Does that make me crazy?

Chomsky goes out of his way to claim that the analysis he presents is not a conspiracy theory. As knowledgeable as he is about propaganda, he must know that by responding to the argument, he gives it some weight. It makes one wonder why people consider his arguments conspiratorial, (in addition to the conspiratorial tone of his writing). I have a theory. Some of his followers may sound more paranoid than he is. This paints him with same brush as crazy.

For example, I’ve been having a discussion with an esteemed colleague (who happens to be a fan of Chomsky) on the merits of free trade, and the validity of Ricardo’s theory in light of what he claims is historical data supporting the protectionist side. To this point in the discussion I don’t believe I had used anything more complicated than Ricardo’s idea of competitive advantage as a positive argument. He writes

“[W]hat do you think is easier 1 – developing a theory that serves your interests or 2 – developing a theory that serves your interests and selecting data that supports. It seems to me that 2 would require more work and make the theory more scrutible, making fraud easier to detect.”

Now, Ricardo’s ideas are essentially the Newtonian system of mechanics, as far as trade goes. They are based on very simple ideas and can be easily understood by college freshman. There may come a Michaelson and Morley who can demonstrate that trade doesn’t generate any benefits near the speed of light, or a Schroedinger who shows that it isn’t true at small scales or near a cat that may or may not be dead. But it seems to work well in many situations. Somehow Ricardo’s ideas are being cast as ideas that serve his interests? Or possibly the claim is that the vast majority of economists who claim to believe that removing tariffs has benefits that outweigh the costs are heavily invested in shipping companies?

I don’t know the total holdings of current economists who tout trade, but Ricardo was wealthy by the time he developed his theory, so he didn’t do it for the money. The non-pecuniary benefits would include respect from peers, but he wouldn’t gain this from propagating an obviously wrong theory. And my colleague seems to be saying that Ricardo was publishing this theory deviously, fraudulently possibly.

Maybe this argument was simply not fleshed out since it wasn’t written for Econometrica this time. But it would take several rather large leaps to believing that someone serving their interest by subscribing to comparative advantage. Or is that just what I want you to think?

Advertisements

4 Responses to “Does that make me crazy?”

  1. Bill Says:

    I thought it was obvious that I was critiquing your statement:

    “The reasons I tend to discount historical data (usually without even seeing it) …”

    As far as I know, Ricardo never made such a claim. I think it is clear that the statement is unscientific.

    Anyway, Ricardo’s argument is valid provided that his background assumptions hold. The problem I have with the your use of his argument is 1) you extend the conclusion beyond his original argument without providing any additional argumentation and 2) you don’t account for when and where the background assumptions hold. If one want to extend the knowledge of the theory, it seems like examining historical data could provide insights into what the underlying assumptions are.

    Calling Ricardo’s argument and its successor the Heckscher–Ohlin Samuelson model seems like a real stretch. I certainly don’t think that Newtonian theory is criticized for it lack of predictive power like the HOS is (http://en.wikipedia.org/wiki/Heckscher%E2%80%93Ohlin_model). I would think that game theory would be a better candidate for the title of Newtonian equivalent, although that would be a stretch too.

    I made didn’t claim that Ricardo’s theory were developed to serve his interests. My question is a reaction to your statement. It really seems that you want to be able to hold to your theory despite any evidence provided. It seems like a way to insulate your position from evidence. Would it matter to you if the position you hold may be based upon theory’s where the background assumptions don’t hold?

    So Ricardo’s wealth doesn’t seem relevant to any argument I made. It would be relevant if he held to his theory despite contrary data, which I don’t think he had at the time. If someone hold a position despite contrary data, then I think it is fair to question if they have incentives that drive them to hold their position.

    So in answer to your question, I think your position on the use of historical data make you unscientific but not crazy. Your attributing arguments to people they don’t make on the other hand …

    • darfferrara Says:

      You’re reference wasn’t clear to me, and there were some verb troubles. Thanks for clearing up.

      I thought it was obvious that I was critiquing your statement:

      “The reasons I tend to discount historical data (usually without even seeing it) …”

      This statement was rather far removed from the claim you made, so I didn’t see the link. To be clear then, I am the one generating a theory that serves my interests, and you (or someone you read who was anti-trade) develop theories that serve your interest is the comparison to b e made?

      As far as I know, Ricardo never made such a claim. I think it is clear that the statement is unscientific.

      To claim that a thinking about a model without data is unscientific throws out quite a bit of science, from Galileo thinking about dropping a large object tethered to a smaller object to the thought experiments of Einstein and beyond.

      Anyway, Ricardo’s argument is valid provided that his background assumptions hold. The problem I have with the your use of his argument is 1) you extend the conclusion beyond his original argument without providing any additional argumentation and 2) you don’t account for when and where the background assumptions hold. If one want to extend the knowledge of the theory, it seems like examining historical data could provide insights into what the underlying assumptions are.

