Spending in the New Jersey gubernatorial race

by

Jono has previously posted on the uncertain relationship between election spending and election outcomes (noting that other factors such as national partisan swings and candidate attributes mean that election spending is not strongly determinative of election outcomes).

Spending reports from last year’s governor’s race in New Jersey, in which the Republican challenger Chris Christie beat the incumbent Democratic governor Jon Corzine, provide an interesting data point.  Reports indicate that Corzine outspent Christie by 2-1. 

Realclearpolitics reports: 

The final tallies are in for the Gubernatorial race in New Jersey last year. Jon Corzine outspent Chris Christie two-to-one the primary and general election campaigns, $31.5 million to $15.5 million. Even those numbers are a bit misleading since Corzine was essentially unopposed in the primary while Christie had a legitimate race on his hands.

Additionally, the Republican Governor’s Association spent another $7.3 million on Christie’s behalf, which was offset by $7.1 million in spending by Democratic groups on Corzine’s behalf.

Despite the $15 million spending advantage, however, Corzine lost by four points.

So when politicians say that they need to impose restrictions on advertising to control the influence of “big money”, it’s worth asking whether there is any cogent evidence that the relationship between election spending and election outcomes is as direct as they claim or whether they are just making assumptions.

Advertisements

6 Responses to “Spending in the New Jersey gubernatorial race”

  1. Michael Appleton Says:

    Well, assumptions are being made on either side, right, cos either way you’re moving from correlation to causation?

    I’m happy to offer an alternative story to yours: Corzine was a terrible Governor with historically low approval ratings in one of the worst-hit states in the worst economic times since the Great Depression – and that it’s *remarkable* he got so close to holding on, and that the money he had to throw around was one of the reasons for him getting closer than he normally would have. (Christie’s massive leads in the polls seem to have been eroded by Corzine’s negative television ads.)

    What we really need is two perfectly matched opponents who have vastly different amounts of money and see who wins. Not obvious how you create such elections, though 🙂

  2. Michael Appleton Says:

    Oh, and I apologise for the somewhat familiar tone of the above comment… I assumed, wrong, that this had been written by my old friend Stephen 🙂

  3. Jesse Says:

    No worries at all Michael. And I’d agree to a large extent. There’s no way to know whether Christie would have had a blow-out win if they had both spent the same amount (or if Corzine could have clung on if he had outspent Christie by 3:1 or 4:1).

    I don’t disagree that money helps. Politicians want to communicate with undecided (or at least persuadable) voters. Most undecided voters probably don’t turn up at rallies or town hall meetings so if you want to communicate with them you have to advertise, or go door to door, or walk around a shopping mall shaking hands with people.

    I just don’t think that money usually makes the difference between winning and losing given how many other issues go into the mix. To take a minor example, it’s likely that the media storm following the crack that Corzine made about Christie’s weight probably cost him some votes. Also, as you say, 2009 was a tough year for incumbents (and the political climate in 2010 looks similar).

    I also tend to think that in most elections there is probably a minimum spending floor for a viable candidacy. I think most reasonably well-established parties can achieve that level of advertising. I also think that diminishing returns kick in pretty hard at a certain level of spending.

  4. Joe Says:

    Where labroatory conditions cannot be created to test causation, the next best thing is to take as many examples as possible and attempt to regress away other factors. If you’re left with a strong correlation that holds accross a large number of examples, you can more safely infer the extent of causation.

    This would serve as a much better starting point for consideration than examples like the one outlined above which is pretty much anecdotal evidence. Jesse, have you come accross much of this sort of analysis in your travels?

    I accept that in plurality systems the chance of the marginal impacts of differences in spending will be unlikely to impact the result of the election, but:

    1) sometimes it may, especially if the election is close, so cannot be dismissed out of hand as a factor (though I don’t suggest that’s what you are doing);
    2) the consequences are wholly different in a proportionally representative system in which marginal differences in vote shares do have a significant difference, and ‘winning’ and ‘loosing’ look quite different to different parties.

  5. Jesse Says:

    Joe,

    1. Yes, I think some studies do things like build in dummy variables for, say, membership of the House Ways and Means Committee to try to account for the effects of seniority.

    2. I agree that the New Jersey result is simply a single example. However, I think it’s interesting because it cuts against what some people regard as a “common sense” view that election expenditure is strongly determinative of election outcomes.

    3. Your point about the impact of election spending in proportional voting systems is interesting. I think that some studies have looked at the connection between election spending and the share of votes won (as opposed to simply winning and losing). However, the only ones I’ve seen relate to essentially two party systems or the vote share of ballot initiatives.

  6. Chris Diack Says:

    Jesse is right. Some of the comments are off the track.

    First. Competitive candidates attract money; money doesn’t make an otherwise uncompetitive candidate competitive. The more competitive a candidate is the more money they can attract (raising money is part context part skill – competitive candidates tend to be good at it).

    The notion that Corzine wasn’t competitive but raised and spent more that his opponent only to lose less heavily that he otherwise would, is nonsense. The US Democratic Party just isn’t that amateurish. The State would have been polled extensively; that in of itself would have attracted donations if it showed their candidate was competitive.

    Second. The electoral system is not determinative of whether a voter is more persuadable by big spending campaign messages. Small shifts of voters in marginal constituencies elected under a plurality system can also have a dramatic effect.

    What the American research shows is the there is a weak co-relation between electoral victory and spending. This is really about the voters. It boils down to this: there are many factors which lead voters to vote a certain way; truth is (the campaign industry don’t necessarily like this) campaign spending is probably the lesser of those factors that influence voting.

    The best way to see this is the Billionaire candidates like Forbes who spent his own money on a platform that was not broadly attractive. If the punters don’t like what you are selling they won’t buy no matter how expensive the pitch.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: