalization altered the flow of state funds, diverting approximately $7 billion annually from formerly overrepresented to formerly underrepresented counties.31 The authors concluded that “as long as people have equal representation, they will get a fair share of public expenditures.”32
Declining partisan bias. By examining the effect of change in the proportion of the vote received by each party on change in their respective shares of House seats, scholars have estimated a pro-Republican bias, in non-Southern states, of about six percentage points as of 1962 (meaning that with 50 percent of the vote for the two main parties, the Republican Party could expect about 56 percent of the seats). After the wave of redistricting in 1966, this bias all but disappeared.33 Southern states, it should be noted, are almost universally omitted from redistricting studies of this period since, at the time, the Democratic Party utterly dominated the vote in the South, general elections there were rarely competitive, and turnout was much lower than in the North. However, once the Republican Party established a presence in the South, evidence of a Democratic partisan bias became clear. Since then, pro-Republican gerrymanders have largely corrected this advantage, though at the state level some have significantly overcompensated. Moreover, the net effect of partisan gerrymandering since 1966 is generally believed to be small and, on the whole, has been found to reduce rather than exaggerate overall partisan biases in the American political system.34
Analyses of the individual redistricting cycles are consistent in finding minimal or no partisan effects, but due to varying methodologies, they often differ as to which party benefited when they do find effects. In examining the 1970 redistricting efforts, one study argues that the status quo was maintained, producing no noticeable partisan advantage, while another found a small pro-Republican bias.35 Similarly, in 1980, one study identifies a pro-Democratic bias, and a second shows a small pro-Republican advantage.36 The 1990 cycle is thought to have either benefited only incumbents or evinced a latent pro-Republican advantage.37 Finally, the available literature on the 2000 redistricting cycle shows no real partisan consequence. While the 2003 out-of-cycle redistricting efforts in Texas and Georgia had substantial pro-Republican effects, they both seem to have worked to offset large preexisting pro-Democratic bias in those states. When the Supreme Court reviewed the constitutionality of the Texas redistricting plan famously orchestrated by former Representative Tom DeLay, in League of United Latin American Citizens et al. v. Perry et al., it confirmed that gerrymandering for partisan purposes is not unconstitutional. With some minor adjustments to correct for violations of the race requirements established in the Voting Rights Act of 1964, the Texas plan was left intact.
Clearly, partisan redistricting has the potential to create a significant and sustainable bias within the political system and overrepresent the interest of one party at the expense of the other. However, since 1966 the most effective partisan gerrymanders have typically moved the state in question from a severe bias favoring one party to a slight bias favoring the other, and collectively, the redistricting efforts of the last 40 years seem to have worked to reduce the overall level of partisan bias within the political system.38
Counting prisoners and citizens overseas. The census data that Congress uses to reapportion seats and that most state legislatures use to redraw district lines is often quite controversial, since nominally technical or administrative determinations have obvious and immediate political impact. The census includes children, noncitizens, institutionalized people, and other groups that are not permitted to vote. It also counts federal employees and military personnel who are overseas, adding them to the totals for their home states and districts, but excludes nongovernmental workers living abroad.
One of the more contentious outcomes of the 2000 census was the allocation of the 435th House seat to North Carolina, which edged out Utah by just 857 residents. In an unsuccessful suit brought in federal court, Utah argued that Mormon missionaries serving overseas should be treated no differently than federal employees working abroad.39 In fact, if Mormons on missions overseas were counted, Utah, with 11,176, would have been awarded the extra seat, since North Carolina had only 107.40 Alternatively, if all residents living overseas were excluded for purposes of reapportionment, Utah, with just 3,545 residents working abroad as federal employees, still would have been awarded the seat, given that North Carolina had 18,360 residents in that category.
