Congressional Careers

Contours of Life in the U.S. House of Representatives

John R. Hibbing

• Publisher: The University of North Carolina Press
• Year: 2017
• Print ISBN: 9780807843406, 9780807819845
• eBook ISBN: 9781469639833
• Topic: Politics & International Relations
• Subtopic: American Government
• Index: Politics & International Relations

According to a Gallup poll, 70 percent of Americans want elected officials to serve only a limited number of terms. Nevertheless, every two years American voters return, on average, more than 95 percent of incumbents to the U.S. House of Representatives. John Hibbing’s book provides unique evidence of the problems that would result from congressional term limitations.

The first scholar to analyze congressional careers using longitudinal data, Hibbing looks at how the career patterns of a typical House member have evolved over the last forty years. By showing that the gap between the legislative contributions of junior and senior members has grown in recent years, Hibbing contends that as members gain in seniority they become more knowledgeable, efficient, and valuable legislators. Thus he argues against congressional term limitations.

Hibbing’s findings illuminate other fundamental and surprising changes. House members now are as electorally secure early in their careers as they are late, and they are less likely to deviate from their previously established roll call voting pattern. Members acquire positions of authority (subcommittee chairs) more quickly than they used to, but these more rapid gains evaporate by the sixth or seventh term of service. Also, House members travel to their home districts less frequently than they did ten years ago.

Congressional Careers is a fascinating portrait of the evolution of American legislators during their congressional service. It is the only study of congressional behavior that is both comprehensive and longitudinal — valuable features in an era when congressional careerism is coming under acute public scrutiny.

1

Introduction: Manny Celler Stayed Too Long

During most of his extraordinary career in the U.S. House of Representatives, Emanuel “Manny” Celler had been the envy of just about all his House colleagues. Despite numerous redistrictings, his Brooklyn constituents remained steadfastly supportive of him. He did not have to bother spending a lot of time doing service work or being “on display” in the district. His experience, wit, and excellent memory caused an unusual number of House members to seek his counsel on various policy and procedural matters. And his Judiciary Committee chairmanship made him one of the House’s “movers and shakers,” as illustrated by the pivotal role he played in the titanic struggles during the early 1960s to pass a civil rights bill. Manny Celler seemed to be living in a legislator’s Nirvana.

But this idyllic situation also created a problem. Manny Celler’s congressional world was too comfortable, and he did not know when or perhaps even how to leave that world. He stayed and he stayed. He stayed for fifty years. He stayed until the psychological distance between him and the people of the district became too great. He stayed until a movement he did not understand overtook him. He stayed until his district and party turned on him in the 1972 Democratic primary. He stayed until he was ignominiously removed from the seat he had held for half a century. He stayed so long that several near-retirement-age House members looked at a slightly befuddled Manny Celler and said they were not going to let “that” happen to them.

Manny Celler spent his entire adult life in the U.S. House of Representatives. Although the length of his stay is somewhat unusual, it is not greatly out of line with the tenure of others in the modern Congress. Carl Vinson (D-Ga.), for example, actually spent a few more months in the House than Manny Celler. Sam Rayburn (D-Tex.) spent the last 47 years of his life in the House. Had not a heart attack cut him short, Rayburn may have made the same mistake as Manny Celler, since they both radiated identical “the House is my life” attitudes. George Mahon (D-Tex.) served 44 years; W. R. “Bob” Poage (D-Tex.), 41; Melvin Price (D-Ill.), 44; and Jamie Whitten (D-Miss.), 50 and still counting, as of early 1992 the holder of the record for longest House service. If the issue is length of a congressional rather than a House career, Carl Hayden makes Manny Celler, Carl Vinson, and Jamie Whitten look like veritable flashes in the pan. Hayden’s 15 years in the House and 42 years in the Senate constitute the record for longest service in Congress. In fact, his retirement in 1968 at the age of 91 marked the first time since Arizona had become a state that it was not represented in the U.S. Congress by Carl Hayden.

Historical Trends in Congressional Careers

The congressional career of Manny Celler would appear much more remarkable in the context of the nineteenth-century Congress. The fact that all the lengthy careers just mentioned were logged by individuals serving well after 1950 is not a coincidence. The likelihood of a new member of Congress staying for an entire career has been much greater in the second half of the twentieth century than at any time previously. Since 1953 the average member of the House has stayed almost eleven years. One hundred years ago, an eleven-year career in the House would have been considered unusually long. As H. Douglas Price remarks, “Up until the 1890s only a handful of men had pursued substantial lifetime careers within the House, and they were often the occasion for puzzled comment” (1971, 16). Price goes on to note that William Holman (D-Ind.) was “one of the few really long-term nineteenth-century members” (1971, 21), yet Holman served barely thirty years, a record of longevity that is hardly impressive by modern standards.

Charles Bullock (1972) once defined a “careerist” in the House of Representatives as someone who had served at least ten terms. In the 92d Congress (1971–72) the number of careerists reached an astounding eighty-seven—indicating that fully one out of five representatives in the House at that time had served over twenty years in the body. The number of careerists has gone down some in the last fifteen years, but still approximately one out of eight members meets Bullock’s criterion.

Information on nineteenth-century Congresses is not as tidy and complete, however there is no mistaking the contrast with modern congressional careers. According to Price’s data (1977, 38), of the 242 House members in the 24th Congress (1835–36), 128 were not even in the 25th. If the analysis is restricted to members elected from districts outside the South, the high level of fluidity is even more apparent: 100 of the 166 nonsouthern members in the 24th Congress were not in the 25th—and 79 of these 100 left voluntarily (3 others died and 18 were defeated). In the 49th Congress (1885–86), there were 325 House members, and only 193 of these answered the call for the 50th. Again, voluntary retirement was the major source of turnover. Throughout most of the nineteenth century, members rarely made it into their second terms, let alone their tenth! Data on average career length have been collected by Stuart Rice (1928) and updated by Nelson Polsby (1968). They show quite clearly that House career length increased during Congress’s 200 years.¹

Why were congressional careers typically short back then? It is not hard to come up with reasonable explanations. At times during the 1800s turnover in many states was pushed higher by the tradition of congressional rotation (see Kernell 1977; Struble 1980; Dometrius and Sigelman 1989). Moreover, for a good deal of that century the nation’s capital was not particularly dignified. Accounts abound of the swampy, mosquito-infested, rural-outpost nature of Washington’s early days. The Congress itself was somewhat raucous. Price writes of “bitter and outrageous language, scathing ridicule, and sarcasm . . . outbreaks of physical violence, guns and knives on occasion being carried into the chamber” (1971, 18). Polsby (1968, 166–67) recounts in vivid detail some of the classic altercations of the nineteenth-century House, including those involving Charles Sumner’s caning, Matthew Lyon’s expectorating, John Randolph’s hunting dogs, and a gunshot wound that was inflicted in the chamber of the House. Former Speaker Thomas B. Reed’s simple description of life in Congress during this period seems an apt summation: “These were not pleasant days” (quoted in Polsby 1968, 167).

Finally, the Congress of that era did not possess a great deal of power. Morris Fiorina writes that it was not until 1890 that the U.S. Congress became more important than the Virginia state legislature (1977b, 6). Price notes that “in many respects the pre-1900 House was similar to the average current state assembly” (1971, 18). Congress was a part-time body with a part-time role in a limited federal government. During this period it was common for even the most powerful members of the House (including Speakers) to leave the body in order to take a chance at becoming a senator or a governor. In short, Congress was neither a pleasant nor a powerful place, so, unlike Manny Celler, most nineteenth-century members did not stay.

But all this is really nothing new. The growing professionalization of Congress over the years has been very well documented. In three separate articles, Price describes the trend (1971, 1975, 1977). Fiorina sets the stage for his indictment of the “Washington establishment” with a brief discussion of the early Congresses (1977b, 1989). Samuel Huntington (1965) sees careerism itself as a major reason Congress is in need of indictment (or at least was in need of indictment in 1965). The thesis of David Mayhew’s (1974a) justifiably well-known essay on congressional structures and behavior is that an enormous amount of analytical leverage is obtained by assuming that members of Congress are interested in nothing except getting reelected, and his first step is to show that in the modern Congress most members do in fact seek and win reelection. Polsby (1968) puts the trend in larger perspective by calling it part of the “institutionalization of the House.” T. Richard Wittmer (1964), Bullock (1972), and Fiorina, David Rohde, and Peter Wissel (1975), using slightly different measures all provide important supportive evidence that, with the of the decades, congressional have become longer. Just about every textbook on Congress and even. many general textbooks on American politics also note the fact that typical congressional careers are now much longer than they used to be and that these lengthy careers can be seen as an important part of the professionalization or institutionalization of Congress Although some slightly peculiar things have happened to this trend since the early 1970s there would be little benefit in providing additional evidence supporting the well-established basic fact that congressional careers have lengthened

Some members may stay too long; others leave before it is too late; but the fact remains that, by historical standards, most modern members stay quite a long time. Lengthy modern congressional careers are now accepted as the norm. The puzzled comment that used to surround those unusual creatures who spent lifetime careers in the House is now reserved for those who turn their backs on further service in Congress. It is widely assumed that a politician who would do this must have ulterior motives. Why would anyone want to leave the House except to make a run for the Senate, and why would anyone voluntarily leave the Senate except to make a run for the presidency? Perhaps, besides seeking higher office, they might leave because of advanced age, poor health, or impending scandal revelation, but certainly not for any other reason. The incredulity with which many voluntary retirement announcements of members of Congress are greeted is perhaps the ultimate testimony to how far Congress and the congressional career have come in 200 years.

Where Research on Congressional Careers Is Lacking

No, we do not need another recitation of how congressional careers have evolved. The general changes are quite well-known and largely unsurprising to anyone who has spent time reflecting on the history of Congress and the United States. But what we do not know is how our national representatives change while they are in Congress. We know more (but not much) about the evolution of legislatures than we do about the evolution of legislators, and this is unfortunate.

Perhaps too many of us have been transfixed by the clarity of the trend toward longer congressional careers, for we have spent far too much time dreaming up new ways to document it. We have used percentage of members in their first term, percentage who have served five terms or more, percentage who have served ten terms or more, mean number of terms served, mean number of years served, average age of members, percentage seeking reelection, percentage replacements, Polsby-Rice percentage first-term, median terms served, and several other perfectly reasonable and relevant measures. We have almost as many ways of documenting the declining membership turnover in Congress and the increasing longevity of congressional careers as we have of measuring the increased electoral safety of incumbent House members!

What we are missing is actual knowledge of how the situations and behaviors of members of Congress change (or do not change) as these members move through what are now frequently lengthy careers. Numerous insights are likely to be generated if the extended congressional career is viewed as an opportunity to observe changes in individual legislative environment and behavior over a several-year period, not as something to discard as soon as we have proven that lengthy careers are in fact quite common in the modern Congress. Career-based research is difficult and meaningless when most members stay in Congress for only a few months. But when a large number of individuals are staying for several decades, career-based research becomes possible and potentially fruitful.

The major reason this research approach holds so much potential is that it reflects how members themselves view the situation—and if one of our goals is to understand congressional behavior, we must approximate as closely as possible the perspective of those doing the behaving. When queried about their electoral situation, members of Congress invariably respond by reciting the support they have received over the years, not by comparing their present electoral status with the other 434 members. When queried about their activities back in the district, they recount all they have done over the years. And when queried about their legislative positions or actions, perhaps on a specific issue, they are more likely to mention what they did on a similar issue years ago than to compare their current predilections with those of other members currently serving. Representatives, like most of us, utilize the past as a bench mark for understanding the present and perhaps the future.

The extent to which past behavior acts as a constraint on the subsequent behavior of representatives is an important piece of information for those prone to reflect upon the nature of Congress and the need for reform. To offer one example, pressure is mounting to pass a constitutional amendment limiting the length of congressional service. This is an issue that will not go away. It has been with us longer than the Constitution (the Articles of Confederation limited stays in the Confederate Congress). Seventy percent of the U.S. population support term limits for members of Congress (see Hook 1990, 567), interest groups now exist for the sole purpose of establishing the congressional equivalent of the Twenty-second Amendment, the popular press of late has bristled with debate on this issue, and some states have now limited the terms of state legislators. But the sad thing is that this debate is taking place at an incredibly low level. Proponents make their case by drawing attention to the length of congressional careers and by offering claims concerning the need for “new blood.” Opponents, who are usually led by sitting representatives and senators themselves, counter by pointing out that even with long careers we still have reasonable levels of turnover (as of 1989, well over 50 percent of House members had begun their service in the 1980s) and by offering equally claims concerning the glories of legislative experience.

Both positions miss the point. The truth is that raw turnover levels do not say much at all about the extent to which Congress is innovative, activist, legislatively sharp and focused, sensitive to constituents, and more generally in tune with the country. To draw conclusions on these issues, we must pay closer attention to how members change throughout their careers. Do they change their roll call voting behavior during a career? Do they change with regard to non-Washington activity, such as taking trips back to the district, doing constituency service, delegating staffers to district rather than Washington offices, and communicating with constituents? If these behaviors do vary with career stage, why? Is it because of career-related variation in electoral support, in formal positions held within the body, or in something else?

In short, the debate on the advantages and disadvantages of legislative experience is taking place without any evidence on what exactly it is that this experience brings. Is there a learning curve? How much would be lost by mandating the removal of senior legislators? If longitudinal research discovers that member behavior is generally constant across a career, it seems to me that serious thought should be given to term limitations. If, however, there are signs that senior legislators are offering something that junior members are not, lengthy careers just may make Congress a better rather than a worse legislative assembly.

Despite the fact that the consequences of careerism should be the foundation for an understanding of the modern Congress, we know next to nothing about them. Most research on individual congressional environments and behaviors has been concerned with cross-sectional rather than longitudinal questions. Why do some representatives support their party more than others? Why do some incumbents attract more votes than others? Why do some representatives run for the Senate while others do not? Why do some members do more constituency service work than others?

On occasion, researchers have attempted to address questions about the effects of tenure on behavior. But often this research is not what it seems. For example, some recent attention has been given to the issue of whether senior members are less likely to do constituency service work than younger members. This may appear to be a tenure-based research question, but in many cases it turns out to be something else. Despite the longitudinal overtones to the issue, it is often tested with cross-sectional data. Typical research on this question may tell us that in a certain Congress members with ten terms of service were less likely to do constituency service work than members with two terms of service, but this really says nothing about tenure-based effects. It may be that those ten-termers have done relatively trivial amounts of constituency service work throughout their entire careers. It may be that those second-termers will continue to do incredible amounts of constituency service even when they are ten-termers. We do not know. The cross-sectional research design, even when it includes tenure as a variable, does not tell us much about change over the course of a career.

Using cross-sectional data to test longitudinal questions usually means it will be impossible to sort out generational, life-cycle, and, for that matter, period effects. In the example presented above, if the ten-termers did in fact always ignore constituency service work, the effect would seem to be generational—the newer generation simply is more likely to do constituency service work because it was socialized into congressional activities at a different time and in a different way. If the ten-termers used to do a lot of constituency service work but drifted away from it sometime during their careers, the effect would indeed be a life-cycle (or careercycle) effect since the stage of the career is what makes the difference. A period effect is one in which all cohorts are moved to a different level by some force. In the case of constituency service, perhaps all cohorts of representatives began doing more of it when the tesources to undertake this work staffers, monetary allocations for trips home, etc.—were increased. But since we do not know how much constituency service work the ten-termers did before they became ten-termers or how much the second-termers did after their second term, data from a single Congress provide an insufficient basis for conclusions regarding the effects of career-cycle forces on congressional behavior.

There are also problems with taking the members of a single class and following them all the way through their careers. Setting aside the possibility that the particular class selected may be atypical of even the classes around it (see Loomis 1988a; Schneider 1989), the key problem with this kind of design is that it is unable to distinguish clearly between life-cycle and period effects. If a change appears somewhere in the course of the careers of many members of this class, is this change the result of a career cycle or is it something that affected everyone in Congress regardless of where they were in their own career cycle? This kind of question cannot be answered with single-cohort data. So even the small amount of research purporting to be interested in longitudinal questions is often much less than it claims.

In this study, I avoid such pitfalls by relying, with one exception, on data structured across individual congressional careers and by including data from numerous cohorts. The task of collecting and organizing the data in this manner confirmed suspicions regarding the virtual absence of previous career-centered approaches to the study of Congress and congressional behavior. Existing data on Congress seldom are structured along individual congressional careers. It is easy to find out what Representative Smith did in the 99th Congress, but generally what that same Representative Smith did in the 100th Congress is treated as an entirely different case, not as a continuation of what that member had done before. If the data do extend over time, more often than not they trace what changed within a congressional district. Again, the individual career fades to the background as several different members may have represented the district during the period in question.

All this is only to point out that relatively few analysts have used appropriate data to pursue questions concerning changes over the course of congressional careers. While there are countless references in the literature to the need to obtain more than just a cross-sectional “snapshot” of congressional behavior, most of these references are followed by research that does in fact present a cross-sectional snapshot. Much w...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Dedication Page
• Contents
• Tables and Figures
• Preface
• 1 Introduction: Manny Celler Stayed Too Long
• 2 The Electoral Career
• 3 The Formal Position Career
• 4 The Roll Call Career
• 5 The Legislative Activity Career
• 6 The District Activity Career
• 7 The Integrated Congressional Career
• 8 Conclusion: Gary Myers Left Too Soon
• Notes
• References
• Index