ó T .... zJJo·' -;Jnio:r.:::; and A1"nerican Eledic:u~: The Politics of Voter IVlobilization in the 2000 Presidential Election Donald W. Beachler ithaca College In 2000, labor unions attempted to enhance their polii"ical iJ~t7uence by il1- creasing voter tumour mnong union manbers ill support of Al Gore. This strat­ egy'l,vas thought to be especially effective in a polity where voter participation is so low. Labor'a political cfforta helped Core to a populm' vote victory, but the union political strategy failed because of the dynamics oj the Electoral College and low union density in many regions of the United States. The Politics of Low Turnout Elections As is well known, American politics is characterized by very low rates of voter participation. The causes of low voter turnout have been fiercely, if inconclusively, debated (Piven and Cloward 1988; Putnam 2001). Less attention has been paid to the political consequences of elections in which so few voters participate. Because of the very low rates of voter pa.rticipation, relatively smnll segments of the population can have great influence over elections if they vote at a high rate. In presidential elections turnout has ranged from 49% to 55% of the eligible electorate in recent years. Off year national elections have drawn about 36% to 39% of the voting age population to the polls. In recent elections, the popular vote for the U.S. House of Rep­ resentatives has been divided near1y evenly between the two parties (Rimmerman 2001). Republicans have retained control of the House since 1994 by winning 1 % to 2% more of the popular vote than the Democrats received. Because of low voter turnout, the votes of .36% to .72% of the eligible electorate can be vital in determining the outcome of elections. Because small increases in voter participation can make a substantial difference in election results, interest groups have greater incentives to mobilize their members into the electorate in a poBty characterized by very low rates of voter participation. This article presents a case study of organized labor, an interest group that has been losing membership and political influence for decades. In recent elections, labor has attempted to compensate for its declining ranks by mobilizing its members to go to the polls in greater numbers and to increase the percentage of union members voting for union endorsed candidates. The article investigates labor's attempt to mobilize its members in a demobilized polity. 23 The Politics of a Declining Union Movement The decline of organized labor in the United States has been much documented and thoroughly debated. The loss of millions of jobs in unionized industries cost organized labor millions of members. Ameri­ can labor law, which is hostile to union organizing, has exacerbated such losses, and they have been compounded further by the reluctance of com­ placent unions to commit resources to organizing new members. It is not possible for this article to delve deeply into the debate over the causes of the decline of labor unions in the United States (Goldfield 1987; Geoghegan 1991). It is important to note that in the 1950s, labor unions represented about one-third of the American workforce. By 2000, roughly 13% of U.s. work­ ers belonged to labor unions. Because of a surge in unionization of pub­ lic employees during the last three decades of the twentieth century, by 2000 just over 9% of private sector workers belonged to labor unions (Greenhouse 2001). Labor suffered a number of legislative defeats even during the osten­ Sibly friendly Clinton admmistratIon. A filibuster in a JJemocratic con­ trolled Senate defeated legislation banning striker replacement. The AFL­ CIO lost bitter battles in opposition to the Clinton administration over the North American Free Trade Agreement (NAFTA) in 1993 and Perma­ nent Normal Trade Relations (PNTR) with China in 2000. While unions did have some political successes during the Clinton administration, they were clearly perceived to be a declining political force (Heberlig 1999). Labor was dealt a major political blow when Republicans won ma­ jorities in both houses of Congress in the 1994 elec lions. MallY Ul llUll members were disaffected by the Clinton administration's support of the NAFTA agreement and the president's embrace of gun control mea­ sures (Reich 1998). Voters from union households constituted just 14% of the electorate. The low turnout of union and other blue-collar work­ ers was one reason for the Republicans' historic victory in 1994. By the 1990s, the Republican Party was much more hostile to labor unions than it had been in the administrations of Presidents Eisenhower, Nixon, and Ford. President Nixon, for example, believed that the social conservatism and nationalism of many union members would lead them to support him as he constructed a silent majority of Americans who objected to the cultural and socialliberalisrn of elements wllhin lhe Dt:UlU­ cratic party. Nixon did not support the desire of unions for changes in American labor law that would make union organizing easier, but he was prepared to accept the status quo in American labor relations in pursuit of his alliance with union members (Safire 1975; Fry-mer and Skrentny 1998). 24 In the 1980s, the Reagan administration was distinctly hostile to the interests of organized labor, signaling a shift in Republican attitudes to­ ward labor. The l\.eagan administration had an ideological anti-union­ ism that brought a missionary zeal to union busting. Drav,Ting support from the conservative Sunbelt regions of the country, Reagan's election signaled a distinctly hostile attitude toward union labor (Gross 1995). A strongly anti-labor faction dominated the Republican Party that as­ sumed control of the House of Representatives and Senate in January of 1995. The Republicans wished to deprive unions of their political funds by passing "paycheck protection" legislation that would have made it dif­ ficult for unions to raise money to spend in political campaigns. The GOP also attempted to lobby the NLRB to restrict what it perceived as the pro­ union bent of the NLRB under the appointees of Bill Clinton (Dark 1999). In 1995 John Sweeney, president of the Service Employees Interna­ tional Union (SEIU), mounted an insurgent campaign that ousted AFL­ CIa president Lane Kirkland. Sweeney's challenge was based on the notion that labor must aggressively organize new members, especially service workers, to reverse the decades long decline in membership that organized labor had experienced. Sweeney also argued that labor must find new ways to expand its political influence. In pursuit of his goals, Sweeney aggre~~ively sought tu Llirect uniun fUIlIi::; tuwaru urgani:dng efforts, to expand labor's alliances with progressive interest groups, and to inject union funds into campaigns in an effort to undo the damage done in the 1994 elections (Sweeney 1996). Faced with a hostile Republican Congress, unions attempted to revive their political influence in the 1996 election with a two-pronged approach. Labor spent $25 million on an independent expenditure campaign of media ads that attacked selected Republican representatives for advanc­ ing a radical right wing agenda (Heberlig 1999). In 1996, labor also ex­ panded its efforts to mobilize and educate its own members about candi­ dates and issues. In the closing weeks of the 1996 campaign, the Repub­ lican Party respundeLl with all $8 milliuIl uullar dU verti::;ing campaign that condemned the political efforts of "union bosses" (Hershey 1997). The renewed electoral efforts of unions met with mixed success. Re­ publicans retained control of the House of Representatives, but labor's efforts were credited with defeating several Republican incumbents. Twelve of the 24 Republican freshmen targeted by the AFL-CIO lost their reelection bids in 1996. Congressional elections scholar Gary Jacobson estimated that without the independent expenditures campaign of orga­ nized labor only two of the 24 would have been defeated (Jacobson 1997; Jacobson 1999). Union voter turnout increased from 19% to 23% even though there was a sharp decrease in the rate of voter participation in the electorate as a whole. 25 The percentage of union household voters who supported Bill Clinton in his successful reelection effort increased slightly from 55% in 1992 to 59% in 1996 (New York Times, 2000). The AFL-CIO also believed that its ads attacking the Republican Congressional agenda were partially re­ sponsible for the moderation in the agenda of the 104th Congress, which increased the minimum wage, enacted a provision facilitating health in­ surance portability, and reversed cuts to Medicare and education. Labor unions were given considerable credit for the Democrats' sur­ prise showmg in the 1998 congressional elections. With the exception at 1934, the party of the president had lost House seats in every off-year election in the twentieth century. The president's party has often suf­ fered exceptionally large losses in the sixth year of a president's