Cato Policy Analysis No. 279 August 20, 1997

Policy Analysis

The End of Representation:
How Congress Stifles Electoral Competition

by Eric O'Keefe and Aaron Steelman

Eric O'Keefe is treasurer of U.S. Term Limits. Aaron Steelman is staff writer at the Cato Institute.

Executive Summary

For the past several decades, the reelection rate for members of the House of Representatives has hovered around 90 percent. In 1996 it exceeded 94 percent. The reasons that incumbents have enjoyed such tremendous electoral success are numerous.

House members frequently engage in pork-barrel politics. Many bring home pet projects that benefit their constituents and, as a consequence, their own chances for reelection. In addition, representatives use myriad taxpayer-funded perquisites to spread their messages and enhance their visibility. Incumbents also are able to help their constituents with various problems that they may encounter with the federal bureaucracy. Such help, known as constituent service, is made necessary by the increasing size of government and can significantly increase an incumbent's approval rating. Moreover, current campaign laws restricting the amount of money that a candidate can raise from an individual or group deter many potential challengers and greatly reduce the electoral chances of those who decide to run. Proposals that would regulate campaign finance even more would only further entrench incumbents.

To counter the many electoral advantages that incumbents now enjoy, a number of reforms could be implemented. Reducing the size of government would shrink the opportunities and necessity for constituent service. Eliminating campaign contribution limits would enable more candidates, particularly those with views outside the mainstream, to wage viable campaigns. Most of all, imposing term limits on members of Congress would ensure that party leaders and committee chairmen would not become part of a permanent ruling class. Since current members of Congress have little electoral incentive to pass such measures, a national movement to reject those seeking careers in Congress will be required.

Congressional Reelection Rates

In 1994 the Republican Party gained control of the House of Representatives for the first time in 40 years. One might think that incumbents seeking reelection that year were extremely vulnerable. However, a closer look shows that such was not the case. In 1994, 387 members of the House sought reelection. Only 4 were defeated in primaries and 34 in the general election, which translates into a 90 percent reelection rate in the most anti-incumbent election year in memory. [1] Moreover, most of those victories were achieved by fairly large margins. On average, incumbents won approximately 64 percent of the vote. [2]

The 1996 congressional elections yielded similar results. Of the 384 House incumbents seeking reelection, 364, or 94.7 percent, won. And according to former Rep. John B. Anderson (R-Ill.), president of the Center for Voting and Democracy, more than half the current members of the House won by at least 20 percent in their last two races. [3]

Members holding leadership positions in their parties and chairmen and ranking members of House committees proved to be invincible in 1996. [4] Those 45 House leaders--23 Republicans and 22 Democrats, including 26 who faced no primary challenge--all won reelection. Perhaps more telling is that, on average, House Republican leaders have been in Congress for 16.6 years, while House Democratic leaders have an average of 19.9 years of congressional service.

What has caused that congressional entrenchment? Why do incumbents, particularly party leaders and committee chairmen, win with such regularity and ease? This paper examines a number of the most important reasons, beginning with a practice in which nearly all members engage: pork-barrel spending.

The Power of the Purse

Instances of pork-barrel spending are commonplace; the most typical is perhaps construction of an unnecessary government building. To cite one example, in 1996 the House appropriated $3.3 million for a courthouse annex in the district of Rep. Phillip English (R-Penn.) just five weeks before English's first bid for reelection. [5] In addition to adding to the budget deficit, what effects do such expenditures have?

One widely held belief is that members able to secure spending for their districts are rewarded at the ballot box. That belief was captured by political scientist David Mayhew when he wrote, "How much particularized benefits count for at the polls is extraordinarily difficult to say, but it would be hard to find a Congressman who thinks he can afford to wait around until precise information is available. The lore is that they count." [6]

To test that hypothesis, in the late 1970s Congress Project researchers took a sample of first-term representatives and grouped them according to the amount of federal spending that went to their districts. "Those with the least federal spending added 4.6 percent to their original victory margins when they sought reelection. The next group, with more federal spending, received a 6 percent hike in the victory margin. The lucky ones who procured the most local public works added a whopping 8.9 percent to their share of the vote." [7]

Robert M. Stein of Rice University and Kenneth N. Bickers of Indiana University have also found that pork-barrel spending can help an incumbent's chances for reelection. In a study of the 99th and 100th Congresses, Stein and Bickers concluded that "voter awareness of new projects in their district increases the probability of voting for the incumbent House member by 9.7 percent." [8]

In another article, Stein and Bickers examined the 100th and 101st Congresses. They found that vulnerable incumbents--specifically those members last elected in an open seat race by a narrow margin--disproportionately secure pork-barrel projects early in their terms. The reason is simple: "While only a minority of incumbents typically face quality challengers, high levels of new awards early in the Congressional term reduce the likelihood that incumbents will face quality challengers in the subsequent primary or general election." [9]

Thus pork-barrel spending not only places challengers at a disadvantage at the ballot box, it also dissuades other potential challengers from entering the race.

Getting the Incumbent's Message Out

In their studies, Stein and Bickers were careful to point out that pork-barrel projects benefit an incumbent only if his constituents know about them. Fortunately for incumbents, House members have a great many tools at their disposal for making sure that constituents are fully aware of their activities.

