Today, Milan speaks with Sumitra Badrinathan and Devesh Kapur about the findings of a brand new survey--the Indian American Attitudes Survey (IAAS)-- that sheds light on the political attitudes of Indian Americans (full disclosure: Milan is a co-author of the new study).
Although Indians in America account for less than one percent of registered voters, this election season they have been actively wooed by both Democrats and Republicans in an unprecedented manner.
Thanks to the increasing political influence of Indian Americans, the camaraderie between Donald Trump and Narendra Modi, and the addition of Kamala Harris to the Democratic ticket, there is a sense that this community’s votes are very much at play.
Today, Milan speaks with Sumitra Badrinathan (University of Pennsylvania) and Devesh Kapur (Johns Hopkins-SAIS) about the findings of a brand new survey--the Indian American Attitudes Survey (IAAS)-- that sheds light on the political attitudes of Indian Americans (full disclosure: Milan is a co-author of the new study).
Milan, Devesh and Sumitra discuss why Indian Americans, contrary to media reports, remain solidly with the Democratic Party and why they are overwhelmingly concerned with kitchen table issues, rather than foreign policy concerns such as U.S.-India relations. They also talk about the impact of Kamala Harris, partisan polarization among Indians in America, and why Republicans face an uphill climb to win over Indian American voters.
Welcome to Grand Tamasha, a co-production of the Carnegie Endowment for International Peace and the Hindustan Times. I'm your host, Milan Vaishnav. We are recording this conversation on October 9. Election Day in America is now less than a month away. One of the most surprising aspects of an already very surprising election year is the fact that Indian Americans have occupied a rather unexpected turn in the spotlight over the past several months. Although Indians in America account for less than 1% of all registered voters, they have been actively wooed by both Democrats and Republicans in a way that I think is truly unprecedented. Thanks to the increasing influence of Indian Americans, the camaraderie between Donald Trump and Narendra Modi, and the addition of Kamala Harris to the Democratic ticket, there's a sense that this community's votes are very much at play this election year. Today, a new survey is being released, the Indian American Attitudes Survey, that can shed light on the political attitudes of Indian Americans. It's a collaborative undertaking of the Carnegie Endowment, Johns Hopkins SAIS, and the University of Pennsylvania. It was a pleasure for me to be a part of this project but an even bigger pleasure to work with two close friends, Sumitra Badrinathan and Devesh Kapur, on this new and I think exciting initiative. Devesh and Sumitra join me today to talk about the first in a series of reports drawing on this new data. Sumitra, Devesh, good to have you on the show.
Thank you, Milan.
So, Devesh, let me start with you. You have been looking at this issue of Indians in America now for many years, you have written extensively about this - you have written perhaps the standard reference for understanding Indian migration to the United States, a book called The Other One Percent: Indians in America. We'll link to that. Before we get into the nitty gritty of the survey, if you kind of step back and approach this at a 30,000 foot level, what do we know to date about the political behavior of Indian Americans? How would you characterize what their leanings have been?
The first survey on this was, I think, the one I did in 2004, which was the first National Survey of Indian American voters. This was just around the time of the transition between the Vajpayee NDA and the UPA in India, and there was a general sense at that time as well in the media that Indian Americans' political inclinations were more to the right. But in that survey, which had a sample of more than 2000, I found two things. One is that the support for Democrats was about five times more than that for Republicans. A large number of the respondents either said that they didn't have any views or were independents, close to half. But the longer they lived in the US, [the more that] their political views formed, and they were more likely to become even more Democrat than remain Republican. So, since then, every other survey from Pew, the [National] Asian American Survey from UC Riverside, each of them has been very, very consistent. Indian Americans are very heavily Democrat voters.
So, we're going to talk about in a second why that's the case. But let me just ask you a question about Indian immigration writ large, because this might not be something that's very well known to a lot of our listeners. What does Indian immigration to the U.S. look like? How big and how fast has this community grown?
So, just a bit of history: there was a little bit of immigration from what was then British India at the end of the 19th century, early 20th century, when the U.S. passed penal laws that basically banned immigration from that part of the world around 1920. At that time, there were about 5000 people from what was then British India, and the numbers dropped so much back in the 1940s and 1950s censuses they weren't even separately recorded. By 1960, there were about 12,000 people from India in the U.S., and from then, and especially after the 1965 Immigration Reform Act, the numbers began to jump every decade. It was about 51,000 in 1970, about 206,000 in 1980, 450,000 in 1990, 1 million in 2002, 2.8 million in 2010, and in 2018, it's about 4.2 million.
So, in other words, the Indian population in America has doubled since the year 2000.
No, it's more than doubled. Two thirds of all Indians in the U.S. have come after 2000.
So, this is not only a growing population, but arguably a rapidly growing population. I think most data we have shows that if you look at the past couple of years, Indians are either the second- or the third-, depending on how you define it, fastest growing minority group in the country. Is that right?
Right. So, basically, they are the second largest immigrant group after immigrants from Mexico, with mainland China being third. If you count Taiwan as part of mainland China, then Indian immigrants from India would rank third.
So, Sumitra, let me turn to you. You have conducted numerous surveys on public opinion in India - we have collaborated on some of those, the three of us have. What is the value-add of doing yet another pre-election survey ahead of the U.S. election? It seems like we're inundated - those of us who are constantly checking the New York Times, 538 - with a new poll every day. What is new and different about this?
Yeah, my social media feed, like yours, is completely saturated with information about polling. And it's funny you mention 538, because I think I checked their tracker yesterday, and they added something like 15 new polls to their model just over the last three days alone. But what's important is that none of these polls are about the demographic that we're discussing here, Indian Americans - we talked briefly about Indian Americans being in the news this season. So, I think it stands to reason that a survey specifically focused on their views would have a lot of value. Now, I think this is important for two distinct reasons. First, there hasn't been a single survey conducted until this one that we know of that studies Indian Americans alone. So, what this means is that all the knowledge we have, everything we know so far about this community, comes from a sub-sample of surveys that generally study Asians in the U.S. writ large. So, suppose a study surveyed about 1000 Asians - it would have to be split into several different nationalities to comprise that sample size of thousand, so Indian Americans end up being 200, maximum 300, of that larger sample size. So, what this means is that you end up drawing inferences about the whole country from a smaller sub sample that you can't guarantee is actually from the whole country. What if that sub-sample was entirely from Texas or California, where there are a lot of Indian Americans? So, that's the first thing. The second thing is that you see this quantity in surveys called the margin of error, which basically describes how sure we can reasonably be that a survey result is close to the true population value. So, we're surveying actually a small sample of the entire population of Indian Americans, so we want to know how true this result is that we got, how close it would be if we actually surveyed the entire population. So, with smaller sample sizes, the margin of error increases, and our confidence in the result decreases. So, what we have over here is, first of all, a nationally representative survey. We've represented Indian Americans for the population from all of the states in which they have a high number in the population. Second, we have a margin of error of about 3.2% overall on a sample size of 936 Indian Americans alone, which means that for most quantities described in this survey, we can be reasonably confident that our results are coming from a sample which we can draw inferences from and that the differences are true differences and not just because of sampling error.
I want to dig into the findings, but before I do that, let me ask you another question about methodology, just to make sure that our listeners understand what went on here. Devesh said earlier that Indian Americans are only around 1% of the population, and that's a statistic you hear quite often. So, that means that this population is actually pretty hard to reach, because 1% is not that big of a number. So, how does one actually survey Indian Americans, then, given the fact that their absolute number relative to the U.S. population is not that big?
Yeah, this is a great question. So, we actually partnered with a survey firm, YouGov, that does a lot of political and other polling in the United States. They have a proprietary online survey panel into which Americans in the U.S. have opted in already. This panel has about 1.8 million U.S. residents who have agreed to participate in YouGov surveys. From that panel, we recruited a subsample of people who identified as Indian American, and then within that, we wanted to match people based on a sampling frame, the gender, age, education, and region that they came from. So, while it was hard, and while it took more time than I think the average survey of all Americans would, we managed to do this using an online proprietary package.
So, getting into the findings here, you know, I think one of the big headlines of this new report is the fact that Indian Americans remain solidly with the Democratic Party. 72% of registered voters surveyed plan to vote for Joe Biden this November while just 22% intend to back Donald Trump. Devesh, there has been quite a lot of talk this election season about the defection of Indians from the Democratic Party and into the arms of Donald Trump and the Republicans. The data that comes out in this report doesn't really seem to provide much empirical evidence for this kind of a shift. Why do you think the conventional wisdom is wrong?
Well, it's wrong because it's intellectually lazy. So, think of the sort of logic that we see that is behind the conventional wisdom. One is that Indian Americans are very strongly pro-Modi, Prime Minister Modi seem to be close to President Trump, and therefore there's almost a transfer of that allegiance from Mr. Modi to President Trump and therefore a shift to the Republican Party. Now, this sort of presumes that Indian Americans decide that they are living in America, this is their country, this is their future, and that they do not care about the hugely important issues that are very much domestic in nature that will shape their future and will shape the future of the United States. So, things like health care, the economy - and above all, also, Indian Americans are immigrants, they are minorities, and so they are obviously very worried about the rhetoric of the Republican party that is very anti-immigrant and also anti-minorities. I think the other thing we should be very careful about is that, you know, we often see that there's a lot of people who focus on visible events like the Howdy Modi event or when Prime Minister Modi came to Madison Square Garden, but one should be very careful in drawing inferences about a population from just looking at a very tiny slice, because that's - essentially, it's what we call "selecting on the dependent variable." You're looking at where there is heat, but often where there is heat, there isn't much light. And I think that's what the difference is between what our survey finds and the conventional wisdom.
You know, it's interesting, just to pick up on one thing that you said - again, contrary to the media narrative, Indian Americans, according to this survey, don't rate U.S.-India relations anywhere close to the top of their priority list when it comes to the issues that really matter to them on election day. So, this survey asked them, you know, "When you go to vote, what is it that you're primarily thinking of?" What came back instead was that bread and butter issues like you mentioned - things like health care, things like jobs, things like the economy - seem to dominate. I'm wondering, Devesh, if this is a surprise to you, or is it just validation of the fact that Indian Americans maybe vote just like everyday Americans?
Well, so I think you've put your finger on the key issue. Why would we not expect the bread and butter issues to dominate? I mean, this is a pretty exceptional year, right? The magnitude of the domestic crisis is so severe. One would have to be very, very surprised if bread and butter issues did not dominate. So, to me, it's pretty obvious that people care about their future, about their family's health, about jobs, all of these issues. It's not that Indian Americans don't care about U.S.-India relations, but relative to what is pressing now, that is clearly - it's not a priority.
Arguably - just to build off of what you just said - it would be even more so this year when you have three simultaneous crises, right? You have a public health pandemic, you have an economic crisis not seen since the Great Depression, and then, of course, you have this set of repeated convulsions over race and social justice and policing and so on and so forth. I want to be careful, though, obviously, not to speak of this community as a monolith. One of the things which comes out in this report is a look at polarization within the Indian American community. You know, this past summer I read Ezra Klein's book, [Why We're Polarized], where the argument basically summarizes a lot of political science literature, saying, "Look, partisan identity in America has now become a central part of our own individual identities." So, Sumitra, the question to you is - people in academia talk a lot about "affective polarization," which is just basically a fancy way of saying that people from different parties don't seem to like each other very much or their leadership. Is that true for Indian Americans as well?
Yes. And you're right that the phenomenon of animosity between supporters and political parties is an effect of polarization. We see a lot of this generally in the American context. Both Republicans and Democrats say that the other party's members are hypocritical, selfish, closed-minded - they don't want to socialize, marry, or even date across party lines. One of the ways we measured this in our surveys is through an item called the feeling thermometer. Respondents basically are asked to rate Democrats and Republicans, or the Democratic and Republican Party, on a scale ranging from 0 to 100, where a number closer to 0 indicates that you feel coldly, numbers closer to 100 indicate that you feel warmth, and the number right at 50 would mean that you're neutral or indifferent. So, in 2016, for instance, researchers found that there was a 40-point difference in the way Democrats and Republicans rated their own party relative to the other party. Now, in our survey, we actually find a pretty high degree of polarization of this kind. So, let me give an example. Democrats in our survey rate the Democratic Party on average 75, which is very warm, but they rate the Republican Party only 39. So, this is a difference of 36 points. And, similarly, for the Republican Party, we find a difference of 40 points. But I'll also say that the highest difference in partisan ratings was saved for President Donald Trump: a difference of 42 points, where Democrats gave Trump 27 lower than even what they rate the Republican Party on average, and even Republicans gave him a slightly lower rating than they do for the Republican Party. So, to sum up, though we have way more Democrats in our sample as compared to Republicans, we see pretty big differences in the way Democrats and Republicans assess their parties and political leaders. And this carries forward to attitudes. We asked respondents their opinions on a variety of policy attitudes, including - you mentioned race - attitudes towards police violence against peaceful protesters. We find that there are pretty big partisan differences with respect to those attitudes, too, mirroring generally the American political behavior context, which is that it feels like partisan socialization, the kind that you just mentioned earlier, is reflected within this demographic.
