Thursday, December 29, 2016

The vote by Miami's Cuban Americans was a referendum on Obama's Cuba policy

My last word on the 2016 election

On December 16, 2016 The Miami Herald published a story with a headline reading as a question: Was vote by Miami’s Cuban community a referendum on Obama’s policy?  that was two separate articles in one. The first part contained statements by elected representatives affirming that the Cuban American vote was a referendum on Obama's Cuba policy. U.S. Rep. Ileana Ros-Lehtinen, a Miami Republican. stated it simply: “Within the Cuban-American community, the presidential election results demonstrate a direct rebuke of Obama’s Cuba policy reversal.”

The rest of the article read like a press release from the anti-Embargo lobby that ignored some essential facts contained within the polls cited. Prior to Donald Trump changing his position on September 17, 2016 from lukewarm support for Obama Cuba policy to promising to roll it back his support ranged from 35 to 37% based on two polls takes months apart. The FIU Cuba Poll was taken between July 11 and August 12, 2016 with a random sample of 1,000 Cuban Americans residing in Miami - Dade County. According to Tim Padgett in his September 15, 2016 article  "FIU Poll Of Miami Cubans Shows Strong Support For Obama Normalization Policy" only 35% of Cuban Americans were going to vote for Donald Trump according to the poll with a margin of error of 3 percent. This number tracks with the 37% support for Donald Trump in the Moreno Poll of 400 likely Miami-Dade Cuban American voters conducted between April 21 - 23, 2016 that had a three point margin of error.

The Bendixen-Amandi poll with +/- 3.99 margin of error conducted between October 15 - 17, 2016 placed Trump's support among Cuban Americans at 47%. I have raised questions in the past about Bendixen - Amandi and their shilling for the anti-embargo crowd, but even here looking at a 10 to 12 point shift demonstrates that the before and after on the Trump Cuba policy shift did have an impact.

Even more telling is what the Clinton campaign did after Trump shifted on Cuba policy airing radio advertisements in early October 2016 accusing Mr. Trump of violating the Cuban embargo.  The Trump candidacy should not be simply compared with previous Republican presidential candidates because the others had supported maintaining sanctions on the Castro regime throughout the entire campaign. Another factor was that with the exception of Kerry in 2004 since 1992 the Democratic candidates for president including Obama in 2008 and 2012 have also supported sanctions. However the actions taken by the Obama Administration on Cuba in October of 2016 placed Cuba on the campaign agenda and Trump's change on Cuba policy created a stark contrast with Hillary Clinton that helped him win more Cuban American votes.

Following the election a precinct analysis looking at where Cuban Americans voted the most in the 2016 election concluded that 58% of Cuban Americans voted for Donald Trump in the 2016 election. There has been an effort by Cuban Americans who have vigorously supported Obama's Cuba policy to downplay the Cuban American vote and the impact of the new Cuba policy on that vote.


  1. John --

    If the 2016 election was a referendum on Obama's Cuba policy then the results were clearly an endorsement of that policy since Clinton's performance in Cuban precincts and neighborhoods was much better than Obama's. (By double digits.)

    Trump was never going to just get 35% of the Cuban vote. Just like Clinton wasn't going to stay at 31%. (Are you suggesting Clinton would have won 70% of the Cuban vote or that a third party candidate was going to win a third of the vote?) You keep making this absurd point, but ignore that the fact that Trump's margin over Clinton never changed before/after changing his position on Obama's policy. It averaged 7-8 points before/after his policy change. If your theory was right, his margin over her would have jumped after the shift, but it didn't.

    If you want to know the reason why both of their numbers were in the mid 30's, it's probably because those polls were taken earlier in the campaign where you have more people telling pollsters that they're undecided, and the FIU poll did not have a likely voter screen question, which would have filtered out more undecideds.

    And, again, that precinct analysis you guys keep citing includes a mathematical error.



    1. Moreno Poll April 2016: Cuban American support at 37%

      FIU Cuba Poll July - August 2016: Cuban America support at 35%
      Trump shifts on Obama Cuba policy on September 17, 2016

      October 14, 2016 Obama issues presidential directive on Cuba calling on US intelligence services to seek areas to cooperate with Castro spy services.

      Amandi-Bendixen Poll Oct 15 - 17: Cuban American support at 47%

      October 26, 2016 Obama Administration abstains on UN resolution against US Embargo on Cuba.