Perhaps the most common way that members of Congress contact voters is by taking advantage of the franking privilege, which allows members to use the mails free of charge for "official business." Over the past several decades that privilege has been increasingly used. In 1954 Congress sent out 43.5 million pieces of mail. By 1992 that number had skyrocketed to 458 million. [10]

Currently, the amount of money a member may spend on franked mail is determined by the number of addresses in his district. The 1995 allowance averaged $109,000. [11] In that year members could also spend an additional $25,000 on franked mail by transferring funds from other office expense accounts. At current bulk mail rates, such an appropriation would allow a member to send nearly 1 million pieces of mail a year.

During the past decade, members have begun patronizing innovative mailing list vendors to ensure that those 1 million pieces are sent to the right people. And, according to authors Larry Sabato and Glenn Simpson, they have used taxpayer dollars to help one vendor make its lists accessible in a more user-friendly format. Sabato and Simpson report that in 1991 the House Administration Committee "granted Aristotle (a private consulting firm) a $250,000 development loan to transfer its database lists onto CD-ROM format. By the following summer fifty-six legislators had bought Aristotle CD-ROM disks for the total sum of $250,000." [12]

The result of such intensified direct mail campaigns has been a greater gap between the number of voters contacted by mail by incumbents and the number contacted by challengers. In 1994, for example, 63 percent of voters received mail from incumbents, while only 25 percent received mail from challengers. [13] In addition, in 1977 members decided that taxpayers would fund unlimited long-distance telephone service for congressional offices. [14] The consequences of that grant of incumbent privilege were predictable. In 1994, 14 percent of voters claimed that they had talked to a member of the incumbent's staff, while only 5 percent said that they had talked to someone working for the challenger. [15]

Congressional candidates also rely heavily on electronic media to convey information to voters. Recognizing that fact, Congress has developed audiotape- and film-preparation facilities that are available to members free of charge. Needless to say, challengers do not have equal--or any--access to those facilities. Again, the result has been to widen the gap between voters' exposure to incumbents and their exposure to challengers. In 1994, 33 percent of voters said that they heard the incumbent on the radio; only 18 percent said that they heard the challenger. The numbers are similar for television. Sixty-one percent reported seeing the incumbent on television; 34 percent said that they had seen the challenger. [16]

Most members have also become aware of the growing importance of the Internet and have used taxpayer dollars to finance the development of World Wide Web sites. The sites contain a vast array of information about the member, including position papers, bills that the member has sponsored, and biographical information. Many of the sites are frequently updated by congressional staff. In 1996 CompuServe decided to offer free sites to incumbents and challengers alike, but the Federal Election Commission said no. In a letter to CompuServe, it stated, "The Commission still concludes that your proposed gift to Federal candidates of valuable services which enable them to communicate with voters and advocate their candidacies would constitute in-kind contributions to those candidates and would be prohibited." [17] Thus, it is legal for incumbents to use taxpayer dollars to "advocate their candidacies" on the Web, but it is illegal for a challenger to receive a contribution that would allow him to do the same thing.

Constituent Service

Updating Web sites is just one task that members of Congress assign to their staffs. A much higher percentage of staff time is spent on "constituent service." In his book Home Style, Richard Fenno provides a good working definition of constituent service:

Many activities can be incorporated under the rubric of "district service," or "constituent service," but the core activity is providing help to individuals, groups, and localities in coping with the federal government. Individuals need someone to intercede with the bureaucracies handling their veterans' benefits, social security checks, military status, civil service pension, immigration proceedings and the like. [18]

Table 1
Personal Staffs of House and Senate Members, 1930-93

Year House Staff Senate Staff
1930 870 280
1935 870 424
1947 1,440 590
1957 2,441 1,115
1967 4,055 1,749
1972 5,280 2,426
1977 6,942 3,554
1982 7,511 4,041
1987 7,584 4,075
1993 7,400 4,138

Source: Norman J. Ornstein, Thomas E. Mann, and Michael J. Malbin,
Vital Statistics on Congress, 1995-1996 (Washington: Congressional Quarterly Press, 1995), table 5-2.

A key to performing constituent service is a large and well-organized personal staff. And over the past 60 years, as Table 1 shows, personal staffs have grown enormously.

In a recent article, syndicated columnist Robert D. Novak mentions the changes he has seen since he first came to Washington in 1957. "Few members of Congress employed a press secretary. Nearly all readily responded to phone calls from a low-level AP reporter without an aide asking what he wanted. With fewer staffers, lawmakers did much of their own work. At night on his typewriter, Sen. Everett McKinley Dirksen wrote summaries of every bill reported by every Senate committee." [19]

Another key to effective constituent service is district or regional offices. Such offices make possible personal interaction between voters and staff and are used almost entirely for constituent service. Like the size of personal staffs, the number of district offices has exploded over the past 60 years. In 1964 only 4 percent of House members listed more than one district office and, of those, only one member listed more than two. [20] Today multiple offices are the rule, and many members of Congress have additional "temporary" or "mobile" district offices that are set up as booths at county fairs and outside sporting events. Moreover, in 1960 nearly 30 percent of all members said that their district offices were open only when they were at home. Today it is difficult to find even a single representative who follows that policy. [21]

With the increased use of regional offices, more personal staff are working on constituent service in districts. In 1972, 23 percent of personal House staff worked in district or state offices; by 1994 that number had grown to 47 percent. [22] To allow members themselves to more frequently discuss matters with constituents face to face, Congress has steadily increased the number of paid trips that a member may take to his district. Table 2 chronicles that trend.

