Are You Smurf Hunting?

Be advised of these FEC data considerations on this Smurfy endeavour.

There has been quite a bit of buzz and excitement this past week with James O’Keefe’s OMG media group knocking on doors and asking people, aka, “Smurfs”, if they are aware that thousands of campaign contributions have been made in their names, and for upward of hundreds of thousands of dollars total in the past few years. Most of these Smurf cases involve donations through ActBlue, but WinRed is also used in the same manner.

Issues with ActBlue questionable practices are not new, this article from 2020 discusses use of BLM donations raised through ActBlue as possible dark money to contribute to Democratic campaigns.

What is a Smurf?

In the OMG media scenarios, a Smurf is someone who launders money by breaking large transactions into many small ones to evade campaign finance regulations. These people whose names are being used aren’t really Smurfs, more like they’ve been Smurfed-they are unaware of the transactions.

Any fan of the original 1981 Peyo smurf cartoons (everything created after that is junk) knows that Smurfs use the word smurf freely as both verb and adjective; e.g., “Have a smurfy day” or “No smurfing!” instead of saying “Have a great day” and “No kidding!” So these mostly elderly people have been smurfed, which is not very smurfy.

Curious to see if Hawaii has our own island smurfs, I started downloading data from the FEC’s campaign contribution database, searching by zipcode, then selecting the names of individual contributors that had 300+ transactions in an election cycle, and then consolidating them using all name permutations that matched their address or addresses. There are a lot of name permutations, which is an indicator of something not-so-Smurfy. I made a list of the top 25, and then expanded the searches back to 2017. Most ActBlue Smurfs have the employment listing of not employed, and the WinRed Smurfs are listed as retired.

What I found shocked me.

One of Hawaii’s top Smurfs, we shall call him ‘Nalo Smurf, is an elderly “not employed” gentleman who lives in a modest residence, yet the data from FEC shows that he made 13,337 contributions between 2017-2023 for a total of $269,832.57.

I needed to verify that I was analyzing this data correctly.

I started checking my smurf data against the ElectionWatch state smurf data, and could not replicate the results for ANY of my Smurfs, including ‘Nalo. My FEC individual data also did not reconcile with the bulk data downloads used by some of the other Smurf hunters in my Smurf hunting online village.

Data must be reproducible to have integrity.

I spoke with some of my fellow Smurf hunters and found they were also having the same issue verifying and reconciling the data. We started smurfing deeper into the data, and more questions than answers were surfacing.

Or should I say, Smurfacing…

This is what we’ve found so far:

• A single contribution from a donor can be earmarked in both ActBlue and WinRed for several different candidates or campaigns, thus splitting one contribution into several. I tested this with a $1 Act Blue Community Form donation plus a .50 tip to ActBlue and was able to create 8 donations from that single dollar. I have also tested this on WinRed using the Team Page feature.

• Tips to ActBlue need to be excluded, as these are not campaign contributions.

• Bulk data downloads vs. data downloads by zip or name produce different results.

• Refunds to donors need to be subtracted from the total sum of all donations (excel handles the -$ subtraction, but you need to also remove these from your total # of donations calcuation).

• Rows of data with an “X” in the memo_code column need to be excluded from the totals (read FEC explanations here and here).

• Rows of data with the receipt_type (column AD) 15J and 24T need to be excluded from the totals.

• Rows of data with the same sub_id are duplicates, and need to be removed from the totals, leaving ony unique row IDs.

Here are examples of some of the data considerations using ‘Nalo Smurf’s data as well as my own donations to WinRed and ActBlue (I about had a smurfattack donating to ActBlue, but it had to be done for the sake of research).

Earmarked donations:

This screenshot from my recent ActBlue $1 donation shows that I made a group contribution to 4 campaigns using a single transaction:

The NY House Majority Makers donation is further split into 3 campaigns, making a total of 4 donations to NY Majority Makers. This wasn’t earmarked by me, NY House Majority Makers does this on their end.

We are up to 7 rows of data in the FEC download from my single $1 donation.

Tips to ActBlue:

After leaving a tip I am up to 8 donations from a single dollar bill. FEC will list this on 8 rows, one for the $1, a row each for the 7 splits with the corresponding $ amount, and a row for the tip.

Instead of showing only one row for $1, the FEC data will download as another 6 rows of earmarked donations adding to $1, plus a row for the tip, which to the undiscerning eye adds up to $2.50 and 8 donations, when the true number is 1 donation for $1.

Rows of data with the same sub id need to be excluded:

Here are 2 donations I made to WinRed in 2021.

Row 2 is a $20 donation and row 4 is a $10 donation. Rows 3 and 4 are duplicate sub ids and need to be excluded from analysis.

If duplicate sub id rows are not removed, this adds up to 4 donations for a total of $60, which is way more than the Hawaii Republican Party deserves given their horrendous treatment of Patriots over the past few years (Precinct Strategy is working to change that but it’s not been very smurftastic, it’s been a mothersmurfer actually).

Now we look to ‘Nalo Smurf for the memo code X, receipt id 15J and 24T examples:

Holy Smurf Balls!

‘Nalo Smurf went from 13,337 donations totaling $269,832.57 down to 3852 donations totaling $80,426.23. Excluding the receipt codes and the sub id duplicates seems to serve the same purpose, but I like to look at both as a double check.

Is he still a Smurf?

Absolutely.

But now if you go to this 81 year old gentleman’s front door and let him know he made $80,426 in donations as opposed to $269,832 it may downgrade his reponse from a heart attack to merely a stroke, neither of which is good.

And if one were to take this to law enforcement, you surely would be better off using the accurate data.

I’m not recommending we do either until even more data deep digging has been done.

I know some of my Hawaii friends following O’Keefe’s work are anxious to go Smurf hunting.

This Smurfing has been going on for years, we can spare a couple or few more weeks of analysis and verification to make sure the data is accurate.

Will analyzing the data correctly downgrade Smurfs with less donations than the mega smurfs into normal donors?

Also absol-smurfo-lutely.

This is an ongoing Smurf Village collaboration with some super smart smurfs, and our knowledge and understanding is evolving as we learn more.

By the time this article is published something new may have been added to the data consideration list.

Who knows what tomorrow will smurf? Until then,