[EL] Convention(al) Wisdom The Myth of the Convention Bump

[David A. Schultz]

Hi all:

I thought all of  you might be interested in this piece of mine as the
Tampa RNC commences. . .or not.

Convention(al)Wisdom
TheMyth of the Convention Bump
 
DavidSchultz
            Conventionalwisdom pervades presidential politics. Many
assume vice-presidential picks matter and can be game changers,leading
presidential nominees to worry about regional balance, ideology,
orfavorite son factors when selecting their Veep. Yet the truth is that
as a rule vice-presidents matters little in termsof affecting voters
decisions.
            Anotherconventional wisdom is about national political
conventions.  Parties seem to believe that the placement ofa national
convention can affecting presidential voting in that state.  The claim
is that there is a convention“bump.”  This bump means that placing
aconvention in a state can flip a state in presidential voting*from
Democrat toRepublican or vice versa*or it can make a state more
competitive.  It does that by rousing the base or perhapseven by
capturing the attention of swing voters, either of whom will be
moredisposed to vote for party holding the convention in that state in
thatelection.  
            Proofof a convention bump is the Democratic National
Committee’s placement of itsconvention in Colorado in 2008.  Here
theDemocrats with John Kerry went from losing the state by 4.7% of the
popularvote in 2004 to George Bush to winning it in 2008 with Barack
Obama by a marginof 8.95% over John McCain* a bump of 13.65%. However,
Obama’s 2008 victory in Colorado is the exception*for the mostpart there
is no bump and in fact, there is evidence that holding a conventionis a
state might actually hurt a presidential candidate.  Look to the
placement of the RepublicanNational Convention in St. Paul, Minnesota in
2008.  Republicans thought Minnesota was a bluestate turning purple and
holding the convention in my hometown might swing thestate to the red
column.  It did just theopposite.  In 2004 Kerry won Minnesota by3.5%,
Obama then won it by 10.24%.  TheRepublicans nearly did 6.75% worse in
Minnesota by holding a convention there.
            Onecan dismiss Colorado and St. Paul as oddities. But more
systematic evidence reveals a mixed picture for the conventionbump. 
Since 1948 there have been 16presidential elections, producing 32
national presidential conventions betweenthe Democrats and the
Republicans.  Ofthose 32, when a party had a convention in a specific
state there was no changein who won the state compared to the previous
election in 22 situations(68.8%). However, there are five instances of
apparent bumps producing aflip.  In 2008 and 1976 the Democratsheld
conventions in Colorado and New York (Carter wins in 1976, Nixon in
1972)and flipped states that the previous election cycle had gone GOP.
In 1968,1952, and 1948 Republicans held conventions in Florida (Nixon in
1968, Johnsonin 1964), Illinois (Eisenhower in 1952, Truman in 1948),
and Pennsylvania(Dewey in 1948 and Roosevelt in 1944),and flipped states
that the previouselection cycle had gone Democrat.  Thus,a total of five
states were flipped (15.6%).
            Conversely,in1980, 1952, and 1948 Democrats held conventions
in New York, Illinois, andPennsylvania and lost those states even though
they had won them the previouselection cycle. In 1964 and 1960
Republicans held conventions in California andIllinois and lost those
states even though they had won them in the previouselection cycle. Thus
a total of five states were lost (15.6%).  Five states flipped, five
states lost, awash.
            Perhapsa better measure is not to look to states flipped,
but look to see if a bumpoccurred.  Did the party holding aconvention in
that state make the election more competitive?  At first blush there is
evidence of abump.  For the 32 political conventions,there is a 1.4
point average gain for party holding convention in a state inthe
presidential popular vote compared to popular vote in that state
theprevious election.  But this bump againis illusionary.  In 18
instances (56%),the party holding a convention in a state received a
smaller percentage of thepopular vote compared to the previous election.
In roughly half the cases a party did better, in half the cases it
didworse.  Again a wash.
            Buteven the apparent 1.4% gain means little and it hides
what reallyoccurred.  For example, in 1968 RichardNixon defeated Hubert
Humphrey in Florida by 9.6%.  The RNC held their 1972 convention in
Floridain 1972 and Nixon beat McGovern in that state by 44%!   Nixon
wins California in 1960 by aboutone-half percent, the Democrats hold
their convention in that state in 1964 andwin it by more than 18%. 
Conventionbumps? No.  There are many instances ofblowout 
elections*Johnson 1964, Nixon1972, Reagan in 1980 and 1984, and
perhaps Clinton in 1996 and Obama and 2008that distort the apparent
bump.
            Theevidence on presidential conventions contradicts the
conventional wisdom*thereis no bump.  Specific factors unique toeach
election affect how a state and its people will vote.  National
conventions are more of a mediafiction or political party myth than
anything else.





David Schultz, Professor
Editor, Journal of Public Affairs Education (JPAE)
Hamline University
School of Business
570 Asbury Street
Suite 308
St. Paul, Minnesota 55104
651.523.2858 (voice)
651.523.3098 (fax)
http://davidschultz.efoliomn.com/
http://works.bepress.com/david_schultz/
http://schultzstake.blogspot.com/
Twitter:  @ProfDSchultz
Named one of the inaugural 2012 FacultyRow SuperProfessors
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20120826/2e5824af/attachment.html>