Why have members spent increasing amounts of time on constituent service? The answer is twofold. First, the expansion of the federal government during the past 60 years, and the resulting bureaucratic problems that many constituents face, has undoubtedly made constituent service increasingly necessary. [23] Second, many incumbents have found that constituent service is an excellent way to bolster their reelection prospects.

Table 2
Paid Trips to a Member's District, 1962-Present

Year Trips to District
1962 3
1966 5
1968 12
1973 18
1975 26
1977 33
1978 and later Unlimited a

Source: Gary C. Jacobson, The Politics of Congressional Elections,
4th ed. (New York: Longman, 1997), p. 30.
a. Since 1978 travel expenses have been included
in an overall expense account that a member may use as he wishes.

In a series of books and articles, Morris P. Fiorina, a political scientist at Harvard University, has argued that those two reasons are not unrelated. "In the postwar period," Fiorina writes, "we have seen both the decline of the marginal district and the expansion of the federal role and its attendant bureaucracy. I believe that these two trends are more than statistically related. An institutional change--the growth of the bureaucracy--has encouraged behavioral change among congressmen, which in turn has encouraged behavioral change among some voters." [24]

Fiorina posits that, like all other people, elected officials wish to improve their current circumstances or, if that it is not possible, make sure that those circumstances do not deteriorate. As a result, they tend to do what will help them get reelected, that is, what will improve their standing with their constituents.

Members of Congress can engage in two different types of activities: the programmatic and the nonprogrammatic. The programmatic is introducing pieces of nontrivial legislation or campaigning for their passage. The nonprogrammatic can be one of two things: pork-barrel spending, which we have already discussed, or constituent service.

By their very nature, programmatic activities are controversial. Unless a member represents a homogeneous district, some voters are going to differ with his stance on any specific issue and hence he will lose support. Moreover, even if he votes for a bill that enjoys widespread support, it is unlikely that he will gain much backing as a result. No member can claim credit for a piece of major legislation that many people have helped to draft and to pass.

In contrast, nonprogrammatic actions are not controversial and usually enjoy widespread support. Members find it easy to claim credit for federal projects awarded to their districts. In those cases, a member's vote is not seen as just 1 vote of 435. Similarly, by helping constituents with their problems, a member can appear to be a helpful advocate without partisan loyalties.

As a result, when drafting bills, members have two incentives to make the legislation broad and vague. First, the broader and vaguer the legislation, the more difficult it is for a constituent to object to its passage. Second, broad and vague legislation presents great opportunities for constituent service. Fiorina explains,

The nature of the Washington system is now quite clear. Congressmen earn electoral credits by establishing various federal programs. The legislation is drafted in very general terms, so some agency, existing or newly established, must translate a vague policy mandate into a functioning program, a process that necessitates the promulgation of numerous rules and regulations and, incidentally, the trampling of numerous toes. At the next stage, aggrieved and/or hopeful constituents petition their congressmen to intervene in the complex (or at least obscure) decision processes of the bureaucracy. The cycle closes when the congressman lends a sympathetic ear, piously denounces the evils of bureaucracy, intervenes in the latter's decisions, and rides a grateful electorate to ever more impressive electoral showings. Congressmen take credit coming and going. They are the alpha and the omega. [25]

In other words, by creating opportunities for constituent service, members of Congress are simply heeding the advice of the 16th-century political theorist Niccolò Machiavelli, who wrote, "A wise prince will seek means by which his subjects will always and in every possible condition of things have need of his government, and then they will always be faithful to him." [26]

Delegation: Congress Creates Imperial Agencies

In recent years Congress has found a new way to expand bureaucracy, and thus opportunities for constituent service, while, at the same time, avoiding any blame: delegation of congressional power. The Constitution is very clear about where the power to make law resides. Article I, section 1, reads: "All legislative powers herein granted shall be vested in a Congress of the United States, which shall consist of a Senate and House of Representatives." The need for such strict delineation and division of power was explained by James Madison, principal architect of the Constitution, in The Federalist Papers: "The accumulation of all powers, legislative, executive, and judiciary, in the same hands, whether of one, a few, or many, and whether hereditary, self-appointed, or elective, may justly be pronounced the very definition of tyranny." [27] George Washington also advised officeholders "to confine themselves within their respective Constitutional spheres; avoiding the exercise of the Powers of one department to encroach upon another. The spirit of encroachment tends to consolidate the Powers of all the departments in one, and thus to create whatever the form of government, a real despotism." [28]

Despite Madison's and Washington's warnings, Congress has ceded its lawmaking power to the executive branch and unelected regulators, who pass laws that are often more complicated and sweeping than what originates in Congress. In congressional testimony, Jerry Taylor of the Cato Institute described the incentives that drive the delegation of lawmaking power:

One of the main reasons Congress delegates is to manipulate voter perceptions. Delegation allows legislators to represent themselves to some constituents as supporting an action and to others as opposing it. As former [Environmental Protection Agency] administrator Lee Thomas described delegation under the Clean Air Act, "Everybody is accountable and nobody is accountable under the way [Congress] is setting it up, but [the legislators] have got a designated whipping boy." [29]

The Electoral Effects of Constituent Service

Just how much do incumbents benefit from constituent service? Apparently, a lot. In Home Style, Fenno repeats what a sitting congressman told him: "What political scientists have to understand is that an incumbent congressmen can get reelected by the services he is in a position to do for people." [30] That sentiment was echoed by senior congressional staffers when in a survey they were asked what the most important factor is in boosting a legislator's political support. Fifty-six percent said that constituent service was most important. The member's legislative record, in contrast, was cited by only 11 percent of respondents. [31] Similarly, in the mid-1980s when members' administrative assistants were asked whether their employers thought that constituent service enhanced election chances, only 2 of 102 replied that service did not have a significant electoral effect. And many offered detailed responses, including the following:

