By: Robert Bryson
This post will be the third in a five-part series examining “voter fraud,” its impact on elections, steps taken to curb it, and it will end with an analysis of Husted v. A. Philip Randolph Institute which is the Supreme Court case scheduled for arguments January 10, 2018 that will determine if Ohio can use voter inactivity to send confirmation notices to that voter under the National Voter Registration Act of 1993 and the Help America Vote Act of 2002 (e.g. if Ohio can purge its voter lists).[1]
Part Three was born out of a thought experiment I started writing on another post until I realized it was taking up way too much space and decided to break out on its own.
A Presidential Example:
How much fraud do you think it take to influence a Presidential election? A senatorial election? A gubernatorial election? The President needs to win a majority of the electoral college.[2] Electoral votes are calculated by combining each States’ (1) representatives (2) senators and the three electors from the District of Columbia[3] for a total of 538. To win, the candidate must secure 270 electoral votes.
Which states?
Put yourself in the shoes of someone trying to manipulate the presidential election. You will need fraudulent votes spread across several states, ideally in “swing” states where you can more easily hide a few thousand that might tip the balance. The swing states of 2016 were Pennsylvania, Virginia, Florida, Ohio, Indiana, Iowa, Arizona, Michigan, Minnesota, North Carolina, and Wisconsin which represent a combined 159 electoral votes.[4] Now, you obviously don’t need agents in every state; so, you spread them across ones you think will be highly competitive and can be swung with a few thousand well-placed voters.
One problem you may notice, we know that Michigan, Minnesota, and Wisconsin were competitive only after the fact. Every expert expected Clinton to win those states, Trump’s win was an upset. Without the benefit of hindsight, which states do you focus on? And, as you have more uncertainty, you have to increase the number of fraudulent votes (the last thing you would want is to pull-off a voter fraud scheme and still lose).
Which counties?
Finally, once you settle on the states to influence, you now need to break your operation down by county. Which county will your voters register in? How many per county? You will need a projected vote breakdown per county. The entire operation would be exposed if you accidentally registered 20,000 fake Republican voters in Broward County[5] , only to realize that it wasn’t nearly enough to swing the county. It also wouldn’t work if you registered 100,000 fake Democratic voters in Bay County, Florida and wildly swung it from heavily favored for the Republican candidate to a blow-out for the Democrat. Do you see? You need to know each county and predict how it will vote for each candidate.
The scale of the crime
Now, we are talking about a scheme that involves thousands of people, spread across 4 or 5 states, engaging in illegal behavior. You’re trying to guess which states are going to have “close” votes, which counties will be close, and you are trying to guess by how much your chosen candidate should win to avoid suspicion. All the while trusting that no one in your massive organization is going to either (1) confess or (2) foolishly tell someone[6]. Big organizations by their nature are “leaky.”
Nothing and no one is immune to this fact of the human experience. The White House regularly suffers from leakers, to the point President Trump directed Attorney General Sessions to investigate leaks.[7] The Paradise[8] and Panama[9] Papers illustrate that secretive international law firms cannot keep information hidden forever. Even the NSA cannot manage to keep its highly dangerous, highly classified hacking tools secret forever.[10], [11]
Concluding thoughts
The bigger the office, the more difficult to defraud because (1) increased awareness (2) increased voter turnout (3) more scrutiny and (4) increased presence of professional vote counters, auditors, and support staff. Accordingly, a coordinated voter fraud scheme is extremely difficult to pull-off in the United States.[12]
[1] http://www.scotusblog.com/case-files/cases/husted-v-philip-randolph-institute/
[2] https://www.archives.gov/federal-register/electoral-college/about.html
[3] As provided by the 23rd Amendment.
[4] https://en.wikipedia.org/wiki/United_States_presidential_election,_2016
[5] http://mcimaps.com/wp-content/uploads/2015/10/1992-Margin.png
[6] https://gizmodo.com/5520438/how-apple-lost-the-next-iphone
[7] https://www.nytimes.com/2017/08/04/us/politics/jeff-sessions-trump-leaks-attorney-general.html
[8] http://www.bbc.com/news/world-41880153
[10] https://www.nytimes.com/2017/11/12/us/nsa-shadow-brokers.html
[11] A leak which dwarfs in scale and damage to the NSA, the United States (including its citizens) and the world, than the Edward Snowden leaks.
[12] Yes, this post is leaving one glaring example of election meddling, namely “hacking” the election computers. However, again, there are big complications in hacking electoral systems. First, the “hacked” votes cannot be added to the totals, they must replace existing voters (or else the total number of votes won’t match the number of registered voters in the county). Second, each state runs its own electoral system, it is inefficient but also very difficult to penetrate. To change a national election, hackers would have to break into multiple states’ election systems. Third, electoral systems are, for the most part, closed from the Internet – requiring direct access to the computers. Fourth, if a discrepancy is identified, election officials can also conduct manual count of the ballots.
This is not meant to minimize the interference of Russian agents on the 2016 election, rather, it is to refocus our attention to the real culprit – fake news sites, fake social media accounts, and leveraging U.S. partisanship to divide us.
Public Interest Advocacy Collaborative 