While in this instance the effect on the overall equality and level of representation in the United States is quite small, the extra seat would have increased Utah’s modest influence in the House by 25 percent and its impact on presidential elections by about 18 percent. Furthermore, to the extent that the interests of the 4.1 million non–federally employed American citizens currently residing abroad are different from those who are counted in the census, not counting them distorts the substantive representativeness of American democracy to some degree.41
A related concern is the Census Bureau practice of counting prison inmates where they are incarcerated rather than in the inmate’s previous place of residence. A disproportionate number of state prisons are located in sparsely populated areas, while the two million inmates they house are predominately from densely populated cities in other electoral districts. Because inmates are not permitted to vote (in all but two states), the ballots cast by citizens of rural districts with large prison populations are effectively given more weight than those of voters in other districts. Fifteen percent of Montana’s 85th state House district, for instance, are incarcerated. One county in Louisiana, a second in Florida, and a third in Texas get more than 30 percent of their population totals from prisons.42
In New York, 91 percent of the state’s prisons are located upstate, but 66 percent of the inmates are from New York City.43 The seven New York State Senate districts with the largest prison populations are rural upstate districts represented by conservative senators, who vote quite differently than those from the city. This pro-rural bias is further exacerbated by the fact that New York City is systematically underrepresented in the Senate, with each of the 29 districts in its five boroughs containing between 1.69 percent and 4.83 percent more individuals than average. If the state’s prison population were excluded, each of the seven rural upstate districts noted above would have between 5.8 and 6.8 percent fewer people than the average district, and all seven of the districts in the Democratic-dominated borough of Queens would have 4.1 percent more people than average. That would raise the deviation between the most and least populous districts beyond the 10 percent limit set by the Supreme Court. It should be noted that in the New York State Assembly, the lower house of the legislature, the situation is somewhat reversed. New York City is overrepresented by just under one percent; if the prison population is omitted, this bias drops to one-half of one percent.
The Electoral College
Another institution that affects the representative character of American government is the presidential Electoral College, adopted in 1789 as part of the grand federal bargain that is the U.S. Constitution. The number of electors in each state is equivalent to the number of representatives a state has in both the Senate and the House of Representatives. As a result, the states with smaller populations receive added weight, and the 10 largest states have 35 fewer electors than they would if the nation’s 538 electoral votes were distributed purely on the basis of population; California alone has 10 fewer electoral votes than it might. In addition, to maximize their influence over the selection of the president, states have an interest in maintaining a winner-take-all system, allocating all their electoral votes to a single candidate; 48 states currently do so. This creates a structural bias that also heightens the influence of “swing states,” where party preferences are about evenly divided, and whose Electoral College votes would essentially cancel one another out if they were allocated proportionately to each candidate.
Four times in U.S. presidential election history (1824, 1876, 1888, and 2000), the winner of the popular vote has failed to win the White House. There have only been 54 presidential elections to date. Therefore, almost 8 percent of the time, the Electoral College system has led to the election of a candidate other than the one who received the most votes nationwide—a modest, but by no means insignificant, proportion. However, candidate and voter behavior under a system without the Electoral College would likely be different in ways that are not entirely predictable. For instance, the perceived closeness of the presidential contest in a particular state currently has a significant impact on voter turnout in that state.44 It is unclear how turnout would be affected if every vote in the country were counted equally.
Even when the structural biases of the Electoral College are not sufficiently significant to change the outcome of the election, they may still have considerable impact on substantive representation. Election-minded first-term presidents, for instance, are much more likely to be responsive to the interests of voters in swing states than those in noncompetitive states. The Electoral College endures, however, because it is generally understood to be a part of the federal inheritance, in which states are seen as historically sovereign entities rather than mere administrative divisions. Proponents, moreover, often argue that the Electoral College obliges candidates to marshal popular support in multiple parts of the country, preventing the victory of those who rely on the exploitation of regional interests or prejudices. The system also discourages third candidates, which, all else being equal, results in more stable, centrist policies than alternative frameworks.
Voting Administration and Equal Access
A major issue of concern is equality of access to and influence over the political process. Despite universal suffrage and the great strides made in protecting individual voting rights, not all citizens in today’s America have an equal voice with which to influence government decision-making. The problem starts with access to the ballot. In recent election cycles, observers reported what are, at best, inconsistent practices regarding the purging of voter rolls and voter registration. The updating of voter lists and the vetting of incoming registration cards are necessary to maintain the integrity and manageability of the electoral process. But the improper purging of voter rolls, inequitable distribution of voting machinery, unreasonable rejection of provisional ballots, and unlawful voter intimidation tactics effectively suppress voter turnout at critical times and places—and do so disproportionately along ethnic, partisan, and socioeconomic lines.