Rep. Henry Hyde (R-Ill.), chairman of the House Judiciary Committee, was equally blunt in his assessment of the importance of constituent service: "Again and again, I will tell you why you have a leg up, good constituent service, accessibility and availability. You ought to have a leg up. You have made an investment challengers never make. I will not apologize for that." [33]

Several recent studies have tried to quantify the advantage of which Chairman Hyde speaks. Many have found it to be quite substantial. In 1988 an anonymous House member, curious about the impact of his casework, provided political scientist George Serra with a list of constituents who had benefited from casework and a list of constituents who had not. That fall Serra contacted 419 of those people and asked them how they felt about the incumbent on a scale of 0 (extremely unfavorable) to 100 (extremely favorable).

For incumbent partisans (respondents with the same party affiliation as the incumbent) who had not benefited from constituent service, the representative's mean score was 66.0; for incumbent partisans who had benefited from constituent service, the representative's mean score was 76.8, an increase of 10.8. For challenger partisans (respondents with different party affiliation than the incumbent) who had not benefited from constituent service, the representative's mean score was 54.7; for challenger partisans who had benefited from service, the representative's mean score was 73.4, an increase of 18.7. For Independents who had not benefited from constituent service, the representative's mean score was 57.2; for Independents who had benefited from service, the representative's mean score was 73.4, an increase of 16.2. [34]

It is important to note that through constituent service the incumbent was able to greatly improve his approval rating among not only constituents belonging to his political party but also constituents belonging to the opposing party and Independents. Indeed, while the average approval ratings of incumbent partisans and challenger partisans were considerably different among those who did not benefit from constituent service--66.0 compared to 54.7, a difference of 11.3--nearly that entire gap was erased if the constituent had benefited from constituent service--76.8 compared to 73.4, a difference of just 3.4. Moreover, challenger partisans who had benefited from casework had a significantly higher evaluation of the representative than incumbent partisans who had not--73.4 compared to 66.0, a difference of 7.4. Thus, it becomes clear that the key to approval is not voting in accord with a constituent's ideology--if that were the case, incumbent partisans would, presumably, always look at the incumbent more kindly than would challenger partisans--but doing casework on behalf of that constituent. [35] Favors, not principles, are the key to reelection.

In his study, Serra also considered whether casework can produce favorable effects beyond its recipients, as had been postulated before. [36] By using an econometric model, he was able to conclude that "respondents who can recall having friends and/or relatives with satisfactory casework experiences are more favorable in their evaluations of the incumbent than other respondents." [37]

Further evidence that incumbents who are reliable providers of constituent service can expect to succeed at the polls was provided by an American National Election Study. Of respondents who contacted their representative for ombudsman service and reported being "very satisfied," 64.7 percent voted for the incumbent; only 3 percent voted against the incumbent, and 32.3 percent did not vote. Even those who said they were simply "somewhat satisfied" voted in a similar fashion. More than 57 percent voted for the incumbent, 11.8 percent voted against the incumbent, and 30.9 percent did not vote. [38]

Constituent service is perhaps the most powerful tool that incumbents have at their disposal. By passing vague laws--or delegating the authority to promulgate regulations that have the force of law--Congress can add to an already large bureaucracy that is both confusing and unresponsive. Unlike large businesses, whose Washington offices are designed to handle such problems (and, indeed, to actually create many of them through lobbying for favorable, but confusing, tax and regulatory statutes), the average citizen must go to his representative to ask for help, for guidance through the great maze of regulations that the incumbent had a hand in creating. When the citizen receives the help he requests, he decides that whatever ideological differences there may be between the incumbent and himself are ultimately unimportant. What matters is that the incumbent will be there the next time a bureaucratic problem arises, which it undoubtedly will. After considering that phenomenon, Fiorina concludes,

Congress does not just react to big government--it creates it. All of Washington prospers. More and more bureaucrats promulgate more and more regulations and dispense more and more money. Fewer and fewer congressmen suffer electoral defeat. Elements of the electorate benefit from government programs, and all of the electorate is eligible for ombudsman services. But the general, long-term welfare of the United States is no more than an incidental by-product of the system. [39]

Campaign Finance Regulations as Incumbent-Protection Laws

Until 1974 campaign finance was only loosely regulated. In that year, however, Congress passed a number of amendments to the Federal Election Campaign Act that greatly circumscribed the ways campaigns could be financed. Among the provisions established by the amendments were the following:

Such provisions were passed for the explicit purpose of protecting incumbents and the two-party system at a time, the Watergate era, when both were in an unusual amount of danger. Indeed, historian John L. Kelley has written, the legislation "recognized the existence of a political duopoly and sought to perpetuate it by discrimination against new entrants to the competition. If the logic of the antitrust movement had been applied to the 1974 statute it is difficult to see how it could have survived even cursory examination." [40] And when in 1976 the Supreme Court deemed only a few of the 1974 FECA amendments unconstitutional, Chief Justice Warren Burger warned, "This Court has, until today, been particularly cautious when dealing with enactments that tend to perpetuate those who control legislative power." [41]

Today, defenders of those provisions argue that regulation is necessary because money is supposedly corrupting the political process. Specifically, it is asserted that campaigns are becoming too costly, elections are being bought, and contributions are influencing the voting patterns of elected representatives. A critique of those assumptions, which have been proven flawed, is outside the scope of this paper. [42] Instead, we shall focus on how the regulations that still stand have been detrimental to the electoral hopes of challengers.

