Robert Wenzel expresses incredulity at my recent post on campaign expenditures. (He doesn’t indicate whether he read any of the papers I linked to.)

Wenzel writes:

Aside from the question of whether empirical studies are the correct method for considering this subject, Brennan takes the ambiguous results and reaches an unambiguous conclusion:

My reading of that lit…is that campaign spending matters much less than most people think…

Wow.

Well, let’s think through that. When political scientists repeatedly try to test the thesis that A causes B, and do so from many different angles using many different data sets, but repeatedly fail to find evidence that A causes B, then this at some point justifies us in believing that A does not cause B.

Wenzel says (I’m paraphrasing here): If campaign expenditures make so little difference, then you must think Soros and Charles Koch are idiots for spending money! But they’re clearly not idiots.*

This kind of incredulity applies to many other findings in political science. Here, read Hans Noel’s outreach piece 10 Things Political Scientists Know that You Don’t. Now, consider:

1. It seems obvious to us that campaigns matter a huge amount. Politicians spend a huge amount of time campaigning, and the media covers these campaigns incessantly. And, yet, when political scientists try to measure what effect campaigns actually have, they routinely find little effect.*
2. It seems obvious to us that people vote their pocketbooks. But when political scientists try to measure how people vote, they find that material self-interest has little predictive power. Mitt Romney made his infamous “47%” remark on the basis of bad information.
3. It seems obvious to us that people have consistent ideologies. Yet when political scientists try to measure political belief, they find that nearly half the country has no real fixed ideology, and instead tends to vacillate between ideas.
4. It seems obvious that politicians who win by a large margin enjoy a mandate that increases their effectiveness in office. And, yet, when political scientists can’t find much evidence of any mandates. The question “Why are political scientists skeptical of the mandate theory of elections?” even finds its way onto the AP Government exam.
5. It seems obvious that Nader cost Gore the 2000 election and that Perot cost Bush I the 1992 election. Yet, when political scientists study this, they find strong evidence that most Nader voters wouldn’t have voted for Gore, but would have stayed home or voted third party. Most Perot voters would have stayed home or were split roughly evenly between Bush I and Clinton.
6. It seems obvious that many people are political independents. But, in fact, most independents vote the same way every time they do vote.
7. It seems obvious that Democrats are better informed than Republicans. But, in fact, highly ideological Republicans do better on civics tests than highly ideological Democrats. Moderately ideologically Republicans do better than moderately ideological Democrats. Weak Republicans do better than Weak Democrats. These effects persist even when we correct for income and education.
8. It seems obvious that if I don’t vote, my interests will be ignored. But then it turns out that the expected values of voting for my least favored candidate and my most favored candidate are approximately the same.
9. It seems obvious that evil plutocrat George W. Bush was more likely than Great Society Lyndon Johnson to side with the expressed political preferences of the 90th income percentile than with the 10th percentile. But in fact it goes the other way.
10. It seems obvious that compulsory voting will tend to benefit left-wing political parties. But, in fact, it appears to have little partisan effect, except that it helps far right wing parties gain a seat or two in proportional representation parliamentary systems.
11. It seems obvious that deliberation makes people smarter and nicer. But, in fact, it tends to reinforce bias and leads to polarization.
12. It seems obvious that perfectly ignorant voters vote randomly. But, in fact, people have a “position bias,” and, all things equal, are much more likely to pick the items at the top or bottom of a list than the middle.
13. It seems obvious that ignorant and well-informed Democrats are a united front. Same with ignorant and well-informed Republicans. But, in fact, well-informed Democrats have systematic ideological disagreements with poorly-informed Democrats. Same mutatis mutandis for Republicans.

And so on. A lot of what people think they know just isn’t so, as far as we can tell. Now, one reason why there is such a vast literature on most of these topics is that political scientists themselves react with incredulity. “We tried to show voters vote their self-interest. But this data set says they don’t. That can’t be right. Let’s do another study.” 100 studies later, you’ve got 100 studies failing to show that voters vote their self-interest. And then political scientists have to start looking for alternative explanations.

*A further point on campaigns and money: If you don’t campaign at all, you aren’t likely to win. But that doesn’t mean the best campaigner wins. Similarly, it does in fact take a huge amount of money to win a presidential election. But that doesn’t mean that the person who spends the most wins because he spends the most. A good analogy for this is football. If you are going to be an NFL center, you need to weigh over 290 pounds. But that doesn’t mean that the heavier you are, the more likely you are to be an NFL center. Nor does it mean the heaviest person gets the job. Most Americans who weigh over 290 pounds have no shot at all at being an NFL center. Money gets you in the game in politics, but it doesn’t win the game for you, as far as political scientists have been able to show.

Two more points:

1. Yes, corporations, unions, etc., buy power. But distinguish post-election lobbying from campaign finance. There’s little evidence that the latter buys power, but much more that the former does. Even in the former case, politicians are much cleaner than most people, especially most libertarians, believe them to be. It may be that lobbyists win not because they spend more, but because they show up. And the only people who show up are people who want the government to spend money on their pet projects. (After all, the benefits of these projects are concentrated, and the costs are diffused.)
2. If you think that the findings I’m reporting above are just stupid, awesome! I await your article forthcoming in APSR showing that of course what we think is obvious really is true, and that David Sears and Tali Mendelberg and Martin Gilens and Stephen Ansalobehere are just fools who can’t figure out how to prove the obvious.

UPDATE: I’m curious what alternative method Wenzel prefers for “considering the subject”. As far as I can tell from reading Economic Policy Journal, his favored method is “Make up my own facts.”

UPDATE 2: I frequently see people from all ideologies, from laypeople to prominent philosophical bloggers, make the following move. “The social sciences suck! They lack proper experiments and controls! [Insert various strawman criticisms.] Therefore, I am free to believe whatever I wish about politics, and I will continue to believe whatever causal claims best support the ideology I developed without any evidence at age 15.” Nope. Even if all your criticisms of the social sciences were correct, what this justifies is agnosticism.