For nearly 100 years after the constitutional enfranchisement of all African Americans and the extension of equal protection under the law to all citizens, the voting power of ethnic minorities throughout the country, and particularly in the South, was suppressed by aggressive intimidation tactics, the widespread use of violence, unconstitutional voting rules, and the lack of effective means of recourse. The passage of the Voting Rights Act in 1965 provided the federal government additional tools with which to enforce the rights guaranteed under the 14th and 15th amendments, and specifically outlawed many voter suppression practices. The statute required large-scale Justice Department monitoring of elections in states with a poor history of ensuring voting rights, and paved the way for legal challenges against those who continued disenfranchisement efforts.
From 1964 to 1974, in the seven Southern states covered by Section 5 of the Voting Rights Act, registration of black voters increased from 29 to over 56 percent, and the disparity between white and black registration decreased from 44.1 to 11.2 percent.45 Over the last 30 years the situation has continued to improve. In Louisiana and Mississippi, the registration rate among blacks today exceeds that of whites. Problems persist nonetheless. In modern presidential elections, about 2 percent of all ballots are spoiled (at the low end of the range in well-established democracies),46 and hundreds of thousands of additional voters are not permitted to cast their vote on election day due to problems in the registration process.47 Thousands more are deliberately targeted in efforts to suppress their vote. The effect on election outcomes is often minute, but the voters in question are disproportionately minority and urban, meaning their collective political voice is muffled and their access to the political process is reduced relative to white, nonurban voters. Because African Americans tend overwhelmingly to vote Democratic when they do vote—and because problems in obtaining government-issued photo identification, rates of arrest and imprisonment, and poverty tend to affect African Americans disproportionately—the purely partisan interests of Republicans in affecting election outcomes assume an unfortunate racial character.
“Majority-minority” districts spawned by the Voting Rights Acts of 1965 and 1982 have added to the presence of minorities in legislatures, as boundaries have been drawn to concentrate minority voters. The 1992 elections, the first based on lines drawn after the 1982 law was enacted, led to a jump in the number of minorities in the House of Representatives, from 38 to 58.48 While this has clearly been beneficial to minorities (and the Democratic Party) at one level of analysis, subsequent studies have raised the question of whether broader minority interests have truly been enhanced, since the districts created have left adjoining districts bereft of the reliable Democratic voters that minority candidates often represent.49
Even after the legislative reforms of the past 50 years, some ethnic minorities, particularly African Americans, continue to face voter intimidation and disenfranchisement efforts, although the methods are often less crude than in the past. In New Jersey in 1981, the Republican National Committee (RNC) set up “ballot security” task forces, exclusively in precincts with a majority of ethnic minority voters, to purge voter rolls of citizens whose mail was returned as undeliverable. It then deployed hired armed guards to polling places on election day. In response, a U.S. District Court issued a ban on efforts to target specific groups for disenfranchisement or intimidation, even if the means are otherwise legitimate.50
Despite the ban, an internal party document indicated that the RNC launched a similar effort in 1986 in order to “keep the black vote down” in Louisiana, Georgia, Missouri, Pennsylvania, Michigan, and Indiana;51 the RNC worked to have 31,000 mostly black voters purged from registration rolls in Louisiana alone.52 Likewise, in 1988 armed guards were placed at polling places in heavily Hispanic precincts in Orange County, California; in 1990, 125,000 postcards were mailed primarily to black North Carolinians, intentionally misinforming them about residency requirements;53 and intimidating mailers, fliers, and signs threatening jail time and deportation for those who did not follow registration requirements were frequently deployed in minority neighborhoods in New York City, Texas, and the Carolinas during the 1990s.54
Following the 2000 presidential election in Florida, there were similar allegations of intimidation and improper purging of eligible voters from registration rolls in heavily African American precincts. That election also brought to light the unequal allocation of voting machinery, and its disparate impact on black voters. Throughout the state, 4 percent of all punch-card ballots were excluded, compared with 1.4 percent of optical-scan ballots.55 Punch-card machines are located disproportionately in African American areas, in Florida and elsewhere.
In response to concerns that the allocation of voting machines and the practice of purging voter registration lists effectively dilute the vote of ethnic minorities, Congress passed the Help America Vote Act (HAVA) in October 2002. The legislation allows voters who have been removed from voter registration lists to cast provisional ballots that may be counted at a later time if they were in fact improperly purged. It also mandates minimum standards for election machinery. Overall, HAVA seems to have significantly reduced the total number of spoiled ballots nationwide. Charles Stewart III, an expert on voting behavior and equipment at the Massachusetts Institute of Technology, estimates that HAVA saved about one million votes across the country in 2004.56
Nevertheless, voters faced familiar problems at the polls that year. A court order to make public a purge list of 47,000 “potential felons” in Florida revealed that thousands of those listed were actually eligible to vote. It appeared that the list, which consisted overwhelmingly of African Americans, had been deliberately cleared of Hispanics, who tend to vote Republican in Florida.57 The list was scrapped nine days after it was made public. The most extensive reports of vote suppression in 2004 were in Ohio. At the center of most of the electoral controversies was Ohio Secretary of State Ken Blackwell, whose position as cochairman of the Committee to Reelect George W. Bush in the state led many to allege a conflict of interest. Blackwell was ultimately named in 16 election-related lawsuits.
In the months leading up to the 2004 presidential election, there was an unprecedented surge in the registration of new voters in Ohio, which was thought to favor the Democratic Party. In a county-by-county analysis, the New York Times found that new registrations between January and September were 250 percent higher than in 2000 in areas that typically vote Democratic, but only 25 percent higher in Republican neighborhoods.58 During this time, Blackwell took several steps to invalidate new voter registrations, purge names from the existing rolls, and establish more scrupulous standards for the state to follow when considering the validity of provisional ballots. Certainly, some such measures are necessary to eliminate erroneous voter registration applications and remove former residents from the current roll, but many in this case were clearly intended to achieve partisan electoral advantage.
For example, citing an arcane Ohio law, Blackwell for three weeks in September instructed county election boards to reject any registration applications submitted on paper thinner than 80-pound stock (most commonly used for postcards). This excluded standard 20-pound copier paper, used by most Ohioans who printed out their applications from the secretary of state’s website.59 In a separate effort to purge registration rolls of likely Democrats, the Ohio Republican Party unlawfully targeted hundreds of thousands of predominantly minority voters by sending registered mail to their given addresses. If the letters were not signed and returned, or were returned as undeliverable, those registrations were challenged.60
Most of the complaints from Ohio concerned the misallocation of voting machines, resulting in four- to ten-hour lines outside dozens of polling places in Cleveland, Columbus, and Cincinnati. Many voters were deterred by the long wait and did not cast ballots. Part of the problem was an overall shortage of voting machines in some parts of the state. Franklin County, for instance, received only 2,866 machines, despite the fact that the Election Board’s own analysis indicated that it needed 5,000.61There was also a bias in the allocation of machines within counties that worked against precincts with predominately minority and lower-income populations. House Judiciary Committee staff members received reports that, adjacent to two heavily African American precincts in Columbus with five- and seven-hour waits, there was a suburban district with just 184 voters per machine and less than a 15-minute wait.62 In the whole of Franklin County, where Columbus is located, there were 262 registered voters per machine in precincts with low proportions of African Americans and 324 registered voters per machine in precincts with a high proportion of African Americans, a difference of 23.7 percent; if calculated using the Election Board’s measure of “active voters,” the disparity is still 13.6 percent.63
With control of both houses of Congress at stake in the 2006 midterm elections, several states passed legislation that significantly hindered voter registration drives. In Ohio, the state legislature passed new laws requiring individual registration workers to personally submit completed forms to an election office rather than allowing the organization for which they work to submit them in bulk.64 The new laws hold the workers criminally liable for any irregularities in the forms. In response to similar laws passed in Florida, the League of Women Voters halted their voter registration efforts in that state for the first time in 67 years.65 However, a federal judge blocked enforcement of the Florida law, which imposed fines of between $250 and $5,000 for the “mishandling” of voter registration cards.
In March 2006, Georgia enacted a law requiring voters to show a passport, driver’s license, or other official form of identification before casting a ballot. In contrast to past elections, for which any of 17 different types of document could be used as proof of identity and residency, the new state law would require Georgians lacking government-issued IDs to purchase a $20 state ID card before voting. A federal judge struck down the law, likening the requirement to a Jim Crow–era poll tax.66 A subsequent version of the law eliminated the $20 fee; the Supreme Court is currently considering the validity of a similar measure in Indiana.
According to press reports in April 2007, a leaked bipartisan study commissioned by the Election Assistance Commission, a government panel created under HAVA, concluded in 2006 that “there is widespread but not unanimous agreement that there is little polling place fraud.” Republicans generally insist that voter fraud is widespread and justifies the voter identification laws that have been passed in at least two dozen states. Democrats, on the other hand, say the threat is overstated and have opposed voter identification laws on the grounds that they tend to disenfranchise the poor, members of minority groups, and the elderly, who are less likely to have photo IDs and are more likely to be Democrats.67
It is a widely held view that money in the American political system is pervasive and problematic, from the consuming race for campaign contributions to the special political access enjoyed by lobbyists and donors, and that contributions and spending associated with politics need to be regulated closely. Yet an equally compelling argument can be made that the amount of money raised and spent in American politics is by itself a red herring. The more important concern is whether political contributions, or services provided to candidates and officeholders, oblige the recipients to depart from the proper performance of their duties.
New policymaking in this area is rare. Indeed, when the Bipartisan Campaign Reform Act, also known as McCain-Feingold (or Shays-Meehan), was enacted in November 2002, it was the first major reform of campaign finance law in 28 years. There is widespread unease about the central role that fund-raising plays in the life of elected officials, and concern about the fairness of citizen access to officeholders for those without the means to contribute. But finding a regulatory framework consistent with the Constitution’s First Amendment guarantees of free speech has proven difficult. The columnist George F. Will, in keeping with America’s libertarian tradition, regularly decries the “attacks on free speech” he sees in campaign finance regulation.68 The Supreme Court, which is currently considered conservative, has acknowledged the validity of viewing campaign donations as a form of free speech, but in a 2006 decision overturning a Vermont law that had imposed a $400 limit on contributions to candidates for statewide office, the court upheld the established principle of permitting contribution limits in the name of thwarting corruption, or the appearance thereof.
McCain-Feingold bans soft money contributions to national political parties, but permits up to $10,000 in soft money contributions to state and local parties. “Soft money” refers to contributions to parties and other political organizations that is not directed to any specific candidate or campaign, meaning it is not subject to older restrictions. The 2002 law also stops so-called issue advertisements, which advocate positions on specific topics, from targeting specific candidates. The ads have been seen as an indirect way of aiding certain campaigns without running afoul of contribution and spending limits. Whether that provision accomplished its main purpose, to curtail the activities of tax-exempt advocacy groups (dubbed 527s after the provision in the tax code that covers them), is disputed.69 Another portion of McCain-Feingold raised the individual contribution limit from $1,000 to $2,000 per election for House and Senate candidates (the figure increased in the 2008 cycle to $2,300 for each candidate in each federal election for individual donors). A “Millionaire’s Amendment” to the law increases the contribution limits for candidates facing wealthy opponents, whose personal spending on their own campaigns remains unrestricted.70 Many advocates assert that the solution to problems associated with fund-raising is mandatory public financing of federal campaigns, especially since the viability of the current system of voluntary, partial public funding of presidential elections appears to have all but collapsed. The top tier of candidates from both parties in 2004 chose to forgo the public financing that was available to them in order to free themselves from the spending limits it entailed, and the same pattern is emerging in the 2008 race. Public financing has become the redoubt of troubled campaigns that are unable to raise funds to compete head-to-head with the front-runners.71 As the costs of airtime on television and radio is often the largest expense associated with campaigning, some have proposed that the airwaves be harnessed to provide equal access to candidates as a public service.72
Voting Rights for the District of Columbia
