STORY UPDATED: check for updates below.
Do eight Iowa counties have more registered voters than there are people eligible to vote in those counties? No, that's not true. Looking at the latest available data in February 2020 there are only five counties where the number of total registered voters exceeds the voting age population. But the number dropped to only one county when just counting the number of "active registered voters," rather than total registered voters. And that single county has a large student population that doesn't always vote where they live.
The story originated from a press release published by the conservative activist group Judicial Watch on February 3, 2020, titled "Judicial Watch: Eight Iowa Counties Have Total Registration Rates Larger than Eligible Voter Population - at Least 18,658 Extra Names on Iowa Voting Rolls." The report then got picked up by other conservative groups:
Judicial Watch sounded the alarm on voter fraud on Monday ahead the Iowa caucuses. Conservative watchdog group Judicial Watch on Monday announced that eight Iowa counties have more voter registrations than eligible voters. There are at least 18,658 extra names on the voter rolls in Iowas, Judicial Watch reported. The watchdog group also reported that ...
Here is the Gateway Pundit picking up the story, for example:
Eight Iowa Counties Have Total Registration Rates Larger Than Eligible Voter Population - More Than 18,600 Extra Names on Voter Rolls
Judicial Watch sounded the alarm on voter fraud on Monday ahead the Iowa caucuses. Conservative watchdog group Judicial Watch on Monday announced that eight Iowa counties have more voter registrations than eligible voters. There are at least 18,658 extra names on the voter rolls in Iowas, Judicial Watch reported. The watchdog group also reported that ...
The press release coincided with the start of the Iowa caucuses, the first real voting test for primary candidates. The Gateway Pundit shared the story with its audience but later revised it after Paul Pate, the state's Republican secretary of state, sought to put an end to the story, tweeting about the misinformation and using the hashtag #FakeNews.
False claim. Here is a link to the actual county-by-county voter registration totals. They are updated monthly and available online for everyone to see. #FakeNews https://t.co/bM1bj51eKA
-- Iowa Secretary of State Paul Pate (@IowaSOS) February 3, 2020
A deeper dive into the eight counties that Judicial Watch examined shows its numbers are off. Judicial Watch explained to Lead Stories it used American Community Survey (5 yr Estimate: 2013-2017) census data together with U.S. Election Assistance Commission (EAC) data released in 2019.
However, we looked at more recently released census data about the 2014-2018 period found here to get the number of eligible voters in each county and compared it to the most recent Iowa Secretary of State's office numbers about voter registration.
Five of the eight counties indeed had a voting age population higher than the total number of registered voters (which includes inactive voters), totaling a 9,138 extra names - not 18,658. But taking a closer look, a more accurate number to use would be "active registered voters."
The Secretary of State's office sends out notices to registered voters periodically and considers voters inactive "who have not returned the card or otherwise responded to the notice." Under the federal election law, 52 USC 20507 (d), these inactive voters can still vote if they update their address information on the day of the election (the exact procedure differs depending on where they live at that moment and where they are trying to vote). Only if they fail to vote in two federal elections after getting the notice are they struck from the rolls.
Having inactive voters on the rolls does not automatically mean there is anything fraudulent going on; it could just mean people moved away and didn't bother to update their registration. Indeed, the precise goal of having a system in place to track inactive voters is to be able to eventually remove them from the voter rolls.
Only one of the counties, Dallas County, had a higher number of "active registered voters" than the voting age population -- 59,667 compared to the voting age population of 57,045. It should be noted that several area colleges are located in Dallas County with students who don't always live where they vote - so the census count would not include them. A similar effect might be at play in Johnson County, which is home to the University of Iowa.
This table provides a look at the numbers for the counties that came under scrutiny by Judicial Watch compared to the numbers for February 2020 provided by Iowa's Secretary of State office.
Citizen Voting Age Population | Total Registered Voters | Active Registered Voters | |
Dallas | 57 045 | 63 409 | 59 667 |
Dickinson | 13 655 | 13 826 | 12 959 |
Johnson Co. | 107 000 | 109 211 | 96 958 |
Lyon | 8 315 | 8 490 | 8 117 |
Madison | 11 705 | 11 630 | 10 858 |
Poweshiek | 14 585 | 14 208 | 12 872 |
Scott | 127 520 | 127 737 | 115 393 |
Warren | 36 785 | 36 358 | 33 644 |
Sources: US Census Citizen Voting Age Population Survey 2014-2018 & Iowa Secretary of State Office, February 2020 numbers. In bold: counts higher than the voting age population.
With the most recent data, we only find five counties with more total registered voters in 2020 than citizen voting age population in 2018 (and only one with more active registered voters). So, the headline from Judicial Watch is not correct in claiming this applies to the present.
In their press release Judicial Watch stated their data sources were the "U.S. Election Assistance Commission (EAC) in 2019 and the most recent U.S. Census Bureau's five-year American Community Survey." The EAC did release the Election Administration and Voting Survey (EAVS) in 2019, but it covered voter registration info from the 2018 election. And at the time when the Judicial Watch report was being created, the latest five-year survey from the U.S. census covered 2013-2017. Looking at that data gives following table:
Citizen Voting Age Population | Total Registered Voters | Active Registered Voters | |
Dallas | 54 705 | 62801 | 58496 |
Dickinson | 13 650 | 13770 | 13012 |
Johnson Co. | 105 890 | 114297 | 97132 |
Lyon | 8 320 | 8530 | 8058 |
Madison | 11 540 | 11826 | 11113 |
Poweshiek | 14 385 | 14680 | 13166 |
Scott | 127 245 | 128308 | 116164 |
Warren | 36 210 | 36394 | 34332 |
Sources: US Census Citizen Voting Age Population Survey 2013-2017 & Election Administration and Voting Survey (about 2018 election). In bold: counts higher than the voting age population.
Using those numbers, there are indeed eight counties that had more registered voters in 2018 than they had citizen voting age population in 2017.
In a statement to Lead Stories, Judicial Watch's Tom Fitton said: "Again, the new Census data and the interim voter registration information does not obviate our study, but is consistent with it. The Secretary of State is playing games and provides no specific numbers, only pointing to a data set and ignoring that the new data set confirms our concerns. We've seen it before on this issue."
Lead Stories does not agree with the claim in the Judicial Watch headline after looking at the specific numbers found in the most recent data set. The statement in the Judicial Watch headline does not apply to the present situation, and the lack of mention of active versus inactive voters in the press release might mislead readers into thinking there is more going on than what is actually the case.
Updates:
-
2020-02-06T14:18:17Z 2020-02-06T14:18:17Z Updated to include table with 2013-2017 census data and 2018 voter registration data. -
2020-02-05T19:04:48Z 2020-02-05T19:04:48Z Updated to include more information about active vs. inactive voters and correcting erroneous statement that inactive voters aren't allowed to participate in elections. -
2020-02-04T18:15:52Z 2020-02-04T18:15:52Z This story has been updated to reflect the further examination after Judicial Watch reached out to Lead Stories and pointed us to the Census data it used in its assessment.