IN DEFENSE OF CALIFORNIA’S VOTING
Little raises my hackles more quickly than politicians misstating reality for partisan expediency. And so it is with respect to the California voting system.
I am not talking about the “jungle” primaries. This is an interesting system in which all candidates run regardless of their party affiliation, and the top two vote getters run off in a two-person election to determine the winner. This can result in candidates from the same party running against each other.
Other states use ranked-choice voting, but that is for another time.
Nor am I writing about Voter ID or proof-of-citizenship for registration. I have written about these in the past, but they are not the focus of this essay.
The most “controversial” piece of the California election is the mail-in ballot voting. It is this aspect that I would like to speak in defense of.
Whether the state sends its voters a mail-in ballot, or whether the state requires the voter to request one in advance, the general process is the same. The ballot is secured at the state level with a bar code printed on the ballot. This bar code connects the ballot to a specific person at a specific address.
The Voter then marks his/her vote, seals it in an unmarked envelope, places that sealed envelope inside a return mailer provided by the state, signs and dates that return mailer, and mails it back to the appropriate location, which may be the central state collection location, or the appropriate local election board.
The outside envelope has the barcode printed on it, so when received, the ballot can be authenticated by the state, and when done so, no other ballot can be authenticated for the same voter, mail-in or in-person. It also has the voter’s signature, so verification is identical to when a voter appears at a polling place to cast their vote. The outside envelope is opened, it is reserved to allow for subsequent voter substantiation, and the sealed ballot is then set aside to be counted. This ensures that the ballot remains anonymous, as required by federal, constitutional, and state law.
There is some discussion that requesting and receiving mail-in ballots provides the opportunity for “ballot harvesting” in which a voter is influenced to cast their ballot for a preferred candidate. This may indeed occur. There IS the potential for inappropriate behavior as the voter could be paid for their vote and be observed marking their ballot before sealing it in the inner envelope. There is less potential for this influence when the voter votes in a private polling booth.
Another issue is that mail-in ballots may be requested by surviving family members of deceased relatives. If the state’s voter registration rolls are not appropriately updated, an unscrupulous individual might request a mail-in ballot for a “voter” who no longer lives, and, therefore, could never show up in person to vote at his/her polling place.
These are legitimate concerns. However, despite extensive research by interested parties on both sides of the aisle, the instances of either of these issues occurring is extremely rare. Rare fraud is a problem that needs context to understand.
It is particularly illuminating to ask the question, “Do states with universal mail-in voting have higher or lower rates of fraud in their elections?”
The answer is that there is no evidence of increased fraud in these states. As with every state, the incidence of fraud is almost non-existent.
The Brookings Institution found a rate of 4 cases of mail-in ballot fraud for every 10 million votes cast across the entire country. Nor have studies found that when states shifted from in-person voting to universal mail-in voting that the rate of fraud increased at all.
For example, in Oregon, with universal mail-in voting, in the 2000 election there were documented around a dozen cases of fraud within over 100 million ballots cast, a rate lower than the Brookings Foundation found for the country in general.
Data modeling in the state of Washington which switched to universal mail-in voting in 2011, showed that the rate of fraud DECREASED after the switch.
In Colorado, which switched to universal mail-in voting in 2013, the rate of fraud was less than 2 cases per 10 million votes cast, again less than the rate in the entire nation.
One can only conclude that mail-in voting is actually more fraud-resistant than in-person voting.
Let us assume that there are some instances of fraud in mail-in voting. What would happen if we were to eliminate mail-in ballots because of the fear of those errors? How many hundreds of thousands or millions of voters would lose their ability to vote? People who are disabled, home-bound, on business or personal travel, or in active military service would lose their ability to vote. This would result in the disenfranchisement of potentially millions of people in order to prevent the potential of hundreds of fraudulent voters. For me, this is an inequitable trade-off.
Now let’s turn to the issue of counting those ballots.
My summer home is in Pennsylvania. In that state, there is one and only one voting day. There is no “early” voting. You vote on the Tuesday that the polls are open, or you don’t vote. The only exception is a mail-in ballot. Perhaps your employment requires you to be at work on voting day; you have no choice because you can’t get to your poll.
Other states are different, for, as the U.S. Constitution states:
Article I, Section 4, Clause 1:
The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof; but the Congress may at any time by Law make or alter such Regulations, except as to the Places of chusing Senators.
What about California?
We are being pounded with the political posturing that elections need to be decided as quickly as possible on election night.
But, as a society we have opted to delay decisions. How? The fastest way to count ballots would be to enter votes electronically into a computer terminal. As soon as polls close, the results would be instantaneously available. Even machine corroboration, done by running a series of control ballots through each terminal to confirm that the votes recorded are exactly the votes that were cast would only delay those results by a matter of minutes. These machines are not connected to the internet and have no internal wireless communications, so they are immune from outside influence. Nevertheless, we have opted as a society to mothball all of our electronic voting machines out of fear that they might be hacked by outside sources intent on influencing elections.
We have chosen to move back to paper ballots. In some cases, we still process those paper ballots through an electronic reader, but in others we actually take the time for humans to count the ballots. Counting paper ballots by machine and then ensuring that the machines are properly calibrated takes time; counting by hand takes even more. This process itself is not error-free. People make mistakes. There is no process that is 100% accurate.
Electronic voting machines are probably the most accurate, although as I said before, we, as a society have chosen to abandon that process because of fears of software malfunctions. Electronic paper ballot counting machines are most likely the next most accurate model, but not all states or municipalities use them.
Now, to mail-in ballots.
I titled this essay, “In Defense of California’s Voting”. I want to take some time to look at how we vote, and how we should want to count those votes.
I believe that every eligible citizen in this country should have the right to vote. (And as I wrote above, it is not the focus of this essay to debate Voter ID or voter registration requirements; I have written about both before.) Not only should citizens execute that right, but they should have as much information available to them as possible before they make their decision.
However, information often comes out in the days just preceding an election. That information should be available to every voter and should be considered prior to casting their vote. I support procedures that allow for voters to vote at the end of the campaign period.
Here are some examples:
Two weeks before the 1880 election a New York newspaper published a forged letter from James Garfield expressing support for employers using cheap Chinese labor to underbid American workers. Those falsehoods were corrected before votes were cast, but if you had voted after seeing the forged letter, but before the correction, you were misinformed.
Five days before the 2000 election, Fox News revealed that George W. Bush had been arrested for DUI in Maine in 1976.
Four days before the 1992 election, Caspar Weinberger was indicted by a federal Grand Jury regarding the Iran-Contra affair, contradicting the timeline provided by George H.W. Bush, damaging his credibility.
A week before the 2012 election, Mother Jones released a recorded video of Mitt Romney claiming that 47% of Americans view themselves as victims dependent on the government.
11 Days before the 2016 election, Director James Comey announced that the FBI was reopening its investigation into Hilary Clinton’s private email server.
In all of these cases, those who voted on election day had different information than those who voted 2 weeks prior to election day.
What about other deadlines?
There are several examples of how we as a country deal with date deadlines. Here is just a sampling:
In corporate finance and M&A, the “Drop-Dead Date” is the defined termination date for the completion of a merger agreement. If the deal does not close by this date, either party has the right to walk away from the deal. Final papers must be mailed by this date.
Tax returns and payments are due on specific dates. There is something that is called the “Timeliness Rule” which uses the postmark date on the material to determine if the return or payment was “timely filed”.
Mutual Fund and Shareholder Mailings have specific SEC regulations that require mailing dates before EDGAR (Electronic Data Gathering Analysis and Retrieval) filing is permitted. The EDGAR filing is due exactly 10 days after this mailing date.
In each of these cases, the Drop-Dead Date is the date that a document must be mailed, not the date that it must be received.
In my opinion, this should be the same rule for Mail-in Voting. For me, there is no difference between someone deciding on the “Tuesday next after the first Monday of the month” to vote in person or to mail in a ballot. In both cases the voters have received the full available information to make an educated decision.
The timing of the vote count is thus affected by the time necessary to receive and process the votes. For in-person voting, this time frame is dependent on the time the poll closes, the nature of the voting machines, the calibration and verification of the machine’s count, the efficiency of the polling station vote counters, and the amassing of final results. This can result in delays of hours or, in some cases days. So, already there are roadblocks to instantaneous results.
The differences in the ability to provide final counts from in-person voting can vary by precinct. Are we so partisan as to reject counts from some precincts that count more slowly than others, particularly if they lean towards the party we oppose?
For those who voted at the exact same time but did so by mail rather than in person at a polling place, I believe their votes are of equal value. The only conclusion is that, if the votes were postmarked on the day of the election, prior to the time that the polls closed, that they have the same right to be counted as those cast in person.
The issue is the timeframe. In California, all ballots postmarked prior to the end of the election day are eligible for counting. This may take weeks, as those ballots are received.
Well, what if we require that only mail-in ballots received prior to that voting date be counted? How many days prior to an election would you need to mail your ballot to ensure that it is received and counted on election night?
Let’s assume that there are 2 parties, “A” and “B”. If Party “B” wants to influence the vote, they can pressure the local postmasters to delay ballots (easily identified by the return envelope provided by the state) that come from zip codes that are historically “A”-leaning, or to speed up delivery of ballots from zip codes that are historically “B”-leaning.
The issue here is that a vote should be vote should be a vote.
If we are willing to wait hours or days for fair and faithful counting of in-person votes, should we not also be willing to wait days or weeks for mail-in votes?
In most cases, the winner can be determined before all votes are counted because the total of all outstanding ballots is less than the number necessary to reverse the results. But, in extremely tight elections, every vote matters, and every vote should be counted.