The Iowa Democratic Party released revised Iowa caucus results on the evening of February 18, following a recanvass of 79 precincts. Recanvass administrators changed delegate allocations in 26 precincts where the precinct chair did not properly apply the party’s rules on February 3, and revised results in three precincts after spotting data entry errors.
The adjustments shrank Pete Buttigieg’s delegate lead over Bernie Sanders to “almost nothing,” a Sanders news release declared: 563.207 state delegate equivalents to 563.127, to be precise. The Sanders campaign will request a recount in several precincts where results were not adjusted during the recanvass.
While the work of tabulating the Iowa caucus numbers nears its end, the work of restoring confidence in the process is just beginning. Events of the past few weeks exposed serious flaws in the party’s operations.
After being chosen to succeed Troy Price as state party chair on February 15, State Representative Mark Smith told reporters, “Priority number one is to get out across the state and to talk to everyday Iowans and restore the faith in the Iowa Democratic Party.” A few places he could start:
1. SWEAR OFF INSIDER DEALS
Price promised on February 7 to arrange for an “independent investigation of what happened” with reporting the Iowa caucus results. It was the right idea. Democrats deserve answers about the multifaceted fiasco.
- Why was an inexperienced but politically connected team picked to design the ill-fated app for precinct chairs?
- Why wasn’t the app properly tested before caucus night?
- Why didn’t the IDP arrange for more phone lines on February 3?
- Why did party leaders release results in a few large batches, rather than frequently updating the numbers as precincts were verified?
- Why didn’t the party quickly correct–or at least commit to correcting–errors in assigning delegates at the precinct level?
Unfortunately, the party fumbled the execution of the “independent” probe.
During a special telephonic meeting on February 13 (after Price had announced plans to resign), the State Central Committee approved a plan to hire former Iowa Attorney General Bonnie Campbell and former U.S. Attorney Nick Klinefeldt to conduct the review. They will be paid up to $50,000.
The Des Moines Register’s Nick Coltrain listened in on the conference call.
With the review, the party aimed to “actually move forward, and looking into how we can regain the trust of Iowa Democrats, electeds and our donors,” said Mary Jane Cobb, the operations subcommittee chair. She introduced the proposal to hire Klinefeldt and Campbell.
“Those individuals have the gravitas, and the understanding of impartiality, and they understand the rule of law, as it were, to be able to do an investigation and provide us the kind of information that we need to move forward,” Cobb said.
One has to question Cobb’s “understanding of impartiality.” Campbell is a former state party chair, as was her late husband, Ed Campbell. Klinefeldt was recommended for his federal post by Senator Tom Harkin and appointed by President Barack Obama.
I’ve been a fan of Campbell’s for many years. When I was between jobs in 1994, I spent about two months volunteering full-time for her gubernatorial campaign. Klinefeldt was well-regarded as head of the U.S. Attorney’s office for the Southern District of Iowa. So I don’t dispute their skills or integrity.
Nevertheless, the optics are terrible. Campbell and Klinefeldt will not be viewed as neutral auditors. By rushing to hire insiders, with no request for proposals outlining the scope of review, the party created the impression they were determined to keep this job in the family.
If Campbell and Klinefeldt produce a report exonerating party officials, saying they did the best they could under difficult circumstances, critics will dismiss their work as a whitewash. If they publish hard-hitting findings about the leaders’ actions before and after February 3, critics will claim they must have swept more damning evidence under the rug.
There was no need to be in such a hurry. The IDP could have opened up the process to others who might bid on the review. Perhaps Campbell and Klinefeldt would have submitted the best proposal.
When some SCC members questioned why the party hadn’t used a competitive bidding process or chosen people with fewer connections, Democratic National Committee member Scott Brennan (himself a former state party chair) shut them down. Coltrain reported that Brennan warned the doubters against forming a “circular firing squad.”
“They are two great lawyers, they are great people, and frankly I am embarrassed that we are even talking about whether they are quality people to be doing this,” Brennan said. “It is appalling to me that we are going down this path.”
He added, “We need to move forward as a party. The whole point of a State Central Committee is to elect Democrats. And frankly, I think some of you all are wasting time, and your focus is not on electing Democrats, it’s on, I don’t know what.”
You can be a great person and not be the right fit for a particular job.
Meanwhile, the whole political world has reason to wonder whether Iowa Democrats can be trusted with the responsibility of starting the presidential nominating process. It’s not “wasting time” for the party’s governing body to demand a credible examination of the failures.
Smith can’t reverse what Price set in motion on his way out the door. Asked on February 15 whether he would pursue the independent review, the new IDP chair told reporters that State Central Committee members made the call, and he has neither the intention nor the authority to change their decision.
Smith could and should pledge that going forward, the party will not spend tens of thousands of dollars on a contract for an essential task without competitive bidding.
2. COMMIT TO TRANSPARENCY
I’ve been stonewalled by a lot of government officials, so trust me when I say that IDP staff are pros when it comes to not answering straightforward questions.
For nearly two weeks, many reporters have been asking when–or whether–the party would correct delegate allocation errors in dozens of precincts (see this New York Times interactive map and the crowdsourced spreadsheet maintained by Jordan Hobfoll). I didn’t get any reply to multiple emails, not even a “no comment.”
Trip Gabriel reported for the New York Times on February 8 that the party’s attorney had declared, “The incorrect math on the Caucus Math Worksheets must not be changed to ensure the integrity of the process.”
I found no attorneys who shared that interpretation of Iowa Code section 43. Was the party really saying it was powerless to fix mistakes if a precinct chair awarded the wrong number of delegates to one or more candidates? Why have delegate allocation rules if precinct chairs can do whatever they want and not apply the mathematical formulas correctly? My follow-up questions went unanswered.
Days passed, and the most obvious mistakes remained on the party’s official results page.
Shortly after the IDP announced on February 12 it had approved the Sanders and Buttigieg campaigns’ requests for a limited-scope recanvass, I (and surely many others) started asking what would happen during the recanvass. Would party officials only compare the caucus math worksheet to the official results, or would they also check and count the presidential preference cards from disputed precincts? (Those were a new feature of this year’s caucuses, introduced to allow for recounts.) Would they correct any calculation errors that may become apparent upon review of the caucus math worksheet? Again, radio silence.
IDP Rules Committee chair Melissa Peterson spoke briefly to reporters after the February 15 meeting where the SCC elected Smith as party chair. Naturally, we all wanted to know how the recanvass set to begin the following morning would play out. Here’s the audio, for those who want to fully experience the frustration.
First question from KCCI-TV’s Beau Bowman: “What exactly is a recanvass?” Peterson said the recanvass would begin at 9:00 am the following day. It had been requested by the Buttigieg and Sanders campaigns. The party granted the request for the precincts specified, and so on. In other words, she dodged the question.
Bowman tried again: “So, what is a recanvass, though? Because you guys have talked about [how] you can’t change the data that’s been reported. So what does that mean?”
Well, first of all, I’d let you know that there was a manual related to recanvass and recount, if anybody really wants to get into the weeds. That was spelled out in the Delegate Selection Plan, which has been approved by the State Central Committee as a body.
Great! I love getting into the weeds.
I went home and searched in vain on the IDP’s website for the recanvass and recount manual. When I couldn’t find it, I wrote to the IDP media contacts that evening asking for a copy. No reply.
I looked in the Delegate Selection Plan (enclosed below as Appendix 1). It didn’t provide any details about the recanvass.
A source sent me a copy of the recanvass/recount manual (enclosed below as Appendix 2). After reviewing it, I had more questions, submitted via email on February 16 and re-upped the next two mornings. No reply.
Back to Peterson’s comments to reporters. She said the committee would “look at the evidence before us,” including caucus math worksheets and delegate summary sheets. But “we will not be altering any of the official record or the math worksheets if we think it would do anything to change the intent of one of the caucus participants.”
