There is understandable controversy over the prospect of more mail-in voting, but it’s reasonable to believe that some additional mail-in or postal voting may be necessary in light of the pandemic. Social distancing reduces the volume of activity that polling places can handle in a single day, and administrative decisions about the voting process can’t be deferred until late October in order to observe the state of the pandemic and make last-minute changes. Most states already permit voters to request a mail-in ballot for a variety of reasons: travel, illness, or other exigencies are usually sufficient, if a reason is even required. In the context of the pandemic, such a request should certainly be granted to those most concerned about contracting the coronavirus. So the option to vote by mail seems reasonable, at least in the abstract, as long as those who prefer to cast their ballots in person can do so.

“Universal” mail-in voting is another story, but the term first requires some qualification. I construe “universal” in this case to mean voting by citizens of the United States, a right protected and reserved to citizens by the 19th Amendment to the Constitution. That also means voters must be registered and must comply with state requirements for identification, if any, before receiving a ballot. In other words, under current state laws, a voter might be required to appear before an election authority to obtain a ballot for return by mail. The proponents of universal postal mail, however, seem to think states should simply mail ballots to the addresses of all registered voters. Many proponents go further, suggesting that all individuals of voting age should be mailed ballots.

The first major problem with a large expansion of postal voting is administrative complexity. It would represent a significant challenge for many jurisdictions to arrange in short order. It’s bound to create major delays in counting and reporting results, and it is likely to create doubt as to the reliability of the official election results. Here are some administrative issues and examples worth considering:

This recent experiment by CBS revealed delays in the official receipt of mailed ballots, a problem that will be more acute given plans in some jurisdictions to send ballots to postal voters only a week prior to the November election. The study also revealed some mis-sorting and misplacement of returned ballots. It concluded that a percentage of voters is likely to be “disenfranchised” by mail-in voting.

In early August, primary balloting by mail in Atlantic County, NJ was said to be especially problematic. Signatures on ballots were difficult to match to DMV records signed on “screen”; there was an extra step in delivering ballots to a central post office location and then on to election officials, causing delays; the voter registration system was plagued by technical glitches related to heavy demand for updated records; and there was insufficient time between sending ballots to voters and the deadline.

New York City’s primary election in June was similarly afflicted with a high rate of invalid mailed ballots. “The city BOE received 403,103 mail-in ballots for the June 23 Democratic presidential primary. … But the certified results released Wednesday revealed that only 318,995 mail-in ballots were counted. … That means 84,108 ballots were not counted or invalidated — 21 percent of the total. … One out of four mail-in ballots were disqualified for arriving late, lacking a postmark or failing to include a voter’s signature, or other defects. The Post reported Tuesday that roughly 30,000 mail-in ballots were invalidated in Brooklyn alone. … The high invalidation rate provides more proof that election officials and the Postal Service were woefully underprepared to handle and process the avalanche of mail-in ballots that voters were encouraged to fill out to avoid having to go to the polls during the coronavirus pandemic, critics said.”

From the New York Times, “In the last presidential election, 35.5 million voters requested absentee ballots, but only 27.9 million absentee votes were counted, according to a study [NYT link is bad] by Charles Stewart III, a political scientist at the Massachusetts Institute of Technology. He calculated that 3.9 million ballots requested by voters never reached them; that another 2.9 million ballots received by voters did not make it back to election officials; and that election officials rejected 800,000 ballots. That suggests an overall failure rate of as much as 21 percent.”

The problem of rejected mail-in ballots is all too common throughout the country. For example, redistricting can cause mail-in voters to cast their votes in the wrong precinct at a higher rate; people move frequently, especially low-income voters, so updating voter rolls is a tremendous challenge; and voters often fail to follow instructions carefully, and there is no one at hand to offer assistance.

Again, these are just the administrative problems. The upshot is that mail-in voting is likely to introduce uncertainties and delays in determining election outcomes, and is likely to result in numerous legal challenges as well.

This piece by Eric Boehm in Reason is skeptical of our ability to vote by mail without major complications of that kind. Boehm then turns to the question of mail-in ballots and fraud, however, quoting a variety of experts who claim that election fraud is a miniscule problem and that fraud has not had a partisan bias in the past. But partisan bias is not really the critical issue… fraud is, party by party, district by district, and state by state.

Despite Boehm’s protestations and widespread denial in the news media, election fraud is a “thing”. More importantly, the risk of election fraud is a thing. It’s instructive that two U.S. Senators (Ron Wyden (OR) and Amy Klobuchar (MN)) have introduced legislation that not only would authorize more widespread voting by mail, but “ballot harvesting” as well. The latter is the practice of visiting homes and “offering” to collect residents’ postal ballots for delivery to collection points. It has been a flagrant form of vote fraud in the past.

So what is our experience with fraud? Here is a “sampling” of 1,290 cases of election fraud, many of which involved absentee ballots and ballot harvesting. Detail on most of these cases can be found here.

The following testimonial reinforces the ease with which fraud can be perpetrated via mail-in voting:”I know because I did it“:

“Last year, a political operative working for North Carolina Republican congressional candidate Mark Harris was charged with fraud for directing a group of people to fill out as many as one thousand absentee ballot requests on behalf of voters — most of whom were unaware the ballots were being requested. … These people then collected the ballots and filled them out themselves. … 

Fraud risk always exists even if detected and proven levels of fraud are low, and the level of risk scales with the extent to which ballots are cast by mail. The sudden, massive expansion in mail-in voting now contemplated by some would create unprecedented opportunities for fraud.

Consider the 28 million mail-in ballots that went missing between 2012 and 2018, roughly 20% of mail-in ballots issued during those years. According to Logan Churchwell of PILF:

“So what do people that really focus on the election process do about that? They go into ballot harvesting. If there’s so many ballots out there in the wind unaccounted for by election officials, surely some manpower could be dedicated to go bring them in. And that’s another part of the system where you have weaknesses and risk.”

It takes only a small percentage of the vote to swing many elections, so ballot harvesting, enabled by more widespread voting-by-mail, is a serious threat to the integrity of the democratic process. The last link cites a few reports that should give mail voting proponents some pause:

“There’s little doubt that as the number of mail-in ballots increases, so does fraud. A 2012 report in The New York Times noted that voter fraud involving mail-in ballots ‘is vastly more prevalent than the in-person voting fraud that has attracted far more attention, election administrators say. In Florida, absentee-ballot scandals seem to arrive like clockwork around election time.’ According to a Wall Street Journal report on voter exploitation in Hispanic communities in Texas, mail-in ballots have ‘spawned a mini-industry of consultants who get out the absentee vote, sometimes using questionable techniques.’ Poor, elderly, and minority communities are most likely to be preyed upon by so-called ballot ‘brokers.’

Concerns about fraud in mail-in ballots were serious enough that a 2008 report produced by the CalTech/MIT Voting Technology Project recommended that states ‘restrict or abolish on-demand absentee voting in favor of in-person early voting.'”

It’s no coincidence that most countries in the European Union restrict mail-in voting to those who are unable to vote in-person, such as those working or studying abroad, as well as the sick and elderly. There are exceptions, of course, but many of these developed countries reject the notion that mail-in voting is worth the risks.

It’s reasonable to expect many cautious voters to request ballots for return by mail. But at a minimum, any large-scale transition to postal voting should be done with care for the security and integrity of the voting process. It is not an exercise to be done in haste, as proponents now demand. The result of such a drastic change would be significant delays, legal challenges, and reduced confidence in the outcome of elections. And there will almost certainly be fraud. As in almost all things, a voluntary option subject to jurisdictional risk controls is far preferable to either mandatory or “universal” postal voting.