Center for Immigration Studies. Center for Immigration Studies. Latinos and the 2002 Election. Republicans Do Well When Latinos Stay Home

Transcription

1 Backgrounder January 2003 Latinos and the 2002 Eection Repubicans Do We When Latinos Stay Home By James G. Gimpe Despite Repubican proposas for increased immigration, amnesty for iega aiens, and new guestworker programs, there is no evidence of a Repubican surge among Latino voters in the 2002 mid-term eections. The principa findings of eection-day pos from 10 states with key races for U.S. Senate or governor are: The Latino vote for GOP Senate candidates was simiar to prior years, at about one-third; gubernatoria candidates fared better, at cose to one-haf. But Latinos who voted in 2002 had higher income and education eves than the Latino eectorate as a whoe. Turnout of ower and midde income Latinos was much ower in 2002 than in Latino voters who identify themseves as independents are, in fact, ikey to vote Democratic. The fact that many of these independents stayed home in 2002 heped Repubicans. There is no Latino voting boc, as such after controing for party identification, income, and education, there is no difference between Latino voting and the voting pattern of non-hispanic whites in either the Senate or gubernatoria races of This is not true of African Americans, who are a distinctive voting boc even after controing for education, income, and party identification. Foowing the 2002 mid-term eections, the strong Repubican showing brought egions of GOP pundits, consutants, and sympathetic journaists forward to crow about the on-going reaignment of Latino voters. Surey the big victories in Latino-heavy states ike Texas (governor, U.S. Senate), Coorado (governor, U.S. Senate) and Forida (governor) were ampe proof that the misceaneous Latino outreach efforts undertaken by the GOP and its candidates had paid off. At first the hype was easy to beieve. It s exciting to think that the traditiona poitica science wisdom about the stabiity of partisan identity and poitica behavior is a wrong (Green, Pamquist and Schicker 2002) that highy paid consutants reay did earn their pay by exercising this magica power to manipuate mass poitica preferences; and that poitica consutants who sod candidates on their power to persuade were not just bowing smoke, but that egions of voters were actuay waiting to be swayed by the next campaign advertisement or catchy sogan. Then the facts began to ro in. In the most competitive U.S. Senate races, Repubicans hardy gained any ground among Latinos, party due to Democratic mobiization efforts. And modest improvements in the Latino GOP vote over the 2000 eection came as the resut of ow turnout and demobiization, not poitica conversion. Since the facts speak so ceary, it made many wonder whether the enthusiastic caims about Latino conversion were merey spin, or whether the purveyors of this myth actuay beieve it to be true. James G. Gimpe is Professor of Government at the University of Maryand, Coege Park. He can be reached at jgimpe@gvpt.umd.edu. Dr. Gimpe wishes to thank 1Professor Daron R. Shaw of the University of Texas, Austin, for graciousy sharing the Fox ews poing data.

2 Consider the resuts from the 10 Fox ews eection-day pos from ten states with key U.S. Senate or gubernatoria races. 1 The states incuded in the po are: Texas, Forida, ew Jersey, ew Hampshire, Arkansas, Coorado, Georgia, Minnesota, Missouri, and South Dakota. These pos hardy represent the nation, but are ikey to represent that segment of the 2002 eectorate that confronted at east one competitive eection (governor or U.S. Senate). The resuts for Democratic and Repubican voting in Figures 1 and 2 show that the share of the Latino vote that went to GOP candidates running for U.S. Senate was no more impressive than in previous years, running about one-third. On the other hand, the Hispanic vote for Repubican Governors was an impressive 46 percent, ostensibe evidence that Repubicans have made in-roads. Of course we shoud keep in mind that a number of these gubernatoria races were not competitive, incuding Coorado where incumbent Bi Owens coasted to an easy reeection. But et s consider the socioeconomic profie of the Latino voters who did show-up. Income and Education The cross tabuation of income by race/ethnicity reported in Tabe 1 ceary indicates that the Hispanics who voted in the 2002 contest had an unmistakaby upper income and higher education profie. They were not average Latino voters; in fact, an unusuay high proportion of the 2002 eectorate coud count themseves among the Latino eite. That the Repubicans woud do we among weathy and weeducated Latinos who turned out in a mid-term eection is not the east bit surprising. But it is hardy evidence that the Repubicans are gaining ground. What it demonstrates is that Repubicans do we when ower and midde income Latinos stay home, or ony the most affuent vote. For instance, Tabe 1 shows that a whopping 24 percent of Latino voters showing up at the pos earned above $75,000 about the same proportion as non-hispanic white voters, and far weathier than the African Americans who voted. Yet it doesn t require intimate famiiarity with the 2000 census to know that in the genera (citizen) popuation Hispanic earnings ag we beyond the earnings of non- Hispanic whites. Latino voters in the 2002 eection were far from representative of the broader Hispanic community of eigibe voters. The education profie is amost as distinctive (see Tabe 1). Among Latino voters in 2002, 32 percent had four-year coege degrees (or more). This compares with 45 percent for whites, and 28 percent for African Americans. But we know that in the genera popuation, Latinos of voting age ag far behind Ango whites in the proportion who have finished coege. The concusion is cear: the 2002 Latino vote hardy represented the Latino citizen popuation, much ess the Latino popuation in genera (incuding permanent residents, iega aiens, and other non-citizens). Figure 1. U.S. Senate Voting in the 2002 Midterm Eections Percent of Vote Voted Democratic Voted Repubican White (on-hispanic) Source: Fox ews Pos Another way of examining the make-up of the Latino popuation in 2002 is to compare its distribution by party affiiation to simiar data from the 2000 Voter ews Service exit pos. Poitica scientists have commony observed that mid-term eections differ from presidentia eections mainy because of the weak presence of independents in the midterms. Ony the most highy partisan voters are activated by the ower-stimuus mid-term eections, whereas independents and nonidentifiers move into the eectorate in presidentia eection years (Converse 1962). This is the phenomenon described as surge- African- American Race/Ethnic Group Latino 2 Party Identification

3 and-decine in poitica science iterature (Campbe 1966; Campbe 1987). If this generaization is aso true of Latinos, we shoud see far more independents in the 2000 resuts than in the 2002 contest (see Figure 3, page 4). ot surprisingy, this is exacty what we see in Figure 3. In 2000, sef-identified Latino independents comprised 25 percent of the Latino eectorate, compared with just 16.6 percent in 2002 an eight percentage-point difference. This cosey mirrors the non-hispanic white drop-off from the presidentia to the off-year, which aso ran about eight points. The difference in 2002, then, is one of turnout, but the story is more nuanced than simpy saying that fewer Latino Democrats showed up at mid-term. Actuay, a sighty higher proportion of them were present in 2002 than in 2000 (see Figure 3). The voters who were present in 2000, but not in 2002 were ower- and midde-income independent-identifying Latinos, with midding education eves, who foat in and out of the eectorate. These independent foaters behave more ike the Democrats around whom they ive and work, so their absence in the 2002 contests meant the Repubicans woud be in a better position to win a arger share of the Latino vote, which they did in some of the gubernatoria races. That the drop in the share of sef-identified independents might hep Repubicans in the midterm is, of course, extremey discouraging news for the GOP. After a, it is independent voters who are typicay targeted for persuasion-oriented advertising and outreach. But the high turnout of Latino independents heps Democrats because more of these Latinos ean toward that party in the first pace, and certainy their friends and neighbors who infuence them do. Patterns of Support As if the income and education profies were not evidence enough of the upper-cass bias of Latino voting in 2002, consider the fact that fuy 49 percent of those Latinos who showed up at the pos in these 10 states had voted for George W. Bush in the presidentia eection. That these same voters woud return in the off-year eection to cast a simiar share of their votes for Repubican gubernatoria candidates is to be expected, and hardy provides evidence of widespread poitica conversion. If Repubican candidates have converted any Latino voters, we shoud see the Latinos who cast baots for Gore in 2000 coming to the Repubican side in Did any of Gore s Latino supporters convert? The data show that precious few Gore voters in the Latino community supported Repubican Senate candidates in Tabe 2 (page 4) shows that Latinos who had voted for Gore were scarcey Figure 2. Gubernatoria Voting in the 2002 Midterm Eections Percent of Vote Tabe 1. Income and Education Profie of Voters in the 2002 Eection in 10 States Less Than $15,000 $15,000-$50,000 $50,000-$74,999 $75,000 and Up High Schoo or Less Some Coege Graduated Coege Voted Democratic Voted Repubican White (on-hispanic) Source: Fox ews Pos. White (non-hispanic) 5.7 % 32.2 % 18.4 % 24.6 % 7, African- American 29.8 % 25.3 % 44.7 % 7,564 Race/Ethnic Group African- American 15.6 % 38.7 % 13.7 % 15.4 % % 26.2 % 27.5 % Latino Latino 13.9 % 35.3 % 17.4 % 24.1 % % 25.4 % 32.4 % 374 Sources: Fox ews Pos for Data are from weighted s. Unweighted s are avaiabe from the author upon request.

