I See 1,800,000 Dead People

From the “Is anybody really surprised about this?” file:

A new report by the Pew Center on the States finds that more than 1.8 million dead people are currently registered to vote. And 24 million registrations are either invalid or inaccurate.

There’s little evidence that this has led to widespread voter fraud, but it has raised concerns that the system is vulnerable.

This is NPR, so I am not surprised at the skeptical tone of the article. Here’s another thought: are we only concerned about “widespread voter fraud” or would should our concern begin with fraud perpetuated by a single voter? After all, as we have seen in election after election in the last few years, it is not at all uncommon for the numbers to be so close–even in statewide races, that a few hundred or more votes separate the winners from the losers. How much voter fraud would be needed in those cases to subvert the legal process? Very, very little.

Moving on, then, we have this from someone whose job description is to ensure the validity and legality of the voting process:

Washington Secretary of State Sam Reed says it’s amazing how many times his state has come across names on the voter rolls that appear to be the same person, but turn out not to be.

“We’ve even had cases, in very small counties, people [with the] same name and same birth dates,” added Reed.

He said that has led to inaccurate reports that “dead” people are voting. He admits there have been a few cases in his state where widows or widowers have cast ballots for former spouses, but he said such fraud is very rare.

Again, dare we ask what “very rare” means? In another article, Reed mentions that there are problems with 1 in 16 registrations. By my basic calculations, that means a bit more than 6% are wrong (person died, moved to another state, etc). That is less than the difference between the winner (Murray) and loser (Rossi) in the 2010 Senatorial Election for Washington State, where Mr. Reed resides.

When it comes to fraud, one does not wait until it can be proven to make changes. Rather, one looks at the probability which exists (that is, how simple is it to act fraudulently within the current system/framework).

It is good that states will be sharing some information with each other to better keep their voter rolls cleaned up. Unless voter ID is required along with such cleanup, one believes that the opportunity for fraud will not be substantively diminished.