The big campaign story for the next few days

by Andrew Gelman on November 2, 2012 · 8 comments

in Campaigns and elections

Thomas Ferguson writes about the campaign:

There’s something chilling that haunts these final days: the Ghost of the 2000 Election. . . . for 2012, the scariest thing about 2000 is the evidence that a flood of highly concentrated Republican money in the very last week of that campaign gave G.W. Bush a decisive edge in the battleground states – and that contrary to reports in the national media, there are signs that history may be about to repeat itself.

The little known 2000 story is meticulously laid out in a study by Richard Johnston, Michael G. Hagen, and Kathleen Hall Jamieson. . . . a natural experiment, in which part of the country was saturated with political money while the rest was only lightly sprinkled.

The outcome was ruinous for Gore. Johnston, Hagen, and Jamieson convincingly show how in non-battleground states, where free media and Gore’s own ads were not overwhelmed by the last minute GOP avalanche, the Vice President preserved his momentum, eventually winning the popular vote. By contrast, in battleground states where the Bush campaign vastly outspent him and the Democrats, Gore’s comeback stalled out. “Where ad volumes – Al Gore’s ad volumes in particular at this point – were mounting, the Democratic candidate held his own for the rest of the month…Where advertising – now overwhelmingly by Bush – was heavy, there was no recovery; indeed in the last week Gore’s share in these places dropped two to three points.”

What about the upcoming election?

This year the gigantic war chests raised and spent by Superpacs have plainly stunned many Americans, making them overwhelmingly receptive to tighter regulation of political money.

Curiously, however, all through the campaign, one commentator after another has derided the idea that big money might be decisive in this election. . . . Some of their skepticism rests, perhaps, on a deep misunderstanding. Big Money’s most significant impact on politics is certainly not to deliver elections to the highest bidders. Instead it is to cement parties, candidates, and campaigns into the narrow range of issues that are acceptable to big donors. . . .

Difficulties with the data:

Let’s look further into why the pundits are confused. The major sources of data on political money are the Federal Election Commission and the U.S. Internal Revenue Service. . . . both agencies routinely accept seriously incomplete reports and obviously inaccurate or misleading reports. . . . As a result, the true influence that large donors wield in American elections is chronically underestimated. . . . Existing data management tools that try to match these up commonly fail to recognize multitudes of contributions coming from the same sources. In turn that nourishes illusions that small donors play bigger roles in campaigns than they really do. Especially where Democrats are concerned, the myth of small donors is a powerful instrument of miseducation.

What about Obama’s army of small donors?

In 2008, for example, the Obama campaign trumpeted such support. Eventually, many analysts caught on and began to question the claims. We have now reanalyzed the entirety of the FEC and IRS data for 2008. . . . Our best estimate – it is only an estimate – is that donors giving less than $250 represent less than ten percent of the record breaking sums raised by the Obama campaign [and similar numbers for McCain in 2008, as well as Obama in 2012].

What happens next?

Since October 17, the big GOP Superpacs appear to be outspending [Obama’s] Priorities USA on media by at least three to one – perhaps a higher ratio than when Bush buried Gore.

2012 could be different from 2000, though:

It can be reasonably objected that the absolute level of spending in the 2012 presidential race will be far above the paltry $343 million reported for 2000. Perhaps the higher level of expenditures will better anchor voter impressions of Obama. It is also true that Obama is not a Vice President, but President, someone much harder to drown out. But how much will these factors matter? No one can say for sure – no one has any experience with political money on this scale.

{ 8 comments }

Felix November 2, 2012 at 11:36 pm

Another reason why 2012 could be different: The Obama campaign bought airtime months ago, when it was much cheaper. So while he is getting outspent, that alone tells us little about how many pro-Obama ads voters actually get to see. See here for more: http://abcnews.go.com/blogs/politics/2012/10/republicans-outspend-democrats-on-tv-advertising-but-have-fewer-ads-to-show-for-it/

Atrox November 2, 2012 at 11:50 pm

They’re a little late. Three days to the election, and today was a great day of polling for Obama. Hard to imagine that Ohio could be bombarded any more than it already is. Romney is going to need a miracle.

Andrew November 3, 2012 at 12:06 am

There is only so much air time too, right? I realize there are a ton of channels on the dial, but there has to be some limit or some threshold where it goes from being effective to drowning yourself out. If Obama bought/buys time as well (and he has his own share of money), then how much air time will actually be available for Romney to buy?

Also, swing states like Ohio, Virginia, and Florida have Senate races as well, as does Pennsylvania, Michigan, and Wisconsin, places where Romney is trying to expand his map. Some of them are still relatively competitive; won’t those candidates invest in ads too, further limiting the amount of time available for the top of the ticket? Has that ever been researched?

Richard November 3, 2012 at 12:19 am

Theory fails to take account of the prevalence of early voting now as compared to 2000, particularly in Ohio, Florida. Nevada and Iowa.

evnow November 3, 2012 at 12:32 am

One thing to note is that in battle ground states, TVs are inundated with ads. Saw a video the other day of some 25 continuous political ads in Ohio.

I wonder what the research shows about effectiveness of ads in that context.

DrunkWino November 3, 2012 at 1:14 am

There’s another side to this. I live in Virginia and I’ve been getting about 4 robocalls a day on average from Republican supporters and the TV ads are basically the only ads on the network channels. I’ve gotten to the point I stopped watching TV and unplugged my home phone. I had already decided my votes two weeks ago and while I was going to vote for only one Republican in a local county race, the sheer annoyance and aggravation of getting call after call after call convinced me to change even that vote.

If somebody is trying to call you that much to try and cajole you into voting for their guy, I look at it as their guy can’t stand on his own and his only chance is to employ the two-year old’s method of just repeating “please,” until his parents give in.

One or two calls, I understand. Twenty in five days turned me way off to Romney.

Christopher Gelpi November 3, 2012 at 7:30 am

I don’t see the evidence that there is any such lopsided advantage for Romney in ad buys. Here are the data through October 27. Romney generally has a modest advantage in most states, but it seems to be fairly even – including in battleground states.

http://www.washingtonpost.com/wp-srv/special/politics/track-presidential-campaign-ads-2012/

scott November 3, 2012 at 9:46 am

In risk management, risks are prioritized according to the probability
of each happening and the severity of the consequences. More people
need to step up to the plate to make sure this election and future
elections are not stolen by computer code that alters the results. It
is extremely plausible that a computer virus could get into the
central tabulators which sum state precinct votes and flip votes from
one candidate to another. Two people whose occupations specialize in
statistics and fraud analysis show that there was vote flipping in the
2012 Republican primaries (benefiting Romney) and vote flipping in the
2008 presidential election (benefiting McCain). See link here:

http://www.themoneyparty.org/main/wp-content/uploads/2012/10/2008_2012_ElectionsResultsAnomaliesAndAnalysis_V1.51.pdf

The severity of the consequences of a Republican presidential
administration is high (climate change, education, health care etc).
More people need to step up to make sure stolen elections do not
happen by analyzing the results.

Comments on this entry are closed.

Previous post:

Next post: