Panel Transcript: GOP Miscalculation

Immigration Promises Not Moving Latino Voters

Related: Report


Mark Krikorian, Executive Director, Center for Immigration Studies


James Gimpel, Professor of Government, University of Maryland, College Park

William Frey, Visiting Scholar, The Brookings Institution

Steven Camarota, Director of Research, Center for Immigration Studies

MARK KRIKORIAN: Good morning. My name is Mark Krikorian. I’m executive director of the Center for Immigration Studies, a think tank here in Washington that examines and critiques the impact of immigration on the United States. All our work, including the report we’re releasing today that you have in your packets, is our website at

Over the past couple of weeks there has been a combined four debates – three presidential debates, one vice presidential debate – and in all of that time there has been one sole question on immigration, asked last night by Bob Schieffer, and it elicited the usual canned responses. There was no follow-up, no challenging of the candidates’ various assertions and claims, and then they moved on to the next issue, allowing the candidates to breathe a sigh of relief. This is despite the fact that immigration policy is central to the issues that actually were discussed in all of these debates: national security, jobs, Social Security, health care, et cetera. But one area where the campaigns have focused their efforts relating to immigration has been, ironically, the least important one: politics. The Bush administration in particular has shaped much of its approach to immigration on the supposed political benefits in getting Hispanic votes by supporting mass immigration, loose enforcement of the immigration laws, and amnesty for illegal aliens.

To report on the success, or lack thereof, of this approach, we’re releasing a paper today, that’s in your packets, by James Gimpel: “Losing Ground or Staying Even: Republicans and the Politics of the Latino Vote.” Professor Gimpel is a professor of government at University of Maryland, College Park. He previously worked on Capitol Hill and the administration, is editor of American Politics Research, a scholarly journal on this subject, and has written a couple of papers previously for the center. He’s the author of several books. His most recent one is Patchwork Nation: Sectionalism and Political Change in American Politics, which I’m sure is on, for those of you who want to buy it.

And Jim is going to kind of give us an overview of the paper first and then we’re going to have two respondents. Going first will be Steve Camarota, director of research at the Center for Immigration Studies and one of the nation’s leading experts on immigration policy and its consequences, both economics, demographics, and politics.

Then following Steve will be William Frey, a professor at University of Michigan. He divides his time between Michigan, where he’s on the faculty of the Institute of Social Research and the Population Study Center there. He divides his time between the University of Michigan and the Brookings Institution here in Washington. He’s written widely on the issue. In fact, in the small group of demographers he’s probably one of the most widely published in the popular media, has written himself or been cited extensively in non-scholarly as well as scholarly publications. And he’ll give kind of a related presentation, not responding specifically to Professor Gimpel’s points but addressing some related issues on the Hispanic vote.

So we’ll start with Jim then Steve and then Bill, and after that we’ll take questions.

JAMES GIMPEL: Okay, thank you for coming. I appreciate your interest. I think I will stand so I can interact with the screen a little bit. As Mark mentioned, this is the third of three studies that I did with CIS on the politics of the Latino vote, and the other two are on the CIS website. Also, the presentation I’ll go through today overlaps somewhat with the tables and figures that are in the report or the piece that I wrote for today, but there are some things here that are new. And so, I’ll make the PowerPoint presentation available at my email address if you want to contact me. I will also see to it that the CIS folks get it, for anyone who would like to download it. [NOTE: You will need PowerPoint to view the presentation]

Key points today: First, Latino voters have not really moved toward the GOP over the last four years. They remain in the Democratic camp. Widespread reports – point two here – of midterm gains did not take into account the lower turnout levels in the midterm election. And Latino voters, we find in this report, are not especially likely to vote based on candidate views on immigration policy, which may be one reason why they haven’t moved. And indeed, this goes to Bill’s very fine presentation today: only one in four Latinos even live in a battleground state, raising further questions about a block vote, okay? So not only do they present a rather heterogeneous mix of public policy views, since only about one in four even live in battleground states, it raises questions about this block, whether it is a block, and how significant it is.

First, this is the steady decline in turnout for all voters 1964 to 2000. And notice the little zigzag pattern. This is what political scientists call surge and decline. The peaks and the little zigzag pattern are the presidential election years. Turnout in the presidential election years moves up to the mountaintop, but in the off years it drops down to the valley. And so we have to be cognizant of who’s showing up in the presidential election and then who is leaving the electorate in the off years. Low salient elections in the off years, no presidential candidates running, many people around the country not facing competitive elections in the off years, and we know that noncompetitive races depress turnout.

And so you see the standard zigzag patterns, and when you see press accounts of how the particular population groups or demographic groups have moved between a presidential and an off year, you have to keep in mind, aha, well, what about lower turnout? Who is missing in the off year? Who is present in the presidential election years? And folks, this is easily a 10-point difference. That’s not a negligible difference. And so, when we look and compare on-year tabulations from polls to off-year tabulations from polls, we need to bear this in mind.

So here we have in Figure 2 Latino and non-Latino voting in the 2002 U.S. Senate elections. Just pay attention to these bars over here on the left side. Okay, 32.4 percent of Hispanics or Latinos voted Republican; 64.7 Democrat in the U.S. Senate elections. Quickly through these – this is the gubernatorial elections: 33.4, 61.4. And these are from the VNS exit polls. When they became available for scrutiny some months after the election, we actually tabulated all of these results to test some of these assertions about Latino movement. So, again, low 30s.

Now look at where Hispanic Republican voting was in the 2000 presidential elections: 35.3. It looks like we’re losing ground here for the Republicans among the Latinos. It doesn’t look like we’re gaining ground – 35.3. And once again, if you go back to the 2002 results, the last two slides, it suggests that the Latino vote was even lower, which is especially alarming given that a lot of Latinos didn’t bother to show up in the ’02 elections. It would be alarming, in other words, for the GOP strategist who is making these claims about Latino movement.

Now, if you go back to the previous CIS report, the one that was issued in early ’03, it suggests that maybe Latinos had gone up to about 37 percent for the Republicans. That was based on the Fox News poll. And what’s interesting and ironic is that a lot of GOP strategists looked at that particular report and scoffed, ironically – oh, Fox, they can’t be trusted, wink, wink – and I think they’re probably better off with the Fox polls, now in retrospect, than they are with the VNS because the VNS shows that Latinos have moved even less than the Fox News polls did a couple of years ago.

One of the things that’s very interesting is the geographic distribution of the Latino population. On of the reasons why you have such a difference in the ’00 and ’02 Latino voter population is because of low turnout in the large cities specifically. These bars show you the location of Latinos who voted and the composition of their block in the 2000 election: 18.2 percent in 2000 came out of large cities – high mobilization in the big cities; 33.6 percent of the Latino vote came out of smaller cities; 30.2 percent of the Latino vote in 2000 came out of suburbs and the small towns and rural areas. Well, note what happened in ’02 – remember, I told you turnout drops in the off years – 14.4 percent is the large-city contribution; 40.1 percent becomes the small city contribution; 36.6. percent the suburban contribution.

So this strongly suggests that the Latino vote actually dropped in the mostly Democratic large cities in 2002. There was less mobilization of cities – (unintelligible) – of the elections and particularly lopsided U.S. Senate and gubernatorial elections conspired to depress turnout in the large cities, and so Latino contributions from the large cities actually dropped.

This is fascinating because this takes all of the ABC News/Washington Post polls, every single one of them post-9/11 out to May, which is the most recent month that they make the data available. They embargo the data for a few months for fear that some enterprising researcher might come along and prove them to be wrong on some timely topic. So CBS News, ABC News, they embargo the polls for a few months until they think they’re irrelevant and then they release them for public consumption. Well, I got them all the way from 9/11 to May charted out here, and what you see here is the job approval rating among Latinos and all respondents. And the blue line happens to be the Hispanics responding in the poll. The gray line is all respondents, and this little arrow here is the week that the president launched his major immigration policy initiative, in January. (Scattered laughter.) And you can see there’s a pretty ironic dip right about that time in approval, but the more general point is that there’s a pretty precipitous decline in the post-9/11 period, and for Latinos it’s even steeper than it is for all respondents as a whole. In fact, I think monthly I calculated it’s about – the president is losing about 3.4 points in his approval rating per month among Hispanics.

Okay, well, next is economic approval, and economic approval was never a tie, but it also declined and declined more among Hispanics than it does against all respondents. And the most recent poll after 9/11 that asks this question was about eight weeks after – I think it’s a November poll – and then we trace it out 140 weeks after 9/11 into May of this year, and we see about a 1 point per month drop among Hispanics, and it is steeper than the overall figures for all respondents. So we see pretty precipitous declines, steady declines, and these declines are not at all abated by the January policy initiative announced by the President.

This is a very fascinating figure that I took from the Pew Hispanic Center, Kaiser Family Foundation 2004 Latino survey. That’s online. You can get a packet of their charts. It’s fascinating material. I encourage you to browse it. And I lifted this right out of their report and cited appropriately. And their tabulation from their 2004 Latino survey, carried out May, June, July, of top issues that will determine the Latino vote in 2004. Check it out. It speaks for itself. Immigration ranks at the very bottom: 11th out of 11 things that they polled. Education, economy, jobs, health care, and the campaign against terror are at the top. This is a fascinating tabulation because it underscores that Hispanics are just like everyone else. Indeed, maybe they’re not really thought of as a block at all. They’re not some especially distinct constituency that can be easily pandered to, and these kinds of things are going to be on the minds of every American voter, Latino or not.

This is a very fascinating tabulation here from the Pew Hispanic Center Kaiser survey, and we actually looked at the data for this on our own – had the data sent to us. And the point of putting this table up is to show that maybe the Hispanic population really does show more of a mix of views on some of these cultural issues than we like to think, or certainly than GOP strategists like to think. And so what we’ve got here is responses of Latinos, over 2,000 of them, to a very standard, neutrally worded abortion question. So this is not a loaded question, folks. It simply asks, “Should abortion be legal in all cases, legal in most cases, illegal in most cases, or illegal in all cases?” And what’s interesting is that you do not see what you might expect if you were imagining Latinos to be homogeneously culturally conservative on the abortion question. It turns out that 36 percent of the Latinos reporting from battleground states actually think that abortion should be legal in most cases. That’s a plurality of Latinos in battleground states.

Does this sound like a population that would easily realign from Democrat to Republican? I don’t think so. I mean, look at the non-battleground states. Now, they may not be that relevant, but the non-battleground states, the population seems to be slightly more conservative. The plurality there is 32.1 percent saying illegal in all cases. So here we’ve got a battleground state population of Latinos that looks to be arguably more liberal on a question like abortion than those who think that this is a culturally conservative population would want you, as the press, to believe.

Check out the next question here: “Do you favor or oppose a constitutional amendment that would define marriage as a union between one man and one woman, thereby prohibiting legally sanctioned marriages for same-sex couples?” That’s a long question but rather neutrally worded; there’s nothing loaded about it. The survey research team is not expecting to evoke a particular response. And while we don’t have big differences here between battleground and non-battleground [states], notice how evenly divided this population is. Okay, again, it suggests that that the Latino population is not unlike any other population in the American electorate – they’re divided on this question: 47.5 percent oppose in the battleground states; 44.8 percent oppose in the non-battleground states. That is, oppose the president’s position in favor of a constitutional amendment. And, again, the population is very mixed on this question. And this, again, I think helps us to explain why it’s not so easy to move this population. They’re just not the culturally conservative, homogenous group of people that they’re often caricatured to be.

Let me just summarize by saying that there’s very little evidence that Republicans have made gains among Latino voters over the last four years. The President’s immigration policies certainly have done nothing to attract Latino voters. And finally, Latinos may not be as uniformly conservative as Republican strategists think.

Now, what about the 2004 election? Will we see an increase in GOP voting among Latinos? There are some circumstances where that could happen. For instance, it could happen if President Bush wins by a lopsided margin among all groups. I mean, it’s certainly possible that the Kerry-Edwards campaign comes completely unraveled in the last three weeks, that we will see a surge in Republican voting among all Democratically inclined groups; we’ll see a bump-up in Republican voting among African Americans, among young voters, among voters in maybe traditionally Democratic areas, and maybe even among Latinos. But, folks, that’s no evidence of conversion. Remember, that’s no evidence that people have been persuaded. That might be evidence that the Kerry-Edwards campaign has run maybe an especially inept effort, or again, it’s come unraveled in the last few weeks. But if President Bush wins by a lopsided margin, we should expect a bump up in the Latino vote.

Another possibility is that if we see low turnout among Democrats, like we did in ’98 and ’02 and other off years; if a lot of Democrats decide to stay home, they don’t see anything much to like in the Kerry-Edwards ticket, then we’re going to see the Hispanic Republicans show up for their folks up and down the ticket, and we will naturally then, as a consequence of the low turnout of Latino Democrats, see an increase in Latino Republican voting. Well, you see, that’s very different than actually saying that you’ve converted Latinos from the other side.

So this is the question then: does it matter whether the GOP vote improves among Latinos based on low turnout of Latino Democrats or actual conversion? And I would insist, just in conclusion, that it makes an enormous difference because whether the GOP vote among Latinos improves based on low turnout or conversion has a lot to do with how an officeholder – President Bush or other Republican officeholders – will interpret their elections for policy purposes.

And so the stories that get told post-election about the results that are coming are ultimately going to be very influential for policy, and what I hope I’ve done here is alert you to some issues and questions that you should bring to your own post-mortem assessments of the election results. And even if you write nothing about what happened here today, you completely blow it off, I hope I’ve planted some seeds of skepticism, that you bring to those post-mortem election discussions and results in a couple of weeks.

These are my research assistants. I had to put them up there.

MR. KRIKORIAN: Thank you, Jim. While Jim and Bill are negotiating their laptop switchover, Steve will offer some thoughts on the subject.


STEVE CAMAROTA: Thank you, Mark, and thank you, Jim, for a very concise but I think illuminating discussion of what is pretty clearly a voting block – or not necessarily a voting block but a lot of consistency in the Hispanic vote in terms of its preference for the Democratic Party, and some explanation as to why that might be, or touch on some of those cultural issues.

Let me try to summarize some other areas where I think why it is that Latino voters have tended to gravitate toward the Democratic Party, and that is unlikely to change in the near future. There are, I think, three fundamental reasons why Democrats have done better historically with Hispanic voters and will continue to do so.

The first is that, as I’m sure you’re all aware, Hispanics tend to be a lower-income population overall. Fifty-three percent of all Hispanics live in or near poverty, with “near poverty” defined as less than 200 percent of the poverty threshold. The comparison figure for non-Hispanics is about 27 percent. Even if we focus on just Latino citizens who are 18 years of age and over, we still find that 40 percent live in or near poverty. The comparison figure for non-Hispanic citizens 18 years of age and over is 25 percent. Again, focusing on only people over age 18 who are citizens, 61 percent of Hispanics over the age of 18 have no federal income tax liability. And I think that’s an important factor because it reflects not just the low income but the larger family size, and it relates to policy directly because the Republicans are the party of tax cuts, especially federal income tax cuts. They’re the party of restraining your taxes. But if a much larger share of a particular population doesn’t pay federal income tax or their tax contributions tend to be more modest . . . and it turns out that the average Hispanic 18 years of age and over who overall pays about 53 percent as much in federal income tax as non-Hispanics 18 and over – and again, we’re only looking here at citizens.

And so, again, if you are the party, in the case of the Republicans, constantly talking about taxes and how they’re a burden and how they need to be restrained, it’s a lot tougher to reach a population which tends to pay more modest tax payments than does everyone else, or just the non-Hispanic population I should say.

Let me give you another statistic. Again, focusing only on people 18 years of age and over, 25 percent of citizen Hispanics 18 years of age and over lack health insurance. The corresponding figure for non-Hispanics 18 and over who are citizens is 13 percent. Again, a very large difference, roughly twice as much, and the Democrats are generally thought of, and perhaps correctly it seems to me, as the party that wants to use the government in a more activist way to try to rectify or try to provide [for] more people, mainly low-income people, with access to health care and health insurance. So, again, that’s likely to give the Democrats some key built-in advantages.

And finally – let me just give you one more statistic – about 28 percent of households headed by Hispanic citizens access at least one of the nation’s major welfare programs. The corresponding figure for non-Hispanic citizens is 15 percent. So this is a population that is somewhat more dependent on government services. So if you hear the Republicans talking about the need to cut back government services, again, it’s not as likely that you’re going to find as receptive an audience in that community if the community has a larger share who access those services.

So I think the bottom line here is that the Democrats, generally speaking – and this has been borne out for many years in polls – have a set of policy prescriptions that are more appealing to low-income voters, those more in need or more likely to access public services, those who don’t have health insurance and so forth. And since Hispanic voters are persons who are citizens over the age of 18 who tend to exhibit those traits, it’s much more likely that they’re going to vote Democratic.

Now, a second issue unrelated to economics or maybe linked to economics in some ways is that the Democrats are the party of race and ethnic-specific policies such as affirmative action and set-asides. Now, since Hispanics can often benefit from those programs, it’s going to be difficult for the Republicans, whose base and most Republican voters are critical in some ways of those programs. The Republicans can, and often do, simply not talk about it, even though their base is clamoring for them to talk about it. Republican leadership can pull back and not talk about affirmative action and set-asides, but the fact is that’s not the same as being the party of affirmative action, which is what the Democrats clearly are. I mean, obviously there are critics within the party, the Democratic Party, and there are supporters of affirmative action within the Republican Party, but in general, the Democratic Party is much more supportive of race and ethnic-specific policies, and that too tends to attract Hispanic voters.

Surveys of Hispanics generally show a fair amount of support for those programs, with critics and diversity of views of course there, but in general that will lead to more votes for the Democrats, and it’s hard for the Republicans to fundamentally undo that because they can’t become the party of affirmative action, partly because their base doesn’t want it, but partly because there is already a party of affirmative action, and that’s the Democrats.

Now, I think a third reason for the stability of Latinos’ Democratic orientation is elite opinion within the community. More than 90 percent of Hispanic elected officials are Democrats right now. When one looks at the writings of journalists, columnists, and intellectuals in the community, we see that in general – though with some very notable exceptions, the community is – the leadership or the elite opinion in the community is overwhelmingly Democratic in its orientation, and as a consequence it is somewhat hard for Republicans to communicate to Latino voters through what is not necessarily a hostile elite but what one described at least as an unsympathetic elite for the most part. Again, there certainly are exceptions; there’s no uniformity of opinion.

So I think these three factors, the fact that the population is low income; the fact that the Democrats are the party of race and ethnic-specific policies, the Republicans are not; and also that coupled with elite opinion within the community make it very unlikely that the Republicans are going to be able to attract a much larger share of Latino voters, at least in the immediate future, and probably going out quite some ways.

So in short, because of this fact you might simply say that it would be unfair to say that immigration is a voter registration drive for the Democrats, but it would also be unrealistic not to recognize that in the long term, immigration does have significant partisan implications for who’s likely to win elections in the future — state, local, and national – given the fact that more than half of newly arrived immigrants in the United States are Latinos, and given the fact that Latinos – there are some built-in reasons why the Democrats have some really fundamental advantages that are not likely to change anytime in the near future.

Thank you.

MR. KRIKORIAN: Thanks, Steve. And now we’ll go to Bill Frey.

WILLIAM FREY: Okay, I’m going to come at this from a somewhat different direction. I’m not a political scientist. I don’t take voter polls and analyze them like Jim does. I’m a demographer. Someone once said, all demographics is local, so I guess it applies to politics too, so I’m going to, with this presentation, rather than say which way Hispanics or Latinos are going to vote, I’m going to try to address how much they matter as opposed to non-Hispanic whites in this current election and what likely change will occur maybe in the next election or in 2012 as a result of underlying demographic shifts and underlying demographic patterns.

I mean, we find ourselves in a situation with the Electoral College that states that tend to be very important strategically are not necessarily representative of the nation as a whole, and from that perspective I think I’m going to try to convince you that the Hispanic vote is not going to be all that important in general just because of the sheer demographic force of non-Latino whites. One of those reasons is that the share of all Hispanics in the United States who actually wind up voting is very small. As you can see from the last bar on this chart, out of every 100 Hispanics in the population, only 40 are voting-age citizens and only 18 are going to vote. Now, the reason only 40 are voting-age citizens is that about 35 of those 100 are below age 18 an another 20-some are people who are above age 18 but are not citizens, and that’s a very important part of the Hispanic population. Actually, the share of voting-age Hispanics who are not citizens has actually gone up from 2000 to 2004 according to the recent Current Population Survey.

Compare that to whites where 76 of every 100 whites are voting-age citizens and 47 out of every 100 whites vote. Compare blacks and Hispanics: twice as many blacks out of 100 blacks vote compared to Hispanics, even though Hispanics are a larger share of the U.S. population. There are almost twice as many black voters in the population.

If you compare the racial composition of the total population in the U.S. with the racial composition of the voter population, about two-thirds of the total population is white but about four-fifths of the voter population is white. And again, you can see that pink Hispanic sliver is about half of the gray or black sliver in the voter population, which is not the case in the total population.

If you look at the share – the two bars shown here for each state are the share of the total population in each state that’s Hispanic and the share of the voting population in each state that’s Hispanic. Now, New Mexico does it pretty well in translating total share to voter share because there’s a high percentage of Hispanics in New Mexico who are long-term residents and citizens, but if you look at Arizona and Nevada, only about less than half the share of the voting population – in other words, 29 percent of Arizona’s total population are Hispanic but only 12 percent of its voting population are Hispanic – a similar situation in Nevada and Colorado. So these are battleground states where there is a large Hispanic resident population but the voting population is not going to be that very large. About 30 states here in the United States have less than 3 percent, and many of them much less than 3 percent of their voting population that’s Hispanic in this next election.

Another way of looking at this, which Jim talked about before, is not only is there a translation problem in terms of the population total moving into voters, but there is also a geographic issue in that most Hispanics tend to live in states that are already classed as red and blue America. This is less the case for whites. So what we see here is that 39 percent of all the whites in the United States live in battleground states. And I’ve classed 18 battleground states. By the way, I have this handout here, which has some of the numbers in it, and on Table 1 you can see what I’ve used as battleground states. About two out of five whites are living in battleground states but only about one out of four Hispanics or one out of four blacks are living in battleground states. The reason Hispanics are in these states is because they’re the states that are the big immigrant gateways. Hispanics tend to stay in states like California, New York, New Jersey and Illinois – blue states – or the red state of Texas. And blacks, by the way, are moving to the South and a large part of the South is red America, and that’s probably going to decrease the black representation in battleground states.

I have some bar charts in your handout here, which might be a good thing to look at – or some pie charts, which might be a good thing to look at compared to these bar charts. What this says is you get increasingly into the more strategic populations for this election – the higher percentage of the voters that are white. So whites make up 68 percent of the total population, they make up 71 percent of the voting-age population, but 79 percent of all voters and 86 percent of all voters in battleground states, and even a little more than 86 percent of all voters in battleground counties in battleground counties in battleground states. So this is part of where we’re seeing the election being really determined by a largely white electorate. This shows you the minority side of things, and you can see that shrinking pink Hispanic share of the voting population as you go from the total population down to the battleground state population.

So that’s part of the translation issue, but I want to address another part of this, which this gets to, okay, whites look like they’re going to be in the driver’s seat for this election, but as we get further on into 2008 or 2012, is this really the last hurrah for aging whites – the title of my presentation here. And to that I want to sort of separate the battleground states into two groups. One I called slow-growing battleground states — and what color are they on here; it’s sort of light yellow – that are around the Great Lakes area – there are 11 of them – and the fast-growing battleground states, which are seven states, Florida plus a bunch of Western states. By the way, the article in your packet from American Demographics does not include Colorado as a battleground state so I updated this analysis, because when I wrote that Colorado was not as much in play.

And then what’s important here to note is there are really two parts of the country that have very different demographics. The fast-growing states have grown by over 20 percent over the last 13 or 14 years. The slow-growing states have grown less than 20 percent, sometimes much less than 20 percent. But what’s even more important are the migration components which are coming around to change the demography of these states. The fast-growing states are states that are attracting lots of immigrants – that white bar there, which is the rate of immigration to those states – as well as domestic migrants. And if you think of Nevada and Arizona and New Mexico and Colorado, these states are getting a lot of migrants coming out of California – middle class, not necessarily professionals, not necessarily high-income people who can’t afford the congested, expensive suburbs of Los Angeles and San Francisco and are moving to these other states. Many of them are elderly, many of them are empty nesters, many of them are young couples. And part of this is perhaps a less upscale group of folks moving to these states, as well as lots of immigrants coming to these areas, largely Hispanic.

And so, while right now we don’t see a high level of Hispanic voting among these folks, they’re helping to put these states perhaps more in play in future elections than are the case today. In the case of Florida they’re getting the domestic migrants from the Northeast Corridor for much the same reasons that the migrants in California are going to the West. Contrast that with the slow-growing battleground states that are losing domestic migrants as a group and they’re gaining immigrants only to a very small degree. It’s a very kind of stable state population.

So, the other way to look at this, although the fast-growing battleground states are growing very fast, they have a lot less electoral clout – only 74 electoral votes for them compared to 115 for the slow-growing states. So therefore, much more strategically important are these slow-growing battleground states. Among things that characterize them are a high percentage of white baby boomers among their voting-age population. A third of the voting-age population in the slow-growing battleground states are white baby boomers and only about a quarter of those in the fast-growing battleground states.

Here’s a comparison. The slow-growing battleground states have a much higher percentage of white married women and a higher percentage of white non-college graduate men than are the case in the fast-growing battleground states. Also, they have an older white population than the fast-growing battleground states. These are groups which are not necessarily representative of the national population in these kinds of numbers, but they are representative of the slow-growing battleground states. This is kind of an important chart to sort of put the two points together: the translation problem of the population to voters and the distinction between fast-growing battleground states and slow-growing battleground states, which are the last two pairs of bars there.

You can see the slow-growing battleground states, the last two bars that are there, 84 percent of the total population is white, so that’s pretty white to begin with, but then when you put in the slight translation problem, 89 percent of the voters in these slow-growing battleground states are white. Look at the fast-growing battleground states. Here the initial total population is 67 percent white, but because of the translation problem of Hispanics, it goes up to 80 percent white of the people who are going to vote on election day. You see how that orange Hispanic bar shrinks when you go from the total to the voter population.

So overall, both the fast-growing and slow-growing battleground states have some pretty white voters, especially the slow-growing battleground states. In the handout you can see in Table 1 where I show this for each individual state, both the battleground and the non-battleground states. And the last column of Table 1 I have something called the white voter advantage, and that’s the white share of the voting population minus the white share of the total population. If you eyeball down that list you can see states like Florida and Arizona and New Mexico show very high white voter advantage. The flip side of that, though, is that as this total population of these fast-growing battlegrounds starts translating into the voter population four years from now or eight years from now, as those younger under-18 Hispanics get to be of voting age and as many more of the non-citizen Hispanics become voters, then these fast-growing battleground states are going to look more like this in terms of their voter population. Now they don’t look like that in terms of their voter population.

So I’ve done a few other things I’m just going to review quickly because I know they want me to get off here, but we can look at battleground counties in battleground states. Here’s Pennsylvania where you see those yellow counties around suburban Philadelphia and the Lehigh Valley in the eastern side of the state and some suburban counties in the Western part of the state. What you can know here is that the racial composition of the battleground counties is just as white as the red counties in this slow-growing battleground state of Pennsylvania. So moving to the battleground states is really – these slow-growing battleground states, the counties that are in play are pretty white. There’s Ohio where the battleground counties are around Cincinnati and Columbus and Dayton and a little bit in the northeast suburbs.

Here again, the battleground counties are pretty white. They’re not quite as white as the red counties but they’re pretty white. In contrast, look at Florida. There’s a lot of battleground counties in Florida. By the way, I base these on the 2000 election, the battleground counties. They’re counties where there was not more than 10 percent difference between Gore and Bush. But here the battleground counties are pretty diverse; more diverse than the red counties, almost as diverse as the blue counties. A lot of that is the kind of I-4 through the middle of Florida, sort of the Tampa-Orlando-Daytona Beach, where there are a lot of Puerto Rican movement and a lot of new minority black movement into that part of Florida.

So here’s a comparison: in the slow-growing – in the battleground counties, the red counties and the blue counties, of the slow-growing battleground states and the fast-growing battleground states – and you can see that in these fast-growing battleground states, the battleground counties as a group are much more diverse than they are in the slow-growing battleground states. These are voters. Now, instead of looking at voters, let’s look at the total population. I’m going to go back and forth on this – (audio break, tape change) – percent of the total population in these battleground counties and the fast-growing battleground states are Hispanics. That goes down to 10 percent when you only look at voters. But what this says is as we get more translation from the total population into the voter population, that these battleground counties, as well as the red and blue counties in these fast-growing battleground states, are going to look much more diverse, maybe not after the 2008 presidential election but maybe at the 2012 presidential election. And if you look at population growth in these three different kinds of counties, they’re much faster even between 2000 and 2003 than they are in the three comparable counties in the older battleground states and the slow-growing battleground states.

So I guess my bottom line in all of this is right now, that 93 percent, those people living in the battleground counties in the slow-growing battleground states that are 93 percent white, they have an inordinate influence in what’s going to happen in this next election, but as we move further down the road, it will matter how many Hispanics vote Republican or Democratic just because of the very change in the demography that’s going to occur.

MR. KRIKORIAN: Thank you, Bill.

Let me just take the prerogative of the chair and just ask kind of a general question for whoever wants to deal with it. Why has immigration been perceived as so important to Hispanic voters and why is it not actually proven to be all that important? I guess the second part of it Jim sort of dealt with, but Hispanic voters for the past several years have been sort of the new version of the soccer moms or NASCAR dads or security moms, or whatever buzzword political consultants have come up with to boost their own business. But what happened? Why hasn’t that really turned out that way?

Whoever wants to deal with that. Jim?

MR. GIMPEL: Well, a couple of things. One is that for registered voters in the Latino population, the immigration experience may well be behind them. I think a second reason is that when asked on surveys, Latinos will, by a fairly sizable majority, say that they are in favor of more open and generous immigration policy, but the question is whether they actually take the opinion on that subject into the voting booth with them. And of course that’s also true of white voters and their immigration opinions – white natives and their immigration opinions, or black natives and their immigration opinions. Do those opinions actually cue their votes? And that’s where I think we don’t find a lot of evidence that they do.

It’s a bit like asking you for your preference of Coke over Pepsi. You’re definitely going to have an opinion on that. You might also have an opinion on strawberry ice cream over chocolate, but having an opinion on the subject and actually using your opinion to guide your voting and your political thinking are two very different things. And I think the intriguing thing about the Pew research and where immigration ranked on that graph was that it was very clear from that graph – those figures from that survey that education and the economy and jobs were far more important issues for cuing the vote than immigration. And I think that’s probably pretty consistent over time.


MR. CAMAROTA: Well, I guess let me start by saying that my read of the polls of Hispanic voters on immigration issues suggests that they’re divided on the issues, so that’s one thing to always think about. There isn’t any kind of set view. In a Zogby poll right after the president began outlining his proposal apparently it was 50-50 in the share saying they liked it or liked the idea of an amnesty and so forth. So it’s not clear, first off. So the community is divided, and as Jim rightly points out, the opinions aren’t that strong. It’s not a top issue.

So I guess the question is, why do people think it is, or why do the Republicans think it is? In other words, advocating an amnesty for illegal aliens, whether you like Bush’s guest worker amnesty or some other ones like some of them that the Republican senators have sponsored in Congress that actually just give a green card . . . I think the reason is that you have to keep in mind certain things. The Democrats have the fundamental advantages with regard to the Hispanic vote that I outlined in terms of the income of the population, in terms of elite opinion within the community and in terms of race and ethnic-specific policies. In effect, the Republicans are sort of casting about for something that they think that they might be able to appeal to the community with, and this is something that the business community within the Republican Party wants. They want open borders; they want an amnesty.

And so, obviously the Republican base hates the idea, and that’s one of the problems the Republicans have, and now you have consultants come along and say, well, if you want to get the Hispanic vote – I know you can’t change these core positions on sort of taxes and nationalized health care and income redistribution or affirmative action, but, you know what, we’ll go with immigration and your business constituency likes it. And I think that’s what it is, and there is also a sense among Republicans that they’re very sensitive to the charge that, look, they’re racially insensitive or indifferent to minorities, and being in favor of open borders as a way – or amnesty for illegal aliens is a way that Republicans say, yes, it’s true, a very tiny share of African Americans vote for us, and yes it’s true that blacks generally do not rate us very well. But, look, we are not bigots; we favor lots of immigration.

And so I think it’s a combination of things that a desire to sort of blunt the charge that they’re racially insensitive, and the problem is they can’t come up with anything else, because as I outlined, the Democrats kind of have the advantage here, and so they’re not sure what to do and this is the best they can come up with, and as I think Jim’s analysis and others’ have shown, it’s just not going to work.

MR. KRIKORIAN: Thanks, Steve. Any questions from the audience? The PowerPoint presentations will be available probably – if we can get them from both speakers we’ll link them on the transcript of this event when we post it to our website.

MR. GIMPEL: Let me just say a comment – make one comment about Bill Frey’s presentation to help you make a linkage between his and mine.

Notice that the heavy concentration of Latinos in non-battleground locations conspires against their mobilization. The fact that the battleground states are so heavily white actually hurts the mobilization of Latinos because we know that presidential electioneering mobilizes people. We know that presidential electioneering educates people about politics. And there are broad swaths of the American electorate that are missing out on the experience of presidential electioneering because they don’t live in the right state. Even negative advertising – we’ve got some political science research to show that even negative advertising teaches people important bits of information about politics. And to me this is worth writing about. A couple of paragraphs of a press story should be written about the concentration of these ethnic minorities in locations where they never see a presidential candidate.


Q: Dr. Frey, I just had a logistical – what was the composition of the battleground states?

MR. FREY: The battleground states were listed here. You mean which states there are?

Q: Yes, which states?

MR. FREY: Yeah, they’re listed here. They’re basically – the fast-growing battleground states are essentially Washington, Oregon, Nevada, Arizona, New Mexico and Colorado. And the others – there’s 11 others which are the usual suspects. Originally, in the middle of the summer, I contacted several well-known political scientists and asked them to give me a list of battleground states and they each gave me a different list. (Laughter.) So I decided to use the list which was used – the states that were close enough after the Gore-Bush election, and I took Tennessee away because at that point it was fairly determined that Tennessee was not in play, although now apparently some people think it is in play. Since then I added Colorado because Colorado definitely is in play. So this was sort of a basis on the judgment of a lot of people that I just kind of averaged and put together.

MR. KRIKORIAN: Could I ask you guys – on the saliency issue, in other words, how important immigration is – this is the graph that you had in your paper, Jim, about immigration being the least important of all the things they asked for Hispanic voters that would determine their vote. Is there comparative research among, say, Republican voters or non-Hispanic voters, or something like that? In other words, I would suspect that it’s actually possible, especially among Republican base, that immigration has a higher saliency, is more important among people who want tough enforcement than it is among people who want open borders. Do you see what I mean? In other words, that people who want tight immigration policies may actually hold that position more strongly than those who want loose ones. Is there anything – I mean, is there any research –

MR. GIMPEL: That’s entirely possible, and of course it’s entirely testable. There are – on some surveys we could probably scratch around for some open-ended questions about the most important problem, and we could rank them for different subgroups of the population and see where the white population was, say, compared to the African American population or to the Latino population. And, yeah, it’s entirely possible that white voters would care more about immigration than Latino voters do.

MR. CAMAROTA: As you may recall, Mark, before September 11th when the president was outlining some ideas when he was meeting with Fox, the center sponsored a Zogby poll on this subject, and I should say that – I don’t have it in front of me but as I recall, one of the key findings is that people who don’t like the idea of an amnesty, they really don’t like it. They were asked, do you think it’s a very good idea, a good idea, a bad idea, or a very bad idea, and the numbers saying it was a bad or very bad obviously outnumbered the numbers who though it was good, but the numbers who thought it was very bad was twice the number that thought it was very good idea to grant amnesty to illegals, and that was especially true among self-identified Republicans and conservatives, and the share of Republicans and conservatives who thought it was a very bad idea, as I recall, was much larger than any share in the Hispanic responses on whether it was a good, bad or – the number saying very bad or very good.

So that would tend to buttress the idea, if that measures saliency, that in general it might be the Republican base who actually cares more about immigration than communities with large numbers of immigrants – the Hispanic community. I think that’s very plausible, and in fact, the reaction to Bush’s proposal has been almost uniformly negative from the rank and file of his own party. And in an election year, for a sitting president to experience that, that’s tough, and you would expect him to sort of just shut-up about it; they didn’t like it. So that would then also tend to buttress the observation that maybe it’s conservatives who care most about immigration.

MR. KRIKORIAN: Actually, I wanted to ask – sort of the flip side of what you answered, Steve, is why the Republican elite thinks that immigration is important. But the other side, related to what we just said, is if the Democrats actually want to win, if they’re actually interested in winning, isn’t there – especially because of the makeup of the battleground states, isn’t there really – wouldn’t the sort of obvious script be to move to the president’s right, if you will, on immigration. In other words, support a tough immigration policy as a way not only to attract voters in battleground states but also to sort of dispirit the president’s base. And why wouldn’t the Democrats, assuming they actually want to win the election – what would inhibit them from coming out with a tough pro-enforcement immigration position, since it seems to me that would be, given the current circumstances, a ticket to almost sure victory? Any ideas?

MR. GIMPEL: Well, first of all, for the simple reason that the Democrats are smart enough to be able to see that the recent immigrant population is heavily Democratic and the Latino population certainly is, so it’s an important part of their base. But note that John Kerry, in the debate last night –

MR. KRIKORIAN: Yes, right –

MR. GIMPEL: – did make a pretty forceful statement about border control.

MR. KRIKORIAN: Right before calling for amnesty.

MR. GIMPEL: Exactly. That’s true. That’s true. But he did take a pretty strong stand on border control, so there must be some attention to polls inside that camp.

MR. CAMAROTA: Part of the reason that the Democrats – not necessarily cynically – they recognize that in the long run, immigration does increase the share of votes that are likely to be Democratic. But also, the Democratic elite, there’s no debate on this issue. There was a Chicago Council on Foreign Relations poll – and we just summarized some of their results in a report – but one of the things they found is that elite opinion in the United States – that is, journalists and editorial page editors but also heads of business groups, church groups, union groups – tend to overwhelmingly not think illegal immigration is a problem and tend to overwhelmingly want more immigration, and the public overwhelmingly wants less immigration.

So, overall, the elites in the United States are overwhelmingly in favor of lots of immigration and non-enforcement of the law, but I think in the Democratic Party it is fair to say there’s virtually no debate on the question of immigration. Large-scale immigration is almost uniformly seen as positive and enforcement of the law is not necessarily universally but often seen as suspect, as motivated by racial or ethnic animus. So it would be very hard – although it probably would win votes even within his own party – for Kerry to move his party that way, given what the leadership and elite of that party thinks.

MR. KRIKORIAN: Well, I’m assuming that our speakers would be happy to be accosted afterwards if you want to come up and ask them some questions personally. Jim’s report is on our website already at We’ll have the transcript and hopefully links to the PowerPoint presentations next week on our website, and thanks for coming.