The Mischief Rule

by

With the release of Power’s Cabinet Paper on Electoral Finance Reform, many commentators are beginning to weigh in on the implications of lifting spending limits on third parties.  Given that some of the coverage has been negative, the Minister is now saying he’s willing to consider imposing spending limits on third parties, if that is what is required to achieve “broad support” (i.e., agreement from political parties).

It’s important to return to basics and ask the question: what is the mischief or defect that the law is trying to prevent through spending limits?

There is simply no firm evidence that an absence of spending limits illegitimates an election.  The United States and Australia do not have spending limits, and yet the integrity of their elections cannot seriously be called into question.

Part of the reason why high levels of electoral spending does not impugn the outcome of an election is because there is very little evidence that spending influences voting patterns. The fact is that if a politician is highly succesful at soliciting votes, he is highly likely to be succesful at soliciting donations (the same is true of a third party – if third parties are effective at securing votes for their favoured platform, then they are more likely to receive money from the supporters of that platform).  Hence, correlation between spending and electoral outcome proves little.

What you need to do is adjust for confounding factors.  When such adjustments are made, there is very little evidence that spending is a major causal factor in how people vote.  One such study found that “campaign spending has an extremely small impact on election outcomes.”  In fact, some studies have even found that incumbent wealth can harm a political cause.

It seems then that prima facie there is no concrete harm that can be demonstrated from a liberal approach to campaign finance.  But, let us assume that higher levels of spending did in fact cause people to support a particular party of platform.  Would this be a bad thing? Some people may choose to vote for a party because of a charismatic leader, because they are bigots, or any other number of questionable reasons.  It is odd to say that someone should not change their mind on the basis of advertising paid for by people who think their cause justifies such sums of money.

Even if the people who change their vote are doing so in response to the volume rather than the content of the advertising, and such voters have real influence, then there is little that limits on spending will do to cure such shallow voters.  As Robin Hanson argues:

Many folks mistakenly assume that distortions from shallow voters stop if [third parties] are silenced.  But not only would that hinder non-shallow voters from getting info from [third parties], the total distortion by shallow voters is not obviously reduced!  Shallow voters who believe whatever side shows the most ads would either be bought by [third parties]  more indirectly, or by other deep pockets more directly.  And the many other kinds of shallow voters, who believe whoever has the funniest ads, or the coolest spokesfolks, or the prettiest candidates, would still cause distortions.

Advertisements

Tags:

16 Responses to “The Mischief Rule”

  1. What do political donors get for their money? « Today's Dissent Says:

    […] post below (“The Mischief Rule”) touches on the first of these assumptions, which he thinks – and I agree – is quite weak.  […]

  2. michael appleton Says:

    congrats to you three on the new blog! Stephen: a question and a comment. If spending had so little influence on voting patterns why would politicians in the united states, for example,spend so much time and effort raising it? Are they all simply mistaken as to what it takes to win an election? There were no doubt many factors that contributed to senator obama defeating senator mccain but i think it would be hard to maintain that his ability to massively outspend his opponent on the ground and in the air in all the major swing states wasnt one of them. And the comment: i am not sure what my ideal campaign finance system would be, but i do know that democracy is premised on each person having the same influence on an electoral outcome as any other – and it seems self evident to me that allowing people to spend whatever they want on campaigning does undermine this premise to at least some extent. Do you disagree?

  3. stephenwhittington Says:

    Taking the question first:

    So far as I see, there are three possibilities:

    First, as you suggest, politicians may just be dumb and are wasting their time. That seems somewhat unlikely, as someone could easily adopt a strategy that didn’t entail trying to raise so much money. Then again, given the common confusion between causation and correlation, perhaps this is having an effect on politicians’ beliefs.

    Second, there may be some other reason why politicians have an incentive to raise large amounts of money. Maybe party institutions reward those who raise lots of money more than those who do not. Maybe politicians who raise a lot of money have better job prospects afterwards in the private market.

    Third, and this seems to me the most likely explanation, there simply may not be all that much a politician can do to convince voters, so even though spending money has such a little impact, it is the best strategy amongst the alternatives. I would put it to you that you will almost never vote for the National party. That is independent of their campaign budget.

    I think the most accurate model of politics is that people’s beliefs are largely fixed. Parties then race to the centre to try to create a winning coalition. Parties like ACT and the Greens are doomed to be minorities specifically because they have core beliefs that they will not reverse. Advertising indicates what parties believe, but as most have some prior indication of this, the marginal effect is not great. You will also notice that the main impact of advertising in those model is informational, rather than propagandistic.

    The Obama example is simplistic again, because his ability to win votes is correlated with his ability to win campaign contributions.

    On the comment, I would need to see more substantiation of the idea that democracy is premised on the idea that each person has the same impact on the electoral outcome as any other. I cannot think of any country in the world where that is true.

  4. Jesse Says:

    Thanks Michael. I disagree that “democracy is premised on each person having the same influence on an electoral outcome as any other”.

    The universal franchise is an important feature of democracy (i.e., one person, one vote). But in a liberal democracy you are free to try to persuade other people how they should cast their votes. Whether the attempt to persuade people involves door-knocking, buying advertisements in the paper, or publishing editorials in a paper – it is inevitable that some people will have a lot more influence than other people.)

    So I don’t agree that your premise is self-evident and I don’t see how it could work in practice.

  5. Michael Appleton Says:

    Stephen: Of course only a certain proportion of voters are “convinceable” in a campaign. (In fact, two of the three general election votes I have cast were done so after a last-minute decision based on campaign events.) Most political scientists who I have seen try to measure it put the figure at somewhere between a fifth and two-fifths of voters. But elections are almost always won by the person/party who wins the greatest share of this bloc of voters. Your claim – that the role of paid campaigning in who this bloc decides to vote is minimal – just doesn’t seem consistent with the way that politicians behave. You almost seem to be saying that advertising/promotion is per se ineffective. If that is true, then there are an awful lot of self-interested corporations out there wasting huge amounts of money every day on promotional budgets.

    Jesse: I did not express well what I was trying to say. In some previous societies, the concept that those who were already in a (relatively) priveleged position would be further privileged by the democratic system was formal and transparent: for example, you had to have certain economic interests to be able to vote, or if you were the member of a certain group of aristocratic families, you had a seat in the Upper House for life. The universal franchise was, to me, a strike for democratic equality. Everyone could vote, and everyone’s vote had equal weight. My concern with the idea that anyone can do and spend anything to convince anyone else as to how they should vote is that you return to a system – albeit in a less formalised way – in which those people and organisations with a great deal of resources are much better able to have their interests protected by the political system than those without. That feels like “mischief” to me – though I concede that as soon as you regulate campaign finance you raise just issues related to freedom of speech just as problematic to resolve.

  6. stephenwhittington Says:

    I don’t buy the figures that suggest so many people are genuinely undecided. If people were not that sure, I hope many of them would be unwilling to vote. I suggest that people who are undecided are often trying to signal that they’re open minded.

    You’re right that this theory does not seem consistent with the way politicians behave…if you assume that politicians have better ways of convincing people. You haven’t responded to that point.

    I also think that it is highly likely that there is a difference between political advertising and commercial advertising. Why? I think that we are much more likely to vote to signal certain things about our beliefs. Because the costs of making a bad political decision are low, we are able to indulge such biases more easily that we are in the commercial area. Ignoring the advertised benefits of airbags will have more detrimental impact on our lives as a matter of expected value than ignoring the socialist calculation debate and voting Communist.

    In other words, politics is not about policy: http://www.overcomingbias.com/2008/09/politics-isnt-a.html

  7. stephenwhittington Says:

    And, I think part of Jesse’s point is that political power may arise by virtue of being attractive, having a firm sounding voice, or being charismatic. By your logic, surely this is mischief too?

  8. Michael Appleton Says:

    Stephen: Your argument that very few voters – and certainly not enough to sway elections – are “convinceable” during political campaigns is not consistent with a whole lot of polling and other social science research. Your only evidence for your assertion seems to be instinct. To take but two examples, the IPSOS exit poll before the last UK general election found 34% of voters making up their mind in the final month before election day; 39% of voters in the 2008 US Presidential election claimed to have decided who to vote for between the conventions and election day. It is also usually possible to see a correlation between the the ups and downs of a political campaign’s polling and good and bad events that have happened on the campaign trail – as “convinceable” voters are swayed back and forth. If you have witnessed focus groups, you will also know that “lines” pushed by campaigns through paid and unpaid media often stick in voters’ minds, and play a significant role in swaying undecideds.

    I grant you (and Jesse) that there are lots of things, other than money, that help some candidates over others – including hairstyles (though John Edwards’ very expensive haircuts only got him so far). And of course it’s not possible to create a completely level playing field, on which the candidate with the most persuasive policy positions will always win. But that doesn’t mean the state should avoid any attempt to tackle those disparaties which are most harmful, and the easiest to counteract.

    (ps i don’t think your notification emails are working…)

  9. Causation and correlation « Today's Dissent Says:

    […] and correlation By Jonathan In response to Stephen’s post, The Mischief Rule, Michael points in a comment (by the way thanks for reading and commenting)  to then Senator […]

  10. stephenwhittington Says:

    Michael, let’s be clear about where we are:

    1) You have not responded to any of the three reasons why politicians spend so much, when it is so ineffecitve. Arguably you have responded to the claim of politician stupidity by pointing out that is unlikely.

