What we learned from the Minnesota primary?

Far less than you think.  Journalists and politicos want to write the big story and find trends.  If there is a special election in one race they see in it a harbinger of a trend.  Think of Eric Cantor losing to a Tea Party candidate and how from one race everyone is saying that the immigration issue did him in and therefore Republicans will refuse to compromise on this topic.  Maybe it was immigration that cost him his seat, or maybe it was that he lost track with his constituents or simply was complacent in his campaigning. This is what did Jim Oberstar in.
    My point is that there are two competing trends in thinking about politics.  One is the classic Tip O’Neill line that all politics is local.  The other is the belief that all politics is now being nationalized.  Evidence exists for both propositions and there are serious questions also whether one can generalize from one event to seeing a trend. 
    Minnesota’s August 12, primary invites this type of speculation.  By now the received wisdom is that endorsements matter and that the turnout was bad.  There is some truth to both but more needs to be said.
    First it appears that all the party endorsed or party favored candidates one–Johnson, McFadden, Otto, Kahn, and Loon just to mention a few.  After what appeared to be a nearly generation-long repudiation of convention endorsements, they seem to have mattered this time and perhaps it is a sign of a trend.  Yes they mattered this time but a trend is not something one can leap to yet.  This primary the endorsements mattered and parties looked strong because the turnout was so bad.  The two are related and when we think about it, the parties do not look that strong.
    One of the classic functions of political parties is voter mobilization and get out the vote.  Strong parties are generally associated with vigorous voter turnout.  But in this primary turnout was 12.8%, more or less confirming a long-term decrease in turnout dating back 30 years in Minnesota when in 1982 31% showed up on primary day.  Since then the number of individuals identifying themselves as a member of a political party has decreased, thereby making it harder for parties to mobilize as many people as before.  Many of those people are also younger voters or Millennials who are less likely to identify with the two major parties compared to the past.  They are also individuals harder to reach by more traditional methods of communication (mail or phone) and instead need to contacted by alternative or social media.  Finally, unlike 30 years ago, third parties such as PACS are more powerful today and they along with interest groups perform many of the functions that parties used to perform. 
    My point is simple.  Low turnout and the appearance of party endorsements as mattering last Tuesday went hand-in-hand.  Because the parties (and the candidates) did such a bad job reaching and mobilizing voters the turnout was so bad.  With turnout so bad we saw young people stay away from the polls and also those who did show up were the hardcore partisans–those whom the endorsement process mattered most to.  Come November when turnout is at about 54% things will look different.  Parties and the candidates will be competing more with third party groups (PACS, legislative caucuses,  and independent expenditures) for influence and they may or may not look as relevant as they did in this primary.
    Now of course there are other reasons why turnout was bad.  As I pointed out four years ago in a study other states that moved from a post to pre Labor Day August primary demonstrated lower turnout.  Four years ago MN had a slight bump in turnout (as did other states with their first August primary) but the longer term trend is for lower turnout.  Few people are thinking of politics n Minnesota except for the hardcore politicos.  Most Minnesotans barely think about elections until the state fair, or after Labor Day when school starts.  Picking an August primary date as opposed to an early one such as June favors incumbents or party endorsed candidates by holding an election when many people are not thinking about voting.  June would be better for an primary but incumbents did not want it because of how close it came to the end or the legislative session (assuming it ends on time) and it would hurt their ability to fund raise and campaign.
    A couple of other thoughts about Tuesday.  This was the first with new excuses early voting.  Many thought this would increase turnout.  It did not for two reasons.  One, few people knew about the change in the law.  Second, the evidence is mixed regarding whether early voting really increases turnout or simply makes voting more convenient for those who do vote.  Additionally, it is not clear  that many candidates knew how to use early voting to their benefit.  I remember talking to Scott Honour at one point and he thought he would win because he had the most money and would use it to get people to early vote.  I asked him how he was taking advantage of early voting and he said he had a link on his web site for people to download an absentee ballot application.  This is hardly  a good use of early voting.  I heard of similar other candidates operating with similar naive strategies.
    Finally, one other argument I had emerging out of this election is that money did not matter–Entenza and Honour lost.  Yes they did but in part because they did not use their money effectively and also because there is something that one needs to consider–they were not quality candidates.  Both faced many liabilities that money just could not overcome.
    So what do we learn from the August 12 primary?  Maybe far less and far different from what the received wisdom is saying

Predicting Turnout and Winners in the 2014 Minnesota Primaries

Getting it Wrong...and Right
Predicting turnout and winners for the August 12, 2014 Minnesota primaries is complicated.  Four years ago when Minnesota held its first August primary I predicted a 12% voter turnout.  This estimate was premised upon two factors.  First, a general downturn since 1982 in primary turnout and a trend from other states of a decrease of 2% when switching from a post to a pre Labor Day  primary.  I was wrong!  The 2010 turnout was 15.94%–an approximate 2% jump in turnout.  I should have anticipated that given the closely contested DFL gubernatorial primary involving Margaret Anderson Kelleher, Matt Entenza, and Mark Dayton.  That race also featured significant political spending.  Couple a closely-contested race with big money and voter turnout should go up.

Table I:  Minnesota Primary Turnout: 1982-2010 (Gubernatorial Years)
Year    Turnout    Contested?
1982    31.08%    Yes
1986    25.69%    Yes
1990    24.28%    Yes
1994    27.17%    Yes
1998    19.79%    Yes
2002    14.93%    Yes
2006    13.8%      Yes
2010    15.94%    Yes

   
Predicting 2014 Turnout
    Given what I learned from 2010, what is my prediction for the 2014 primary turnout?

    The prediction for turnout in the 2014 primary can be expressed as an equation:


T= C +$ +A + P
Where:

T = August 12, turnout
C = Contested election
$ = Money spent by candidates, parties, and political organizations to encourage turnout
A= No excuses absentee voting
P = Party turnout


Let me explain each variable.  Contested election refers to whether there is a seriously contested statewide election on the ballot that should drive turnout.  In 2014 there are at least two such races.  The first is the Republican gubernatorial election featuring Jeff Johnson, Kurt Zellers, Marty Seifert, and Scott Honour in a close race.  The second race is the DFL primary for state auditor between Rebecca Otto and Matt Entenza.  Both races will ensure that party regulars will be excited about these elections, that the candidates are working hard to get out the vote, and that the potential closeness of these elections will entice voters to cast ballots.  On balance, I would argue that the level of contestation here is not as high as 2010, thereby suggesting a slight decrease in turnout in comparison.

Money spent refers to amount of money spent by candidates, parties, and political organizations on  advertising, voter mobilization, phone banks, and get out the vote to encourage turnout.  Here again  one can anticipate that several candidates, such as Scott Honour and Matt Entenza will spend significant amounts of their own money to support their campaigns and that in turn will force their opponents to also spend more money.  Of course, not all spending will be directed toward voter mobilization, but in general the more money spent the greater the turnout.  Here less money is being spent compared to 2010, again suggesting a lower turnout.

No excuses absentee balloting refers to how the change in laws affecting early or absentee voting will impact voter turnout.  By that, in general no excuse voter turnout has lead to more people across the country voting before election day and there is also some evidence that it has increased turnout.  The issue here for the August 12, primary is to ask whether this new law will lead to an increase in turnout or simply to a shift in more voters casting ballots before election day. Assume no excuses absentee turnout slightly increases turnout from 2010.

Finally, party turnout refers to which party or parties have a more competitive turnout.  The DFL generally enjoys a greater percentage of the population voting in their primary as opposed to the Republicans.  This reflects the fact that more people in the state consider themselves Democrats than Republicans.  In 2010 74% of the primary voters for governor voted DFL.  This high turnout was  in part due to a close DFL race and an almost pro forma GOP primary.  In most years the DFL share of the primary vote is lower than this.  Given that there are close statewide races for both parties, my estimate is that about 60-65% will vote DFL and the remainder Republican.  Even though the DFL State Auditor’s race is contested and will drive turnout for that party, it is not generating the same excitement as a governor’s race.  Thus, the main race driving the 2014 primary is the GOP gubernatorial race.  Given that, I anticipate a slight decrease in enrollment.

Taking all of the variables together, what can we predict?  As of May 1, 2014, the Minnesota Secretary of State estimates the total number of registered voters in Minnesota is approximately 3,116,899.  My estimate is a turnout of 14% or 436,000 voters.  Of those, the estimate is that 275,000 will vote in the DFL primary and 161,000 in the GOP primary.

Winners and Losers
One final prediction: In the Republican gubernatorial primary I am estimating that the winner will receive approximately 35% of the vote.  If that is correct, it will take about 57,000 votes to win the Republican gubernatorial primary.  Put in perspective, the winner will receive approximately 2% of the vote in the state.  This proves that every vote counts.  If I had to make a prediction I would argue that the GOP primary is between Jeff Johnson and Kurt Zellers.  This pits party endorsement and resources against name recognition.  This is too close to call but the guess goes to Johnson.  In the Otto-Entenza race, party mobilization for Otto and backlash against Entenza should help her win.


Upset Prediction
Finally, a note of caution: I have note described the race between and Mike McFadden and Jim Abler as a contested one.  But after what happened to Eric Cantor in Virginia complacency is a clear vice to avoid.  Jim Abler is the underdog but he has the endorsements of the Star Tribune, Governor Al Quie, and Senator Durenburger.  If I had to pick one possible upset this is it.

Come August 13, we shall see how wrong I am!