term. In 1998, Democrats were further burdened with a presidential scandal that would shortly lead to the impeachment of President Bill Clinton. De­ spite predictions of political disaster, the Democrats held their 4:5 Senate seats and gained five seats ih the House of Representatives. Labor fo­ cused on mobilizing its members to vote, and the percentage of the elec­ torate that resided in a union household increased from 14% in 1994 to 22%. in 1998 (Heberlig 1999). Sixty-four percent of the union electorate reported voting for the Democratic candidate for the U.S. House of Rep­ resentatives as opposed to 60% in 1994 (New York Times 1998; Heberlig 1999). The marginal improvement in the vote of union members for Demo­ cratic House candidates and the substantial improvement in voter turn­ out among unionists and their families gave labor reason to hope that it could be even more successful in 2000. 2000: Labor Attempts to Elect Al Gore The AFL-CIO endorsed Al Gore in the fall of 1999. The fecleration's support· for Gore came at a time when he appeared vulnerable to the challenge of former New Jersey Senator Bill Bradley. Labor provided vital assistance to Gore as he extinguished Bradley's quest for the nomi­ nation with victories in the Iowa caucus and the New Hampshire pri­ mary. During the nominating process, some unions, most notably the Teamsters and the United Auto Workers, withheld support from Gore because of their disagreement with the trade policies of the Clinton ad­ ministration~ Even after Gore had secured the Democratic nomination, some labor leaders publicly flirted with the notion of remaininls neutral or endorsing Ralph Nader's Green Party candidacy (Corn 2000). After the Democratic convention in August 2000, however, virtually the entire labor movement united to support Gore's quest for the preSidency (Beachler 2001). 26 Table 1 Labor Union Household Voting in Presidential Elections, 1992-2000 Presidential Vote of Union Members % of electorate residing in Democratic union households Democrat Republican Other* Margin** 1992 19 55 24 21 31 1996 23 59 30 9 29 2000 26 59 37 3 22 * The other vote was for Ross Perot in 1992 and 1996 and primarily for Ralph Nader in 2000. ** The Democratic margin was calculated by subtracting the Republican percentage of the union household vote from the Democratic percentage of the union household vote. Table 1 shows the voting patterns.of union household members in the presidential elections of 1992,199(1, and 2000. It also lists the percentage of the electorate residing in union households. The percentage of the electorate residing in union households increased substantially in each of the three elections examined in Table 1. The focus on increasing labor's share of the total electorate was very successful as union turnout increased from 19% in 1992, to 23% in 1996, and 26% in 2000. Table 1 illustrates that, while labor union members supported Demo­ cratic presidential candidates by large margins, unions cannot be said to have had unqualified political success in convincing their members to vote for the Democratic nominee. The historical pattern of many union household residents voting for the Republican nominee persisted dur­ ing this period. In 1992, Bill Clinton took 55% of the union household vote to 24% for Bush and 21 % for Perot. In 1996, Perot was a weakened political force and received just ~% of the union household vote, while Clinton took 59% and Dole won 30%. Clinton's margin among union voters was re­ duced by 2% from 1992 to 1996. In 2000, labor made a major political effort on behalf of Al Gore's bid for the presidency. In the 2000 election, third party candidates were less of a factor. Despite a platform that included opposition to trade treaties strongly opposed by many unions, and a call for labor law reform that would facilitate union organizing, Ralph Nader scored only 3% of the vole among union households. Nader's showing among the union elec­ torate was only marginally better than his percentage among non-union voters. 27 Al Gore won the same 59% of the union electorate that Clinton re­ ceived in 1996. George W. Bush improved on Bob Dole's performance among union voters and received 37% of the union vote. In three con­ secutive presidential elections the Democratic margin over the Republi­ can candidate among union voters has actually declined. The Demo­ cratic advantage over the Republican candida~e declined by 9% from 1992 to 2000. Put another way, in 1992, the percentage of the electorate that was from a union household and voting Democratic was 10.45% and the percentage of the electorate that was union household and vot­ ing Republican was 4.56%. In 2000, 15.34% of the electorate was union household and voting Democratic, while 9.62% of the electorate was union household and voting Republican. From 1992 to 2000, Republicans actu­ ally gained support among the union electorate at a faster rate than Demo­ crats. By increasing the union percentage of the electorate in 1996 and again in 2000, the labor movement clearly aided the Democratic presidential nominee. The impact of the labor vote would be even greater if the unions were able to persuade more of their members to vote for the union en­ dorsed candidates. It should be noted that the inability to turn out a higher percentage of voters for the Democratic candidate is not a new problem for unions. Gore's 59% of the vote was exceeded by th4? Demo­ cratic nominee in only three presidential elections since World War II: 1948, 1960, and 1964 (Leroy 1990). Table 2 demonstrates that labor was of vital importance to Gore's ef­ forts in several key states. Two of the most heavily contested \;tates in the election were Pennsylvania and Michigan. While unions made a major effort to mobilize their members for the Democratic ticket, in both battle­ ground states, they faced difficult challenges. The Clinton-Gore campaign's embrace of trade policies that imperiled manufacturing jobs dampened enthusiasm for the Democratic ticket among some union members and union officials (Greenhouse 2000a). The Bush campaign believed that because there were no real differ­ ences between Bush and Gore on the trade issues that were so important to autoworkers, steelworkers, and other workers subject to international wage competition, the Republicans could appeal to the conservative so­ cial values of some union workers (Dreazen 2000). Many union mem­ bers have conservative views on social issues. For example, in 199434% of union voters in Pennsylvania supported the political goals of the Na­ tional Rifle Association (NRA) (Clark and Masters 2001). In 2000, the NRA countered labor's efforts among blue-collar workers, by distribut­ ing literature arguing that Gore was a threat to the rights of gun owners (Eilperin and Edsall 2000). In response to the NRA campaign, labor dis­ tributed literature in several battleground states that asserted 1/ Al Gore 28 Table 2 Labor Voting in Selected States in 2000 Presidential Election % of electorate Union Non-Union residing in union households households households Democrat Republican Democrat Republican Michigan 43 62 35 42 55 Penn~ylvania 30 65 32 44 54 Wisconsin 32 55 39 44 52 Minnesota 30 56 35 44 50 Washington 27 60 33 46 47 Oregon 22 58 38 43 51 California 28 5.9 35 50 46 Illinois 33 62 36 52 46 New Jersey 32 64 32 52 44 Maryland 24 65 33 54 42 Nevada 33 51 43 42 53 West Virginia 33 51 47 43 54 Doesn't Want to Take Away Your Gun. But George W. Bush Wants to Take Away Your Union" (Greenhouse 2000b). The principal focus of the union effort in several key states was an intensive get out the vote effort that emphasized mobilizing union mem­ bers and their families to come to the polls and vote for Gore. In 2000, labor expanded its efforts to build voter turnout by having union activ­ ists contact other members of their unions. In 1998, when labor used what AFL-CIO political director Steve Rosenthal called, liThe Program," 75% of the union members contacted voted for the union endorsed can­ didate. In 2000, more than 500 organizers were trained at AFL-CIO head­ quarters and dispatched to 25 states targeted by the AFL-CIO. Speaking of the AFL-CIG' s massive get out the vote effort, Rosenthal claimed, "Ba­ sically, it is a throwback to what we were doing in the 1930's and 40's when we were at our peak. It is a culture change" (Dreyfus 2000). Table 2 demonstrates that labor was able to meet its goals in Michigan and Pennsylvania. In Michigan, 21.5% of the workforce belonged to unions in 1999. In the 2000 presidential election, 4:3% of Michigan voters resided in a union household. Twenty-seven percent of the voters in 29 Michigan Wf're actual union mf'mbf'fs. V\Thf'n Wf' consiof'r thM not f'very member of the electorate is employed, it is obvious that unions turned out their members in numbers far greater than their percentage of the voting age population. It is not possible to determine precisely what portion of the eligible electorate was in the workforce and what percent­ age of union members were employed in November 