So, let me ask you about one particular leader who has gotten maybe disproportionate attention, which is Kamala Harris, given that she is partially of Indian origin. You know, there has been quite a lot of speculation in the media and on social media about whether she is going to energize the Indian American community because of her heritage, is she going to turn them off because in the past she has made what some would perceive to be critical statements about the Modi government and some of its policies, or would it just kind of be a wash and not really matter all that much. So, I'm wondering, what do you think we can say from the survey about her net effect to the Democratic ticket?
So, the survey approached this issue through two different lenses. First, we asked respondents whether Biden's selection of Kamala Harris would make them more or less likely to turn out to vote in the upcoming election. Second, we asked whether the selection of Harris would make them more or less likely to be enthusiastic about the candidacy. So, first, with respect to turnout, we actually find that 45% of all of the respondents in our sample say that Harris' selection makes them more likely to vote in November while only 10% say that it would make them less likely to vote. But it's important to note that a huge chunk, 40%, of our sample said that it would make no difference either way. So, in all, choosing Kamala Harris does seem to have galvanized a section of Indian American voters with respect to actually turning out to vote. Now, on the other hand, turning out to vote could mean that you would turn out to vote either in favor of Harris and Biden but also maybe in favor of the Republican Party. So, we actually find that the mobilization effect is actually working in favor of the Democratic ticket - in our sample, 49% of respondents say that the nomination made them more enthusiastic toward the Biden candidacy and only 15% said it makes them less enthusiastic. Like the previous question, a pretty large chunk, 31%, actually said that it would not make difference either way. And the final thing I'll say on this is that amongst respondents who said that Kamala Harris, would make them more enthusiastic about the Biden candidacy, not surprisingly, we find that our Indian American respondents say that the reason they're more enthusiastic is because Kamala Harris is Indian American.
You know, one of the things that some research by a political scientist named Sara Sadhwani has pointed out is that, when you look at Asian Americans in particular, Indian Americans seem to be more mobilized to actually vote in favor of one of their own than even other Asian Americans if there is an Indian American on the ballot. So, I think this is a kind of interesting test of that. Devesh, let me circle back to something that we talked about before. In your book, which came out a few years ago, you posed the question of why Indian Americans are so tied to the Democratic Party, and in one of the answers you gave before, you pointed out the issue of immigration. Now, I think a lot of people look at the Indian American demographic and say, "Wait a second, they are doing very well in terms of median household income, they're doing very well when it comes to educational attainment..." I mean, literally, if you plot them on a chart, they're almost falling off the chart. So, we would actually expect them perhaps to be more Republican than they are. What do you think the survey tells us about why Indian Americans seem to be put off by the Republican party? In other words, the Republican party seems to have an image problem. When it comes to Indian Americans, is it just immigration, or do you think there are a broader set of forces at work?
So, you're absolutely right, Milan. Given their income, which is twice the median for Americans, you would think they would want a party that supports lower taxes; many are entrepreneurs, so they would support a party that calls for less regulation; many are doctors who like tort reform. All of these are Republican Party issues. So, you would think that economic self-interest would drive them to the Republican Party, and the fact is that it's the opposite. We should ask why. And so that's what our survey also shows - I think in the Republican Party, it's pretty obvious in this election [that] it's more than just immigration. Of course, Indian Americans have been hit [with respect to immigration, too]. The way that so many Indians have come in the last 20 years is the H1-B visa, and this government, the Trump administration, including a few days ago, has tried its best to squeeze down H1-B visas. But I think there are two other broader issues that our survey highlights. One is the way the Republican Party, around race, seems to be becoming a white-only party, [and] they are very worried about its stance on minorities. As we said earlier, they are 1% of the of the population, they are a small minority. All minorities are worried about the attitude of the Republican party, and Indian Americans are one of them. The second thing that comes out is the apprehension about the growing hold of the Evangelical Christians on the Republican party. It's something we had noticed earlier through interviews and focus groups, but now we see very clear survey evidence. Most Indians in the country are not Christian - they're Sikhs, they're Muslims, they're Hindus. And so a growing hold or influence of Evangelical Christians on the Republican Party, in addition to its stance on minorities - I think these are two very compelling reasons why they are turned off by the Republican Party.
One of the most interesting findings to me is on the issue of foreign policy. So, the report clearly finds that, for most survey respondents, when you ask them to pick from a list of competing priorities, U.S.-India relations kind of floats towards the bottom. But the survey does elsewhere ask Indian Americans about their views on Narendra Modi. And I think that most of us believe, at least anecdotally, that the Prime Minister of India is extraordinarily popular amongst the diaspora in America. It's not just Howdy Modi, it's not just Madison Square Garden - I think in most of our conversations, if you were to kind of take an average internally, you sort of get that sense. But Sumitra, what does this survey actually tell us about the Prime Minister's stature here in America among Indian Americans?
This is a good question. So, based on the data that we have, I wouldn't conclude that the sample of Indian Americans was overwhelmingly pro-Narendra Modi. Actually - we talked about the feeling thermometer scale earlier - we asked this question for Prime Minister Modi, too, and we found that on average Indian Americans rate him at 55 on a 0 to 100 scale. So that is slightly on the warmer scale, but closer to 50, which is actually indifferent, which kind of places him somewhere between what they rated Biden and Trump on average, but closer to indifference. But, for me, the most interesting finding was when we disaggregated this result by partisanship. So, we actually find that Trump voters give Prime Minister Modi the highest thermometer rating, at 70, and Biden voters give him the lowest thermometer rating, at 52. So, it averages out to about 55 because of the larger number of Democrats in the sample. The other thing I'll mention is that the survey also asked respondents whether they approve of the job that Prime Minister Modi is doing as a prime minister. We find that 48% of Indian Americans say they approve, 32% disapprove, and 20% express no opinion. However, Indian Americans who approve of Trump's job performance are also more likely to approve of Modi's performance.
You know, what you see from these numbers is, yes, he is reasonably popular among Indian Americans, but less popular than Prime Minister Modi is in India. If you see Prime Minister Modi's ratings in India on his job performance, and his ratings in the U.S. among Indian Americans, very clearly, his rating here is less than his rating within India itself.
Devesh, as a quick follow up, one of the unusual hallmarks of U.S.-India relations - going back to the year 2000, which is really when the strategic partnership kind of started to lift off - is the sense of bipartisan consensus, both in India and in the United States, about the need for closer U.S.-India ties. So, in other words, there was a consensus between the BJP and the Congress in India and between the Democrats and the Republicans in the United States. Do you worry, as you look out into the future, that there is a risk of fracture there? In other words, that Democrats may take a less favorable stance towards India under Modi because of their concern over democratic backsliding, over minority rights, over the situation in Jammu and Kashmir, maybe other aspects of economic policy as well - does India risk, in your view, becoming a bit of a political football?
I think much less so than some people fear because there are much bigger forces at work that will influence the shape of U.S. foreign policy, and that, frankly, is the specter of the rising China. And so, they may not particularly like the current regime in India, but they also have to face a much harsher and a much more difficult global political landscape for the U.S. And so, to that extent, I don't think there will be a fracture. It may be less ideal, perhaps. But on the other hand, as you know, on trade issues, on H1-B visa issues, the Trump administration has caused a lot of problems with India - on things like climate change... India and a Biden administration will have much more in common than India and the Trump administration. So, yes, on a few things, it will be less, but on a few things, actually there will be much more agreement.
One of the survey findings we haven't talked about: respondents are asked which party, the Democratic Party or the Republican party, does a better job of managing U.S.-India relations, and by a two-to-one margin, respondents select the Democratic party. Now, in part that's based on the fact that, you know, they're more Democratic-leaning generally. But it does give you a sense that, again, some of the media narrative about there being a lot of fear in the American community if the Democrats were to come to power, because they might not be as good toward India, doesn't seem to get backed up by the data. I want to end by asking two questions about the future, and Devesh, let me start with you. You have pointed out that there are about 150,000 Indian Americans who become newly eligible to vote each year. Now, most of this comes from younger Indian Americans who turn 18, and then they become eligible to cast their vote for the first time. But about a third of it comes from naturalization - immigrants becoming citizens. But, however you slice or dice it, they remain as what you talked about before, one of the fastest growing human groups in the country. What do you think the survey tells us about their particular place in America's political dynamics? How would you place them on a spectrum when we think about the white population, other minorities - how do they fit in?
Three features of this group, looking into the next few years, stand out. One, because a greater fraction are those born in the U.S. - this group, our survey shows, is more liberal than naturalized Indians, which were the dominant Indian Americans in the past - the community as a whole, its political preferences' center of gravity will become even more liberal. Second is not just their numbers or views but also their location. In the past, it was concentrated in places like the northeast and California. These were reliably blue states, so their marginal impact was lower. But now you see a considerable growth in states like Georgia, Florida, swing states like Michigan, Pennsylvania, and also Texas, which is turning somewhat purple. So, to the extent that they are, and the numbers are growing enough in closely tied elections or in close elections, these numbers will begin to matter much more. So, it's the overall numbers, the fact they're becoming more liberal, and the states where they are growing more - all these three combined. As we see in our surveys, they're roughly between Hispanics and African Americans in how pro-Democratic Party they are, and that dynamic is likely to continue into the near future.
So, this report is part one of a larger exploration of Indians in America. The next study is going to look at how Indian Americans view India. This study just touches on it tangentially but doesn't really get into to the wide array of survey questions that have been collected. Tell us a little bit more about why that's an important or interesting question to look at - what Indian Americans think about what's happening in India?
First, let me say that our original survey had about 1200 respondents, both citizens and non-citizens, and we dropped the non-citizens for this analysis because it's relevant just to the U.S. election, but we did want to include non-citizens because we have a whole module on how Indian Americans view India. We've discussed a lot about the conventional wisdom about Indian Americans and about the diaspora's attitudes towards Indian politics, but the truth is, we don't actually have data or numbers to put to these hypotheses and conventional wisdom. So, that is our aim with this section. So, in that section of the survey, we actually have a very vast module on attitudes toward policy issues and current affairs issues in India. So, for example, in that section, we have questions about what people think about the suppression of press freedom in India, what people think about the All-Indian National Register of Citizens and the passage of the Citizenship Amendment Act, both of which had prompted pretty vast protests across the country before the lockdown happened. We have questions on whether affirmative action is appealing to people or not, and we also ask questions about police and law enforcement using physical violence and tear gas, which we saw in response to the citizens to the citizenship laws. And finally, we also asked about religious majoritarianism in the country and whether people think that's a threat to democracy or not. Now, what's interesting about these questions is that we have pretty similar questions about the U.S. context, where instead of asking about religious majoritarianism, we asked about white supremacy, or instead of asking about affirmative action as it relates to caste, in the U.S. context, we asked about it as it relates to race. So, one of the hypotheses that I think we're interested in studying is whether people in our sample view the same issues differently in the Indian context as opposed to in the American context. So, I think I and all of us are excited to dig into this data because it promises to be pretty telling.
My guests on the show today are Sumitra Badrinathan at the University of Pennsylvania and Devesh Kapoor of Johns Hopkins SAIS. The three of us are co-authors of a new study titled "How will Indian Americans Vote?" It presents new data and new evidence from the 2020 Indian American Attitudes Survey. The report is out now. We look forward to hearing from you and your comments and your questions. Sumitra and Devesh, thank you so much for taking the time to break down and explain the survey findings. And I hope that you guys will come back in a month or two when we have part two of this report on how Indian Americans view India to talk about our findings on that one.
If you'll have us.
[Laughing] It depends how much work you're going to do on that report. But yes, we would love to have you both back.
Thank you, Milan.