Friday, August 8, 2008

Is there a case to be made against advance voting?

Advance Voting has now been around since the 1996 election cycle. This year, a total of over 23,000 Johnson County voters cast their ballot early -- 30% of primary voters -- a new record in terms of the percentage of voters voting early and total advance ballots returned. While the 22.52% overall turnout was fairly low for a primary, the trend is clearly for more and more voters to vote early. Brian Newby of the Johnson County Election Office recently said that his office was encouraging more and more voters to vote early.

Is this really a good thing?

The conventional wisdom is that it is. Some will say that it increases the likelihood of participation in the political process. Others will say that the more convenient, the better.

Kaw & Border is not so sure.

First of all, we believe that there may be a Constitutional issue here. Take a look at Article 4, Section 2, which deals with the General election in Kansas:
2: General elections. General elections shall be held biennially on the Tuesday succeeding the first Monday in November in even-numbered years. Not less than three county commissioners shall be elected in each organized county in the state, as provided by law.
Notice how it says the general election shall be held on the day. Right now, as a result of advance voting, it occurs on that day -- and the 20 days proceeding election day as well. To our knowledge, this has never been challenged. It may still be okay -- for it doesn't prohibit advance voting, just says that there has to be an election day. For primary elections, which are set by the legislature, there would be no constitutional conflict whatsoever. But, perhaps it was envisioned by our state (and federal) constitutional framers that elections were meant to be on a select day. Why is this? this leads to our second question:

2. Should elections be a snapshot? Political campaigns are continually putting out information, whether they be for or against a candidate or for or against a certain ballot initiative. Due to the nature of campaigns, some information comes out late in the game. This is often the case with ballot initiatives. As such,is it good policy to therefore encouraging people to vote so early in the process (up to 20 days early) that they may miss information that could have led them to a different conclusion?

In the primary this past Tuesday, we have a ready made example of this: The Johnson County Forever Sales Tax. Information against the tax -- including newspaper editorials -- didn't come out until after voting had begun. There was a clear trend against the tax, shown by the results Tuesday evening:

Advance Voting: Yes, 14,036. No, 8606 -- a difference of 5430 votes. (61% - 39%)
Total Votes: Yes, 40183. No, 35,189 -- a difference of 4994 votes. (53%-47%)

Do the math.
On election day, the Forever Tax LOST 26,583 - 26,147 (50.4% - 49.6%).

That's a HUGE swing -- from 39% voting no to over 50! Though the majority of votes were on election day (70%), the 5430 margin win for the tax in advance voting was enough to overwhelm the small 400+ vote No margin on election day. However, there was a clear trend against the tax. A simpleton may say, well, there is no guarantee that those voting Yes early would have voted No on election day, if we had no advance voting.

This might be true -- as the margin was razor thin on election day. But is it such a stretch to think that 2715 advance Yes voters (enough to make it a tie in advance voting) would have changed their mind, given the trend and information coming out against the tax? Had that occurred, the tax would have failed.

What we had here was a rather innocent ballot question, seemingly, asking about public safety and such, plus a taxpayer-funded "information" piece. This drove up the advance number. But, over the last three weeks of the campaign, you had newspapers coming out against the tax, and a lot of buzz about how the tax was a bad plan. On election day, the tax failed, despite the wording of the ballot and the fact that there were no mailings against the tax -- vs. 3 or 4 by the proponents. Point is -- information got out, and the voters, on election day, said no.

But this doesn't just impact ballot questions where information is slow to get out. It also impacts races for office. In races across Johnson County and the state of Kansas, much of the informative mail and information came out in the last week, after a 1/3 of the voters had voted. In Senate District 10, for example, check out the difference:

Advance Voting: Mary Pilcher Cook, 52% - Sue Gamble, 48%
Election Day Voting: Mary Pilcher Cook, 59% - Sue Gamble, 41%

Mary ended up with 57% overall, and would have won in either case, but that 7 point swing is rather large. Could this have been because voters responded to Mary's positive campaign and rejected the KTRM hit pieces, which mostly came in the last two weeks?

If you look at races in the county and statewide, in this cycle and past cycles, you will consistently see a large gap between advance voting and election day voting numbers. Now, some of this is simply because a candidate or party has a better advance voting operation or got out earlier. But, some of it is related simply to money -- challengers to incumbents and new candidates for open seats have a harder time raising money, delaying their ability to get their message to the voters. This leads us to the third question:

3. Does Advance Voting give incumbents and well known candidates an unfair advantage?

The answer here is -- yes and no.

In general elections, the answer is largely no -- as in 2006, Democrats consistently out performed against incumbent Republicans in advance voting, while the Republican overwhelmed the Democrat in election day voting. In a few cases, you'd see the Democrat WIN in advance voting and the Republican win handily in election day voting. In two races, that resulted in the Democrat winning -- District 18 and District 16 -- by very narrow margins (2 votes in District 16!) . This would reflect a superior advance voting operation by the Democrats. Not only that, any challenger would have had significant time -- during the summer and fall, to get their message out.

In primary elections, the answer is largely yes. With the filing deadline on June 10, there is a small window of just 40 days before voting starts. That's not enough time to walk a district, and a limited amount of time to raise enough money to pay for mailings.

Take the race in the District 8 Senate race. Though an open seat, Rep. Tim Owens was from the District 8 area while Rep. Ben Hodge was from an area in Olathe and had recently moved. Check out the numbers:

Advance Voting: Owens, 1552 - Hodge, 795, a margin of 757 votes, or 66-34.
Total Voting: Owens, 4455 - Hodge, 3010, a margin of 1445 votes, or 60-40.
Election Day Voting: Owens, 2903 - Hodge, 2215, a margin of 688 votes, or 56-44.

Now, Owens would have won anyway, probably due to the factors stated above, but the 66-34 - 56-44 difference is a large swing. Hodge was walking, KTRM's hit pieces were backfiring, and it was clear that Hodge was trending up because the voters were getting more information and thus casting more informed votes. Now, does this mean that advance voters are all misinformed? No -- but many just vote to vote early, and wouldn't have had the full set of information in any event.

Another factor here is that it does drive up the cost of campaigns. Many pundits and citizens alike complain about the huge costs of campaigns, running now into the tens of thousands of dollars for state legislative races. Is a system with two election days contributing to this?

So what's the solution? We believe it doesn't have to be an all or nothing approach. Perhaps the advance voting period should simply be shrunk from 20 days to 10 days? That would lessen the impact of the margins above -- and still give people the ability to vote early in case of being gone, and also start advance voting AFTER the deadline to register to vote, which would seem to make logical sense.

The unfortunate probability here is though that no change is likely to occur, as any change would have to be passed by the legislature, and few politicians are likely to vote for a plan that would open them up to a charge that they are restricting voting access, despite the arguments made above.

However, we here at Kaw & Border wanted to pose the question.