Confessions of a Polling Truther

Do we have to spend the entire election season acting like it’s tin foil hat material to suggest that billion dollar corporations might manipulate publicly available data to protect their profits like fossil fuel companies with climate change, or big tobacco with cancer?

— little jon quijote (@brotherbeat) September 26, 2019

*****

In a Los Angeles Times survey of Bernie Sanders’ 2016 donors they found that the single largest group were those without a job. In September, as his campaign reached the million donor mark, 5 months earlier than in the 2016 cycle, we got a glimpse into Bernie’s current support. Pew Research released the results of its poll of Democratic voters which included the following graphic detailing the supporters of the major candidates

Open Secrets also released an interactive graphic that allows tracking of donors by profession. The professions with the strongest support for Bernie are teachers, nurses, farmers, truck drivers, and food service workers. The biggest employers for his donors were Starbucks, Amazon, and Walmart.

*****

Democrats represent 41 out of the 50 richest congressional districts, and all of the top 10. Democrat represented districts account for 61% of the GDP, Republican led districts account for 38%.

*****

• In June, in an analysis of Federal Reserve data, Matt Bruenig of The People’s Policy Project found that over the last 30 years the top 1% had increased its wealth by $21trillion, while the bottom 50% lost $900 billion in wealth.
• 40% of Americans earn less than the 1968 minimum wage, adjusted for inflation
• 50% of Americans make less than $30,000 per year
• $15/hr represents a raise for 40% of Black workers
• 78% of citizens live paycheck to paycheck
• 40% of citizens can’t afford a $400 emergency

*****

In September Berkley Economics professors Emmanuel Saez and Gabriel Zucman analyzed Bernie Sanders’ and Elizabeth Warren’s wealth taxes. They found that Sanders’ wealth tax was more aggressive on the ultra-rich and raised 68% more revenue than Warren’s.

While Bernie has the support of no billionaires, at least 20 have given to Warren over the last two years. Her pledge to reject big donor fundraisers don’t apply to money raised in those fundraisers for her senate seat transferred to her presidential campaign, or after the primary. That pledge also doesn’t preclude asking billionaires directly for money. In essence her pledge isn’t about who she’ll take money from, or who might buy her influence, it’s about how they might buy her influence.

*****

According to the media, Warren has been rising in the polls, even when she wasn’t. And Bernie has been losing support, despite maintaining steady poll numbers. Now, Warren appears to be first or second depending on the poll you look at. My Twitter timeline is full of people declaring Warren’s current position in the polls the end of Bernie’s campaign, 5 months before voting. Looking at the polling as a Bernie supporter it’s tempting to give in to a sense of despair or declare the polls irrelevant and ignore them. But it’s useful data. Let’s assume that polling accurately reflects the groups that those polled represent. It’s worth asking how well each poll reflects our actual populace. When considering the polls, statistical data, I like to also consider Bernie’s historic levels of volunteer and donor numbers, raw data, to ask if the statistical data reflects it. The intent of relying on polling without questioning or unpacking methodology is to give the impression that it’s reflective of reality. I’d suggest that it doesn’t tell a complete story for a number of reasons. If you feel that the media’s attempt to create a self-fulfilling prophecy has worked on the basis of polling there are important questions to ask. Does the polling reflect this:

This is an infographic detailing the concentration of the candidates’ donors. In order to get a sense of national support for candidates not Bernie, they had to remove his donors to see the rest.

When I suggest that the polling data doesn’t reflect these donors, the response is that Bernie’s million donors are a pittance compare to the size of the voting populace. This is true, but moot.

Is it likely that another candidate’s many fewer donors represent much deeper support just because of polling, or might it call for some skepticism of the polling?

I’d intended to review the methodology of all the polls used by Real Clear Politics to make up its average, widely seen as representative of candidates’ level of support. That quickly became boring. It’s much more interesting to consider what data is structurally excluded by polling and what pollsters decide to exclude. Polling is often focused on likely rather than registered voters or legally eligible voters. This is usually determined on the basis of past voting behavior. Not voting in the election between the two most detested candidates in history might make you a less likely voter and essentially excluded polling noise.

Even when the poll includes all registered voters there are entire segments excluded. Polling considers the class interests of those polled narrowly, if at all. When earnings are considered, voters are grouped like this:

Does the idea of grouping voters like this contain the recognition that half of Americans earn less than $30,000? Are any polls 50% comprised of people making 30k or less? Do food service and retail workers get polled in accordance with their raw numbers? Do Amazon fulfillment center and Wal-mart workers appear in the polling at all?

Granted, many low wage workers and impoverished people don’t participate in the political process. It’s usually unresponsive to their needs. They are as invisible to the process as they are to the polling.

A common refrain among Democrats is that poor white voters are voting against their interests when they vote for Republicans. The assumption at the heart of this statement is that voting for Democrats, who represent 4/5 of the richest districts, is voting in their interests. I can’t remember the last Democratic proposal for workers made once a democrat had gained office on the basis of campaign promises to workers. I do remember a wave of housing foreclosure under Obama and Republican sponsored welfare reform under Clinton. Focusing on likely voters, even all registered voters, offers an incomplete picture. In addition to the millions of voters who don’t appear in polling, there’s no precedent for accurately considering the impact of having a trusted, credible candidate whose platform is focused on the 50% of Americans after decades of politicians ignoring them.

Back in April I suggested that Biden’s numbers were hollow and more a result of sentiment than actual support, predicting that he’d rather not run and tarnish his image. I was right on behalf of his image, though the prediction was wrong. I still stand behind the suggestion that his numbers are hollow. It’s possible that the campaign’s strategy of avoiding live town halls, maintaining a light schedule, and hiding from the press is enough to counter the spontaneous bleeding eye, incredibly toxic policy history, forgetting names of candidates and his locations, or there are other possible explanations. He’s been frontrunner for the same reason that Warren appears to be rising, it’s true among the people represented by those surveyed. Those surveyed over-represent their distribution in the populace. Those polled have skewed older and wealthier.

How much can we trust the polls showing Warren gaining on and surpassing Biden? How likely is it that the rich white woman who claimed to be a racial minority with no history of working in or for any Black community suddenly has more Black support than Bernie just because she started putting ‘black’ and ‘woman’ together after announcing? I would propose that her standing in the polls is as hollow as Biden’s. Rather than a sign of strength, her $10 million ad buy in the early states is an attempt to secure the aesthetic gains of the press acting as PR. Honestly, desperation is the only way to explain such a big investment on highlighting corruption a week after The New York Times highlights that her pledge against the influence of money was meaningless. and as more video is disseminated making clear in yet another way that she lied about her heritage for personal gain. I’m truly frightened by the idea that her numbers are somehow real and this is a sign of her political instincts from a place of strength. It’s almost as self damaging as letting the president troll her into doing race science to protect a lie.

reprinted from Medium