The 585,000 residents of the District of Columbia do not enjoy the same political rights as other U.S. citizens. There are also a number of small U.S. island territories that have limited access to the federal political process, including American Samoa,Guam, , and the U.S. Virgin Islands. Puerto Rico is the largest, though its citizens have consistently voted to retain the island’s current intermediary status as a commonwealth rather than seek full statehood or independence. As U.S. territories that are not states, these five jurisdictions fall into a constitutional lacuna that leaves their residents with many of the obligations of citizenship but without a say in the election of full voting members of Congress. Their elected delegates have long been permitted to participate in the deliberations of the House and to vote in House committees. In 1993 and again in 2007, Democratic majorities in the House adopted a procedural rule that permits the five delegates to vote with the full House, but their votes count only if they have no effect on a measure’s ultimate outcome. None of the five territories are represented in the Senate.Puerto Rico
The District is noteworthy in this group because it is the seat of the national government and because, unlike Puerto Rico, its citizens do not control their own destiny. Unfortunately, the intertwined issues of race and partisanship that arise in other aspects of the American political process are evident in this issue as well. Since the population of the District is mostly African American and overwhelmingly Democratic in its political preferences, consideration of how to provide equitable political representation to these citizens is often hampered by partisan wrangling.73 Fortuitously, the grievance felt by predominantly Republican Utah, which narrowly missed receiving the 435th seat in the House of Representative following the 2000 census, as discussed earlier, provided the basis for a compromise in 2007 that would have extended full voting rights to a representative from the District. The proposal would have enlarged the House of Representatives to 437 and given one seat each to the District and Utah. However, the measure was blocked in the Senate by the threat of a filibuster and appears to have died for the time being.74
In 1961, the 23rd Amendment to the U.S. Constitution was ratified, granting the District electors in the Electoral College. District residents have thus participated in presidential elections since 1964. In 1978, Congress passed another constitutional amendment that would have given the District its own voting members of Congress, making it virtually a state. However, the amendment was given a seven-year time limit for ratification by the states, and only 16 states gave their approval in that time, far short of the three-quarters (currently 38) required.
District sympathizers have tried to craft solutions that would pass constitutional muster and also avoid inflaming partisan or other passions. Congressman Ralph Regula, a Republican from Ohio, believes that the time is long past to fix the “appalling breach in our democracy” that deprives District residents of voting representation in the House and Senate. His solution is to shrink the District down to a tiny enclave of federal buildings and return the office and residential areas to Maryland, of which the territory was originally a part. This notion is often referred to as retrocession. “Retrocession would allow D.C. to use Maryland state facilities and all other state-funded institutions,” Regula has said. “The District could also finally be run like any other major city in the U.S.”75
There are enduring problems of inequity in the U.S. political system that lead to diminished access for certain groups of voters, frequently the poor and people of color, to full participation in the political process. Efforts to correct these injustices are ongoing, though the federal nature of the American system sometimes disperses decision-making and responsibility and makes resolution more difficult.
Longevity in office and apparently uncompetitive elections are an affront to a significant minority, abetted by gerrymandering, the capture of public resources, and other advantages of incumbency. Yet, as has been discussed, the nature of the problem is frequently misstated, and the potential virtues of long incumbency in the legislature are often overlooked. For instance, a secure district enables a lawmaker to challenge the president, or leaders of his own party, on important issues, and lengthy service allows legislators to develop areas of expertise that enhance the quality of their oversight of the executive branch.
Debate about regulatory limits on contributions to and spending by candidates will surely continue. The corruption of public officials, in which they take action—whether spending public money, enacting laws or regulations, or wielding some other form of influence—in return for personal payments or campaign donations, occurs regularly and in both parties, but does not describe the behavior of the vast majority of elected officials in America. Indeed, many incur financial hardship or forgo other opportunities in order to serve in public office.
Today’s American is able to choose the country’s leaders in increasingly fair processes. U.S. citizens have demonstrated again and again that they can oust incumbent officials at all levels whose performance is not satisfactory. Despite its flaws, the system fundamentally works, providing a substantial measure of democratic accountability and regular rotation in government.