      November 8, 2016 exit polls place Cuban American support for Trump at 54% and precinct analysis of actual votes places that support at 58%.

      The charge that the math on precinct analysis is flawed needs to be demonstrated.

      Anecdotal evidence of Cuban American voter who had voted for Obama in the past and flipped to Trump in 2016 appears in Babalu Blog on November 28, 2016

      Finally the abandonment of Republicans of anti-communist approach to China led to their support among Chinese Americans, especially after 1989, to collapse to the 30s. If Trump had maintained his luke warm support for Obama's Cuba policy he would have done worse among Cuban Americans.

  2. John,

    None of this is personal, so don't take it the wrong way, but making the same points over and over again doesn't make them correct.

    This is all basic math.

    There were three polls with significant Cuban-American samples taken before DJT changed policy and three taken after:
    - May: DJT 37%, HRC 31% (+6 Trump / Moreno, GOP)
    - May: DJT 41%, HRC 29% (+12 Trump / Bendixen, Dem)
    - Aug: DJT 35%, HRC 31% (+4 Trump / FIU, Ind.)


    - Oct: DJT 47%, HRC 41% (+6 Trump / Bendixen, Dem)
    - Nov: DJT 54%, HRC 41% (+13 Trump / Edison, Ind.)
    - Nov: DJT 52%, HRC 47% (+5 Trump / LD, Dem.)

    As you can see, the results of the six polls are virtually identical.

    (You're not really supposed to take averages of percent since each poll has different sample sizes, etc., but for our purposes, we'll just run a straight average of those numbers...)

    AVG BEFORE: DJT 37.67%, HRC 30.33% (Trump + 7.33)
    AVG AFTER: DJT 51%, HRC 43% (Trump +8)

    You're arguing that by changing his position on Cuba policy, Trump's numbers went from the mid 30s to the high 40s / low 50s. But you ignore the fact that neither Clinton or Trump was ever going to stay in the 30s and that those polls naturally had higher percentages of undecided voters who weren't going to stay undecided. This is normal in polls taken earlier in an election cycle.

    Regardless, let's look at the change for both candidates:

    BEFORE / AFTER: DJT + 13.33, HRC + 12.7

    If you round, BOTH TRUMP AND CLINTON'S SAW THEIR NUMBERS GROW BY 13 POINTS. You can't say "Oh but look after changing his Cuba policy, Trump's numbers grew by 13 points" and then ignore the fact that Clinton's grew by the same exact amount.

    If you compare their 13-point growth to where they started, Clinton's surge was larger than Trump's. Her share of the Cuban-American vote grew by 42% (not points, percent), while Trump's grew 35% (again, not points, percent).

    As far as the precinct analysis: There are a lot of things wrong with it. First off, there is no such thing as a homogenous Cuban precinct. Second, those precincts are mostly in the western suburbs of Miami-Dade and not necessarily representative of how all Cuban-Americans vote, and finally, like I've said, your math is wrong. (Whoever performed that calculation did one of two things wrong to reach 58%, they either ignored votes for third party candidates, which artificially inflated Trump's share or they took an average of the percentages, which s I explained earlier, isn't kosher in math).

    Regardless, even if we use the absurd precinct analysis and compare it to how those precincts voted four years ago, you see an 11-point growth for Clinton vs. Romney. If, as you argue, Obama's Cuba policy hurt Clinton, you wouldn't see Democrats making gains in the most heavily Cuban-American and reliably GOP areas precincts of the County.

    Furthermore, if you take an average of how Westchester, Hialeah, and West Miami voted (where 3/4 of voters are Cuban-American), you get Clinton 48% and Trump 50%. Compare that to how Obama and Romney fared in those three places four years ago, and it's a 10-point improvement for Clinton over Obama.

    So, again, there's no evidence that Obama's policy hurt Clinton in any way. There's also no evidence that by doing a 180 on Cuba, Trump improved his standing among Cuban-Americans. And even if he had consistently opposed Obama's policy, there's no evidence that his performance would have been any better. Three out of four Cuban-Americans who are self-described embargo supporters are also self-identified Republicans. Those voters were going to come home for him regardless of his stance on the issue because by and large, Cuban-Americans are not single-issue voters.



  3. Forgot...

    As far as your GOP / Chinese-American voters argument: That's a nice theory. Prove it.

    I don't think most people, including Cubans or Chinese, are single issue voters. Decades of research has shown that people vote based on a variety of factors, including personal values and how they feel the respective parties' nominees reflect them. People are usually not single-issue voters. It's really tough to make an argument that an ethnic minority would feel embraced by the GOP at this time, and I don't think a hardline stance on China would change that.

    John, I commend you for your human rights activism. Honestly, I mean that. You should stick to those kinds of arguments and leave the electoral ones alone. This isn't your strength and you're arguing yourself into a corner. If you want to say a policy should be XYZ for human rights / moral reasons, that's perfectly fine and debatable. But basic math doesn't support your electoral argument; it just isn't there.

    All the best,


  4. Sorry, typo...that should have said "an 11-point growth for Clinton vs. Obama..." not "vs. Romney..."

    Okay, I'm done now.

    Best wishes,


  5. Most folks are not single issue voters but some issues way more heavily then others. I am not the only one that has observed that Republicans got a large share of Chinese American and Vietnamese American votes due to their anti-communism. In both cases, coincidentally, after Republicans jumped on the normalization campaign that support collapsed.

    With regards to working class Cuban American voters in Hialeah do you think that Cuba policy is not an important issue to them?

    Furthermore comparison between Romney & Trump support among Cuban Americans should also factor in that Romney maintained a harder line on Cuba throughout the campaign whereas Trump only took that position in mid-September 2016.

    Not to mention the Clinton campaign's radio ad blitz against Trump in late October on violating the Cuba embargo issue is also telling. If this was not an issue that would impact voters then why spend the money?

    Finally, why where press accounts fearful that Obama's Cuba directive and UN abstention on Cuban Embargo resolution vote were going to hurt Hillary?

  6. John,

    I don't know how important Cuba policy is or isn't to working class voters in Hialeah. All I know is that the polling consistently shows that Cuban-Americans aren't monolithic in their support of those policies, and that their stances on Cuba policy isn't necessarily the best indicator of how they'll vote in elections. That's why I think this whole argument that Trump's 180° on Cuba helped him is silly.

    The Cuban-American vote for POTUS has been consistently trending Democratic since 2008, while at the same time, members of Congress and Senators who support the embargo have won strong majorities of the Cuban-American vote. It's unclear what, if any, role US-Cuba policy is playing in how Cuban-Americans vote. By the way, some perspective is much needed here: the fact that voters' views on specific policies isn't consistent with how they vote in elections isn't a unique phenomenon to Cuban-Americans and U.S.-Cuba policy. (See: views on gun safety, universal healthcare, entitlements, tax cuts, etc.)

    The differences between Romney and Trump's support levels among Cuban-Americans should factor in a lot of variables, not just U.S.-Cuba policy. Three quarters of embargo supporters were self-described Republicans. In other words, Trump was already going to win over the vast majority of people who support that policy. And if your theory was right, his margin over Clinton would have gotten a bounce after changing his position on the issue, and they didn't.

    The community is changing demographically, too. My family is a perfect example of this. In the last 13 years, all of my grandparents have passed away, unfortunately, and they were solid GOP voters. Concurrently, my cousins and I have all become of voting age and we vote Democratic. I'm not in the crowd that says that Cuban-Americans are voting more Democratic because they've changed their minds on U.S.-Cuba policy. I think most of what's driving the shift in views on U.S. policy is the same thing that's driving the change in voting patterns: demographics.

    As far as radio ads: Low and mid-level staffers write scripts for radio local radio ads. Maybe they wanted to keep Trump on defense? Who knows. What I will say is the fact that his numbers didn't drop after that story came out goes to show that this issue isn't as potent as you may think it is. As far as news stories: who knows. Maybe Capitol Hill Cubans planted those stories? Maybe it was lazy reporting? Maybe it was speculative reporting?

    What we know from empirical evidence and hard data is that Obama's policy didn't hurt Clinton and Trump doing a 180 on it didn't help him either.

    I don't believe Democrats should take it for granted that Cuban-Americans will continue to trend in their direction, and it's foolish for Republicans to think they can stop bleeding or win them back via U.S.-Cuba policy appeals. The bulk of persuadable Cuban-American voters who are truly up for grabs in an election tend to be younger and English-speaking. They want to hear about bread-and-butter issues, not rhetoric on Cuba.



  7. Sorry typo should read "some issues weigh more heavily than others."

  8. Are there any available data describing the Cuban-American voter turnout by age in the 2012 and 2016 elections?