Forgive me for obsessing on the topic, but when our foils at the Seattle Times editorial board and our friend Joel Connelly at the Seattle P-I are both editorializing in favor of changing the voting deadline from postmarked on election day to received by election day, you just know there’s gonna be another legislative move afoot to do exactly that. And with the facts firmly on my side, I just can’t let this one go.
Both Joel and the Times complain that ballot counting in Washington state is too damn slow, and both point to first-in-the-nation all-vote-by-mail Oregon and its received by election day standard as a model for how to do these things right, so you might reasonably assume that Oregon counts its ballots considerably faster.
Well… not exactly.
It’s hard to do an apple to apple comparison, what with last Tuesday having been our first all vote by mail general election, while Oregon didn’t have a 2009 general election at all, but a quick look at King County Washington’s performance during 2009 versus Multnomah County Oregon in 2008 bears some mixed results.
Of the 366,948 ballots cast in Multnomah in November 2008, 133,908 were tallied and reported by the end of election night, or roughly 36.49%. Of the approximately 600,000 ballots projected to have been cast in King in November 2009, 254,261 were tallied and reported on election night, or roughly 42.4%.
That’s right… on election night, slow as a snail King reported a larger percentage of the total ballots than did supposedly speedy Multnomah.
From there on, Multnomah jumps out ahead, tallying 60.69% of the total ballots cast by Wednesday night, and 94.3% by Thursday, compared to a relatively paltry 51.4% and 62.9% respectively for King. But how much of this advantage was due to Multnomah having all the ballots in hand by 8pm Tuesday? Not much.
Unlike King, Multnomah elections appears to have labored around the clock during the first few days following the election, generating 29 reports between 7:41 PM Tuesday and 4:40 PM Thursday, and at all hours of the day and night. KCE, not so much, generating just three reports during its equivalent three day period, working what appears to be a daily, eight-hour shift. And it really does take Multnomah a three-day, round-the-clock effort to push its way through 94.3% of the ballots it has on hand.
So what if King were to invest in the same sort of effort?
Well, as it turns out, KCE reports a daily estimate of the uncounted ballots it has on hand, and when you add those to the daily totals, King could have conceivably tallied as much as 72.9% of ballots by Wednesday night, and 85.9% by the end of the day Thursday. And by Friday night, when Multnomah had tallied 95.7 of its ballots, King already had 94.1% of projected ballots either tallied or on hand.
Thus it isn’t a lack of ballots that slows the counting process in King, but rather the lack of sufficient manpower and infrastructure to count them as the ballots come in. Indeed, moving the ballot deadline without dramatically increasing KCE resources would not have sped up the tallying process at all, as KCE barely got through the election day ballots on hand by the end of Friday’s first shift.
The point is, tallying mail in ballots takes time — much more time than polling place voting machines, which tally the ballots as they are cast — and given the rules that govern our elections, no all mail-in election is going to produce the near-complete election night totals we see from other states. And that is what the Oregon example really proves.