Contribution Limits

Limiting how much individuals may give to a candidate either directly or indirectly (through a party committee or a political action committee) has prevented challengers from raising money quickly from a few dedicated supporters. Unlike an incumbent who already has an operational campaign organization and a database of former contributors from whom to solicit, a challenger may need a few wealthy patrons to fund his campaign in order to even get it off the ground. If, for example, restrictions on donations by individuals had been in place in 1968, it is unlikely that Sen. Eugene McCarthy, whose bid for the presidency was funded primarily by a few donors, would have been able to mount a challenge to President Lyndon Johnson in the Democratic primaries. Moreover, there is much recent evidence that spending limits have influenced who has, and has not, run for office.

Jack Kemp, the Republican vice-presidential candidate in 1996, stated repeatedly that one of the reasons he did not run for president was that he thought soliciting a sufficient number of $1,000 donations was like "trying to fill a swimming pool with a teaspoon." Ironically, Steve

Forbes--who would have backed Kemp--did seek the Republican nomination because, under FECA, he was able to spend unlimited amounts of money on his own campaign. Similarly, it was only because of his own personal resources that Ross Perot was able to mount the most successful third-party presidential campaign in a generation. [43]

Although there are undoubtedly other factors at play, there can be little question that individual contribution limits have helped to keep people from running as well as helped cause the underfinancing of most campaigns by congressional challengers. While that might sound good to Rep. Bill Clay (D-Mo.), a 15-term congressman from St. Louis, who has argued, "If there's one thing we don't need, it's more candidates running for Congress," it is not a good thing for a healthy democratic republic. [44] Indeed, it is hard to see how anybody except congressional veterans benefits from the fact that in 1996 fewer than one-fifth of the 384 House incumbents who ran for reelection faced challengers with at least half the incumbent's resources. [45]

Individual contribution limits place an especially high burden on candidates with views that are out of the mainstream. To raise enough $1,000 contributions to mount a credible run for the House, one has to have very widespread support. A candidate with a visionary but unconventional platform is unlikely to attract that kind of backing. Thus, like constituent service, individual donation limits tend to make questions of ideology and principle less important. In the case of constituent service, principle does not matter because people tend to vote for the candidate who can help them with the bureaucracy, not the candidate they think has the correct platform. In the case of donation limits, platforms do not matter much because candidates with policy positions that point in new directions from the consensus are not usually able to raise enough money to fund credible campaigns.

In business, entrepreneurs with innovative ideas frequently bring their product to market with the backing of just a few investors, having been turned down by hundreds of others. In American politics, however, there is no such opportunity. The candidate is denied the chance to present his ideas, and the voters are denied the chance to hear them. Individual contribution limits have done what they were intended to do--limit challenges to incumbent power. The result has been banal campaigns centered on trivia waged by candidates seeking power instead of promoting an agenda.

Spending Limits

Absolute ceilings on the amount that can be spent on a campaign are once again being debated in Congress, even though the Supreme Court in 1976 stated that they violated the First Amendment and hence struck them down as unconstitutional. To get around that prohibition, campaign finance reformers have proposed limits that are ostensibly "voluntary." [46] In addition to being an affront to the First

Amendment, those limits would greatly improve the reelection chances of incumbents, who, as we have seen, are already at a great advantage.

The House campaign finance bill receiving the most attention would cap total spending on House races at $600,000. Happily for incumbents, in 1994 and 1996 challengers who spent less than that sum won only 3 percent of their races, while those who spent more won approximately 40 percent of theirs. [47] An analysis of races from 1986 to 1994, adjusting for inflation, yielded similar results. [48] Thus, empirically, the reason that incumbents would wish to impose ceilings at, say, $600,000 is clear.

Campaign spending is subject to diminishing returns--the more dollars spent, the less gained by each additional dollar. Incumbents, as we have seen, exploit their official "perks" to the fullest extent possible. By the time the election comes, most voters know who they are and what they stand for. As a result, their marginal benefits from campaign spending are exhausted relatively soon.

Indeed, there is an inverse relationship between incumbent spending and electoral success. [49] That is, the more money an incumbent spends, the worse he does on election day. That seeming anomaly is easily explained: high campaign spending is usually a sign of weakness rather than strength. If an incumbent cannot win a race by simply exploiting all the "free" electoral benefits he acquires by holding office, he is probably too vulnerable to win; nevertheless, he will try to make up for his vulnerability by increased spending.

In contrast, most challengers enter their races as complete or relative unknowns and, thus, profit greatly from campaign spending. In most cases, their marginal benefits from spending are still relatively high by the time they hit any spending ceiling, but especially one as low as $600,000.

Explaining the problems with campaign spending ceilings, political scientist Larry Sabato writes,

The question is, who would determine the ceilings? The Congress, of course--a body composed of 535 incumbents who are fervently convinced of the worthiness of their own reelections. It is in their electoral interests to set the ceilings as low as possible. After all, the incumbents already have high name recognition, purchased with lavish spending during previous campaigns, and with hundreds of thousands of dollars of taxpayers' money (via congressional staffs, the frank, mobile offices, constituency services, and so forth during their years in office). The average challenger, then, begins his or her campaign perhaps millions of dollars behind the incumbent in overall real spending. Major expenditures are necessary to compensate and to compete. [50]

Gerrymandering for Reelection

The "vanishing marginal seat"--the phenomenon of House seats being won not only by the same person year after year but by increasingly large margins--has one other explanation: gerrymandering. In 1964 the Supreme Court ruled that the principle of one person, one vote must be applied to congressional districts and that state legislatures must ensure that all districts contain roughly equal populations. [51] That rule has led to the political manipulation of districts.

The party that controls the state legislature has a substantial incentive to draw U.S. House districts that will give it an advantage. By creating districts that contain significantly more voters of their own party than their opponents', the majority in a legislature can ensure that most districts will not be marginal for many elections to come. The district's majority party candidate will win year after year. And since it is very rare for a House incumbent to face a serious primary challenge, not only will the same party usually win year after year, the same person will.

The U.S. House of Representatives plays a substantial role in perpetuating that process. By doling out billions of dollars to state governments each year, a state's congressional delegation exerts tremendous influence over the state legislature. When, every 10 years, it comes time to redraw districts, state representatives are inclined to remember which senior members of the House were able to bring home major projects to the state. In addition to wanting to give their party an advantage at the polls, state representatives also want to give those senior members an advantage so that federal dollars will continue to flow home. A related point is that many members of Congress started their political careers in state legislatures with an eye toward higher office. As a result, current like-minded state legislators have an additional incentive to create safe congressional districts. "A major element in the job security of incumbents," concludes political scientist Edward R. Tufte, "is their ability to exert significant control over the drawing of district boundaries; indeed, some recent redistricting laws have been described as the Incumbent Survival Acts." [52]

Proposals for Reform

Incumbents have attempted to rig the game by passing measures that put their opponents at an almost insurmountable disadvantage. Indeed, the mountain of legislation that has been passed for the express purpose of incumbent protection has prompted Gordon Tullock, one of the pioneers of public-choice analysis, to wonder why "the Federal Government is never mentioned in debates about campaign finance as the largest campaign contributor. A congressman gets over $500,000 a year to maintain his office. He gets plane trips to his constituents, a TV studio in the Capitol basement, and other perks. Most important, he can pressure government agencies to do special favors for his constituents. An old woman who finds that her pension comes much faster after she calls her congressman is a pretty secure vote." [53] Reforms that would lessen or eliminate the advantages that incumbents enjoy should be aggressively pursued. The first, and perhaps most obvious, reform would be to dramatically scale back the size and scope of the federal government and, as a result, the chances for incumbents to perform constituent service.

Perhaps the two agencies that cause the most bureaucratic problems for citizens are the Internal Revenue Service and the Social Security Administration. Congress should pursue reforms that would lead to their elimination. [54] Replacing the federal income tax with a flat-rate national sales tax would yield substantial gains. Not only would a national sales tax promote economic growth by reducing the bias against work, savings, and investment, it would also be much easier to administer and to comply with than is the income tax. [55] Privatizing the current Social Security system by allowing workers to invest in private individual retirement accounts would produce a larger capital stock and a higher national income. [56] Privatization would also greatly reduce bureaucratic problems for employees and employers.

A third measure that Congress could take to reduce the burden of government is to stop delegating its lawmaking powers. Last year Congress enacted, as part of the Small Business Regulatory Enforcement Fairness Act of 1996, a procedure that would make it easier to revoke a law promulgated by an agency. While that is a step in the right direction, the proper end is complete compliance with the Constitution. The Constitution requires the House, the Senate, and the president to enact a new law, while the new procedure requires the House, the Senate, and the president to revoke a new law. Delegation is a principal generator of constituent service; it is also unconstitutional.

Congress should also deregulate campaign finance, not increase the regulatory burden that is borne almost entirely by challengers. Ending all limits on campaign contributions and dismantling FECA would increase the number of challengers as well as the viability of their campaigns. Moreover, deregulation would lead to the introduction into the political process of ideas that currently are not given a hearing.

The most substantial reform that Congress could implement would be to pass a constitutional amendment to limit the number of terms that a member may serve--specifically, three in the House and two in the Senate. Term limits would dramatically change the incentives of the political system. [57] If the seniority system were eliminated, more potential citizen-legislators would be inclined to seek office. There are many people who care about their government and would be willing to spend a few years in Washington fixing it, if they believed that career legislators--party leaders and committee chairmen--did not completely control the agenda.

Once in office, it is likely that citizen-legislators would behave much differently than those currently in power. Since they would have little desire and, indeed, no opportunity to spend their lives in Congress, citizen-legislators would have no incentive to pursue activities--such as pork-barrel spending, constituent service, and regulating campaign finance--that, as one of their principal effects, entrench legislators in office. Moreover, term limits would do much to overturn Washington's "culture of spending," as James Payne has put it, and thus shrink the federal bureaucracy. [58] The benefits of such a change are clear.


Congressional leaders have proven themselves adept at pursuing reforms that stifle electoral competition and perpetuate their own rule. Is it, then, a bit naive to expect them to repeal those measures and open up the political system? Yes. The current system serves them well. They have what most rulers only dream of--immense power with nearly complete protection from competition and accountability.

Incumbents frequently propose modest reforms to win editorial praise. But, as the past several years have shown, there is little chance that veterans will voluntarily surrender their power by passing a term-limits amendment. And the other reforms mentioned above are unlikely to pass if they are recognized as contributing to the revitalization of truly representative government. The increased power those initiatives would give to voters would be taken directly from members of Congress.

Restoring representative government in the United States will require a fundamental change in the composition of Congress, from career rulers to citizen-legislators. To achieve that change, voters must initiate a national movement to reject those seeking a career in Congress in favor of those pledged to using their short service in Washington to return the federal government to its limited, constitutional role. They must, in short, honor the words of James Madison: "If it be asked, what is to restrain the House of Representatives from making legal discriminations in favor of themselves and a particular class of society? I answer: above all, the vigilant and manly spirit which actuates the people of America--a spirit which nourishes freedom, and in return is nourished by it." [59]


[1]. Gary C. Jacobson, The Politics of Congressional Elections, 4th ed. (New York: Longman, 1997), p. 21.

[2]. Ibid., p. 23.

[3]. John B. Anderson, "Participation Crisis: Take Action Now to Make Voters Vote," Roll Call, November 21, 1996, p. 5.

[4]. Leadership positions are speaker, majority and minority leader, and majority and minority whip.

[5]. Ronald G. Shafer, "Washington Wire," Wall Street Journal, October 4, 1996, p. 1.

[6]. David Mayhew, Congress: The Electoral Connection (New Haven, Conn.: Yale University Press, 1974), p. 57.

[7]. Quoted in Mark Green, Who Runs Congress? 3d ed. (New York: Viking, 1979), p. 272.

[8]. Robert M. Stein and Kenneth N. Bickers, "Congressional Elections and the Pork Barrel," Journal of Politics 56, no. 2 (May 1994): 392-93.

[9]. Kenneth N. Bickers and Robert M. Stein, "The Electoral Dynamics of the Federal Pork Barrel," American Journal of Political Science 40, no. 4 (November 1996): 1300.

[10]. Jacobson, The Politics of Congressional Elections, p. 31.

[11]. National Taxpayers Union, Press release, July 10, 1996. The document can be accessed at

[12]. Larry J. Sabato and Glenn R. Simpson, Dirty Little Secrets: The Persistence of Corruption in American Politics (New York: Times Books, 1996), p. 233.

[13]. American National Election Studies, 1990 and 1994, and Senate Election Studies, 1988-1992, cited in Jacobson, The Politics of Congressional Elections, p. 100.

[14]. Ibid., pp. 29-30.

[15]. American National Election Studies, 1990 and 1994, and Senate Election Studies, 1988-1992, cited in ibid., p. 100.

[16]. Ibid.

[17]. The Federal Election Commission's statement is quoted in Steven Hayward and Allison R. Hayward, "Gagging on Political Reform," Reason, October 1996, p. 26.

[18]. Richard F. Fenno Jr., Home Style: House Members in Their Districts (Boston: Little Brown, 1978), p. 101.

[19]. Robert D. Novak, "What a Change 40 Years Makes," Washington Post, May 12, 1997, p. A19.

[20]. Morris P. Fiorina, Congress: Keystone of the Washington Establishment (New Haven, Conn.: Yale University Press, 1977), p. 58.

[21]. Ibid., p. 59.

[22]. Norman J. Ornstein, Thomas E. Mann, and Michele J. Malbin, Vital Statistics on Congress, 1995-1996 (Washington: Congressional Quarterly Press, 1995), tables 5-3 and 5-4.

[23]. For a quantitative description of the growth of government in the 20th century, see Stephen Moore, Government: America's #1 Growth Industry: How the Relentless Growth of Government Is Impoverishing America (Lewisville, Tex.: Institute for Policy Innovation, 1995).

[24]. Morris P. Fiorina, "The Case of the Vanishing Marginals: The Bureaucracy Did It," American Political Science Review 71, no. 1 (March 1977): 177. Emphasis in original.

[25]. Fiorina, Congress, p. 48.

[26]. Niccolò Machiavelli, The Prince and the Discourses (New York: Modern Library, 1950), p. 39.

[27]. James Madison, "Federalist no. 47," in The Federalist Papers (New York: New American Library, 1961), p. 301.

[28]. George Washington, "Washington's Farewell Address," in George Washington: A Collection, ed. W. B. Allen (Indianapolis: Liberty Classics, 1988), p. 521.

[29]. Jerry Taylor, Testimony before the Subcommittee on Commercial and Administrative Law of the House Committee on the Judiciary, September 12, 1996. For more on the problems caused by delegation, see David Schoenbrod, Power without Responsibility: How Congress Abuses the People through Delegation (New Haven, Conn.: Yale University Press, 1993).

[30]. Fenno, p. 101.

[31]. The results of the survey are discussed in Richard H. Shapiro, Frontline Management: A Guide for Congressional District/State Offices (Washington: Congressional Management Foundation, 1989), p. 94, and cited in Eric Felten, The Ruling Class: Inside the Imperial Congress (Washington: Regnery, 1993), p. 192.

[32]. Quoted in Bruce Cain, John Ferejohn, and Morris Fiorina, The Personal Vote: Constituency Service and Electoral Independence (Cambridge, Mass.: Harvard University Press, 1987), pp. 78-79.

[33]. Congressional Record, March 29, 1995, p. H2906.

[34]. George Serra, "What's in It for Me? The Impact of Congressional Casework on Incumbent Evaluation," American Politics Quarterly 22, no. 4 (October 1994): 403-8. For more of Serra's work on the subject, see two articles he coauthored: George Serra and David Moon, "Casework, Issue Positions, and Voting in Congressional Elections: A District Analysis," Journal of Politics 56, no. 1 (February 1994): 200-213; and George Serra and Albert D. Cover, "The Electoral Consequences of Perquisite Use: The Casework Case," Legislative Studies Quarterly 17, no. 2 (May 1992): 233-46.

[35]. An excellent analysis of how term limits would produce legislators with ideological views more reflective of those of their constituents can be found in Einer Elhauge, "Are Term Limits Undemocratic?" University of Chicago Law Review 64, no. 1 (Winter 1997): pp. 114-18.

[36]. See, for example, Diana Evans Yiannakis, "The Grateful Electorate: Casework and Congressional Elections," American Journal of Political Science 25, no. 2 (May 1981): 568-80.

[37]. Serra, p. 412.

[38]. The study is cited in Douglas Rivers and Morris P. Fiorina, "Constituency Service, Reputation, and the Incumbency Advantage," in Home Style and Washington Work: Studies of Congressional Politics, ed. Morris P. Fiorina and David W. Rohde (Ann Arbor: University of Michigan Press, 1989),pp. 25-26.

[39]. Fiorina, Congress, p. 49.

[40]. John L. Kelley, Bringing the Market Back In: The Political Revitalization of Market Liberalism (New York: New York University Press, 1997), p. 128.

[41]. Buckley v. Valeo, 424 U.S. 1, 819 (1976) (separate opinion of Burger, C.J.).

[42]. See Bradley A. Smith, "Campaign Finance Regulation: Faulty Assumptions and Undemocratic Consequences," Cato Institute Policy Analysis no. 238, September 13, 1995. See also Bradley A. Smith, "Real and Imagined Reform of Campaign Corruption: A Review of Dirty Little Secrets: The Persistence of Corruption in American Politics by Larry J. Sabato and Glenn R. Simpson," Cornell Journal of Law and Public Policy 6, no. 1 (Fall 1996): 141-56.

[43]. It should be noted that the 1974 amendments to FECA did limit personal spending in House and Senate races to $25,000 and $35,000, respectively. That provision was struck down as unconstitutional by the Supreme Court in 1976.

[44]. Clay is quoted in Amy Keller, "Why Fund Study on Why Good Candidates Don't Run?" Roll Call, June 30, 1997, p. 8.

[45]. Common Cause, Press release, November 7, 1996.

[46]. In H.R. 493 there are penalties for violating the "voluntary" limit. If a House candidate spends more than the voluntary limit, his opponent may increase spending to match that of the noncomplying candidate. In addition, the noncomplying candidate's opponent is allowed to accept larger contributions. For example, since the bill also would limit individual contributions to $250, instead of the current $1,000, the noncomplying candidate would be restricted to receiving $250 donations while the opponent could accept donations up to $2,000.

[47]. Bradley A. Smith, "The Sirens' Song: Campaign Finance Regulation and the First Amendment," Brooklyn Journal of Law & Public Policy, forthcoming.

[48]. Jacobson, The Politics of Congressional Elections,pp. 40-41.

[49]. Ibid., p. 40.

[50]. Larry J. Sabato, Paying for Elections: The Campaign Finance Thicket (New York: Priority, 1989), p. 22. For more on the benefits for incumbents of spending ceilings, see James L. Regens and Ronald Keith Gaddie, The Economic Realities of Political Reform: Elections and the U.S. Senate (New York: Cambridge University Press, 1995), pp. 95-96; Frank J. Sorauf, Inside Campaign Finance: Myths and Realities (New Haven, Conn.: Yale University Press, 1992), pp. 210-12; and Gary C. Jacobson, "Practical Consequences of Campaign Finance Reform: An Incumbent Protection Act?" Public Policy 24, no. 1 (Winter 1976): 1-32.

[51]. Wesberry v. Sanders, 376 U.S. 1 (1964).

[52]. Edward R. Tufte, "The Relationship between Seats and Votes in Two-Party Systems," American Political Science Review 67, no. 2 (June 1973): 551.

[53]. Gordon Tullock, "The Real Issue," Letter to the editor, National Review, May 19, 1997, p. 6.

[54]. Congress should not limit its cutting of spending and bureaucracy to those two agencies. Every program in the budget should be subjected to very careful scrutiny.

[55]. For a discussion of the merits of a flat-rate national sales tax, see Stephen Moore, "The Economic and Civil Liberties Case for a National Sales Tax" in Frontiers of Tax Reform (Stanford, Calif.: Hoover Institution Press, 1996). For a discussion of how such a tax could be implemented, see David R. Burton and Dan R. Mastromarco, "Emancipating America from the Income Tax: How a National Sales Tax Would Work," Cato Institute Policy Analysis no. 272, April 15, 1997.

[56]. See Martin Feldstein, "Privatizing Social Security: The $10 Trillion Opportunity," Cato Social Security Paper no. 7, January 31, 1997; and Peter J. Ferrara, "A Plan for Privatizing Social Security," Cato Social Security Paper no. 8, April 30, 1997.

[57]. See Edward H. Crane, "Term Limits and the Need for a Citizen Legislature," in Cato Handbook for Congress, 105th Congress (Washington: Cato Institute, 1997), pp. 55-61.

[58]. James Payne, The Culture of Spending: Why Congress Lives beyond Our Means (San Francisco: ICS Press, 1991).

[59]. James Madison, "Federalist no. 57," in The Federalist Papers, p. 353.

© 1997 The Cato Institute
Please send comments to webmaster