30 Aug 2017

If Nobody Gets My Analogies, Whose Fault Is It?

Scott Sumner 17 Comments

This post is really only directed at those of you who are faithful readers of Scott Sumner. If you don’t follow him, obviously you can feel free to chime in, but I don’t expect you to get the point here.

Anyway, Scott recently criticized David Glasner’s definition of “currency manipulation” because it relied on the intentions of the policymakers, and (according to Scott) we as economists can’t be mind readers. According to Scott, any concept that requires an analyst to get inside the mind of the government official is nonsense.

I thought this was a very surprising move for Scott to take. So I left this comment:


On this post, I’m not trolling you. (In contrast, I *am* going to troll you a bit on your super-neutrality post.)

David is making an obvious point. If a government is intentionally lowering the real exchange value of its currency in order to gain an export advantage, then it is engaging in currency manipulation. A la your point about Bastiat, we don’t need to *care*, but the concept is coherent.

To push back against that, you write:

Surely if a concept of currency manipulation has any coherent meaning, it cannot depend on the motive of the policymakers in a particular country? We aren’t mind readers.

Well, what about this claim?

“Surely if a concept of tight money has any coherent meaning, it cannot depend on the mental framework of the policymakers. We aren’t mind readers.”

Do you agree with that? For example, if Fed officials take some actions during the day and we see interest rates go up, surely that’s all we need to know if we’re going to classify it as “tight” or “loose” money, right? Somebody might suggest that it depends what the motivations were–maybe the Fed officials didn’t even care about interest rates, and were just trying to get NGDP growth to be according to their subjective target–but you would say that’s nonsense because we can’t read minds?

I’m truly not being facetious here, Scott. If you had asked me (before I saw your current post here about Glasner) which monetary economist stressed the importance of evaluating policy in terms of the intentions of the policymakers, I honestly would have said, “Scott Sumner has made me appreciate this more than anybody else.”

Then, to be kind of a wiseguy but also to isolate just how silly Scott’s stance was, I left another comment:

David Glasner: In order to classify Dick Cheney as an attempted murderer, we need to know that he intended to shoot his hunting buddy.

Scott Sumner: What are we, mind readers? I can’t believe the legal concept of “homicide” involves intentions. This is nutty. What if a tree fell over and killed his buddy, would the tree be a murderer now too?

In my mind, I won the internet. Sure, Scott might push back and say my cutesy analogy leaves stuff out, but surely he would get what my point was, right?

Nope, not at all. Scott had no idea what I was talking about, and this other guy accused me of making a freshman error regarding monetary policy.

If somebody can tell me, “Ah, OK Bob I get what you were trying to do, but here’s why nobody else did…” I am all ears. I honestly get baffled when stuff like this happens. To be clear, I’m not expecting people to say, “Drat! We would’ve gotten away with it too, if it hadn’t been for your awesome analogy,” but I expect people to get the point I’m making. That’s the whole rationale for using analogies.

17 Responses to “If Nobody Gets My Analogies, Whose Fault Is It?”

  1. E. Harding says:

    I’m with Sumner here. Murphy, your analogies only work half the time, and this ain’t one of them.

    • Bob Murphy says:

      OK because you have no clue what I’m talking about, or because you get what I meant, and think I failed to hit the mark?

      • E. Harding says:

        A country can attempt to manipulate one’s currency while failing to do so. A country can manipulate one’s currency while not intending to do so. Regardless of intentions, the effects on U.S. trade remain the same. Thus, intentions should not matter to policymakers when deciding whether to label a country as a currency manipulator.

        Of course, it’s true that your (Bob’s) definition of currency manipulation would make sense if it was explicit, market-understood policy, but it rarely is.

    • Stephen Dedalus says:

      No: stupid people only get half of Bob’s analogies, and this is one of the times they didn’t.

    • Craw says:

      Why not? The point is that any concept which is inherently about instrumentality is inherently about intent. “Currency manipulation ” might in fact be impossible, but the idea one might try to affect currency markets to indirectly achieve another aim involves intent, and is a coherent concept. The question whether Cheney attempted to murder his companion likewise involves considering Cheney’s intent. It’s likewise a coherent concept.

  2. DesolationJones says:

    Regarding you second comment, do you believe intentions matters enough in every single type of action in the world that we should have unique words and phrases for every action depending on the intent? Like should we have a unique word/phrase for randomly throwing a feather that accidentally hits another person, and another phrase/word for throwing a feather with the intent of it hitting someone. Maybe we’d have a more complex and richer language, but is it useful to have it?

    Sumner thinks definitions should be useful to have any logical meaning. Your analogy implies that Sumner is arguing intent never matters. I don’t think that was his point. My guess is he thinks intent does matter in certain actions (homicide being one of them), but in this case, he just thinks intent does not add anything useful to whatever “currency manipulation” means for various reasons he outline in his post.

  3. Tel says:

    Anything other than open slather, hands off, free banking is currency manipulation… pretty much by definition.

    If government takes any action whatsoever that has some effect on currency valuation, then it has manipulated the price of the currency. No different to my hand holding a jug of beer… if the beer raises or lowers as a result of my action, then I’m manipulating the beer.

    I’m sure that’s not the way either Trump or Scott Sumner see it… but I thought I should stake out a position.

    Now onto the business of being a “mind reader”… well suppose a husband and wife were overheard loudly arguing and screaming at one another, then the husband runs over the wife in the car… that’s known during the trial as evidence. Similarly, let’s suppose China announces that their policymakers have made a policy to peg their currency to the US dollar… think about it, one currency would not automatically remain pegged to another just by random market forces, of course this is an open statement that they intend to manipulate! No mind reading required, because they announced as much.

    Mind you, pegging your currency to gold is also manipulation, by much the same logic.

  4. RPLong says:

    Full disclosure: I almost never agree with Scott Sumner, so my comment is not coming from that direction.

    I find myself in a bit of a pickle here, because I think I agree with both of the following statements:


    If a government is intentionally lowering the real exchange value of its currency in order to gain an export advantage, then it is engaging in currency manipulation. A la your point about Bastiat, we don’t need to *care*, but the concept is coherent.

    And this:

    For example, if Fed officials take some actions during the day and we see interest rates go up, surely that’s all we need to know if we’re going to classify it as “tight” or “loose” money, right?

    But I’m not sure if this is because the analogy you’re making doesn’t work, or because I’m weird and I allow for both possibilities. I think it’s possible that a policymaker could attempt to engage in currency manipulation, but for that attempt to fail. (Maybe the policymaker chose an impotent policy, for example.)

    I think motives are important, but I think when it comes to analyzing whether a policy was expansionary or contractionary (or neutral), we ought only be concerned with the effects, not with the motives. To me, it’s two different questions, although both are important.

    • Craw says:

      The effects of the policy might not correspond to the observed change in the world. Bob’s boat sank, but that doesn’t mean he wasn’t bailing. And I don’t conclude that Bob’s bailing was a pro-sinking policy.

      • RPLong says:

        My problem with that objection is that it assumes an analytical blind spot. If Scott Sumner fails to put all the relevant variables in his model, then that’s the problem. The problem isn’t that his benchmark for analysis is wrong, only that he’s missing something.

        But, assuming all the relevant factors are in the model, then the effects of the policy should be clear: The policy might not be as effective as intended, but the effects of the policy don’t change just because some other factor was too large to be counteracted.

        So, is Bob’s claim that Sumner fails to put everything relevant into his model and then suggests that expansionary policies were really contractionary?

  5. Stephen Dedalus says:

    Scott’s “we aren’t mind readers” is absurd: as highly social creatures, we are mind readers par excellence. One of the very first things we learn to do is read our parents’ minds!

    • Tel says:

      All communication would be impossible unless we had some kind of shared experience and a plausible idea of how the other guy is going to interpret the message.

  6. Andrew Keen says:


    I understand your analogy up until the last sentence:

    What if a tree fell over and killed his buddy, would the tree be a murderer now too?

    What is this in reference to? I don’t get how this part of the analogy aligns with what Scott is saying.

    • Craw says:

      If a tree falls and kills the only person in the woods, is it a silent killer?

    • Craw says:

      Is intent part of the definition of murder? If we cannot make sense of ideas that involve intent, because we aren’t mind readers, then how come you know the tree isn’t a murderer?

  7. Transformer says:

    I think the problem with the analogy is that Scott was not saying that looking for intentions in any field is bad but specifically within the field of economics.

    If Scott had written a post arguing that drinking beer while jogging is a bad idea, it would not be much of a rebuttal to reply with an analogy where you imagine Scott arguing that drinking beer while in a pub is a bad idea.

  8. Giovanni says:

    I use analogies and reductio ad absurdum every day when I try to explain myself to other people. I honestly think everything makes sense and is totally correct and understandable most of the times, but almost never people understand my analogies. Most of the times I hear back that my analogy has nothing to do with the actual case we were talking about.

Leave a Reply