08 Dec 2015

Scott Sumner, the Rortyian Historian

Scott Sumner 38 Comments

Scott Sumner has flirted with the notion that there is no such thing as truth, that “truth” is whatever your colleagues let you get away with. (I’ve dug up this post of him merely talking about it, but I’m pretty sure there are others where it looks like Scott is receptive to the idea.) And he has also written that he used to use arguments to support his position that he knew were a little bit suspect, but thought he could get away with, until the sharp commenters on his blog forced him to raise the bar. (If someone can find that post–I think it was on EconLog–I’d appreciate it. People will think I’m putting words in his mouth, but no, Scott really did say that.)

So in that spirit, it occurs to me that maybe Scott doesn’t think anyone will bother to go read the text of the recently passed (in the House) Fed Oversight and Modernization Act (FORM Act). Now if you click the link, under the Definitions in the beginning you will find:

(9) Reference Policy Rule

The term “Reference Policy Rule” means a calculation of the nominal Federal funds rate as equal to the sum of the following:

(A) The rate of inflation over the previous four quarters.
(B) One-half of the percentage deviation of the real GDP from an estimate of potential GDP.
(C) One-half of the difference between the rate of inflation over the previous four quarters and two percent.
(D) Two percent.

OK, so if you’ve studied economics, you know that the above is a version of the Taylor Rule, with equal weight placed on hitting the 2% price inflation target and in minimizing the output gap.

In contrast to the “Reference Policy Rule,” earlier in the text they define:

(2) Directive Policy Rule

The term “Directive Policy Rule” means a policy rule developed by the Federal Open Market Committee that meets the requirements of subsection (c) and that provides the basis for the Open Market Operations Directive.

OK, so the FORM Act lets the Fed come up with its own Policy Rule. However, the Fed can’t just issue any old rule, like, “We are going to try super duper hard to foster a good economy.” In the section of the bill explaining the requirements of an acceptable Policy Rule, it says:

(c) Requirements for a Directive Policy Rule

A Directive Policy Rule shall—
(1) identify the Policy Instrument the Directive Policy Rule is designed to target;
(2) describe the strategy or rule of the Federal Open Market Committee for the systematic quantitative adjustment of the Policy Instrument Target to respond to a change in the Intermediate Policy Inputs;

(5) describe the procedure for adjusting the supply of bank reserves to achieve the Policy Instrument Target;
(6) include a statement as to whether the Directive Policy Rule substantially conforms to the Reference Policy Rule and, if applicable—
(A) an explanation of the extent to which it departs from the Reference Policy Rule;
(B) a detailed justification for that departure;
and
(C) a description of the circumstances under which the Directive Policy Rule may be amended in the future; …

In the block quotation above, I left out (3) and (4) for brevity, but I retained (1), (2), and (5) so you’d get the gist of it. Look now at (6). It is saying that if the Fed’s proposed rule differs from the Taylor Rule (with equal weights on inflation and output gap), then the Fed needs to justify to Congress in detail why they are doing so.

Because of this item, Narayana Kocherlakota–who isn’t some punk journalist at Bloomberg, but is the president of the Minneapolis Fed–recently said in a speech: “The U.S. House recently passed a measure, the Fed Oversight Reform and Modernization Act, that would enshrine the Taylor Rule as a key benchmark for monetary policy.”

Pretty straightforward commentary on the FORM Act, right? I mean, if you get a memo from your boss saying, “Employees are allowed to propose their own lunch hour, but anyone proposing a bloc different from 11:30am – 12:30pm must provide Human Resources with a detailed justification for departing from this time frame,” then your co-worker would quite reasonably conclude, “They are setting 11:30am – 12:30pm as the benchmark for lunch breaks.”

But Scott Sumner likes the FORM Act–it’s a move in the direction toward a “discretion-free, let-the-markets-target-NGDP-growth” rule–and so he doesn’t want anybody criticizing it. So in response to Kockerlakota’s statement, Sumner wrote:

Unfortunately, Kocherlakota is flat out wrong about the recent House bill, it does not “enshrine the Taylor Rule as a key benchmark for monetary policy”. Not even close. It asks the Fed to come up with an explicit monetary rule. I suggest NGDPLT, target the forecast.

Notice that Sumner didn’t merely say, “I quibble with such a strong term as ‘key benchmark.'”

No, Sumner he said it’s “not even close,” and that all the bill does is ask the Fed to explicitly say what its rule is. If you didn’t go read the bill yourself, and relied on Sumner to tell you what it says, you would be horribly misinformed.

Now those of you who think Bernanke was too tight, go ahead in the comments and tell me, “Come on Bob, all Scott is saying is…” But when Krugman does stuff like this, and I call him out on it, most of you give me a high-five.

Last thing: Part of the reason I’m making this point is that I’m going to soon review Scott’s book on the Great Depression. I’m sure it will be a fine piece of scholarship, and will present all sorts of evidence that others were too busy/lazy to research. But I am expecting that I will often have to go dig up the original sources on some key issues, rather than trusting Scott’s paraphrase of what “it means.” This episode with Kocherlakota is a good example of why I feel I have to do that.

38 Responses to “Scott Sumner, the Rortyian Historian”

  1. Tel says:

    Sounds like off-the-shelf Keynesian “dual mandate” stuff to me… never works but other than that hardly controversial.

    I mean, if you get a memo from your boss saying, “Employees are allowed to propose their own lunch hour, but anyone proposing a bloc different from 11:30am – 12:30pm must provide Human Resources with a detailed justification for departing from this time frame,”

    Kind of similar, but 11:30 am is fairly well defined and can be repeatedly measured (give or take some government department making arbitrary changes to Daylight Savings) while things like “Output gap”, ” Potential output “, ” Natural rate of unemployment ” are meaningless concepts. There’s no theoretical justification for any linear trend growth, no way to measure how many people would be employed under some different economic conditions.

    All just made up seat of the pants stuff, and never will be more than that.

    • E. Harding says:

      Good point. The Fed could simply tinker with its estimates of potential GDP to arrive at its desired rate.

      • Major.Freedom says:

        You guys are both right, but to Bob’s point, the ability of the Fed to “fudge” potential GDP is I am sure you will agree limited enough to constrain the Fed to following a rule closer to the Taylor Rule set out in 9. (A) – (D), as compared to further away from it whereby the fudge factor makes the entire policy amendment moot.

        • Major.Freedom says:

          And even more to Bob’s point, Bob is right that this policy amendment is not “Fed, just pick ANY target and stick with it” as per Sumner’s interpretation, but rather “Fed, pick THIS target and stick to it.”

          Whether or not there is an unintentional escape hatch for the Fed to print as much as it wants by fudging potential GDP, does not turn Sumner’s interpretation into an accurate one.

  2. E. Harding says:

    I think Bernanke was too tight, and I agree with Bob here.

    “I’ve dug up this post of him merely talking about it, but I’m pretty sure there are others where it looks like Scott is receptive to the idea.”

    -Yup. Scott described Rorty as his favorite philosopher.

    • Major.Freedom says:

      My wife thinks the manufacturer of Prada handbags was too tight.

      You should advocate for the state to monopolize purses so that we never have to feel there is not enough ever again.

      • Major.Freedom says:

        I say this only to try to win an argument and “get away with it”.

  3. Transformer says:

    ‘if you get a memo from your boss saying, “Employees are allowed to propose their own lunch hour, but anyone proposing a bloc different from 11:30am – 12:30pm must provide Human Resources with a detailed justification for departing from this time frame then your co-worker would quite reasonably conclude, “They are setting 11:30am – 12:30pm and the benchmark for lunch breaks.” ‘

    If your boss’s memo said ‘We are reviewing our lunch hour policy to allow more flexibility as the current one has been criticized. If you have an idea to propose that differs from the current 12:20-1:30 rule then please provide Human Resources with a detailed justification for your suggestion:”, then you might think that at least your company is genuinely open to discussing a change.

    Is it possible that Sumner thinks that this is what this proposed act represents , rather than your view ?

    • Bob Murphy says:

      Transformer there is no wiggle room on this. The Act calls the Taylor Rule the “Reference Policy Rule” and Kocherlakota calls it a “key benchmark.” Sumner doesn’t just say, “Nah, it’s a reference, not a benchmark,” instead he doesn’t mention the “Reference Policy Rule” and “detailed justification” at all, but calls Kocherlakota’s summary “not even close.”

      Anyone relying on Sumner’s description would have no idea there was even a “Reference Policy Rule” in there.

    • Bob Murphy says:

      Let me answer you this way, Transformer: Even if Sumner thinks that is what the act represents, then Kocherlakota’s view is still totally reasonable. Sumner is very misleading to call his statement “not even close.”

      In your scenario, if a worker referred to the 11:30-12:30 rule as a key benchmark but you disagreed, you would have to get subtle and say it wasn’t a “key” benchmark but instead was just the default position, blah blah blah. If you instead objected in a way that led your other co-workers to not even realize you had to write a justification to Human Resources, you would be misleading them.

      • Transformer says:

        Suppose my boss really did want to consider changing the lunch break process and sent a memo that referred to the current 11:30-12:30 lunch time as “the reference lunch break” and asked people to propose “directive lunch breaks” and explain how they differed from “the reference lunch break”.

        One of my coworkers said “don’t trust that memo – its just an attempt to enshrine 11:30-12:30 lunch break as a key benchmark”, I think I would be justified as saying they were “not even close” to understanding the memo.

        • Transformer says:

          Oh, no sooner had I hit enter than I think I got your point…

          The very fact that any “Directive Policy Rule” has to explain how it differs from the “Reference Policy Rule”, makes the latter a benchmark – even if it is never actually used to decide monetary policy anymore. That’s kind of the meaning of “benchmark”, right ?

          • Bob Murphy says:

            Transformer wrote:

            The very fact that any “Directive Policy Rule” has to explain how it differs from the “Reference Policy Rule”, makes the latter a benchmark – even if it is never actually used to decide monetary policy anymore. That’s kind of the meaning of “benchmark”, right ?

            I can’t tell if you’re trolling me at this point.

            Yes, that was my point. My next move in our argument, Transformer, was that I was going to look up “benchmark” in the dictionary / thesaurus and see if they used the term “reference point.”

            Like I said in the OP, if Sumner had said, “You know, I think it’s a bit much to call it a key benchmark, and in any event my inside sources assure me the ‘detailed justification’ is just a formality that will be rubber-stamped so long as the Fed clerk types up some mumbo-jumbo about efficiency properties…” OK fine. But Sumner not only denied that it was a key benchmark, he said it wasn’t even close. What more could that mean, except, “I have no idea what is even in this bill that made Kocherlokato say that. He must have been hitting the bong that day.”

            • Transformer says:

              No intent to troll – just somehow took Kocherlokato to mean “the Taylor rule will be enshrined as a key driver of monetary policy” rather than a mere “benchmark for monetary policy”

              Not much of an excuse – but there is a typo in your OP ““They are setting 11:30am – 12:30pm and the benchmark for lunch breaks.” (“and” rather than “as”) that I think caused me to miss the point, when I initially read it.

              • Bob Murphy says:

                OK sorry for the typo, I’ll fix it.

                But, more generally in response to your question, when Kocherlakato (I spell his name differently each time I type it) said the Taylor Rule would be enshrined as a key benchmark, I took him to mean it would be a key benchmark.

              • Transformer says:

                yes, I’m sure that is what he meant – my earlier comments were based on a misreading of your post.

  4. Transformer says:

    Perhaps I an naive in the ways of politicians, but at face value the act seeks to at least consider replacing the current “Reference Policy Rule” with a “Directive Policy Rule”, and asks that any suggestions for a “Directive Policy Rule” explain how they differ from the existing “”Reference Policy Rule”, which seems a reasonable requirement even if somewhat bureaucratically stated in the verbiage you quote .

    If the act really is serious about looking for replacements for the current “Reference Policy Rule” then Kocherlakota is indeed “not even close” to categorize it as enshrin[ing] the Taylor Rule (which I take to be the current “Reference Policy Rule”) as a key benchmark for monetary policy”

    And (unless the act is entirely disingenuous) I don’t see why supporters of NGDP targeting shouldn’t see this as an opportunity to propose their favorite policy be put forward, which would explain Sumner’s enthusiasm.

    • Bob Murphy says:

      OK Transformer, what is the bare minimum the Act would have to do, for it to be “even close” to suggest that it had enshrined the Taylor Rule as a “key benchmark”?

      In other words, you think I am being way too harsh on Sumner for saying K.’s description was “not even close.” I’m curious what the Act would have to say, at the point where you would agree, “OK yeah Bob, Sumner was out of line for criticizing him like that.”

      To repeat: Sumner didn’t say something like, “Yeah sure, there’s a ‘Reference Policy Rule’ and you have to justify in detail why you’re deviating from it, but come on, is that really a ‘key benchmark’? Not even close.”

      No, Sumner didn’t talk about the “Reference Policy Rule” stuff at all. You would think from his description that Congress was merely saying, “Guys, just tell us what rule you’re using so we can tell if you’re hitting your own target.”

      • Transformer says:

        Taken literally the “Requirements for a Directive Policy Rule” in the act clearly do allow the rule to deviate from the Taylor rule as long as an explanation is given – so at level Sumner is correct.

        Had it gone beyond requiring an explanation and said something like ‘The directive rule must be closely based on the reference rule’ – then he would be wrong.

        Its also possible that while the act on paper allows any rule to be proposed and accepted insiders like Kocherlakota know that the Directive Policy Rule actually accepted would be certain to be a Taylor-like rule and not NGDPT. That would make Sumner wrong too – but that’s not your argument is it ?

        • Transformer says:

          And its not really clear to me how saying someone “is wrong” and “not even close” differ that much. Degrees of wrongness are entirely subjective and scarcely worth arguing about.

        • Bob Murphy says:

          Transformer wrote:

          Taken literally the “Requirements for a Directive Policy Rule” in the act clearly do allow the rule to deviate from the Taylor rule as long as an explanation is given – so at level Sumner is correct.

          !!!! Transformer you don’t seem to get my point. Sumner did not say, “All you have to do is explain why your rule differs from the Taylor Rule.” He didn’t mention that the bill said anything about the Taylor Rule at all. If you just read Sumner, you would think the bill merely said, “Guys, pick a rule and tell us how to evaluate your performance. We’re all ears, we don’t care which rule you pick, just pick one.”

        • Bob Murphy says:

          Taken literally the “Requirements for a Directive Policy Rule” in the act clearly do allow the rule to deviate from the Taylor rule as long as an explanation is given – so at level Sumner is correct.

          Another problem with your defense here, Transformer, is that the bill says justify, not merely explain. *Those are different.

          * Well it might use nouns, not the verbs, but you get my point I hope.

  5. RPLong says:

    Bob, i am inclined to agree with you, but if I give Sumner as much benefit of the doubt as I am humanly capable of, then I reach a slightly different conclusion: The fact that Sumner can see this so differently from Kocherlakota shows how even the clearest bureaucratic rules can be twisted by motivated reasoning to mean the exact opposite of what they do mean.

    If we pull this discussion out of the realm of monetary policy and into the realm of bureaucracy and/or statism in general (and why would we? but hear me out) then I think this is a poignant example of how any game played with the state is heads I win, tails you lose, where “I” is whoever’s calling the shots today.

    Or maybe i’m just being cynical – but that’s how it always seemed to me when i was working with bureaucrats and state agencies.

    • Bob Murphy says:

      I think we can ultimately blame Sumner’s misleading post on racist cops in Texas, right RPLong?

      (Inside joke everyone, chill out.)

  6. Daniel Kuehn says:

    Aside from the poor handling of Rorty at the beginning, this is a nice post.

    You are his peer (as are his commenters, in the Rortian sense at least), you (and they) are not letting him get away with this, so this is not anything like truth. It’s precisely what does not make the cut of truth for Rorty. The point is not that that we can just make things up as we go, it’s that there’s no independent foundation for establishing a truth outside of the relevant audience of our claim.

    • Gene Callahan says:

      So if I claim, “Fire does not burn humans,” this is true as long as my audience lets me get it away with saying it? There is no independent foundation, like, say, whether or not fire ACTUALLY burns humans, for establishing the truth of my claim?

      • Daniel Kuehn says:

        It seems to me whether fire actually burns humans is precisely what you’d use to judge the claim, right?

        Don’t we think it’s true that fire burns humans because everyone agrees that fire actually burns humans? Don’t we think it’s not true that a glass of lukewarm water burns humans because everyone agrees that a glass of lukewarm water doesn’t actually burn humans?

        Isn’t the verdict out on the sentence “God exists” as a statement about the world for society as a whole because we can’t seem to agree amongst ourselves whether it’s true or not but it carries more (or less) weigh in certain communities who are convinced?

        You act as if things that happen have no bearing on truth claims which is clearly not what’s been said at all. The point is that although nature presents us with all source of stuff it doesn’t present us with truth claims. Truth claims are the stuff of human brains (actually I don’t know if that’s strictly true – let’s just say they’re the stuff of smart ape brains).

        • Transformer says:

          “Don’t we think it’s true that fire burns humans because everyone agrees that fire actually burns humans?”

          If we saw fire burning a human we would probably start believing that fire actually does burn humans even if no-one else believed it, wouldn’t we ?

          • Daniel Kuehn says:

            Yes

            • Craw says:

              Didn’t you just lose the argument?

        • Bob Murphy says:

          In my view, I have a sick burn of Sumner in this post, but my peers don’t seem to agree.

        • Major.Freedom says:

          “Don’t we think it’s true that fire burns humans because everyone agrees that fire actually burns humans? ”

          I think it is true even if nobody else agrees, so agreement is not a factor. Knowledge of truth requires only one consciousness.

          “The point is that although nature presents us with all source of stuff it doesn’t present us with truth claims. Truth claims are the stuff of human brains…”

          Actually our brains inherently present us with some truth. There is an “overlap” between what external nature presents us, and what our own natures present us. That overlap is what we can know as truth without anyone else’s agreement.

          The very basis of you making the claim “nature presents us with all source of stuff it doesn’t present us with truth claims” rests on a self-reflective understanding of your own nature.

        • Craw says:

          How do we agree on new things then? Everyone thinks all dinosaur lineages died out. How does that change? The first dissenter was the only one.

      • Tel says:

        So if I claim, “Fire does not burn humans,” this is true as long as my audience lets me get it away with saying it?

        For sufficiently large audiences, and suitable context awareness, I’d agree on that. It sounds better when you say it like this:

        You can fool all the people some of the time, and some of the people all of the time, but you can’t fool all the people all the time.

        As for how important context sensitivity is, I regularly ride or drive around in vehicles powered by internal combustion (that’s fire BTW) and cannot remember ever being burned by it. I travel in electric trains power by a very large coal furnace and those are no problem. I go to the beach under a really massive ball of fire, and now and then might come up a touch pink the next day, but no big deal.

        I’d say from a statistical basis humans are pretty dang good at handling fire without being burned. But possibly you had something else in mind.

        Once you allow for context of course, there’s no other way to evaluate this concept of truth without an audience. For certain types of truth we could go to war over it I suppose, last man standing can say what he likes… some might consider that to be cheating.

    • Craw says:

      On what basis do we not let someone get away with a claim? On the basis of a loaded gun? Or on the basis of pointing out a statement is false?

  7. Dan says:

    You shouldn’t have given people the actual bill’s language. You should’ve just asked people to comment on what the bill allows based off of Sumner’s views. It’s easier for people to see how misleading someone was if you don’t explain why what he said was misleading upfront. Let them be misled, say their incorrect view out loud, and then correct them. Give people enough rope to hang themselves but no more.

    • Bob Murphy says:

      I understand what you’re saying Dan, but I didn’t anticipate anybody would disagree with me on this one. I am naive.

Leave a Reply to Transformer

Cancel Reply