Many say California Republicans’ anti-immigration ballot initiatives in the 1990s lost them the Latino vote and set the party on the road to ruin. Is Trump leading national Republicans down the same failed path? Iris Hui finds that the propositions were not the tipping point, with Republicans starting to lose ground beforehand and feeling the brunt of their shifts only with the next generation. Joshua Zingher finds that California Republicans did lose Latinos without gaining whites, but he says anti-immigration politics in Arizona provides Republicans with a better national model: they gained white voters faster than they lost Latinos. Western state politics provides a glimpse of the national future.

The Niskanen Center’s Political Research Digest features up-and-coming researchers delivering fresh insights on the big trends driving American politics today. Get beyond punditry to data-driven understanding of today’s Washington with host and political scientist Matt Grossmann. Each 15-minute episode covers two new cutting-edge studies and interviews two researchers.

You can subscribe to the Political Research Digest on iTunes here.


Grossmann: This week on Political Research Digest, anti-immigration politics, is California’s past, America’s future? For the Niskanen Center, I’m Matt Grossman. Many say California Republicans anti-immigration ballot initiatives in the 1990s lost them the Latino vote and set the party on the road to ruin. Is Trump leading National Republicans down the same failed path? New research finds that the propositions were not the tipping point. With Republicans starting to lose ground with Latinos beforehand and feeling the brunt of their shifts only with the next generation.

I talked to Iris Hui of Stanford University about her new study with David Sears, published in Political Behavior called, “Reexamining the Effect of Racial Propositions on Latinos’ Partisanship in California”.

Another new study maintains the conventional story on California, but says Arizona’s experience might hold more promise for National Republicans. I also talked to Joshua Zingher about his recent study with Gregory Robinson, Jonathan Krasno and Michael Allen, published in Politics, Groups and Identities, called “Creating a Racially Polarized Electorate”. They find that California Republicans lost Latinos without gaining whites, but anti-immigration politics in Arizona gained Republicans white voters.

California Prop 187, in 1994 denied public services to those who immigrated illegally. Prop 209 in 1996 limited affirmative action. And Prop 227, in 1998, limited bilingual education. All passed in quick succession, but Hui collected all field polls and found little evidence that the proposals immediately changed Latino partisanship.

Hui: A lot of people say 187 was the turning point in California politics because of the way that the Republican party come out in support of the 187, so people believe that cost the Republicans support in the state of California. But then, what we find is quite the opposite, so actually among the registered voters, we didn’t really find a much, significant effect on the partisanship. Republican support in the state, especially among Latinos has been decreasing over time. And that decreasing started before 187.

The heyday for the Republican Party was around the Gulf Wars, the senior Bush administration, and then it starts declining, well before 187. So 187 is just kind of like one of the steps in the process that led to a kind of a decline in the Republican Party in the state of California, but it is not the tipping point in the changing or the transition.

Grossmann: They’re arguing against a strong conventional wisdom that blames these polarizing initiatives.

Hui: We believe that because of the rhetoric, and so the radical rhetoric during the campaign. So we think it has a mobilization effort, and then we also see that during that time there were a lot of protests, either mass scale demonstration to support or either support or against 187. And so, the conventional wisdom was that their proposition must happen, can apply, be a tipping point or a beckoning event. So but, mobilize the Latinos for political participation.

Grossmann: But based on long running political science, partisanship should not be easy to change.

Hui: We know that partisanship is stable throughout one’s lifetime. So, we know that partisanship started very early on, some of them, people learn from their parents, from vocalization, and that they develop at adolescence and then as young adults. And then once they start voting in elections, their partisanship tends to become stable over time. And after a couple of elections, we know that party ID, pretty much set in one’s life.

So even though there is still being the arena’s argument that party ID can be a running tally, so people can update their party ID. But then it seems like, from the work by [Eric Scheckler 00:04:10] and [Tom Green 00:04:10], we know that party ID is mostly stable throughout one’s lifetime. And so, if that is the case, if party ID remains stable throughout one’s lifetime, then the trouble comes in.

Why do we see Latinos suddenly switching from being a Republican to a Democrat? That shouldn’t be the case. That’s kind of like how we started the project.

Grossmann: Hui agrees that in the long term there may still have been an effect on future voters.

Hui: One question is that, did Proposition 187 per se, led to a tipping point? Did the changes occur before and after 187? And I think the answer is definitely no. We didn’t see this 187 as a major event that changed partisanship. Rather what we argue, especially if you look at [inaudible 00:05:03] one and two, is this an over time trend that was kind of like partly a result of demographic movement, demographic shift in the population and part of the result of people responding to political parties, how they shift in their ideology over time.

So what we are arguing is that if you look at Republican Party and probably understand their fortune in the state of California, they have their heyday in the Bush administration during the Gulf War. But then, since then the economy kind of tanked and then you started to see the demise of the Republican party in the state. And then also because of the Republican party has started to switch from more, move to more social concerns in their positions that they think will resonate with the voters in the state. And so what you get is that people are moving away from the Republican party well before 187.

Grossmann: The next Latino generation was quite a bit more Democratic, though still not as much as Latinos were in the 1970s and 1980s in California.

Hui: There is a generational gap, and the generational gaps ended up widening over time. And then you also see Latinos, the younger ones less likely to identify with the Republican party over time. And that is a composition fact that we see in California.

Grossmann: William Sears looked at three different data sources to assess the impact of the propositions.

Hui: You see that the points actually jump up and down a lot. So we worry that that may be because of the inaccuracy of the data or maybe insufficient sample size in the data. So we may not be capturing the right estimate. So that’s why we moved to the two other data sources. One is the exit poll and then the other is the statewide database of registration data. So we wanted three, we wanted to triangulate our data to see whether we’d find the same new facts with the two other data sets.

So what we see is that with the exit polls, we compared the California estimates with the adjacent states and then we don’t really see a big difference in the party registrations. And then when we look at the statewide database data, it’s even more telling because the statewide database data actually gives a couple older registered voters in the state. And what we find is actually a pretty stable results over time. We don’t see 187 as a tipping point.

Grossmann: They also looked at adjacent states finding similar patterns in Latino partisanship.

Hui: We tried to do a limited adjacent states studies, because we want to have a natural experiment. So the idea is that if we see an impact in California, the impact would be strongest in California and that it will have a have a relatively weaker residual impact in the adjacent states. So what we see is not the case. We didn’t really find much of a difference, compare California with adjacent states.

Grossmann: Hui says there’s also not much evidence of effects on white voters. Republicans had already mobilized their base.

Hui: Wilson was actually pretty successful in mobilizing the hard core white Republican base. And then we do see strong support among the base and then that support didn’t decrease after 187. So the story for the Republican Party is that it’s fairly important to rally that base with some controversial issues.

Grossmann: And white voters changed little after the ballot propositions.

Hui: We notice that the white voters was not turned away by 187, so there was some suggestions in the convention with them that because of the radical rhetoric, the Republican Party actually turned away the white voters in the state of California. And that is not true. This is not what we find in our data. So I think that in that perspective that our result actually collaborate with our their stories, that we don’t see especially bad in California.

Grossmann: But Joshua Zingher found a different pattern comparing California and Arizona. Arizona anti-immigration politics took the form of SB 1070, a bill passed by the legislature to enable police to enforce federal immigration laws. Latinos reacted similarly in both states. But whites reacted very differently.

Zingher: Groups react in the ballot box. Groups respond in the ballot box to legislation or ballot initiatives that they feel target them unfairly. And I think there’s clear evidence that this happened in California. And I think there’s clear evidence that this happened in Arizona.

Now the effects, in terms of partisan ballots in the state, is really what separates California from Arizona. In California in the early 1990s was largely considered a swing state and post Proposition 187 in 1994 and several subsequent ballot propositions after that. The state has a combination of increased Latino voting power and a white electorate that is a little more Democratic than the national white electorate has moved into a state that is no longer a swing state and is a solidly Democratic state. Presidential elections, statewide elections, up and down the ballot.

Arizona’s story is a little different. Here, post SB 1070, which passed in 2010, Latinos in Arizona did move towards the Democratic party. But we saw, unlike we saw in California, we also saw whites towards the Republican party, which has the effect of largely canceling each other out and Republicans stay in control of state politics. You see the Republicans continue in statewide elections.

Grossmann: Zinger says most California research does show effects, but they mostly took the form of registration and new voters, and changes in presidential voting rather than partisanship.

Zingher: There’s a consensus in most, but not all research, regarding the effects of Proposition 187, that Proposition 187 and Proposition 209 which followed it, helped to push Latinos into a Democratic column. And there’s a number of pieces of evidence you can point to that speaks to this. Latinos preferences for voting for Democrats increased, as did turnout, and as did registration. So you kind of have a three fold break of Latinos toward the Democratic party. More were registering to vote, more were actually turning, more voters who registered were actually turning out and those who did register were voting for Democrats. I think there’s a number of pieces of evidence there that are quite clear.

If you look even between 1992 and 1996, and Proposition 187 was passed in 1994 there was 500,000 newly registered Latino voters between those two elections, which is a really, really big increase. Bigger than the population of some states. So you think, there’s a lot of evidence that Latinos responded to this ballot proposition.

Grossmann: Normally, we think of partisans reacting to national politics, but Zingher says state politics influence is possible for new voters.

Zingher: States I think is hard to separate what’s going on at the state level from what’s going on at the national level. Normally say, we think of politics, United States federal system kind of moving top down, right? You form preferences about presidential candidates, then these preferences are form your opinions of people lower down on the ballot.

While there’s evidence that this can go both ways, since state politics can color our perceptions of the political parties on the national level. And I think this is especially true in the case of Latinos, many of whom are recent voters, when you have a lot of recently, recent entrance into American politics, whether they’re first generation immigrants, second generation immigrants. And I think this is consequential because we have lots of evidence assessing these groups partisan attachments aren’t as nearly well formed as native born whites or African Americans or other groups that have been in the United States for a long time.

Grossmann: Looking to the California experience, Democrats had expected big Arizona gains, but so far they’ve failed to materialize.

Zingher: There’s a lot of optimism within the Obama campaign in 2012, and I think there was some also in the Clinton campaign in 2016, that Arizona might be in play. And you look at some of our simulations, you basically can take, they can be boiled down to this: If you assume that there’s going to be an increase in Latino turnout and there’s going to be a greater likelihood that Latino’s vote for the Democratic party in response to SB 1070. And you take into account that Latinos are a continually growing size of the state population, and you simulate those into the future, you can assess well, what’s the likelihood that Arizona will go blue?

What we found is, basically, the future of Arizona is not necessarily going to be dictated by as much Latinos as it is amongst whites. Even with the rapid population growth of Latinos, they’re not, the state is not going to be a majority Latino state for the foreseeable future. Certainly, your or I lifetime, it’s probably not going to hit that point.

So the question is, well, what will whites do over the long term in Arizona? If whites vote like they have in 2004, 2000, 1996, if whites in Arizona return to those levels of Democratic support, then the state could easily go blue. If they stay where they were in 2012, when I [inaudible 00:15:16] moved away from the Democratic party to the Republican party, there’s essentially no hope for the Democrats to win Arizona.

Grossmann: Republican turnout advantages in Arizona, kept Democrats from winning.

Zingher: Yes, they are winning more Latinos. Yes, Latinos are, have become more likely to vote, but no, Latinos are not, still not nearly as likely to vote as whites. Even as the group size increases, that turnout issue is still a big one. They’ve become more likely to turn out, but not nearly likely as whites, especially older whites. And if you do the math, if you can turn out at a much higher rate than the opposition, you can win despite having smaller group size.

Grossmann: Nationally, Zingher says Republicans might have similar advantages to those in Arizona.

Zingher: Whites nationally are more likely to vote for the Republican party than white people in California. That’s an automatic. Trump is sitting in a better position than Pete Wilson was twenty years ago, probably. If you do the math, going into the foreseeable future, Republicans can win if they win enough white people. It might be tough, but it can happen.

The counter argument there is Trump essentially threaded a needle. If you look at, how difficult it is to win a presidential election, while losing the popular vote by three million votes, that takes a really favorable distribution of votes across states to win the electoral college. It’s something he managed to do. I don’t think he’ll be able to win in 2020 with 45% of the popular vote, and so he’s going to have to come up with another source of votes from somewhere and where that somewhere is … well, I’m glad I don’t have to answer that question, it’s not my problem.

Grossmann: But Hui says, California’s experience might suggest long term effects that are difficult to reverse.

Hui: I can see the younger generation coming in, will be less and less likely to identify with the Republican party. And then, think of this cohorts ten years from now, they will comprise the bigger portion of the electorate. What we will see is a gradual decline, because of Trump, because of the Republican Party. either positions on the immigration issues, you’ll see a decline in the general support of the Republican Party.

And I think this is kind of also like a point of no return. It started way in the 90s and I don’t think there is even the possibility of reversal.

Grossmann: And Zingher agrees that Trump is taking Republicans down a high risk path.

Zingher: Trump is essentially trotting out Pete Wilson’s strategy on the national level. If you look at Pete Wilson, his strategy was a factor. He won reelection in 1994. The long term story is not kind for Pete Wilson, whose saying he’s the man who was the architect of the Democratic majority in California.

What will happen to the Republican Party post Trump? If you look at demographic trends, the electorate becomes less white every year. And even if you assume whites turn out at a higher rate than non-whites, all of a sudden, the window becomes narrower each subsequent election. And so you think, if the electorate was about 28% non=white in 2016, maybe we’re headed to 30% or 31% in 2020.

Grossmann: Hui says it wasn’t the ballot propositions, but the wider positioning of the Republican Party that brought it’s California downfall.

Hui: We cannot really blame 187 as a critical point in the decline of the Republican Party. Rather, I think it’s a story about the general composition of the state as it’s changing. And also the national party changing their positions to more socially conservative positions that don’t really resonate with voters in California.

Grossmann: And Zingher says the policies don’t actually have to hold up in court or be enacted for the symbolism to have an impact.

Zingher: It was passed in 1994 by the voters, immediately got challenged in the courts, a number of different legal challenges. These things played out, really for 4 years and Greg Davis was elected governor in 1998 and one of the first things he did as governor is told the state attorneys to stop defending Proposition 187 in court and the courts essentially struck it down and it was null and void.

And so the legacy of Prop 187 is a loss in the courts, yet we still saw all of these electoral random [inaudible 00:19:43] that I would argue electoral ramifications, from this ballot proposition.

SB 1070 has had better luck in the courts, although certainly not perfect. The question is, is that enough to produce a change? I think it’s really hard to wait for voters to attribute specific policy outcomes to specific parties or specific parties positions at this particular point in time. It’s really easy to identify the tenor and the tone of the messaging.

A couple of examples: Trump’s transgender ban in the military or the ban on immigrants coming in from specific countries that are linked to terrorism or supposedly linked to terrorism. You can think, alright, it might be less important whether these policies actually get implemented than singling to voters about where I stand on these particular issues.

Grossmann: But Hui reminds us not to believe the conventional wisdom without consulting the research.

Hui: You know, it’s important to count and you go back check the data. And I think what we are trying to do is use three different sets of data, use three different sets of methodology. And then we try to ask the same questions and then they both, all, come to a consensus that there wasn’t much of a change. And I think this is an important lesson learned, for me, is to count. Don’t always believe in the conventional wisdom.

Grossmann: There’s a lot more to learn. Political Research Digest is available bi-weekly from the Niskanen Center and on I-tunes. I’m your host, Matt Grossman. Thanks to Iris Hui and Joshua Zingher for joining me. Join us next time to find out whether public employee unions are tackling inequality or loading up states with debt, and why they’re influential even in red states.