For the past few months, the Forward has been working on a project to quantitatively analyze Jewish political activity in the Democratic presidential primary. Our first article in this series, examining Jewish donations to Democratic presidential candidates, can be read here.
Which donors did we track?
Every three months, candidates for president and Congress have to declare their fundraising information to the Federal Election Commission – how much they received and from whom, and how much they spent and on what. Most of that data is then made available on the FEC website.
Specifically, campaigns must publicly disclose what are called “itemized contributions” – every donation made by someone who has given them a combined total of $200 or more since the campaign began.
So if John Doe wrote a $199 check to President Trump’s reelection campaign, Doe and his donations wouldn’t show up in Trump’s data. But if Doe followed that up by mailing a $1 bill, the campaign would have to disclose the $199 donation, the separate $1 donation, and the fact that they’ve received a cumulative total of $200. Donors also have to disclose their address and their occupation and employer.
For our project, the Forwarded downloaded and analyzed 21 Democratic presidential campaign’s quarterly report of itemized contributions – the 20 remaining candidates who have qualified for a Democratic presidential debate, plus Rep. Seth Moulton (New candidate Tom Steyer has not yet had to declare donation information, and 89-year-old former Senator Mike Gravel does not seem to actually want to be president, despite the best efforts of his teenage campaign managers).
Only being able to look at itemized contributions meant that we couldn’t get a complete fundraising picture for the campaigns – particularly candidates like Bernie Sanders who rely on a large volume of small-dollar donations. But, as mentioned above, even small-dollar donors will show up as itemized contributors if their donations add up to $200 or more.
Note: American citizens (and permanent residents with “green cards”) are allowed to donate up to $2,800 for a campaign to use in the primary, and another $2,800 to be used only in the general election. There is no limit on the number of campaigns someone could donate to – indeed, we saw several examples of donors, like former Dreamworks guru Jeffrey Katzenberg, giving the $2,800 max to multiple candidates.
How did we identify who is a Jewish donor?
After downloading the data of every donation between January and the end of June from every itemized contributor (that is, the donation that took them over the $200 mark and every subsequent gift), we then ran a series of searches designed to determine the likelihood that a donor was Jewish.
So how did we indicate whether we thought someone was Jewish? We did so if they were marked in any one of three fields.
First, the Forward marked anyone who indicated that they worked in Jewish institutions – anyone who, when they were filling out their online donation form, used the following words in the “employer” or “occupation” field: Rabbi, Jewish, Judaism, Hebrew, Temple, Synagogue, Israel, Congregation, Beth, Day School, JCC, Federation. We then manually checked each entry, in order to eliminate people who worked for non-Jewish day schools, or for historically-Jewish hospitals that are now secular.
Then, we cross-referenced the donations with a dataset called the Distinctive Jewish Names list. The DJN list, which was originally created in the 1940s and has since been refined by University of Miami professor Ira Sheskin, among others, is a list of more than 700 surnames and first names that researchers use to track changes in a city’s Jewish population. Think surnames like Cohen, Goldberg and Mizrahi, and first names with Yiddish and Hebrew origins like Yisroel or Aviva. (Specifically, we marked anyone who fit DJN Groupings 1-3 and/or Sheskin’s list of first names and Sephardic surnames)
“What people say is the following: ‘I know a Chinese-American person named Chin who’s Jewish, and I know a Schwartz who’s not Jewish.’ And all that is true. The use of Distinctive Jewish Names does not do a 100% perfect job when you try to go name by name,” Sheskin admitted to the Forward. But in the aggregate, he said, it’s been shown to be an accurate tool, and is usually relied upon when academics or Jewish Federations want to do community censuses. “It’s not perfect, but it’s good enough that everyone is still sampling that way,” he said.
Finally, we did a search for anyone who had ever given an individual donation in a multiple of $18. The number 18 means “life” in Hebrew numerology, and Jews often give donations in that amount to charities they support. More specifically, we tracked if anyone gave a donation of $18, $36, $54, $72, $118 (which isn’t actually a multiple of 18, but is 18+100), $180, $360, $540 and $720. We did not track people who gave $1,800, because we discovered that it was common for donors to reach the $2,800 maximum by writing two checks, one for $1,000 and another for $1,800.
People with names on the DJN list were also more likely than people with non-DJN names to donate in multiples of 18, which would indicate that both are reasonably accurate as Jewish identifiers.
There was one complicating factor: At first glance, it appeared that Sanders had the most donors with Jewish identifiers than any other candidate - the Vermont independent had 1,407 such contributors, with Pete Buttigieg receiving support from 1,394. However, a closer glance showed that nearly half of those Sanders donors were categorized as Jewish because they had given in a multiple of 18 rather than because they had a DJN name – a much higher proportion than any other candidate. And more than half of those donors had given $54, not $18 or $36. Sanders famously proclaimed in 2016 that the average contribution from his army of small-dollar donors was $27 – and 27x2=54.
When eliminating the 18-multiples and only looking at people with DJN names or Jewish occupations, Sanders’s tally of Jewish donors sank to Kamala Harris’s level, leaving Buttigieg the clear winner in terms of number of donors (as well as total donations received). You can see from the following two graphs that none of the other candidates’ stats changed much using this method:
The evidence seems to indicate two possibilities: Our original tracking method (using 18-multiples) overstates support for Sanders because of the $54 phenomenon; or, less likely but still plausible, Sanders’ Jewish donors are much less likely to have Jewish names or work in the Jewish world than every other candidate’s Jewish donors.
This process is not perfect – and we welcome any suggestions on how to improve it. But we think it is a reasonable way to track relative levels of Jewish financial support for political candidates – both because it builds on methods that have long been used by social scientists to study the Jewish community, and because its initial results square with the expectations of seasoned political observers.
Our Jewish Donor Tracking Methodology, Explained