Would they look at the preference cards to see if the numbers matched the worksheets? “We are in a recanvass.” They would be looking only at the worksheets and summary sheets.
If there are obvious errors on the math worksheet, will those be fixed? “We will not be changing anything that would affect the intent of the caucus participant in the room.” Peterson gave an example of a mistake on calculating the viability threshold. They won’t change those numbers, because you can’t determine how people would have realigned if they had been rightly declared non-viable. She didn’t address other kinds of errors that would be easy to fix.
So is it possible for a change to come out of the recanvass?
It is possible that some of the calculations would change, again, if they do not presume an assumption about something that happened in that room that would have altered the outcome or interfered with someone’s participation.
What we were all trying to get at was, how would the recanvass committee handle it if a given caucus math worksheet clearly didn’t reflect what happened in the room, because of a computational error? I tried to bring up a specific precinct, Des Moines 14 in Polk County, where 50 caucus-goers for Sanders and 31 for Elizabeth Warren got converted into two delegates for Warren and one for Sanders, instead of the other way around. Before I could finish the question, staff cut me off and ended the media availability.
When Smith came out to speak to us a few minutes later, he was asked about problems with the official results. “A recanvassing is occurring right now. Let’s wait until we hear the results of the recanvass.” Would mistakes identified on caucus math worksheets be corrected during that recanvass? “I want to wait and see what we get from the recanvassing before I make any decisions in that regard.”
The IDP revealed on the evening of February 18 that the SCC had approved a resolution on February 13 to allow certain types of corrections during the recanvass. If the caucus chair “misapplied the rules” by making rounding errors or awarding the wrong number of delegates, the recanvass committee would apply the delegate selection rules properly to the final alignment numbers from that precinct.
Good grief, why didn’t they just say so? We were asking that very question repeatedly.
The process they used contradicted the bizarre legal opinion Price had quoted in a February 8 message to SCC members, suggesting that Iowa law did not permit the party to correct any caucus math worksheets. Gary Dickey and other attorneys told me that was illogical; of course the SCC had the power to direct that mathematical errors be fixed to get an accurate count. In fact, that’s exactly what happened.
For what it’s worth, the party’s manual didn’t call for those kind of corrections during the recanvass: “In reviewing Caucus Math Worksheets and Reporting Forms, inconsistencies are declared if results on official precinct or satellite forms are different from the results reported through the Caucus Reporting Tool or by telephone.”
Incidentally, multiple sources on the SCC told me the body never voted to approve the recanvass and recount manual, which is dated January 31. IDP executive director Kevin Geiken told SCC members on February 11 that when they voted in October to approve the Delegate Selection Plan, that action was “sufficient” to create the recount manual. That’s not what the Delegate Selection Plan states (page 12): “The Iowa Democratic Party State Central Committee will approve a Recount and Recanvass Procedure Manual to be publicized with the Call to Caucus no later than November 5, 2019.”
Smith needs to tell every staffer who interacts with the media to stop ignoring questions from reporters. Volunteers who occasionally speak for the party should also be encouraged to be more informative and less evasive. Journalists were not playing “gotcha.” We were trying to nail down the plan for dealing with the many demonstrable miscalculations in official results.
Speaking of which, Smith should put one more item on his to-do list.
3. MAKE THE IOWA CAUCUS RESULTS AS ACCURATE AS POSSIBLE
Well before February 3, the Iowa caucuses were on shaky ground, with some prominent voices highlighting barriers to participation and the relative lack of diversity here, compared to the U.S. as a whole.
Now, on top of defending this system that excludes tons of eligible voters, we’ve raised legitimate concerns about whether reported results are consistent with caucus-goers’ preferences.
Smith said on February 15 that as a member of the DNC, he’ll make “a strong and loud argument” to keep the caucuses first on the nominating calendar. Iowans are “committed to this process, we study the candidates very well” and give them the opportunity to hone their message.
He will be a more effective voice if he can vouch for the validity of all results.
During the recanvass, the IDP corrected three data entry errors and 26 delegate allocation errors (details here). A few more numbers may change if the Sanders recount request is approved. But the party looked only at precincts where presidential campaigns asked for and paid for a recanvass.
There are many more problematic precincts than that. Buttigieg and Sanders weren’t the only candidates affected. Naturally, other campaigns have moved on and have no incentive to spend resources to pick up a few extra county delegates. Regardless, Iowa Democrats need to stand up for accuracy as a core value.
Imagine if Secretary of State Paul Pate said after an election, we know vote counts are wrong in more than a hundred precincts scattered across Iowa, but we won’t change them unless campaigns pony up for a recount. We’d be outraged.
Not all of the mistakes are correctable. Todd Dorman wrote about a “messy” situation in Marion precinct 5 (Linn County). The Warren group was one person short of viability after the first alignment. The precinct chair wrongly said they could not attempt to become viable by picking up supporters from other groups that fell below the threshold. I know of two other precincts (Davenport 44 in Scott County and Des Moines 66 in Polk County) where the same egregious mistake likely cost a candidate a delegate. We can’t turn back time and figure out how people would have realigned in those precincts if the volunteer chairs had been competent.
That doesn’t mean we should give up on identifying fixable errors. The New York Times found 79 precincts where “recorded vote totals do not add up.” One was Waukee 3 in Dallas County, where official results show 204 voter preferences were recorded on first alignment and 254 on the final alignment. That should not be possible, if the chair followed the rules. The IDP’s central committee should authorize some group to consult the preference cards here to figure out whether delegates were properly assigned.
Ethan Corey and Daniel Nichanian have repeatedly pointed out that numbers from Des Moines County precinct 3 were never entered into the system. On the official results page, totals from that precinct are identical to those reported from Winneshiek County’s precinct 3. The Des Moines County co-chair told Corey “there were about 140 fewer voters” in the precinct than the official results show.
Because the discrepancy didn’t affect delegates awarded to Buttigieg or Sanders, neither campaign put Des Moines County 3 on their recanvass list. But the state party must stand for correcting all data entry errors.
Who’s going to pay for it? The SCC just approved $50,000 for an independent review. Spend part of that money on checking results in every precinct. Smith needs to be able to show Iowa Democrats made every effort to report caucus results accurately.
Thousands of volunteers across the state “did their jobs incredibly well” on February 3, Smith told reporters last weekend. Many had positive experiences with “this wonderful, unique method of democracy that we have here,” he added. My contacts around the state largely agreed that their precinct caucuses ran smoothly. Bleeding Heartland authors Amber Gustafson, Ira Lacher, Mike Jacobsen, Laura Hubka, Andy Johnson, and Shawn Harmsen likewise described harmonious and uplifting caucuses in their areas, later overshadowed by the reporting delay and other controversies.
Party insiders are understandably focused on “moving forward” now. But Iowa Democrats need more than happy talk from Smith as he works to rebuild trust.
UPDATE: The IDP announced on February 19 that the Sanders campaign requested a recount of ten precincts, and the Buttigieg campaign requested a recount of 54 precincts. One was on both lists, so the request affects 63 unique precincts. A statement said the party “will review the requests to determine whether each meets the required standard of evidence suggesting that caucus errors would change the allocation of one or more National Delegates” and announce its decision within 48 hours.
Scanning the official results after the recanvass, Corey found that five counties are still awarding more county county convention delegates than they should be, indicating that one or more precincts assigned too many delegates on February 3. Twitter user Prodigy listed the problem precincts here and here. The IDP should fix those mistakes, even though the precincts in question weren’t part of the recanvass request.
Appendix 1: Iowa Democratic Party’s final Delegate Selection Plan for the 2020 cycle
Appendix 2: Iowa Democratic Party’s 2020 Caucus Recanvass and Recount Manual, version dated January 31, 2020
Top image: Mark Smith with Paula Martinez at the February 15 meeting of the Iowa Democratic Party’s State Central Committee. Martinez nominated Smith to serve the remainder of Troy Price’s term as state party chair.