The United States (along with many Latin American nations) has a congress; the United Kingdom (along with most Western European nations) has a parliament. A hint as to the difference between the two kinds of legislatures can be found in the original meanings of the words. Congress derives from a Latin term that means "a coming together," a meeting, as of representatives from various places. Parliament comes from a French word, parler, which means "to talk." There is of course plenty of talking— some critics say there is nothing but talking—in the U.S. Congress, and certainly members of a parliament represent to a degree their local districts. But the differences implied by the names of the lawmaking groups are real ones, with profound significance for how laws are made and how the government is run. These differences affect two important aspects of lawmaking bodies: how one becomes a member and what one does as a member. Ordinarily, a person becomes a member of a parliament (such as the British House of Commons) by persuading a political party to put his or her name on the ballot. Though usually a local party committee selects a person to be its candidate, that committee often takes suggestions from national party headquarters. The local group selects as its candidate someone willing to support the national party program and leadership. In the election, voters in the district choose not between two or three personalities running for office, but between two or three national By contrast, a person becomes a candidate for representative or senator in the U.S. Congress by running in a primary election. As we discussed in Chapter 9, parties may try to influence the outcome of primary elections, but they cannot determine them. As a result of these different systems, a parliament tends to be made up of people loyal to the national party leadership who meet to debate and vote on party issues. A congress, on the other hand, tends to be made up of people who think of themselves as independent representatives of their districts or states and who, while willing to support their party on many matters, expect to vote as their (or their constituents') beliefs and interests require. Once they are in the legislature, members of a parliament discover they can make only one important decision-whether or not to support the government. The government in a parliamentary system such as that of the United Kingdom consists of a prime minister and various cabinet officers selected from the party that has the most seats in parliament. As long as the members of that party vote together, that government will remain in power. Should members of a party in power in parliament decide to vote against their leaders, the leaders lose office, and a new government must be formed. With so much at stake, the leaders of a party in parliament have a powerful incentive to keep their followers in line. They insist that all members of the party vote together on almost all issues. If someone refuses, the penalty is often drastic: The party does not renominate the offending member in the next Members of the U.S. Congress do not select the head of the executive branch of government—that is done by the voters when they choose a president. Far from making members of Congress less powerful, this makes them more powerful. Representatives and senators can vote on proposed laws without worrying that their votes will cause the government to collapse and without fearing that a failure to support their party will lead to their removal from the ballot in the next election. Indeed, despite record levels of party unity in recent years, members of both parties have rebuked their leaders and yet remained in office. For example, conservative Republicans effectively forced Republican John Boehner to resign as Speaker of the House in 2015. While Boehner-and his successor as Speaker, Paul Ryan—may have been unhappy with this behavior, they could not remove these Republicans from office or really do all that much to punish them). Congress has independent powers, defined by the Constitution, that it can exercise without regard to presidential preferences. Political parties do not control nominations for office, and thus they cannot discipline members of Congress who fail to support the party lead-ership, Because Congress is constitutionally independent of the president, and because party discipline is highly imperfect, individual members of Congress are free to express their views and vote as they wish. They are also free to become involved in the most minute details of law-making, budget making, and supervising the administration of laws. They do this through an elaborate set of committees and subcommittees. A real parliament, such as that in Britain, is an assembly of party representatives who choose a government and discuss major national issues. The principal daily work of a parliament is debate. A congress, such as that in the United States, is a meeting place of the representatives of local constituencies-districts and states. Members of the U.S. Congress can initiate, modify, approve, or reject laws, and they share with the president supervision of the administrative agencies of the government. The principal work of a congress is representation and action, most of which takes place in committees. What this means in practical terms to the typical legislator is easy to see. Because members of the British House of Commons have little independent power, they get rather little in return. They are provided a modest sal-ary, have a small staff, are allowed only limited sums to buy stationery, and can make a few free local telephone calls. Each is given a desk, a filing cabinet, and a tele-phone, but not always in the same place. By contrast, a member of the U.S. House of Repre-sentatives, even a junior one, has power and is rewarded accordingly. For example, in 2017, each member camed a substantial base salary ($174,000) plus generous health care and retirement benefits, and was entitled to a large office (or "clerk-hire") allowance, to pay for about no dozen staffers. (Each chamber's majority and minorry leaders carned $193,400 a year, and the Speaker of tie House carned $223,500. Each member also receive individual allovances for travel, compurer services, and the like. In addition, each member could mail newsleter and certain other documents to constituents for fire using the "franking privilege." Senators, and representatio with seniority, received even larger benefits. Each sens-tor is entitled to a generous office budge and legislative. assistance allove a generous off hire as many saf mei. bers as he or she wishes with che money. These examples are not given to suggest that members of Congres are over-rewarded, but only that their importance as individuals in our political system can be inferred from the resources they command. Because the United States has a congress made up of people chosen to represent their states and districts, rather than a parliament that represents competing political par-ties, no one should be surprised to learn that members of the U.S. Congress are more concerned with their own constituencies and careers than with the interests of any organized party or program of action. And because Congress does not choose the president, members of Congress know that worrying about the voters they represent is much more important than worrying about whether the president succeeds with his programs. These two factors taken together mean that Congress tends to be a decentralized institution, with each member more interested in his or her own views and those of his or her voters than with the programs proposed by the president. Inded, the Founders designed Congress in ways that almost inevitably make it unpopular with voters. Americans want government to take action and follow a clear course of action. Americans dislike political arguments, the activities of special-interest groups, and the endless puling and hauling that often precede any congressio-na decision. But the people who feel this way are deeply dided about what government should do: Be liberal? Be conservative? Spend money? Cut taxes? Support abortions? Stop abortions? Because they are divided, and because nembers of Congress must worry about how voters feel, it is inevitable that on controversial issues Congress will engage in endless arguments, worry about what interest groups (who represent different groups of voters) think, and work out compromise decisions. When it does those things, however, many people feel let down and say they have a low opinion of Congress. Of course, a member of Congress might explain all these constitutional facts to the people, but not many members are eager to tell their voters that they do not really understand how Congress was created and orga-nized. Instead, they run for reelection by promising voters they will go back to Washington and "clean up that mess.
————- The Framers chose to place legislative powers in the hands of a congress rather than a parliament for philosophical and practical reasons. They did not want to have all powers concentrated in a single governmental institu-tion, even one that was popularly elected, because they feared such a concentration could lead to rule by an oppressive or impassioned majority. At the same time, they knew the states were jealous of their independence and would never consent to a national constitution if it did not protect their interests and strike a reasonable balance between large and small states. Hence, they created: bicameral (two-chamber) legislature-with a House of Representatives, whose members are elected directly by the people, and a Senate, consisting of two members from each state who are chosen by the legislatures of each state. Though "all legislative powers" were vested in Congress, those powers would be shared with the president (who could veto acts of Congress), limited to powers explicitly conferred on the federal government, and, as it turned out, subject to the power of the Supreme Court to declare acts of Congress unconstitutional. For decades, critics of Congress complained that the body cannot plan or act quickly. They are right, but two competing values are at stake: centralization versus decen-tralization. If Congress acted quickly and decisively as a body, then there would have to be strong central leader-ship, restrictions on debate, few opportunities for stalling tactics, and minimal committee interference. If, on the other hand, the interests of individual members-and the constituencies they represent-were protected or enhanced, then there would have to be weak leadership, rules allowing for delay and discussion, and many opportunities for committee activity. Though there have been periods of strong central leadership in Congress, the general trend for much of the 20th century was toward decentralizing decision making and enhancing the power of the individual member at the expense of the congressional leadership. That said, the recent rise in polarization has somewhat reversed that trend, though leaders today are less powerful than those of the late 19th and early 20th centuries (the apogee of the speaker's power). This decentralization may not have been inevitable. Most American states have constitutional systems quite similar to the federal one, yet in many state legislatures, such as those in New York, Massachusetts, and Indiana, the leadership is quite powerful. In part, the position of these strong state legislative leaders may be the result of the greater strength of political parties in some states than in the nation as a whole. In large measure, however, it is a consequence of permitting state legislative leaders to decide who shall chair what committee and who shall receive what favors. The House of Representatives, though always power-ful, often has changed the way in which it is organized and led. In some periods, it has given its leader, the Speaker, a lot of power. In other periods, it has given much of that power to the chairs of the House committees. In still other periods, it has allowed individual members to acquire great influence. To simplify a complicated story, the How Things Work box staring on page 298 outlines six different periods in the history of the House. The House faces fundamental problems: it wants to be big (it has 435 members) and powerful, and its members want to be powerful as individuals and as a group. But being big makes it hard for the House to be powerful unless some small group is given the authority to run it. If a group runs the place, however, the individual members lack much power. Individuals can gain power, but only at the price of making the House harder to run and thus reducing its collective power in government. There is no lasting solution to these dilemmas, and so the House will always be undergoing The Senate does not face any of these problems (we review important House/Senate differences in Table 13.3). It is small enough (100 members) that it can be run without giving much authority to any small group of lead-ers. In addition, it has escaped some of the problems the House once faced. During the period leading up to the Civil War, it was carefully balanced so that the number of senators from slave-owning states exactly equaled the number from free states. Hence, fights over slavery rarely arose in the Senate. From the first, the Senate was small enough that no time limits had to be placed on how long a senator could speak. This meant there never was anything like a Rules Committee that controlled the amount of debate. Finally, senators were not elected by the voters until the 20th century. Before that, they were picked instead by state legislatures. Thus senators often were the leaders of local party organizations, with an interest in funneling jobs back to their states. The big changes in the Senate came not from any fight about how to run it (nobody ever really ran it), but from a dispute over how its members should be chosen. For more than a century after the Founding, members of the Senate were chosen by state legislatures. Though often these legislatures picked popular local figures to be senators, just as often there was intense political maneuvering among the leaders of various factions, each struggling to win (and sometimes buy) the votes necessary to become senator. By the end of the 19th century, the Senate was known as the Millionaires Club because of the number of wealthy party leaders and businessmen in it. There arose a demand for the disect, popular elecion of senators. Nacurally the Senate resisted, and without its aproval the necessary constitute resisten, endment could nor pas Congress. When some states threatened to demand a ner constitutional convention, the Senate feared that such a convention would change more than just the way in which senators were chosen. A protracted struggle ensued, during which many state legislatures devised ways to ensure that the senators they picked would already have won a popular election. The Senate finally agreed to a constitutional amendment that required the popular election of its members, and in 1913 the Seventeenth Amendment was approved by the necessary three-fourths of the states. Ironically, given the intensity of the struggle over this question, no great change in the composition of the Senate resulted; most of those members who had first been chosen by state legislatures managed to win reelection by popular vote. The other major issue in the development of the Senate was the filibuster. A filibuster is a prolonged speech, or series of speeches, made to delay action in a legislative assembly. It had become a common—and unpopular —feature of Senate life by the end of the 19th century. It was used by liberals and conservatives aike and for lofty as well as self-serving purposes. The fist serious effort to restrict the filibuster came in 1917, after an important foreign policy measure submitted by President Wilson had been talked to death by, as Wilson put it, "eleven willful men." Rule 22 was adopted filibuster An attempt to defeat a bill in the Senate by talking indefinitely, thus preventing the Senate from taking action on the bill. by a Senate fearful of tying a president's hands during a wartime crisis. The rule provided that debate could be cut off if two-thirds of the senators present and voting agreed to a "cloture" motion (it has since been revised to allow 60 senators to cut off debate). Two years later, it was first invoked successfully when the Senate voted cloture to end, after 55 days, the debate over the Treaty of Versailles. Despite the existence of Rule 22, the tradition of unlimited debate remains strong in the Senate, and examples of famous filibusters abound. One—by former South Carolina Senator Strom Thurmond—lasted for more than 24 hours (Thurmond was filibustering a proposed Civil Rights Act). We discuss the contemporary effects of the filibuster later in the chapter.
————- With power so decentralized in Congress, the kind of person elected to it is especially important. Since each member exercises some influence, the beliefs and interests of each individual affect policy. Viewed simplistically, most members of Congress seem the same: the typical representative or senator is a middle-aged white Protestant male lawyer. If all such persons usually thought and voted alike, that would be an interesting fact, but they do not, and so it is necessary to explore the great diversity of views among seemingly similar people. Gender and Race Congress has gradually become less male and less white. Berween 1950 and 2017, the number of women in the House increased from 9 to 83 (plus 4 delegates, who represent U.S. territories or Washington, DC, as well as the resident commissioner of Puerto Rico) and the number of African Americans increased from 2 to 47 (plus 2 delegates). There are also 38 Latino members (plus 1 delegate and the Resident Commissioner of Puerto Rico), and 2 Native American members. Until recently, the Senate changed much more slowly (see Figure 13.1). Before the 1992 election, there were no African Americans and only two women in the Senate. But in 1992, four more women, including one African American woman, Carol Mosely Braun of Illinois, were clected. These numbers have gradually increased over time, and today in the 115th Congress, 21 women, 3 African Americans, and 5 Latinos serve in the U.S. Senate. Part of the increase in African American and Latino members of Congress comes from the creation of majority-minority districts. In such districts, a majority of residents are racial or ethnic minorities. These districts are designed to allow said minorities to elect candidates of choice, and were created as a result of litigation surrounding the Voting Rights Act. Most often, the candidate of choice is someone from their racial or ethnic group: for example, districts with a majority of African American voters typically, though not always, elect an African American candidate. Such districts have played a key role in bringing more racial and ethnic minorities into Congress. As a result, such districts certainly increase descriptive representation, when a minority officeholder represents minority constituents. Such descriptive representation is valuable because someone from a minority group will typically be best positioned to understand and represent the needs of that group.' Yet some scholars claim that such districts may inadvertently harm minority interests. To create majority-minority districts, many racial and ethnic minorities need to be packed into a single district, and as a result, surrounding districts typically have fewer racial and ethnic minorities. This has two consequences. First, members in surrounding districts, because they have fewer minority constituents, have less incentive to respond to the needs of minority voters.' Second, these surrounding districts become less likely to elect Democrats to office. 13 This follows because these districts have fewer racial and ethnic minorities, who typically strongly support Democratic candidates. For example, evidence shows that the creation of new majority-minority districts following the 1990 census helped to elect more Republicans to Congress.' Because Democrats often, but not always, support policies that are more in line with the preferences of most racial and ethnic minorities, a Congress with fewer Democrats is less likely to pass legislation favored by racial and ethnic minorities (for example, on policies such as affirmative action). This suggests that while such districts increase symbolic rep-resentation, they might decrease substantive representa-tion: the ability of voters in this case, minority voters) to elect officials who will enact policies in line with their preferences. In the end, then, which is preferable? Is it better to elect more racial and ethnic minorities to Congress, even if it means also passing fewer policies supported by racial and ethnic minorities? Or instead should we aim to elect fewer racial and ethnic minorities to Congress, but disperse minority nod eaio minories to Congress, members who (on average) are more supportive of policies favored by minorities? The answer is unclear, and depends on how one views the relative importance of descriptive and substantive representation. However one feels about descriptive versus substantive representa-tion, majority-minority districts have increased the power of African American and Latino members in another way. Because such districts are typically quite electorally safe, their members often become senior leaders in Con-gress, especially on com-mittees. For example, in 1994, African Ameri- cans chaired four House committees and Latinos chaired three. When the Democrats retook control of Congress in 2007, African Americans chaired five committees and Latinos chaired two more. Some of the committee chair-persons— such as Charles Rangel and John Conyers-have become very powerful members of Congress. Similarly, the first woman to become Speaker (Nancy Pelosi in 2007) was a Democrat, and the increase of women in Congress after 1970 has been led by Demo-crats: in the 115th Congress that began in 2017, 16 of the 21 women in the Senate, and 62 of the 83 women in the House, were Democrats (plus three delegates). Among the most notable members first elected in 2016 is Senator Tammy Duckworth of Illinois, a disabled Iraq War Veteran who is one of two Asian American women elected in 2016 (the other is Senator Kamala Harris of California; three Asian American women, all Democrats, now serve in the U.S. Senate). The year 2016 also saw the election of the first Latina Senator, Democrat Catherine Cortez Mastro of Nevada. Middle-aged white men with law degrees are still prevalent in Congress, but as Table 13.4 shows, compared with the makeup of the 102nd Congress that began in 1991, the 115th Congress that began in 2017 had not only more women, blacks, and Latinos, but also fewer lawyers, fewer persons who had served in the armed forces, more businesspeople, more people over the age of 55, and more members (about one in six overall) serving their first term. Incumbency The recent spike in first-termers in Congress is interest-ing, but the most important change that has occurred in the composition of Congress has been so gradual that most people have not noticed it. In the 19th century, a large fraction-often a majority-of congressmen served only one term. In 1869, for example, more than half the members of the House were serving their first term in Congress. Being a congressman in those days was not regarded as a career. This was in part because the federal government was not very important (most of the interesting political decisions were made by the states); in part because travel to Washington, DC, was difficult and the city was not a pleasant place in which to live; and in part because being a congressman did not pay well. Furthermore, many congressional districts were highly competitive, with the two political parties fairly evenly balanced in each. By the 1950s, however, serving in Congress had become a career. Between 1863 and 1969, the proportion of first-termers in the House fell from 58 percent to 8 percent.' As the public took note of this shift, people began to complain about "professional politicians" being "out of touch with the people," and some pushed for term limits. The issue had a brief burst of popularity in the mid-1990s, and numerous states passed laws to limits the terms of members of Congress. In 1995, however, the Supreme Court ruled that any such term limits on federal legislators could only be imposed by a constitutional amendment,' and efforts to pass one in Congress failed (some states do have term limits for state legislators, however). The issue largely vanished from popular discussion until Donald Trump resurrected the issue during the 2016 election. After Trump's win, several legislators introduced constitutional amendments to impose term limits, but the issue faces a long and uphill battle. Term limits remain popular with thepublic: a2013 Gal. lup poll found that 75 percent of the public would impose Torm limits if given the chance. Scholars who are stdid state legislative term limits have found that they have lite ste on who get cleted to ofyin, hey do not increase citizen legislators.' Second, if anything, they find that term limits decrease politicians responsiveness to public opinion (since term-limited legislators know they cannot run for reelection) and tend to shift power to the executive branch and the bureaucracy: As in many areas, reforms designed io solve one problem can create others! In more recent years, political forces did what leg.-isation could not-they brought new faces to the capi-tal. In 1994, Republicans retook the House majoriy from the Democrats, who had held it since 1952. In so doing, they brought many new members to power who had not previously held elected office. Recent elections have continued this pattern. The November 2012 elections brought 75 first-term members to the House, 58 new House members won election in November 2014, and 59 new members won in 2016. In the 115th Congress (2017-2019), 50 percent of members had less than 8 years of experience.21 But these periodic power shifts accompanied by the arrival of scores of new faces in Congress should not obscure an important fact that was documented decades ago by political scientists and is still true today: Even in elections that result in the out party regaining power, most incumbent House members who seek reelection not only win, but win big, in their districts.2 And while Senators have been somewhat less secure than House members, most Senate incumbents who have sought reelection have won it by a comfortable margin. Figure 13.2 shows the 1964-2016 reelection rates for incumbent House and Senate members who sought relection. Over that span of more than two dozen elec-rions, the average reelection rate for House incumbents vas 93 percent and the average relection rate for Senate incumbents was 82 percent. As Figure 13.2 demonstrates, redecion rates have been consistently high throughout this period. Even in years characterized by an anti-incumbent mood, the vast majority of House and Senate incumbents were typically reelected. In the 2010 midterm election, despite polls showing mass disaffection with Congress and a strong "anti- incumbent" mood, 85 percent of House incumbents who sought reelection won it (53 House incumbents who sought reelection lost), and 84 percent of Senate incumbents who sought reelection won it (4 Senate incumbents who sought reelection lost, 2 in primary elections and 2 in the general election). And 2014 was another year characterized by anti-incumbent sentiments. The largest-ever number of voters told pollsters that their own member did not deserve reelection (35 percent), which many took to mean a deeply dissatisfied electorate would vote many members out of office.23 While some highly notable incumbents were defeated, such as Senator Kay Hagan in North Caro-lina, 95 percent of House members who sought reelection won, as did more than 80 percent of Senators. In 2016, 97 percent of House incumbents, and 87 percent of Senate incumbents, who sought reelection won. Year in and year out, most members of Congress are reelected. House incumbents who seek reelection normally beat their opponents by 10 points or more. Political scientists call districts that have close elections (when the winner gets less than 55 percent of the vote) marginal districts and districts where incumbents win by wide margins (55 percent or more) safe districts. By this standard, in the 2014 election, only 11 percent of House seats were marginal. But perhaps we should use a stricter definition of safety: winning with 60 percent or more of the vote. Even here, the majority of incumbents would be considered safe. Since the 1970s, more than 60 percent of House members—in some years as high as 80 percent-have been reelected with at least 60 percent of the major-party vote.? By contrast, over the same period, less than half of all Senate incumbents who won reelection did so by such a wide margin. Safe states are far less common than safe districts. Why congressional seats have become less marginal- that is safer— is not entirely clear, and a number of factors thag have been proposed. Some of the most prominent ones focus on the resources of incumbents. Incumbents, as we explained in Chapter 10, have a large fundraising advantage over challengers. Further, incumbents are simply much better known than challengers, so they have a built-in advantage in terms of name recognition. Incumbents also do much to deluge the voters with free mail-ings, they can travel frequently (and at public expense) to meet constituents, and they can get their names in the headlines by sponsoring bills or conducting investigations. Simply having a familiar name is important in getting elected, and incumbents find it easier than challengers to make their names known. Further, incumbents can use their power to get programs passed or funds spent to benefit their districts-and thereby to benefit themselves." They can help keep an army base open, support the building of a new highway (or block the building of an unpopular one), take credit for federal grants to local schools and hospitals, make certain a particular industry or labor union is protected by tariffs against foreign competition, and so on. They can also provide individual services to their con-stituents, helping them locate a lost Social Security check or provide help with a federal agency, such as the IRS or the Department of Veterans Affairs. If a member helps out a voter this way, then that voter is more likely to support the member in his or her next election. 7 Finally, incumbents over time have learned to behave as if they are at risk even when they are not.? No one thought Eric Cantor, then House Majority Leader, would lose his primary to a virtually unknown economics professor in 2014, but he did. While losses like Cantor's are relatively rare, close elections are not-many members have had an uncomfortably narrow election win, and even if they have not, they know someone who has. These sorts of unexpected losses and near-losses lead members to always be wary, and to act as if they are not safe, even if they are. So members work hard to raise money, increase their name recognition, and provide services to their constitu-ents, which increases their safety. Probably all of these factors make some difference and help to explain why districts are so safe today. This has two important implications. First, as we discussed in Chapter 10, there is an incumbency advantage, whereby incumbents do better than challengers (for all of the reasons we discussed above). Second, this incumbent advantage means that in ordinary times no one should expect any dramatic changes in the composition of Congress. Even when elections effect a change in party control in one or both cham-bers, even when new leaders are in charge and new members abound, many old hands will still be on hand in Congress. Party Forty-three Congresses convened between 1933 and 2017 (a new Congress convenes every two years). The Democrats controlled both houses in 27 of these Congresses and at least one house in 31 of them, and they controlled the House continuously from 1952 10 1994. How scholars predicted the 1994 Republican victory and many at the time thought that Democrats would con. mol the House well into the future. Since 1994, Repub. licans have been in power more ofien than Democrats Democrats controlled the House from 2006 to 2010 but otherwise it has been in Republican hands. What explains these patterns of control? In particular, why did Democrats control the House for so long, and why hav the House become more competitive in recent years? A key part of the reason that the Democrats con. trolled the House for so long is that the Democratic Patty throughout much of the 20th century, was really two separate parties operating under a common name: a more liberal Northern wing and a more conservative Southern wing. While they did not agree on many policies, they did share a common party label and hence formed a large partisan bloc, which allowed them to be the majority party in the House for many decades. While Northern and Southern Democrats aligned to maintain majority control of the chamber (and with it, control of congressional committees and the legislative process), they typically parted ways when it came to policy. Southern Democrats often would vote with the Republicans in the House or Senate, thereby forming what came to be called the conservative coalition. During the 1960s and 1970s, that coalition came together in about one-fifth of all roll-call votes. When it did, it usually won, defeating Northern Democrats. But since the 1980s. and especially since the watershed election of 1994, the conservative coalition has become much less important. The reason is simple: Many Southern Democrats in Congress have been replaced by Southern Republicans, and the Southern Democrats who remain (many of them African Americans) are as liberal as Northern Democrats. This change was an important contributor to the growing levels of polarization we observe in Congress today. But this factor alone does not explain why the Democrats controlled the House for so long, or why the proces is now more competitive. One popular theory among the public, though not really among scholars, is that gerry. mandering is to blame. As we discussed in Chapter 10, gerrymandering is the process of drawing districts to faror one party or the other. Those who favor this explanation suggest that when Democrats are in power, they tend lo draw districts favoring Democrats and vice versa when Republicans are in power. As we discussed in Chaper I0, gerrymandering does influence congressional elections and it has contributed to che con safery. But it eft is modest rather than massive. Several studies of recit elections, such as 2006 asd 2012, note that relisit ing affected the outcome, but only to a modes degree. Summing up much of the scholarly literature, one study found that "Virtually all the political science evidence to dare indicates that the electoral system has little or no par-risan bias, and that the net gains nationally from redis- cal limits as well. First, drawing congressional districts is the duty of the states, most typically, of the state legislature (though some states use some sort of commission). So for one party to really stack the deck in its favor, it needs to control the state legislature of many states, which is difficult to do. Many commissions are nonpartisan, or receive input from the governor. Because of the political sensitivity of congressional districts, many district boundaries are ultimately decided by the courts, which adds another layer of complexity to the process. Many different actors contribute to drawing congressional boundaries, making it hard for one party to really gain an advantage solely due to redistricting. The effects of gerrymandering are also constrained by relevant state and federal laws. Federal law requires that districts have equal population, and the courts have interpreted this rather strictly, rejecting even modest deviations in population across congressional districts." Further, as we discussed above, the Voting Rights Act established major-ity-minority districts, which requires many states to have districts predominantly comprising racial/ethnic minori-ties. Many states also have relevant state laws that require districts to be contiguous and geographically compact, as well as to respect political boundaries and communities of interest. Even when legislators want to engage in ger-rymandering, their ability to do so is constrained by other factors. Even if there were no gerrymandering, many members of Congress would easily be reelected to Congress. While redistricting alone typically does not give an advantage to one party or the other, the Republicans do have a small but persistent advantage in contemporary House elections that stems from geography. Simply put, Republican voters are more evenly spread across districts, whereas Democratic voters are more heavily concentrated in certain districts. Democrats win a large share of voters from racial and ethnic minorities, young people, and liberals, who tend to be clustered in cities. As a result, Democrats tend to carry overwhelmingly districts located in urban areas. While Republicans currently do better in mostly rural districts, those districts are not as skewed toward Republicans because even rural areas tend to have pockets of Democrats (in, say, a college town or a former industrial city). For example, Philadelphia, Pennsylvania (the nation' it-largest city), contains two congressional dis-trices (PA-1 and PA-2), both of which the Democratic incumbents won with more than 80 percent of the vote in 2016. But even in the most rural parts of the state, Republicans do not win so overwhelmingly. Democrats have more votes that are "wasted" by being packed into overwhelmingly Democratic urban districts. As a result, Democrats must win a larger share of the vote to win the same number of seats as Republicans. While this is a boost to Republicans in contemporary elections, Democrats certainly can overcome it, as they did when they recaptured the House in 2006. Today, unlike in earlier generations, there is strong competition between the parties for control of Congress. Representation and Polarization In a decentralized, individualistic institution such as Congress, it is not obvious how its members will behave. They could be devoted to doing whatever their constituents want or, because most voters are not aware of what their representatives do, act in accordance with their own beliefs, the demands of interest groups, or the expectations of congressional leaders. You may think it would be easy to figure out whether members are devoted to their constituents by analyzing how they vote, but that is not quite right. Members can influence legislation in many ways other than by voting: they can conduct hearings, help mark up bills in committee meetings, and offer amendments to the bills proposed by others. A member's final vote on a bill may conceal as much as it reveals; some members may vote for a bill that contains many things they dislike because it also contains a few things they value. There are at least three theories about how members of Congress behave: representational, organizational, and attitudinal. The representational explanation is based on the reasonable assumption that members want to get reelected, and therefore they vote to please their con-stituents. The organizational explanation is based on the equally reasonable assumption that because most constituents do not know how their legislator has voted, it is not essential to please them. But it is important to please fellow members of Congress, whose goodwill is valuable in getting things done and in acquiring status and power in Congress. The attitudinal explanation is based on the assumption that the many conflicting pressures on members of Congress cancel one another out, leaving the members virtually free to vote on the basis of their own beliefs. Political scientists have studied, tested, and argued about these (and other) explanations for decades, and nothing like a consensus has emerged. Some facts have been established, however, in regard to these three views. Representational View The representational view has some merit under certain circumstances-namely, when constituents have a clear view on some issue and a legislator's vote on that issue is likely to attract their attention. Such is often the case for civil rights laws: representatives of districts with significant numbers of black voters are not likely to oppose civil rights bills; representatives of districts with few African Americans are comparatively free to oppose such bills. Until the late 1960s, many Southern because the African Americans in their districts were resentatives without black constituents have supported civil rights bills, partly out of personal belief and partly perhaps because certain white groups in their districts-organized liberals, for example-have insisted on such support. From time to time, an issue arouses deep passions among voters, and legislators cannot escape the need either to vote as their constituents want, whatever their personal views, or to anguish at length about which side of a divided constituency to support. Gun control has been one such question and the use of federal money to pay for abortions has been another. Some fortunate members of Congress get unambiguous cues from their constituents on these matters, and no hard decision is necessary. Others get conflicting views, and they know that whichever way they vote, it may cost them dearly in the next elec-tion. Occasionally, members of Congress in this fix will try to be out of town when the matter comes up for a vote. You might think talection who come fis who won a close race in the last lection Who come from a man. sinal distric would be especia so farr to voce thc vay their constituents want. Research so far has shown that is the generally the case. There seem to be about as many independent-minded members of Congress from marginal as from safe districts. Perhaps it is because opinion is so divided in a marginal seat that one cannot please everybody; as a result, the representative votes on other grounds. The limit to the representative explanation is that public opinion is not strong and clear on most measures on which Congress must vote. Many representatives and senators face constituencies that are divided on key issues. Some constituents go to special pains to make their views known (these interest groups were discussed in Chapter 11). But as we indicated, the power of interest groups to affect congressional votes depends, among other things, on whether a legislator sees them as united and powerful or as disorganized and marginal. But when public opinion is strong and clear, members do respond to it. A recent study nicely illustrates this point. The researchers gave some legislators-but not oth-ers— information about public opinion in their districts toward a proposed spending bill. Those who received the information were much more likely to support the position favored by their constituents." This fits with broader studies that show that legislators are highly attuned to public sentiment in their districts, and try to vote in ways that reflect their constituents' views.35 Why does constituent opinion exert such a strong effect on member behavior? Because voting counter to the wishes of your constituents put members at grave risk of being voted out of office. If a member is repeatedly our of step with public opinion in his or her district, then challengers will leap on this pattern of votes in the next elec-tion. While most voters do not know how their member of Congress voted on various pieces of legislation, challengers will pounce and exploit votes taken by a member that many constituents would oppose. If a member is too liberal or conservative for their district, they will ypicaly be defeated. " Indeed, even one vote against the constit ency's wishes can be fatal, especially if it is on a highly salient picce of legislation such as Obamacare." Thr members who vote against the district's wishes pilly find themselves out of a job. Organizational View When voting on matters where constitueng interest u opinions are not vitally a stakeomembers of Congai respond primarily to cues provided by the colleagues. This is the organizational explanation of their votes. The principal cue is party no other factor explains as much of a member's behavior in office. Even when a Democrat and a Republican represent the same district, with the exact same voters, they will often vote differently (note the parallel to the power of party in shaping voters' views, as we discussed in Chapter 7). But do not think that members blindly adopt their parry's position on the issues wich little or no thought far from it. Nor does the power of party simply reflect the power of party leaders to whip members into adopting the party line. While leaders do have some powers to reward and punish members, & those powers are relatively constrained." Rather, the effect of party reflects different values of Democratic and Republican members. A member's party reflects his or her beliefs about how the government should be run in today's Congress, those who want to see a more active role for government are by and large Democrats, and those who want to see the government do less are typically Republicans. Further, Democratic (Republican) members of Congress have similar constituencies to other Democratic (Republican) members, and similar interest groups support them. It is the power of these other influences the constituents, supporting interest groups, and political values-that lead Democrats and Republicans to vote differently in Congress. Another influence-closely related to party-could be the view of an important ideological group within the House. A number of groups on both sides of the aisle represent various points of view in the various ideological debates in Congress. On the left are groups like the Congressional Progressive Caucus, and on the right, the House Freedom Caucus. But party and other organizations do not have clear positions on all matters. For the scores of votes that do not involve the "big questions," a representative or senator is especially likely to be influenced by the members of his or her party on the sponsoring committee. It is easy to understand why. Suppose you are a Democratic representative from Michigan who is summoned to the floor of the House to vote on a bill to authorize a new weapons sys-tem. You may well not understand the bill in any detail, since you are not a member of the authorizing committee. There is no obvious liberal or conservative position on this matter. How do you vote? Simple. You take your cue from Several Democrats on the House Armed Services Com-mitee that handled the bill. Some are liberal; others are nore moderate. If both liberals and moderates support the bill, you vote for it unhesitatingly. If they disagree, your rote with whichever Democrat is generally closest to your on political ideology. If the matter is one that affects your state, you can take your cue from members of your state's delegation to Congress. Attitudinal View Finally, members' own ideologies influence their behav-ior. This should hardly be surprising. As we saw in Chapter 7, political elites think more ideologically than the public. And as we saw above, it is a member's personal views-their ideology and values-that shapes why party is such a powerful influence. But, as we suggested at the start of this chapter, Congress has become an increasingly ideological organization, that is, its members are more sharply divided by political ideology than they once were. Today, all of Congress's most liberal members are Democrats, and all of its most conservative ones are Republicans. Why attitudes have hardened along ideological and partisan lines in Congress is a topic of much scholarly debate. Many different factors have contributed to Congress becoming more polarized, and we lack the space to discuss all of them. We discussed a crucial factor above— conservative Southern Democrats gradually became conservative Southern Republicans over the second half of the 20th century. Another factor is that those who are the most involved in politics (the activists) tend to be those with the strongest views, as we discussed in Chapter 7. Most Americans, unlike members of Congress, remain relatively moderate and nonideological. But among those who participate the most, there tends to be more division and ideological thinking. This division in the electorate supports congressional polarization. As we discussed in earlier chapters, members of Congress respond to those who participate. If those who vote, donate money, and volunteer for campaigns are more extreme, this influences the positions taken by members of Congress. Further, while most voters prefer compromise and bipartisanship, these activists do not; they instead want their members to stand firm for his or her ideological principles, adding further fuel to the polarization fire. This stands in stark contrast to most ordinary Ameri-cans, who are moderate, are largely nonideological, and like compromise and consensus. Unfortunately, most ordinary Americans are also not terribly politically interested, and are less likely to turn out and vote or to participate in politics in other ways. For example, one study found that, of political moderates, only about 20 percent are politically attentive, while the rest are largely disengaged from politics." Given this, members of Congress tend to pay these voters less heed, unless someone organizes them to make their voice heard. While overall district sentiment tion will change—if at all-in the years to come remains to be seen. 13-4 The Organization of Congress: Parties and Interests Congress is not a single organization; it is a vast and complex collection of organizations by which the business of Congress is carried on and through which members of Congress form alliances. Unlike the British Parliament, in which the political parties are the only important kind of organization, parties are only one of many important units in Congress (though they are one of the most important). Party Organizations The Democrats and Republicans in the House and the Senate are organized by party leaders, who in turn are elected by the full party membership within the House and Senate. The Senate The majority party chooses one of its members—usually the person with the greatest seniority—to be president pro tempore of the Senate. This is usually an honorific position, required by the Constitution so that the Senate will have a presiding officer when the vice president of the United States (according to the Constitution, the president of the Senate) is absent. In fact, both the president pro tem and the vis to resident usuallyasign phe tedious chore of presiding to hiunior senator. the real leadership is in the hands of the majoriy and minority leaders. The pines of taek of the majority leader is to schedule the business of the Senate, usualy in consultation wich the minority leade. A majority leader who has a strong personality and is skilled at political bat-gaining (such as Lyndon Johnson, the Democrat' leader in the 19505) may also acquire much influence over the substance of Senate business. A whip, chosen by each party, helps party leaders say informed about what the party members are thinking rounds up members when important votes are taken, and. attempts to keep a count of how voting on a controversial issue is likely to go. Several senators assist each party whip. Each party also chooses a policy committee comprising a dozen or so senators who help the party leader schedule Senate business, choosing what bills will be given major attention and in what order. For individual senators, however, the key party organization is the group that assigns senators to the Senatés standing committees: for the Democrats, the Steering and Outreach Committee; for the Republicans, the Committee on Committees. For newly elected senators, their political careers, opportunities for favorable publicity, and chances for helping their states and constituents depend in great part on the committees to which they are assigned. Achieving ideological and regional balance is a crucial-and delicate-aspect of selecting party leaders, making up important com-mittees, and assigning freshmen senators to committees. The House of Representatives The party structure is essentially the same in the House as in the Senate, though the titles of various posts are different. But leadership carries more power in the House than in the Senate because of the House rules. Being so large (435 members), the House must restrict debate and schedule its business with great care; thur leaders who manage scheduling and determine how the rules shall be applied usually have substantial influence. The Speaker, who presides over the House is ti most important person in that body and is cord by whichever party has a majority. Unlike the president pu tem of the Senate, this position is anything but honorth. for the Speaker is also the principal leader of the mine Sarty. Though Speakers as presides are exper be li. Speakes as pary lodes ar exped o s the pue to help pass legislation favored by their party. On helping his of her party, the Speaker has some important formal powers. He or she decides who should be recognized ro speak on the floor of the House, rules whether a motion is relevant and germane to business at hand, and decides (subject to certain rules) the committees to which new bills shall be assigned. He or she influences what bills are brought up for a vote and appoints the members of special and select committees. Since 1975, the Speaker has been able to select the majority-party members of the Rules Committee, which plays an important role in the consideration of bills. The Speaker also has some informal powers. He or she controls some patronage jobs in the Capitol building and the assignment of extra office space. Though now far less powerful than some of his or her predecessors, the Speaker is still an important person to have on one's side. In the House, as in the Senate, the majority party elects a floor leader, called the majority leader. The other party chooses the minority leader. Traditionally, the majority leader becomes Speaker when the person in that position dies or retires—-provided, of course, that his or her party is still in the majority. Each party also has a whip, with several assistant whips in charge of rounding up votes. For the Dem- ocrats, committee assignments are made and the scheduling of legislation is discussed in a Steering and Policy Committee chaired by the Speaker (or minority leader, depending on which party is in the majority on the com-mittee). The Republicans have divided responsibility for committee assignments and policy discussion between two committees. Each party also has a congressional campaign committee to provide funds and other assistance to party members running for election or reelection to the House. The effect of this elaborate party machinery can be crudely measured by the extent to which party members vote together in the House and the Senate. A party vote can be defined in various ways; naturally, the more stringent the definition, the less party voting will occur. Figure 13.3 shows party voting in the House of Representatives since the end of World War II. Most scholars say a party vote occurs when at least 50 percent of the Democrats vote together against 50 percent of the Republicans; this is the definition we use in Figure 13.3 (though some insist on a stricter definition, when 90 percent of Democrats vote against 90 percent of Republicans; by this defi-nition, there are obviously fewer party votes). Figure 13.3 shows a striking trend: since the 1970s, there have been more and more party unity votes as a fraction of all votes cast. That is, more votes (today, nearly 7 in 10 votes) are party unity votes. Given that political parties as organizations do not tightly control a legislator's ability to get elected, this high level of party voting is surprising Congressional members of one party sometimes do vote together against a majority of the other party, for several reasons. First, members of Congress do not randomly decide to be Democrats or Republicans; at least for most members, these choices reflect some broad policy agreements. By tabulating the ratings that several interest groups give members of Congress for voting on important issues, it is possible to rank each member of Congress from most to least liberal in many policy areas, including economic affairs, social issues, and foreign and military affairs. Democrats in the House and Senate are much more liberal than Republicans across nearly all issues. This has been true for many years, and as we discussed elsewhere in the chap-ter, the gap between Democrats and Republicans on the issues has been increasing. In addition to their personal views, members of Congress have other reasons for supporting their party's position at least some of the time. On many matters that come up for vote, members of Congress often have little information and no opinions. It is only natural that they look to fellow party members for advice. Furthermore, supporting the party position can work to the long-term advantage of a member interested in gaining status and influence in Congress. Though party leaders are weaker today than in the past, they are hardly powerless. Sam Rayburn reputedly told freshman members of Congress that "if you want to get along, go along." That is less true today, but still good advice. In short, party does make a difference-though not as much as it did at the turn of the 20th century and not as much as it does in a parliamentary system. Party affiliation is still the single most important thing to know about a member of Congress. Because party affiliation in the House today embodies strong ideological preferences, the mood of the House is often testy and strident. Members no longer get along with each other as well as they did 40 years ago. Many liberals and conservatives dislike each other intensely, despite their routine use of complimentary phrases. Caucuses Congressional caucuses are another set of important organizations in Congress. A caucus is an association of members of Congress created to advocate a political ideology or to advance a regional, ethnic, or economic interest. In 1959, only four such caucuses existed; by the early 1980s. there were more than 70. There are several types of caucuses in Congress. First, there are the ideological caucuses that unite members around a set of beliefs; examples of these include the Congressional Progressive Caucus on the left and the House Freedom Caucus on the right. Sec-ond, there are regional caucuses, which bring together members from a common geographic region to work together on issues of concern to that area. One example is the Northeast-Midwest Congressional Coalition, which brings members from 18 Northeastern and Midwestern states from both parties to discuss areas of common concern to their districts. Third, there are caucuses devoted to particular issues, such as the Congressional Diabetes Caucus, which seeks to address diabetes-related issues. Finally, there are caucuses that advocate for those from particular racial or ethnic groups; the most famous of these is the Congressional Black Caucus. The activity level of these caucuses varies widely, with some being very active and pressing an agenda on many issues, whereas others remain more behind the scenes. The Organization of Congressional Committees The most important organizational feature of Congress beyond the parties is the set of legislative committees of the House and Senate. Most of the power of Congress is found in the chairmanship of these committees, and their subcommittees. The number and jurisdiction of these committees are of the greatest interest to members of Congress because decisions on these subjects determine what groups of legislators with what political views will pass on legislative proposals, oversee the workings of agencies in the executive branch, and conduct investiga-tions. A typical Congress has, in each house, about to dozen committees and well over 100 subcommittees. Periodically, efforts have been made to cut the number of committees in order to give each a broader jurisdiction and to reduce conflict between committees over a single bill. But as the number of committees declined, the sumber of subcommittes rose, leaving matters much as they had been. Three kinds of committees exist: standing committees (more or less permanent bodies wich specific legislative responsibilities), select committees (groups appointed for a limited purpose, which do not introduce (gadain and which xit for only a few years), and joint committees on wiien both representatives and senions sene). An especially important kind of joint committee i the conference committee, made up of representatives and senators appointed to resolve differences in the senate and House versions of a bill before final passage. Though members of the majority party could in theory occupy ill he seats on all the committees, in practice they take the majority of the seats, name the chairperson, and allow the minority party to have the remainder of the seats. The number of seats varies from about 6 to more than 50. Usually the ratio of Democrats to Republicans on a com-mitte roughly corresponds to their ratio in the House or Sen-at. Sanding committees are more important because, with a few exceptions, they are the only committees that can propose legisla-tion by reporting a bill se out to the full House or Senate. Each member of the House usually serves mm app es in on two standing commit-tees, but members of the Appropriations, Rules, Ways and Means, Energy and Commerce, or Financial Ser-vices Committees are limited to one committee. Each senator may serve on two major committees and one minor commit-tee (see Table 13.5), but this rule is not strictly enforced. In the past, when party leaders were stronger, committee chairs were picked on the basis of loyalty to the leader. When this leadership weakened, seniority on the committee came to govern the selection of chairpersons. While the seniority system still largely governs which members become committee chairs, seniority is no longer sacrosanct. In 1971, House Democrats decided in their caucus to elect committee chairs by secret ballot; four years later, they used that procedure to remove three committee chairs who held their positions by seniority. Between 1971 and 1992, the Democrats replaced a total of seven senior Democrats with more junior ones as committee chairs. When Republicans took control of the House in 1995, Speaker Newt Gingrich ignored seniority in selecting several committee chairs, picking instead members who he felt would do a better job. In this and other ways, Gingrich enhanced the speaker's power to a degree not seen since 1910. Throughout most of the 20th century, committee chairs dominated the work of Congress. In the early 1970s, their power came under attack, mostly from liberal Democrats upset at the opposition by conservative Southern Democratic chairs to civil rights legislation. The liberals succeeded in getting the House to adopt rules that weakened the chairs and empowered individual members. Some of the key changes included electing committee chairs by secret ballot within the majority party, banning committee chairs from blocking legislation by refusing to refer it to a subcommittee, requiring public meetings in all committees and subcommittees (unless the committee has voted to close them), and electing subcommittee chairs by a vote of committee members. When the Republicans took control of the House in 1995, they made further changes. They eliminated some committees, and they also changed the powers of Anadolu Agency/Getty Image: IMAGE 13-9 Deputy Attroney General Sally Yates and FBI Director James Comey testify before Congress. committee chairs. Some of these reforms strengthened the power of the committee chairs for example, chairs were allowed to hire subcommittee staff while others limited chairs in other significant ways, such as imposing tem limits on committee and subcommittee chairs (three consecutive terms, or six years, and banning proxy voting (i.e, allowing the chair to cast an absent member's vote by proxy). The Senate has seen fewer such changes, in large part because individual senators have always had more power than their counterparts in the House. That said, in 1995, senators also imposed six-year term limits on their committee chairs and voted to elect chairpersons by secret ballot of the committee members. Despite these new rules, the committees remain the place where the real work of Congress is done. These committees tend to attract different kinds of members. Some, such as the committees that draft tax legislation (the Senate Finance Committee and the House Ways and Means Committee) or that oversee foreign affairs (the Senate and House Foreign Relations Committees), have been attractive to members who want to shape public policy, become experts on important issues, and have influence with their colleagues. Others, such as the House and Senate committees dealing with public lands, small business, and veterans' affairs, are attractive to members who want to serve particular constituency groups. 43 For example, a member from a district with a great deal of agricultural land might want to serve on the House Committee on Agriculture, or a member from a district with a large military base might want to serve on the House Armed Services Committee. Doing so will allow those members to gain expertise on policy areas relevant to their districts, as well as to provide benefits to their constituents. Such knowledge and benefits in turn further a member's reelection chances.* Indeed, many members choose to serve on committees that are relevant to their districts' economic interests. The Organization of Congress: Staffs and Specialized Offices In 1900, representatives had no personal staff, and senators averaged fewer than one staff member each. By 1979, the average representative had 16 assistants and the average senator had 36; the total number of individuals employed by Congress as staff persons was nearly 27,000 (that number includes member's personal staffs, plus committee staffs, the staff for support agen-cies, and other miscellaneous staf). Today, despite the world, and the federal government, growing more com-plex, Congress has reduced in at all staf to around 19,600, a decline of more than one-quarter.S Starting in the mid-1990s, Congress began to reduce its staff as a cost-cutting measure. Later in the chapter, we will see that some have argued that was a penny wise but pound foolish decision. Regardless of the number of staff, they perform a variety of important tasks. Some staff persons work in a member's home district, meeting with constituents and fulfilling requests for assistance with the government (this is a component of the incumbency advantage we discussed earlier in the chapter). The legislative function of congressional staff members is also important. With each senator serving on an average of more than two committees and seven subcommittees, it is virtually impossible for members of Congress to become familiar with the details of all the proposals that come before them or to write all the bills that they feel ought to be introduced. The role of staff members has expanded in proportion to the tremendous growth in Congress's workload. The orientation of committee staff members differs. Some think of themselves as and to a substantial degree they are—-politically neutral professionals whose job it is to assist members of a committee, whether Democrats or Republicans, in holding hearings or revising bills. Others see themselves as partisan advocates, interested in promoting Democratic or Republican causes, depending on who hired them. Those who work for individual members of Congress, as opposed to committees, see themselves entirely as advocates for their bosses. They often assume an entrepreneurial function, taking the initiative in finding and selling a policy to their boss—a representative or senator-who can take credit for it. Lobbyists and reporters understand this completely and therefore spend a lot of time cultivating congressional staffers. The increased reliance on staff has changed Congress, mainly because the staff has altered the environment within which Congress does its work. In addition to their role as entrepreneurs promoting new policies, staffers act as negotiators: Members of Congress today are more likely to deal with one another through staff intermediaries than through personal contact. Congress has thereby become less collegial, more individualistic, and less of a deliberative body.* In addition to increasing the number of staff mem-bers, Congress also has created a set of staff agencies that work for Congress as a whole. These have come into being in large part to give Congress specialized know!-duge equivalent to what the president has by virtue of his or her position as chief of the executive branch. One of the, the Congressional Research Service (CRS), is part of the Library of Congress and employs about 600 people; it is politically neutral, responding to requests by members of Congress for information and giving both sides of argu-ments. The Government Accountability Office (GAO), once merely an auditing agency, now has about 3,000 employees and investigates policies and makes recommendations on almost every aspect of government; its head, though appointed by the president for a 15-year term, is very much the servant of Congress rather than the president. The Congressional Budget Office (CBO), created in 1974, advises Congress on the likely impact of different spending programs and attempts to estimate future economic trends.