2000. Still it is clear that union over-representation was quite substantial. Not only did labor turn out its members in Michigan, but they voted for Gore by a 64% to 32% margin. Gore also won 59% of the vote among those voters who resided in a union household, but were not themselves union members. The strong labor turnout in a state where the most re­ cent U AW contract with thp rl11tomohilp m::Jnufacturers made election day a holiday, enabled Gore to win Michigan by a margin of 5%. In the battleground state of Pennsylvania, labor started with a some­ what smaller base, as 17.4°/;'" of Pennsylvania workers belonged to unions in 1999. Residents of labor union households 'accounted for 30% of the electorate, with union members themselves constituting 17% of the vot­ ers. vVhile the union voter turnout equaled the percentage of the labor force that belongs to unions/ it is again important to recall that not all of the electorate is employed. In Pennsylvania/ 19% of the voters were over the age of 65 and another 9% were between 60 and 64 years of age. Pre­ sumably a significant portion of this population was not employed. While the data unfortunately do not permit a predsf' c::Jkll1anon of thp df'3rf'f' that labor was over-represented in the electorate in Pennsylvania, it is clear that union members turned out at a higher rate than all eligible voters. In Pennsylvania/ union members voted for Gore by a margin of 67% to 29%. Those who resided in a union household/ but were not union members themselves supported Gore by a 62% to 36% margin. Labor's vigorous efforts in Pennsylvania delivered another key state to Gore. Table 2 also indicates that unions were crucial to Gore's victories in Minnesota, Oregon, Wisconsin, and Washington. In each state, Gore won the union household -vote, but lost the non-union household vote. It is also reasonable to assume that union voters wl"rp oVl"r-rl"prf'sf'ntf'd in the electorate in each of these states. Labor was clearly harmed by the Electoral College system because in some states with a huge union turnout, Gore won by large margins. For example, in New York, 39% of the electorate reported living in union households. Gore won the union electorate by 65% to 30% and also cap­ tured the non-union electorate by a 57% to 39% margin. Union voters helped Gore to his remarkable 1.7 million popular vote win in New York. (Gore's 60% to 35% victory in New York was slightly better than Bush's 59% to 38% win in Texas). In New Jersey, 32% of the electorate resided in 30 union households and these voters bv a tv,iO to one margin. • u The Democratic candidate also carried the non-union electorate by 8%. Similar pattems occurred in California, Maryland: and minois where Gore won both the union and non-union electorates. Core's very strong shmv­ ing in union states meant that he won by very large popular vote mar­ gins in these states. Gore's substantial margins of victory in these states enabled him to win the popular vote nationally. Unfortunately for Gore, and. fur labor uniuns in the United States, the candidate with the most popular votes does not necessarily win the election. The Regional Structure of the American Labor Movement Labor's ability to assist Gore was further restricted by the geographi­ cal disparities in labor union density in the United States. Table 3 lists the states won by Gore and Bush and the percentage of the work force in each state that belonged to labor unions. Gore won only two states, Ver­ mont and New Mexico, where the union density percentage was signifi­ cantly below the national average of 13.9%. Vermont is an unusual state that elects Congress' only avowed democratic SOCialist, and has moved leftward as a result of population influx (Barone and Cohen 2002). In New Mexico, an electorate that was 32% Latino aids Democrats. In 2000, Gore won 66% of this Latino vote as he eked out a 500 vote victory over Bush in New Mexico. Gore also won the District of Columbia, which has a union density of 13.1 %, but the huge black electorate in the nation's capital aided him there. Bush was able to win several states where the union percentage was above the national Bush won several high union density west­ ern states (Alaska, Nevada, and Montana), where issues such as land use and gun control have alienated a majority of voters from the national Democratic Party (Barone and Cohen 2002). Bush also won in some high union density states such as Ohio, Indiana, and Missouri, which are per­ ceived to be more culturally conservative than the Northeast and West Coast states where Gore ran well. Finally, Bush won West Virginia, one of the most Democratic states in the country in recent presidential elections. Prior to 2000, West Virginia had voted Democratic in eight of ten presidential elecbons, a record of Democratic fealty exceeded only by Minnesota. Bush's win in vVest Vir­ ginia illustrates the difficulties unions have in maintaining a coalition with environmentalists and liberal social issue groups in the Democratic party. In West Virginia, the union vote was essentially a dead heat. Union vut~rti were turn~d off by Gore's environmentalism, which they feared would cost union mining jobs. West Virginia is also a morally conserva­ tive state wher~ there is strong opposition to gun control. Furthermore, 31 Table 3 Gore and Bush States by Union Density States Won Union by Bush Density Alaska 20.4 Nevada 19.5 Ohio 17.9 IndIana 15.7 Missouri 15.3 Montana 15.3 West Virginia 15.2 Kentucky 11.6 Alabama 11.0 New Hampshire 10.6 Kansas 9.7 Colorado 9.4 North Dakota 9.3 Idaho 9.2 Wyoming 1.).1 Nebraska 8.8 Oklahoma 8.1 Louisiana 8.1 Tennessee 7 . .5 Arkansas 7.5 Georgia 7.3 Arizona 6.7 Virginia 6.6 Florida 6.5 Utah 6.4 Mississippi 6.2 Texas 6.0 South Dakota 6.0 South Carolina 3.5 North Carolina :1.2 32 States Won Union by Gore Density New York 25.3 Hawaii 23.2 Michigan 21.5 Washington 20.7 New Jersey 20.5 Minl1esola 19.3 Connecticut 18.2 Wisconsin 18.1 Illinois 18.0 Rhode Island 17.4 Pennsylvania 17.4 Maine 17.1 California 16.6 Massachusetts 16.2 Oregon 15.2 Maryland 15.0 Iowa 13.8 Delaware 13.8 District of Columbia 13.2 New Mexico 9.9 Vermont 9.7 .. National percentage of union workers = 13.9% **Gore states' average union density = 17.1% ***Bush states' average union density = 9.91% steel jobs in northern \-Vest Virginia were threatened b"y the free trade policies of the Clinton administration. Given the demographies and po­ litical climate in V\Test Virginia in 2000, Bush was able to split the union vote, carry the state, and in effect, win the presidency (Barone and Cohen 2002). Bush mainly carried states where the union density was below the national avcragc. Thc states Bush won had an Llvcrage union density of 9.9%. Average union membership in Bush states was about 60°,{) of the average union density of 17.1 % in states won by Gore. If labor is to have more influence in presidential politics, it will need to expand its mem­ bership in states where a disproportionate share of the workforce is un­ organized. The significance of unions in state electoral outcomes in the 2000 presidential election is evident in Table 4. Table 4 Union Density and Race in the 2000 Presidential Elections Gore Popular Vote by State Black Population Latino Population Union Density Constant Adjusted R2 .336** .106 .372 .188 .116 .194 .962*** .180 .622 28.13~** 3.128 .384 The numbers in each variable column are the regression coefficient, standard error and the standardized regression coefficient. ** si):Jnificant at .01 level *** significant at .001 level Table 4 is a regression that seeks to measure union influence on state results in the 2000 presidential election. The dependent variable in the regression was the percentage ot the popular vote received by AJ Gore in each state. To control for racial differences in voting, which are very significant in the United States, independent variables were included for 33 the African-American and Latino percentages of each state's population. The third independent variable was the percentage of the state's labor force that belonged to a labor union. The results of the regression presented in Table 4 indicate that union density is an important predictor of a state's percentage voting for Gore. While African-American population was positively corrdated with the vote for Gore, it is a large union workforce lhal conlribuled IIlU::;t ::;jg-nifi­ cantly to a larger vote for Gore. When we control for the impact of race in each state, union density was a strong positive predictor of support for Gore in the 200U election. The regression in Table 4 demonstrates the price that labor has paid for being a highly regionalized factor in American politics. Based on the 2000 election, it is clear that an expanded union membership would ben­ efit the Democratic Party. Table 4 also demonstrates that if labor wishes to hp ;m pven stronger influence in presidential politics, it must expand union density in at least some states where it has been weak. Conclusion In many senses, labor's massive voter mobilization strategy was a success. It provided Al Gore with a popular vote victory in the nation. Unions helped Gore win the crucial battleground states of Michigan and Pennsylvania and were also essential to Gore's wins in 'Minnesota, Wis­ consin, Washington, and Oregon. Labor's turnout strategy was limited by several factors. First there is a rather seVPTP 3PogrPlphicailimit to labor's ability to influence the re­ sults of presidential elections. Labor's influence is, for the most part, concentrated in the Northeastern, upper Midwestern and Pacific coast states of the country. In most of the so-called Dush 1/ red zone" of states in the South, Great Plains, and Rocky Mountain West, union density is well below the national average. With the exception of California, all of the states gaining House seats, and therefore electoral votes, as a result of the 2000 census were carried by Bush in 2000. Unless they can expand union membership in low density states, unions' electoral power will be further circumscribed in future presidential elections . . Union turnout strategy is further limited by the inability of labor to convince a grestpT nllmber of union members and members of their households to vote for union endorsed candidates. While it is true that empirical research has indicated that union members are far more likely to vote for Democrats than are other workers, unions must do a beller job of convincing more unionized workers to vote for union endorsed candidates. The Democratic margin over Republicans in union house­ hold voting actually declined in each of the three presidential elections 34 between 1992 and 2000. If AFL-CIO political director Rosenthal is cor­ rect in his assertion that 75% of union members contacted by activists using "The Plan" vote for the union endorsed candidate, then we must conclude that a significant number of union workers are not being reached hy union po1iiical activists Without revprsing thi.s trend, labor will have even less opportunity to determine the outcome of presidential elections. Union political power would be enhanced by electoral reforms that are unlikely to occur. The Electoral College's "winner-take-all" tradition weakens union political influence because in many states union turnout efforts increase the margin by which the Democratic candidate wins the state. This effect is compounded by the fact that many of the states in which union organizing is easiest, such as California and New York, have been among the strongest Democratic states in recent elections. In the Electoral College, of course, the margin of victory in a state is irrelevant. The abolition of the Electoral College and the implementation of a direct popul, November 2, A28. 2001. "Unions Hit Lowest Point in Six Decades," The Ne,o York Times, February 12, A12. Gross, James A. 1995. Broken Prmnisp: The SlIh71P1'.c;inl1 nf flu.> IJS Lahor Relatio11s Policy, 1947-1994. Philadelphia: Temple University Press. Heberlig, Eric S. 1999. "Coordinating Issues and Elections: Organized Labor in the Republican Era." American Revie'w of Politics 20: 163-80. Hershey, Marjorie Randon. 1997. "The Congressional Elections." In The Election of1996: Reports and Interpretations, ed. Gerald Pomper. Chatham, NJ: Chatham House Publishers. Jacobson, Gary. 1997. "The 10Stil Congress: Unprecedented and Unsurprising." In The Election of1996, ed. Michael Nelson. Washington, DC: Congressional Quarterly Press. , 1999. "The Effect of the AFL-CIO's Voter Education Campaigns on the 1996 House Elections." Journal of Politics 61(1): 185-94. Leroy, Michael H. 1990. "The 1988 Elections: Reemergence of the Labor Bloc Vote." Labor Studies Jour11aI15: 5-32. Meyerson, Harold. 2001. "California's Progressive Mosaic." The American Pros­ pect 12 (June): 17-23. New York Times. 1998. II A Look at Voting Patterns of 115 Demographic Groups in House Races." (November 9): A20. New York Times. 2000. "Exit Poll." Week in Review section, (November 12): 4. Piven, Frances Fox, and Richard A. Cloward. 1998. Whil Americans Don't Vote. New York: Pantheon. - Putnam, Robert. 2001. Bowling Alone: The Collapse ami Revival of American Com­ mU11lty. New York: Fantheon. Reich, Robert. 1998. Locked il1 the Cabil1et. New York: Vintage. Rimmerman, Craig. 2001. The New Citizenship: Unconve11tional Politics, Activism, and Service. Boulder: WestVIew Press. Safire, William. 1975. B~fore the Fall: An Inside View of the Nixon White House Before Watergate. Garden City: Doubleday & Co. Sweeney, John. 1996. America Needs a Raise. Boston: Houghton Mifflin. 37