      Calling Ricardo’s argument and its successor the Heckscher-Ohlin Samuelson model seems like a real stretch. I certainly don’t think that Newtonian theory is criticized for it lack of predictive power like the HOS is (http://en.wikipedia.org/wiki/Heckscher%E2%80%93Ohlin_model). I would think that game theory would be a better candidate for the title of Newtonian equivalent, although that would be a stretch too

      .

      Let me say that I frame the argument in terms of trade. Not international trade necessarily, just trade. And comparative advantage has a huge amount of data that supports it. Did you grow your own breakfast this morning, or did you purchase it from the store with money you earned from a job that you can do better than 99.9% of the population? If you did the latter, you are seeing the benefits of comparative advantage.

      The fact that a particular numerical model doesn’t match real world data doesn’t concern me any more than the fact that an object dropped out of an airplane takes much longer than predicted by Newtons laws to hit the ground. “But there was air resistance, obviously this wasn’t a pure experiment!”, you cry. Well, yes. But that’s the point. Economics rarely has pure experiments, which is why I would tend to discount heavily claims that “Ricardo was proven wrong”.

      I made didn’t claim that Ricardo’s theory were developed to serve his interests. My question is a reaction to your statement. It really seems that you want to be able to hold to your theory despite any evidence provided. It seems like a way to insulate your position from evidence. Would it matter to you if the position you hold may be based upon theory’s where the background assumptions don’t hold?

      Luckily Don Boudreaux has answered this today.

      So Ricardo’s wealth doesn’t seem relevant to any argument I made. It would be relevant if he held to his theory despite contrary data, which I don’t think he had at the time. If someone hold a position despite contrary data, then I think it is fair to question if they have incentives that drive them to hold their position.

      So in answer to your question, I think your position on the use of historical data make you unscientific but not crazy. Your attributing arguments to people they don’t make on the other hand …

  2. Bill Says:

    There is a lot to you post I should address, but this statement :

    “To claim that a thinking about a model without data is unscientific throws out quite a bit of science, from Galileo thinking about dropping a large object tethered to a smaller object to the thought experiments of Einstein and beyond.”

    really seems to miss the point.

    The scientific method is usually represented as some variation of the following steps:
    1. Ask a question.
    2. Do some research
    3. Form a hypothesis/ predictive model and predict
    4. Gather data and test the hypothesis
    5. Analyse the data/update hypothesis
    Bayesians may add the qualifier to step 4 that the hypothesis must be compared to another hypothesis to be meaningful. However, my point is that you seem to be rejecting at least step 4 when you say.

    Note that Galileo made the argued for the importance of testing hypothesis in his book “Two New Sciences.” Einstein updated his theory in due to evidence presented by Georges Lemaitre. Karl Poppers great insight into the philosophy of science is that the possibility of falsifiability a necessary component of science.

    So I don’t see how my criticism that your “discounting historical data” discards science. Note I am not criticizing scientists who primarily work on step 3. Nor am I criticizing mathematicians who make no claim that there work applies to the real world.

    I can understand why you would be unwilling to give up on a hypothesis like HOS theory, if there weren’t competing models that better predicted that data. However, given that there are other hypotheses, shouldn’t the question be which model makes the best sense of the data? Further even if one hypothesis better fits the data, if the model has poor predictive power. One should be reluctant to make unqualified recommendations based on that model.

    I suppose this shouldn’t bother me, but it does. I still don’t see why you were confused about what referring to in my email.

    In your earlier post you make a statement followed by a list of supporting reasons. In response, I produced my own list and counter statement preceded by “The main topic of your email(post) is why you feel free to ignore historical data.” Followed by my list (which you quoted). The list was a point-by-point response to your list. I even mirrored your form of outlining.

    I still don’t see where I was unclear. (As I don’t see how my criticism would undermine scientific endeavors.

  3. darfferrara Says:

    At this point I’m not sure if we’re discussing the same thing. You keep talking about HOS. My claim (on the phone, I think) was that increasing tariffs would not improve allocation. You seem to believe this is wrong and that there are models that fit the data better than HOS. I’m not wedded to HOS, but what models do you have in mind, and what makes you think that increasing tariffs would improve allocation in these other models?

    Do you believe that Galileo’s thought experiment about tethering a smaller object to a larger one is science? I do. It really doesn’t lead to any sort of numerical prediction about how fast or slow something will fall. I don’t think it fits well into the Popper falsifiability scheme. But if someone claims that heavier objects fall faster than lighter ones, I wouldn’t need to look at empirical data. It has been refuted by mere thought.

    Economics doesn’t yield the certainty of mechanics, but comparative advantage is pretty close to the tethering example. While it is usually presented with simple numerical examples, there are several important unstated assumptions that go into it. Probably the most important of these have to do with individual preferences which manifest by the price mechanism.

    It looks like HOS sweeps this under the rug in order to get a clean model. Wikipedia lists nine major assumptions made by the model. These would be fine if it was predictive, but apparent it’s not. That doesn’t show that impediments to trade are helpful.

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: