Conveniently, Thoma's article provides some opportunities to juxtapose the influences of approach, ideology, and evidence on our beliefs. In the first paragraph of Thoma's piece he says:
Economists who work for businesses also have a tendency to present evidence more like a lawyer advocating a particular position than a scientist trying to find out how the economy really works.This, I think, is a confusion of a difference of approach for a difference of ideology. I would wager that very few professional economists start from a position of advocacy in forming their beliefs about how the world works. Instead, many economists Thoma disagrees with simply take a different approach. They, in fact, take very seriously the notion that we should be "find[ing] out how the economy really works." A previous column by Thoma purportedly laid out the case against free markets, leaning heavily on what Harold Demsetz called the "Nirvana Approach." I responded to Thoma's earlier column here, but the gist is that the Nirvana Approach compares the real world to an theoretically-contrived ideal. Demsetz argued that a "Comparative Institutions" approach was better for comparing two real-world systems. (This is not to say that theoretical abstraction is itself bad, but that it can be misapplied.) In my experience, many economists whose beliefs line up with Thoma's don't heed Demsetz's warning about the proper approach to take, whereas those who disagree with Thoma take Demsetz's critique to heart. (You can read some critiques of the Nirvana Approach in agricultural economics here.)
Thoma goes on to make the case that evidence is really what drives economists beliefs about the world. I agree that evidence is a big factor, but his evidence is lacking. He cites a survey of economists on two questions. Here's the first:
Question A: Because of the American Recovery and Reinvestment Act of 2009, the U.S. unemployment rate was lower at the end of 2010 than it would have been without the stimulus bill.It's simply not possible for evidence to be the driving factor for economists' beliefs about this question. The problem lies in this phrase "than it would have been." We don't know what would have happened without the stimulus bill. Macroeconomics doesn't lend itself to controlled experiments. So all that is left are ideology and approach. Again, I think this is about approach. Economists who see the world through the lense of Keynesian demand management will tend to hold the view that the stimulus was necessary and helpful. It may have something to do with their ideology but, strictly speaking, it has nothing to do with the evidence itself. Thoma asserts that the economists in the survey represent a wide range of political views, but I highly doubt that this small pool of economists located primarily in the northeast represent anything close the diversity of political perspectives we find in the U.S.
Another couple of quotes will help drive home the point:
Ideology certainly influences which questions academic researchers believe are the most important, but there is nothing wrong with that. What’s essential is that they follow the evidence wherever it might lead. I have no doubt that, despite a few bad apples, the vast majority of economists honor the evidence over their prior beliefs when the evidence speaks clearly on an issue.I agree, but we can't leave out approach. Without, say, the Public Choice Revolution, the contributions economists have made to our understanding of political institutions simply wouldn't exist. Even though many Public Choice economists have similar views about the markets vs government question, their approach is at least as important as their ideology for determining the questions they study.
Our reputation with the public is bad enough as it is, and our best hope of changing that is to be honest about the evidence, and follow it wherever it might take us.The problem here is that the way we interpret the evidence is dependent upon the way we approach the question. It's not just about high quality data and advanced techniques. Because our approach (or interpretive framework) determines the way we perceive the evidence, data itself can't be used to overturn a given approach.