    2) I provided a further reason why studies of voter indeterminacy are suspect. You said this was not backed up by the polling (the pollling which I was questioning).

    3) You have moved from your initial position of saying that limiting contributions is good, to arguing that in a perfect world the Government would centrally determine potential politicians’ haircuts.

    If you want to know the relationship between spending and outcome, look at the studies.

    Or, put another way, what evidence would convince you?

  11. Michael Appleton Says:

    Stephen,

    Let me go through your comment point-for-point so I can’t be accused again of ignoring arguments you’re making!

    “1) You have not responded to any of the three reasons why politicians spend so much, when it is so ineffecitve. Arguably you have responded to the claim of politician stupidity by pointing out that is unlikely.”

    The first reason – that politicians are dumb and wasting their money – is “somewhat unlikely” you said, so let us leave that to one side for now.

    The second reason – that party institutions reward those who raise lots of money more than those who do not, or politicians are seeking to set up post-politics jobs – seems implausible to me, based on my personal experience of watching and talking to politicians who fundraise, and reading about the process. The vast majority hate it, and would avoid it if they felt they could do so without hurting their electoral fortunes. The common expressions used by politicians to describe fundraising is “degrading” and “like begging”. And post-politics job prospects would seem to be more likely be enhanced by political figures using their positions to get to know prominent people in the field they wish to work in than through an exchange of money. But I grant you that in certain situations politicians benefit institutionally from their ability to raise money – and this can occasionally explain their fundraising. (An obvious example is some of National’s financial backers announcing, through the National Business Review, that their money would dry up if Bill English were not replaced by Don Brash.)

    The third reason which you call “the most likely explanation”, is that “there simply may not be all that much a politician can do to convince voters, so even though spending money has such a little impact, it is the best strategy amongst the alternatives.”
    I am willing to accept this explanation – that spending money is the best of a lot of bad options to try and convince people. But that admission of yours – that spending money on campaigning is one of the best tools that politicians have to persuade (even if it’s the best of a bad bunch) is all I am really trying to argue. My argument is basically: 1) Some voters are persuadable during a campaigning (though the majority – perhaps the vast majority – are not for other, structural reasons); 2) In some elections, this pool of voters will determine the result; 3) Campaigning costs money; 4) In an ideal world, parties/candidates’ abilities to persuade this pool of voters would not be significantly determined by how much money they have; 5) I would be concerned if campaign finance laws did not seek to acknowledge and regulate to some extent campaign spending in recognition of 1)-4).

    “2) I provided a further reason why studies of voter indeterminacy are suspect. You said this was not backed up by the polling (the pollling which I was questioning).”
    Your reason was: “I suggest that people who are undecided are often trying to signal that they’re open minded.” So what you’re basically saying is this: Although politicians believe and act as if some voters can be persuaded through campaigning; although many voters believe and act as if they are persuadable through campaigning; although every focus group ever conducted has shown that some voters make their decisions based on campaign activities; and although everything we know about public opinion research tells us that voters can be swayed back and forth during an election campaign, you reckon people are just trying to strike an open-minded pose? But on what basis do you think that? Alas, but I’m not sure many voters are as contrarian as you and I – and that seems a throughly unconvincing explanation for voter behaviour during campaigns. Indeed, if this were all just posturing, it would be possible to predict elections before the formal campaign period, based on underlying structural factors (incumbent approval ratings, unemployment rates and the like) – but in fact sometimes the campaign does matter.

    “3) You have moved from your initial position of saying that limiting contributions is good, to arguing that in a perfect world the Government would centrally determine potential politicians’ haircuts.”
    Actually, I was just making a joke about John Edwards – his several-hundred dollar haircuts were declared campaign expenses. Perhaps with more stringent limits, he would not have spent this money?

    “4) If you want to know the relationship between spending and outcome, look at the studies. Or, put another way, what evidence would convince you?”
    I agree that we’re both hamstrung, because there are so many confounding factors here and the leap from correlation to causation is a difficult one to make – and no two campaigns are exactly the same, just with different spending levels (despite Levitt’s attempts to simulate them).

    You asked me how I would be convinced. I would be convinced by politicians who started acting as if money for campaigning didn’t matter. I’d be convinced by minor parties with next to no money (perhaps, say, because their core supporters are all poor) were as successful as minor parties with much more money (perhaps, say, because their core supporters tended to be wealthy). I would be convinced bu political parties spending no money on polling and focus grouping because they understood that campaigning didn’t really matter that much. And I’d be convinced by public opinion research which demonstrated your contention that virtually no-one is persuadable during the campaign period.

    What would it take to convince you? Someone coming to you with a campaign brochure, waving it around, and saying, “Stephen, this has changed my mind!”

    Okay, I have just read Jono’s post, and the many others above. I didn’t realise you had made a submission on the electoral finance legislation! Sounds like I need to cut my losses 🙂

    Again, cool website…

  12. Alex Says:

    Cool discussion.

    I’m not satisfied with Michael’s response to Steve’s point 3).

    It seems like in Micahel’s fourth stage of argument, ” In an ideal world, parties/candidates’ abilities to persuade this pool of voters would not be significantly determined by how much money they have”, he only cares about money altering voting behaviour.

    However 1) and 2) make it seem like the mischief to be tackled is simply that of some elections being determined by factors to which we are not all equally endowed – money.

    But why just money? Why not skill in journalism – or public speech? Haircuts seem easy to dismiss. But earlier Michael says he would like the state to attempt to tackle disparities which are most harmful and easily addressed. It doesn’t seem obvious at all that campaign spending is more harmful than the public speaking or writing skill of a particularly vocal political commenter or party proponent.

    Then if the harm from these disparities is high Michael ought to want to equalise them “in a perfect world” if it were easy enough to do?

    I find the case Michael makes quite convincing but I am worried because I cannot see why the logic invoked ought to stop at just money.

  13. Michael Appleton Says:

    Alex: Well, when deciding to regulate something that is tilting the political playing field, presumably you would ask such questions as:
    1) Is this disparity something that is outside the control of the players in the political contest? I would argue that fundraising disparities are, to some degree, because of who has money and what sort of policies tend to attract money. Things like haircuts and the way people talk and present themselves are within the control of candidates/parties to a much greater extent – and thus not in the same need of regulatory remedy.

    2) How intrusive is this regulation on a candidate/political party? I don’t consider a particular monetary limit to be that intrusive – so long as the limit provides for parties to spend sufficient resources to get information out to the public. I would consider forcing politicians to look and speak the same to be more intrusive, by several degrees of magnitude.

    But, to be honest, I have only really considered campaign spending – because that’s what Stephen’s post was about. If you have other ideas for regulated campaigning activity, I am all ears…

  14. Alex Says:

    Political commentators may be quite outside the control of players inside the political contest. At least some may plausibly be as far removed as many big campaign donors.

    Intrusiveness may be a good way to limit the application of your logic to money but not to other disparities outside political candidates control such as journalistic skill.

    I don’t seriously want to convince anyone that these other disparities ought to be regulated – I was simply highlighting that if the logic that leads us to want to regulate spending also makes us want to regulate other disparities such as charisma or education of non-party affiliated candidates then we may want to question our logic.

  15. stephenwhittington Says:

    I’ll just reply briefly to Michael.

    1) We can probably agree on the first point.

    2) I think you are putting too much weight on what is said and written. If I was a politician who enjoyed fundraising, I would write that I hated it, if only to seem like I opposed special interests.

    3) Interesting, and I accept the point you make. However, at what point is something both the most effective mechanism to convince people, but also so absolutely ineffective as to warrant ignoring it? That is essentially my reading of the evidence. Yes, it is a relatively good way to convince people, but a very tiny majority are convinced. In that case, and given all the other number of shallow voters out there, why bother?

    4) Seldom am I on the wrong end of a revealed preference argument, but essentially, I think many of those things (especially the focus group and polling aspects), are more likely to be used for signalling. It is specifically because the instrumentality of voting is so low than increases its likelihood of being used to signal (and also decreases its value as a signal, as it is so cheap).

    5) Let’s look at your most interesting point:

    “You asked me how I would be convinced. I would be convinced by politicians who started acting as if money for campaigning didn’t matter. I’d be convinced by minor parties with next to no money (perhaps, say, because their core supporters are all poor) were as successful as minor parties with much more money (perhaps, say, because their core supporters tended to be wealthy). I would be convinced bu political parties spending no money on polling and focus grouping because they understood that campaigning didn’t really matter that much. And I’d be convinced by public opinion research which demonstrated your contention that virtually no-one is persuadable during the campaign period.”

    The problem with the first piece of evidence is that I do not believe money doesn’t matter – I believe it matters very little, but compared to alternatives matters more. The problem with the second is that most people in the Western World are not poor. That is why all political parties with support – be it from poor or wealthy communities – are able to raise money. The problem with the third piece of evidence is that I believe polling and focus groups make total sense, because the best way to win votes is to move to the centre. On the issue of persuadability, I belive the revealed preference is with me on it. What percentage of “undecideds” two weeks out end up voting?

  16. How much money was spent by electoral candidates in 2008? « Today's Dissent Says:

    […] and Michael are having in the comments section in response to a one of Stephen’s earlier posts, Michael comments that he would be more convinced that spending plays a limited role in elections […]

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: