Tuesday, September 15, 2020

The Absurdity of Widespread Mail-In Voter Fraud

 Mail-in voting is now under great scrutiny from politicians and the media, particularly the fear that some people will vote multiple times. One politician went so far as to explicitly tell his supporters to vote twice. I worked in the County Elections Office in Santa Barbara, California in the 2002 election, and know that such activity would be extremely unlikely to succeed, especially at a scale large enough to sway any but the closest and most local election.


In the first section of this piece I will lay out the exact procedure for three kinds of voting -- two are in-person, and one is absentee. In the final section I will demonstrate how extremely difficult it would be to a) successfully vote more than once without incurring prison time or fines, and b) significantly impact an election.


The Mechanics of Voting 

Early and In-Person Voting - These two voting procedures are virtually identical. A voter comes into their polling place (or elections office) and identifies themselves in the required way. Some states require that they show ID, while in states like California they need only give their name, address, and identifying information. Once the voter is identified, and confirmed not to be an absentee voter (this part is critical for reasons we’ll see later), they sign the voter rolls, and are given a ballot which they vote on the spot. 


If the voter is determined to be absentee, or they show up in-person on Election Day and the records show they voted early, they’re given a provisional ballot. This is placed in a sealed envelope and signed (in some cases multiple times) in the presence of the election worker, and dropped into the ballot box to be processed later (see below).


Mail-In Voting (sometimes called absentee) - This is where a voter receives their ballot in the mail, votes, and then sends it back, usually with one/more signatures and other identifying information required written on the envelope. It’s important to note that voter files record if a person is registered absentee, and as stated above, if they are registered absentee, they’re not allowed to vote a normal ballot if they show up in person.


I’d also note here that absentee ballots also have an additional security measure not applied to normal ballots: each one is checked individually by an elections worker for matching signatures and any other required identifying information. If the ballot doesn’t have a matching signature, it’s set aside to be reviewed by a second worker, or in some cases a pair of workers, to confirm that it should be thrown out. Between ~1-5% of ballots fail this check and are deemed invallid and discarded.


Provisional Ballots - As mentioned above, if something goes wrong, and a voter shows up in-person on Election Day when they were registered absentee, they’re given a provisional ballot, which is placed in a bright yellow envelope marked “Provisional.”


In all cases, these provisional ballots are set aside for counting until the very end of the process, after all other ballots (in-person, absentee ballots, etc) have been counted. The process for checking these ballots is extremely similar to mail-in ballots, with additional checks on signatures, but with one additional step: The voter’s record for that election is checked, and if it shows that any other ballot (of any type) was received or counted, the provisional envelope is discarded as invallid.


As the absentee (and eventually provisional) ballots are counted, members from the major parties, or any member of the public, can walk into an elections office at any time and demand to observe workers while they’re processing ballots. This means exactly what it sounds like. Someone standing over your shoulder, staring at your computer screen, scrutinizing every decision you make. 


At any time, these observers can challenge a decision, whether it be accepting a ballot they think is suspect, or rejecting a ballot they deem to be valid. These challenged ballots are set aside to be examined by multiple workers, often with one or more observers. While this sounds like an avenue ripe for corrupt intent, in my observation challenges from either party were fairly rare (well under 5%) for the main reason that anyone who’s worked at a bank will attest: signatures are pretty darn unique. In 96-98%+ of cases when you looked at an envelope, that signature looked exactly the same as it did on the image capture from the voter registration database.


Why Voter Fraud is Rare

Given the mechanics of voting as described above, it’s clear that it is exceptionally difficult to successfully vote more than once. Moreover, a further deterrent is the fact that voting more than once in any election is a federal felony, punishable by 5 years in prison and a $10,000 fine. In 28 states, it’s also a felony according to state law, with similar lengths of imprisonment and fines; (in Georgia you can get 10 years and a $100,000 fine). 


Now that we’ve made clear this is a serious crime with extremely strict penalties, usually at both state and federal levels, let’s talk logistics: As I laid out above, most states have systems in place to prevent double voting (any second attempt will get you a provisional ballot that won’t be counted if you’ve already voted.) Some states, like Pennsylvania, have automated systems in place that literally won’t allow a second vote from any voter to be counted. Thus, if a voter tries to vote early, then absentee, then in person, the mechanics of the system will ensure that (at most) only one of those votes will be counted.


So, one might ask, how can we steal an election? The answer will look different, depending on whether we’re trying to do so in-person or through mail-in voting.


In-person voter fraud

For in-person fraud, we would begin by hacking into the elections database to find voters' names, addresses, and signatures. Then, we would go to each different polling place (because you could never visit the same polling place twice, unless you wanted to be recognized by the poll workers there), and pretend to be each voter, reciting the memorized names/addresses. Here we are hoping that said voter has not already shown up to vote, since that would immediately identify you as a felon (likely on both a state and federal level). We would need to bring fake ID for each voter (in states where ID is required), or be able to fake each person’s signature on the rolls (in states where it isn’t). Once done, we would drive to the next polling location and repeat the process. I’d also note that each one of these stops is likely taking us 15-20 minutes, longer if there’s a line, and travel time between locations adds time.


Each time we do this we’d be subject to being caught, as polling locations close to each other sometimes share workers. Also remember, that the later in the day we go, the more likely we are to be impersonating someone who has already voted, thus increasing our chance of being identified as a fraudulent voter. Each stop also adds a possible state and federal felony charge if caught, meaning that if we did this for 12 full hours, possibly impersonating ~20 voters, we could easily be looking at potential life imprisonment.


If a potential life sentence in federal prison seems absurd for ~19 additional votes, you’d be right, and that number of additional votes would never sway an election of any size. Thus, to make this work, you’d have to recruit a small army of friends, giving them their own list of voters, their own fake ids or forged signatures, and their own willingness to commit several dozen felonies all in one day. I’m not sure about you, but I don’t have any friends that I’d be comfortable “bringing in” to such a scheme, for fear that each person brought in would increase the likelihood the entire operation would be revealed to authorities.


Mail-in voter fraud

I laid out the absurd scenario above in order to illustrate the preposterousness of the idea of widespread in-person voter fraud. Mail-in voter fraud, as I’m about to demonstrate, would be nearly impossible to execute.


First, an additional piece of context to our discussion is necessary: Mail fraud is a federal felony punishable by up to 20 years in prison and up to a $250,000 fine. Mail theft is also a federal felony, punishable by up to 5 years in federal prison and up to a $250,000 fine.


To begin this process, we would need to hack into the voter database to obtain names, addresses, and signatures of said voters. We’d then need to monitor the mail for each voter, waiting for the day(s) when they get ballots. We’d then have to collect each ballot without being caught (mail theft), bring it home, fill it out, forging signatures and required identifying information, and then send it back (mail fraud). We’d also have to hope that no voters call the elections office to report that their ballot never arrived, since they would potentially come into the office in person and vote early, preventing any mail-in ballots later received from being counted. They could also request another ballot be sent, which would mean that two ballots would be received in the elections office and only one would be counted (giving us only a 50% chance of swaying that one additional vote). 


A quick assessment of risk: Using Georgia as an example, if caught at any step, we would be looking at a potential 25 years in federal prison, and a $500,000 fine, as well as state penalties of 10 years in state prison and a $100,000 fine. I’m not sure how the average criminal would feel about it, but 35 years in prison and $600,000 in fines seems like a lot to risk for one additional vote that may only have a 50% chance of being counted in some cases.


The risk/reward calculation above is why almost all cases of voter fraud (which are exceedingly rare in nature) don’t involve mail-in ballots. The penalties and prison time are far higher for mail-in voter fraud than in-person voter fraud.


Conclusion

Voter fraud is rare: It entails high fines, long prison sentences (at both the state and federal level), and logistical hurdles that make it exceptionally difficult to pull off at any scale more than a small handful of votes. 


Even assuming someone was willing to put in all that time and energy (committing crimes), wouldn’t that same amount of time and energy be better spent simply walking door to door and getting out the vote? 


Thursday, August 13, 2020

Thoughts on Executive "Orders"

I want to address these point by point, but let me first explain why they're "orders" and not just orders. Two of the four main items weren't even orders, they were memoranda. Think of them as fancy office memos. The other two are pretty limited in scope, and it's fairly easy to argue that all four will be moderately effective at best in helping anyone.

Let's look at each of these in turn and explore why they're all fairly toothless and ineffectual.

1) Payroll Tax Deferral - Likely Legal

This one doesn't really do much, except defer when the taxes will be collected. There is no "cut," since the executive holds no power to levy or reduce taxes. He can only control "the execution of the laws," and so in this case is telling the IRS to collect the tax later.

Only a few problems:

  1. Payroll companies and HR departments aren't set up for this. It would require them to change their software to track what they're supposed to collect, but to not collect it now. Then later (at some unspecified date), they'd collect it. It's not legally clear when this later collection would have to happen, but bleh...
  2. People aren't paying less in taxes, they're just paying them a month or two (or 6) down the line. 
  3. This has NO EFFECT on people who are unemployed, and are thus suffering the most, since they don't have paychecks, and thus don't pay payroll taxes

I think it's safe to say that #3 is the biggest problem, and the one people are talking about the least.

2)  Unemployment Aid Extended - Possibly Legal

In this case the president is pulling from FEMA disaster funds right before hurricane season kicks off, so sounds like a great idea. That being said, it has other problems too:

  1. It maxes out at $400, 2/3 of the current benefit (before that benefit expired two weeks ago). So, in effect, it's a cut from where people were that was keeping them afloat.
  2. It only works if the state kicks in extra money first. Want the full $400? Your state first needs to authorize giving you an extra $100 from their pocket, and the Feds kick in the other $300. Think states will be authorizing extra money in this environment? Yeah, think again.
  3. It only has enough funds for 5 weeks. Yeah, you read that previous sentence right, though because of the problem laid out in #2, maybe the funds will last far longer because no one can use them because states won't kick in any money to start with.
Who knows, but the above problems are real problems, and even if states do give extra money(#2), look at #3 again. It's not good.

3) "Halting" Evictions - Definitely Legal

This is possible the stupidest and most useless of the four. It order federal agencies to think about whether to stop evictions in some cases.

Yeah, let's think about it. Maybe we should. Who knows, but it's safe to say that thinking about it won't do much on a massive scale. 

This one is definitely legal, hands down, precisely for the reasons that it's so fucking stupid and toothless.

4) Student Loan Forgiveness - Definitely Legal

This actually seems like the most useful, and definitely in the legal camp. It extends the 0% interest rate on student loans and pauses payments until the end of the year. This is well within the purview of the Feds, and doesn't seem to be pushing any major legal envelopes. 

My one critique of this? Why not go further and actually cancel some amount of student debt? It wouldn't cost that much in the grand scheme of things, and could have profound consequences on individual people's lives. Maybe wipe out $10,000 or $20K of people's debt. For lower income folks that extra couple hundred bucks a month could have a really positive impact, both in their lives, and in the extra money that would be plowed into the economy in their local communities.

Conclusion

Trump's overall actions seem mostly legal. They also seem mostly ineffective or, at best, moderately effective in some limited cases ($400 a week isn't nothing, even if it does only last 5 weeks).

The problems we're facing are much bigger than any we've seen since the Great Depression and WWII. This is not the time for small solutions. In a future post I hope to share a few solutions that would be drastically better and more effective, but that (unfortunately) would absolutely require buy in from Congress. 

That last part is why we're all fucked....