4 more ikey to vote for a GOP U.S. Senate candidate (3.7 percent) than African Americans (3 percent). To be sure, Latinos who had supported Gore were more favorabe toward the GOP gubernatoria candidates, but this shoud come as no great surprise given the non-competitive nature of the gubernatoria contests in severa of the states where the pos were conducted. If these eections woud have been more competitive, and Latino turnout higher, many of these easy GOP victories woud have been narrower. The mora of the data is cear: Repubican candidates do we among Latino voters if Latino turnout remains ow. But gaining ground through the ow turnout of Democrats, and gaining ground by Latino poitica Tabe 2. Did Gore Voters Convert in 2002? Support for Repubican and Democratic Candidates in 2002 Among Gore Voters by Race/Ethnic Group Voted Democratic U.S. Senate Voted Repubican U.S. Senate Voted Democratic Governor Voted Repubican Governor Figure 3. Partisan Profie of Latino Voters in the 2000 and 2002 Eections in 10 States Percent of Vote White (non-hispanic) 93.4 % 6.6 % 2, Gore Voters in % 19.1 % 1,748 African- American 97.0 % 3.0 % 396 Sources: Fox ews Pos for 2002 from 10 states. Data are from weighted s. Unweighted s are avaiabe from the author upon request Repubican Independent/Other Democrat Party Affiiation Sources: Fox ews Pos for 2002; VS Exit Pos for % 6.7 % conversion, are two entirey different goas. Repubicans wi continue to strugge in attaining the atter. Texas and Forida What about the Forida and Texas governorships? Didn t Hispanic Democrats surge into Repubican ranks in these two states? ot according to the FOX ews pos. In Texas, amost no Latinos who had supported Gore in 2000 cast votes for GOP Senate candidate John Cronyn. And in the governors races, about 8 percent of Latinos who had supported A Gore cast votes for Rick Perry and Jeb Bush a respectabe improvement, but no evidence of a surge. In Forida, Jeb Bush poed much worse among Latinos in 2002 (57percent) than he had in his narrow oss to Lawton Chies in 1994 (71 percent). If I earned first grade mathematics correcty, these figures are headed in the wrong direction surprising Latino 96.3 % 3.7 % % 15.5 % 142 given that 2002 found the President s brother a we-entrenched incumbent whereas 1994 found him a reative neophyte. Moreover, the Latino Democrats who voted for Perry and Bush ook very much ike Repubicans, and most of them voted Repubican in the 2000 eection so there is meager evidence of Latino poitica movement between 2000 and The consutants who consider themseves so adept at manipuating voters aegiances are iving in a dream word. The evidence strongy supports the conventiona view of poitica science that partisan commitments and poicy preferences are highy stabe, and campaign messages matter much ess than poitica consutants woud have guibe poiticians beieve (Green, Pamquist, and Schicker 2002). The Poor and Dropouts We can ampify our points about the Latino vote by comparing the behavior of this group to other socia groups in the mid-term. By examining the percentage of Latinos who cast GOP baots in comparison to those

5 who cast baots in ow income and ow education groups, we can evauate the extent to which Hispanic voters are a distinctive boc in the gubernatoria and U.S. Senate races (see Tabe 3). What the figures derived from statistica anaysis revea is that even though an estimated 47 percent of Latinos supported the GOP s gubernatoria candidates in 2002, a simiar percentage of ower income voters (earning ess than $20,000) did aso, as did a majority of those with ess than a high schoo education. One wonders if this makes the case for aggressive Repubican outreach to high schoo dropouts and ow income voters. Most GOP strategists woud hardy take this suggestion seriousy, but it makes about as much sense as focused Latino outreach does. Indistinct Boc The resuts reported in Tabe 4 indicate that after controing for party identification, income and education, there is no difference between Latino voting and the voting pattern of non-hispanic whites in either the Senate or gubernatoria races of This raises serious questions about whether Latinos are a distinctive voting boc at a. Whie back distinctiveness persists even after controing for education, income, and party identification, Latino distinctiveness ceary does not. After we account for their education and income eves, the Latino vote disappears as such. This too is a conventiona finding among socia scientists who have discussed the ack of a pan-ethnic identity across the various Latino nationaity groups (DeSipio 1996; Jones-Correa and Lea 1996; Kaufmann 2002; Lopez and Espiritu 1990). This coection of resuts raises a very important question. Why do parties and candidates want to treat Latinos as if they were ike African Americans, a distinctive voting boc? If Latinos cannot be poiticay distinguished on the basis of ethnicity after we account for their income and education eves, why do we persist in the beief that we need to reach them with a distinctive set of poicy proposas on immigration, iega-aien amnesties, guestworker programs, or any other set of issues taiored ony around their interests? The findings presented here strongy suggest that there is 5 no them at east in mid-term contests. Given their heterogeneity, the idea that they can be moved by ethnicay-specific issue appeas has itte merit. Genera themes designed to promote upward mobiity and educationa opportunity are ikey to work as we as they do with any subset of voters in an economicay disadvantaged position. But poverty and access to education are not uniquey Latino probems. The characteristics that do distinguish Latinos as an ethnic group: sharing a common anguage, their predominant Cathoicism, and an Iberian infuenced heritage, have no obvious and consistent poitica ramifications. I am eft with much the same concusion that I have expressed in previous research on this subject (Gimpe and Kaufmann 2001). Paying ethnic poitics with a group this diverse wi continue to be a waste of time, and is as ikey to insut and turn off Latinos (to say nothing of non-latino voters) as it is to turn them on to Repubican party poitics. And the process of turning a predominanty Democratic group into a Repubican one wi take a ot of time because younger Latinos continue to earn from their eders that their economic interests are generay better served by the Democrats. And once individua Latinos have acquired a firm partisan identity, they are no more persuadabe than any other American citizen. Looking to 2004 Finay, a fascinating question paced on the FOX ews pos asked whether voters woud reeect George Tabe 3. Estimates of the Percentage of Each Group Voting Repubican in the 2002 Midterm Eections Latino Voters African-American Voters Less Than High Schoo Coege Educated High Income Low Income Repubican U.S. Senate Candidates 32.9 % 7.7 % 37.6 % 49.7 % 54.5 % 41.0 % Repubican Gubernatoria Candidates 47.1 % 13.6 % 53.4 % 56.2 % 60.9 % 46.9 % Estimates from mutivariate ogistic regression anaysis of Fox ews eection pos from 10 states. Compete statistica resuts are avaiabe from the author upon request.

6 W. Bush in The data te a cautionary tae whie suggesting the behaviora simiarity of Latinos to non- Latinos. Support for the President s reeection dropped precipitousy among Latinos with grim perceptions of nationa economic conditions. Athough the President remains popuar among a soid majority of American voters, incuding those of Hispanic ancestry, this support is far from anchored in a firm commitment to re-eect him. Aside from being more cognizant of turnout, taking measures to ensure a strong economy wi be the best path to sustained popuarity and a successfu reeection bid. Since the persuasive power of campaigns is greaty exaggerated, Repubicans woud be better off Tabe 4. Logistic Regression Estimates of the Repubican Vote Among Latinos in the 2002 Midterm Eections, Controing Party Identification, Income, and Education Latino Voters African-American Voters Repubican Identifiers Democratic Identifiers Less Than High Schoo Coege Educated High Income Low Income Repubican U.S. Senate Candidates (odds ratios) a a a a a Repubican Gubernatoria Candidates (odds ratios) a a a a b in the short-term focusing their efforts on targeting and mobiizing Latino voters who have voted their way in the past. This is not so much a matter of adopting specific poicy stands as it is an effort to contact and mobiize peope with the appropriate voting history. This invoves turning to voting ists and studying the geography of Repubican-incined Latinos. Simpy put, the turnout efforts in areas of Latino Repubican concentration must be superior to the determined mobiization efforts of Democrats in the Latino Democratic areas. If the 2002 mid-term eection reminds us of anything, it shoud remind us that turnout decides eections far more than persuasive appeas directed toward voters who are paying scant attention. We can a be reieved that fundamenta poitica commitments are not neary so voatie and subject to manipuation as poitica consutants want their candidates to beieve. But the emperor is wearing at east one shred of cothing a t-shirt with the word turnout written across it. LR x 2 Significance Pseudo R 2 4,775 2,997.8 p< ,030 1,928.7 p< Interpretive ote: The odds ratio reported shows the number in each category that voted Repubican for every one that voted Democratic. Hence, the African-American figures show that 16.5 African-American voters cast Repubican baots for U.S. Senate for every 100 that voted Democratic, controing for income, education, and party identification. For Latinos, however, the odds ratio was not statisticay significant. Technica otes: Estimates from mutivariate ogistic regression anaysis of Fox ews eection pos from 10 states. Statistica significance eves: a p<0.05; b p<0.01; are based on og-odds coefficients and standard errors. Compete statistica resuts expressed in og-odds coefficients and standard errors are avaiabe from the author upon request. 6

7 End ote 1 The Fox ews Eection Day Pos were conducted by Opinion Dynamics Corporation. In each state, a random sampe of approximatey 900 voters and ikey voters were contacted via teephone on Eection Eve (Monday, ovember 4) and throughout the day and eary evening on Eection Day (Tuesday, ovember 5). In addition, absentee and eary voters were identified and queried by a ovember 2-3 weekend po. The reative proportion of these voters was weighted in the fina sampe based on previous years resuts. Aso, surveys were further weighted to approximate demographic proportions evident in previous eections. The margin of error for each state po is +/- three percentage points. For more on the Fox ews pos see their website at: In previous years, the source of eection day poing data by state was the Voter ews Service, a research organization created as a coaborative effort of the four major networks, C and the Associated Press, and based in ew York. On Eection Day Sources Campbe, Angus Surge and Decine: A Study of Eectora Change. In Eections and the Poitica Order, eds. Angus Campbe, Phiip E. Converse, Warren E. Mier and Donad E. Stokes. ew York: John Wiey and Sons. Campbe, James E A Revised Theory of Surge and Decine. American Journa of Poitica Science 31: Converse, Phiip E Information Fow and the Stabiity of Partisan Attitudes. Pubic Opinion Quartery. 26: 4: DeSipio, Louis Counting on the Latino Vote. Charottesvie, VA: University of Virginia Press. Garcia, F. Chris, ed Latinos and the Poitica System. otre Dame, I: University of otre Dame Press. Gimpe, James G. and Karen M. Kaufmann Impossibe Dream or Distant Reaity? Repubican Efforts to Woo Latino Voters. Center for (ovember 5, 2002) VS announced that it was not satisfied with the accuracy of today s exit po anaysis and wi not be in a position on eection night to pubish the resuts of state and nationa surveys of voter attitudes. Athough the VS computer system was redesigned after the 2000 eection in which networks used VS data to incorrecty ca states, incuding Forida, VS officias were not sufficienty confident in the accuracy of the information produced by the new computer system to reease the data for media use. Actuay, the VS 2000 pos were not deepy fawed, exit pos simpy have imitations of which most non-experts are unaware. Exit pos can never definitivey predict the resuts of eections in the very tightest races due to the typica margins of error (3 and 5 percent). The 2000 eection debace was much more of a probem with news anchors being too hasty to ca races, not a probem with the pos per se. Much of the 2002 data VS coected may eventuay be avaiabe for schoary research, and VS did generate eection night information for purposes of decaring winners and osers. Immigration Studies Backgrounder. Washington, DC:. Green, Donad Phiip, Bradey Pamquist and Eric Schicker Partisan Hearts and Minds: Poitica Parties and the Socia Identities of Voters. ew Haven, CT: Yae University Press. Jones-Correa, Michae and David L. Lea Becoming Hispanic : Secondary Panethnic Identification among Latin American-Origin Popuations in the United States. Hispanic Journa of Behaviora Sciences 18: Kaufmann, Karen. n.d. Cracks in the Rainbow: Group Commonaity as a Basis for Latino and African- American Poitica Coaitions. Forthcoming in Poitica Research Quartery. Lopez, David and Yen Espiritu Panethnicity in the United States: A Theoretica Framework. Ethnic and Racia Studies 13: 2:

8 O-PROFIT U.S. POSTAGE PAID PERMIT #6117 WASHIGTO, DC Backgrounder Latinos and the 2002 Eection Repubicans Do We When Latinos Stay Home By James G. Gimpe K Street W, Suite 820 Washington, DC (202) center@cis.org Despite Repubican proposas for increased immigration, amnesty for iega aiens, and new guestworker programs, there is no evidence of a Repubican surge among Latino voters in the 2002 mid-term eections. The principa findings of eection-day pos from 10 states with key races for U.S. Senate or governor are: The Latino vote for GOP Senate candidates was simiar to prior years, at about one-third; gubernatoria candidates fared better, at cose to one-haf. But Latinos who voted in 2002 had higher income and education eves than the Latino eectorate as a whoe. Turnout of ower and midde income Latinos was much ower in 2002 than in Latino voters who identify themseves as independents are, in fact, ikey to vote Democratic. The fact that many of these independents stayed home in 2002 heped Repubicans. There is no Latino voting boc, as such after controing for party identification, income, and education, there is no difference between Latino voting and the voting pattern of non-hispanic whites in either the Senate or gubernatoria races of This is not true of African Americans, who are a distinctive voting boc even after controing for education, income, and party identification. 2-03