You can’t have it both ways

In two recent Econlog posts (here and here), I pointed out that a wise man or women should always have two levels of belief.  One is their own view of things, independently derived from their own research.  This is the view from within your skin.  The second level of belief is the awareness of the wisdom of crowds.  The awareness that an index fund is likely to do better than a fund that you personally manage. An awareness that the consensus view of the true model of the macroeconomy is likely to be better than your own model of the economy.  This is the view from 20,000 miles out in space, where it’s clear that you are nothing special.

In the comment section, Philo suggested:

In most things, you’re admirably sensible (insightful, etc.). In philosophy . . . well, better stick to your day job!”

He likes my market monetarist view of things, but not my philosophical musings.  But you can’t have it both ways.  If my philosophy is wrong then my market monetarism is equally wrong.  Either the wisdom of the crowds is true, or it isn’t.

(As an aside I’m aware that the wisdom of the crowds might be slightly better if the views are weighted by expertise, but that has no bearing on my claim.  Even if you think I have a bit more expertise that the average economist, the entire weighted sum of non-Scott Sumner economists is, objectively speaking, far more qualified than I am.)

In this blog I am normally giving you my views on the optimal economic model from the “within the skin” perspective, because otherwise I am of no use to society.  I’d be just a textbook.  In contrast, I give you my views on where markets are heading from the 20,000 miles up perspective, because that’s the most useful view for me to communicate the intuition behind market monetarism.  You don’t care where I personally think the DOW is going, and you should not care.

It is the job of the economics profession to weigh my arguments, and the arguments of those who disagree with me, and reach a consensus.  That consensus is not always correct, but it’s the optimal forecast.  Unfortunately, at the moment the optimal forecast is that I’m wrong about monetary offset, but I’ll keep arguing for monetary offset because that’s the view I arrived at independently, and I’m of no use to society unless I report that view, and explain why.

When I talk to philosophers about epistemology, they often mention concepts like “justified true belief” which seems question begging to me.  I’m certainly no expert on the subject, but I can’t see how the EMH is not right at the center of the field of epistemology.  If back in 1990, we wanted to know whether there were Higgs bosons or gravity waves, the optimal guess would not have been derived by asking a single physicist, but rather setting up a prediction market.  Yes, traders know less about physics than the average MIT physicist, but traders know whom to ask.

Many worlds vs. Copenhagen interpretation? Perhaps it can’t be tested.  But if it could, then set up a prediction market.  Robin Hanson’s futarchy is a proposal to have public policy based on society’s best estimate of what is true—derived from prediction markets.  He wants us to vote on values and bet on beliefs.  Richard Rorty might go even further, and have us bet on values, where the outcome of the bet depends on a poll of values 50 years in the future.  Rorty would say values are no more subjective than science.

I think the EMH is basically what Rorty meant when he said truth is what my peers let me get away with.

PS.  My two types of beliefs do not have a rank order; they are incommensurable concepts.  Both are essential, and one is not more or less important than the other.  There’s no answer to “What do I really believe about monetary offset?”  I believe different things, at different levels of belief.


Tags:

 
 
 

59 Responses to “You can’t have it both ways”

  1. Gravatar of E. Harding E. Harding
    27. May 2016 at 14:30

    “the optimal guess would not have been derived by asking a single physicist, but rather setting up a prediction market.”

    Are prediction markets even all that reliable? I don’t see what makes them magic to you, Sumner.

    “That consensus is not always correct, but it’s the optimal forecast.”

    -Define “optimal”.

    “I believe different things, at different levels of belief.”

    -Huh? Pardon me, Scott, but this sounds like theological BS.

  2. Gravatar of Rajat Rajat
    27. May 2016 at 14:34

    Hang on… there’s a big difference between “the wisdom of crowds” as expressed through prediction markets as to compared to the consensus view of economists. Re monetary offsett, I am happy to defer to a liquid prediction market, but not to the views of most of the profession, assuming this can be accurately measured in some way. Professionals of any stripe – especially academics – can have a lot of ego bound up in, and hence can be quite attached to their views (eg Krugman & co’s reaction to the 2013 austerity ‘test’).

  3. Gravatar of Gary Anderson Gary Anderson
    27. May 2016 at 14:41

    Philosophers are not going to heaven. They even know they aren’t going to heaven. Take Camus, who said it was better to roll that stone up a hill than to become unaware of anything. And “I think, therefore I am” just never did anything for me. I am hungry and I eat. What philosophical complexity.

    I worry a lot about economists because they are good at what they do and they try to make sense out of the economy. Pure torture.

    The futurists may as well had never been born. They are wicked. Many of them are Eugenicists and they want to kill little children.

    Just think, if central banks didn’t buy bonds, they could just issue helicopter money with no bonds being issued, no debt whatsoever. The problem with the Fed is that it bought bonds instead of handing out helicopter money in the first place. Now it is screwed and if it raises interest rates it will have to pay the banks instead of taxpayers. The Fed already has one foot in hell.

  4. Gravatar of Vladimir Vladimir
    27. May 2016 at 15:01

    I think you’re a bit confused about how a market should factor into your beliefs. An ideal Bayesian should include *all* information when doing the calculation. For example you can’t say that you believe one particular theory of monetarism but then decide to bet by aggregating the prediction market of monetarists. There is only 1 level.

    Your belief in your specific monetarist theory should take into account the current prediction market, which should be weighted fairly heavily depending on how smart you think they are, and then it is adjusted after your own independent research depending on how good your theory is in comparison (or vice versa).

    I think your confusion comes from the artificial examples given in “wisdom-of-crowd-scenarios”, where 100 independent estimates of how many beans are in the jar are better than a single estimate. In that artificial scenario, nobody has access to the other answers, whereas in the real world it’s more like the Price is Right opening game where you see everyone else’s guess and you *definitely* need to take that into account however it shouldn’t necessarily be your final belief based on how well calibrated you think the prediction market is.

  5. Gravatar of Jim S, Jim S,
    27. May 2016 at 15:15

    I am aware of studies that show that groups are better than individuals at judging how things are in the present, from the weight of a cow, to the number of jelly beans in a jar, to the value of a stock. I don’t see how that extends to knowing the future. I think the future is unknowable.

    There are also interesting studies showing that the judgement of groups can be swayed by a variety of methods, hence I believe in a “weak” wisdom of crowds as opposed to the “strong” wisdom of crowds. Bubbles do happen, and are evident to the unbiased observer, to deny they exist is indicates strong prior beliefs are leading one overrule the evidence of history

  6. Gravatar of Benjamin Cole Benjamin Cole
    27. May 2016 at 15:16

    Here is what I think about what I think about what think: I think too much.

    What I know is that modern-era populations are happy when there are “labor shortages.”

    What should we think about that?

  7. Gravatar of jonathan jonathan
    27. May 2016 at 17:15

    Aren’t your “two levels of belief” just what people mean by the outside view vs. the inside view?

    And when you talk about the deep philosophical meaning of the EMH, are you talking about Bayesian updating on the evidence of other peoples’ beliefs, as illustrated by (e.g.) the Aumann agreement theorem?

  8. Gravatar of Major.Freedom Major.Freedom
    27. May 2016 at 17:28

    “In contrast, I give you my views on where markets are heading from the 20,000 miles up perspective, because that’s the most useful view for me to communicate the intuition behind market monetarism.”

    You do not have any 20,000 miles up perspective. You only have an “in your own skin” perspective.

    Perspective is from within only you. You do not have the mind of God.

    In most cases thinking of aggregate concepts for humanity poses no issues in reasoning, but there is a risk of making a mistake. This is the mistake of ignoring, forgetting, or otherwise contradicting the fact that such aggregates are not in any ontological way separate or distinct from the individual humans of which the aggregate concept refers.

    In reality, there is only acting individual men. The concepts “NGDP” and “RGDP” have no reality apart from acting individual men.

    Unfortunately, you treat NGDP as a causa sui, directing the course of affairs of all individual men from without, in short, you hpostatize the concept.

  9. Gravatar of Gary Anderson Gary Anderson
    27. May 2016 at 17:54

    “And when you talk about the deep philosophical meaning of the EMH, are you talking about Bayesian updating…”

    Logic verses personal belief. However, personal belief and God given faith are two different things. The Apostle Paul never considered personal belief except to agree with the view that it isn’t of the man who runs or the man who wills, but of God that shows mercy. Real faith is counted as EVIDENCE of things unseen.

    As for evidence of Scott being right about his personal economic views, only time will reveal that. But then, if no one ever tries to implement Scott’s ideas, how will people add to the evidence?

  10. Gravatar of Vladimir Kovac Vladimir Kovac
    27. May 2016 at 18:02

    An ideal Bayesian should include *all* information when doing the calculation. For example you can’t say that you believe one particular theory of monetarism but then decide to bet by aggregating the prediction market of monetarists.

    Your belief in your specific monetarist theory should take into account the current prediction market, which should be weighted fairly heavily depending on how smart you think they are, and then it is adjusted after your own independent research depending on how good your theory is in comparison (or vice versa of course).

    jonathan: Inside view refers to a cognitive bias in “Thinking Fast and Flow” I think. The point is that you should always be taking into account the outside view when making predictions and avoid the inside view.

  11. Gravatar of ssumner ssumner
    27. May 2016 at 18:39

    Rajat, You said:

    “Re monetary offset, I am happy to defer to a liquid prediction market, but not to the views of most of the profession, assuming this can be accurately measured in some way.”

    Very similar to what I said:

    “Unfortunately, at the moment the optimal forecast is that I’m wrong about monetary offset, but I’ll keep arguing for monetary offset because that’s the view I arrived at independently, and I’m of no use to society unless I report that view, and explain why.”

    The only difference is that I’m not even willing to defer to a prediction market. I’m an insider, and will continue to form independent judgements.

    Jonathan, That’s for you to decide, I’ve never studied philosophy. I’ll say this, however, you said “just” as if the idea is obvious. But over at Econlog lots of people treated this as bizarre ideas that they’d never heard of.

    I don’t think the point I’m making is Bayesian, although I am certainly fine with Bayesian reasoning.

  12. Gravatar of Dan W. Dan W.
    27. May 2016 at 18:42

    Scott,

    The phrase “Wisdom of crowds” is an oxymoron. If one looks at the choices made by many crowds one might conclude there is little evident wisdom. Consider the nomination of Donald Trump? Is that a wise choice? But it is the one the crowd is making.

    Then one must consider who exactly is the crowd. As it concerns political primaries the crowd is large but it is only a portion of the American population. And of that voting crowd how many members are knowledgeable about their choice?

    Consider, how many expert investors would it take to fairly price a company? 2? 5? 50? 500? Would it take more than 500?

    Might it be that the crowd has little real wisdom? This doesn’t disprove EMH but it invites more inquiry into who is actually making the market.

  13. Gravatar of ssumner ssumner
    27. May 2016 at 18:46

    Vladimir, You missed the point. I’m arguing that these two levels of belief cannot be combined without destroying useful information. I’d suggest reading my two Econlog posts. I understand what Bayesians believe, but that’s not relevant to the issue I’m looking at here, where there are two levels of belief.

    Jim S. You said:

    “I think the future is unknowable.”

    Yes, unknowable in the sense that we cannot know for certain, right now. But I believe a prediction market would give you an optimal forecast of the future. If not there’d be lots of $100 bills on the sidewalk.

  14. Gravatar of ssumner ssumner
    27. May 2016 at 18:47

    Dan, You are in way over your head. Please don’t bother.

  15. Gravatar of Dtoh Dtoh
    27. May 2016 at 18:59

    @Scott
    You’re overthinking. It’s much simpler.
    Crowds are good at empirical estimation. They suck on theory.

  16. Gravatar of Lorenzo from Oz Lorenzo from Oz
    27. May 2016 at 22:00

    “the failure of epistemology made postmodernism possible, and the failure of socialism made postmodernism necessary”

    The thesis of a great book by philosopher Stephen Hicks (which I recommend to all and sundry) , a quick review by another philosopher here:
    http://www.philosophyblog.com/2004/08/hicks-explaining-postmodernism.html

    Longer review by me here.
    http://lorenzo-thinkingoutaloud.blogspot.com.au/2009/02/explaining-postmodernism.html

  17. Gravatar of Dan W. Dan W.
    28. May 2016 at 04:32

    Scott,

    You titled this post “You can’t have it both ways” yet yourself want it both ways! How can you defend the wisdom of the crowds as it concerns financial prediction markets yet bemoan and berate the wisdom of crowds as it concerns democratic elections? If the crowd can be disastrously wrong when it comes to choosing presidents how can it be trusted to be right when it comes to predicting NGDP?

    Research shows that for the “wisdom of crowds” to work there must be accommodation for independent and diverse opinions. Democratic elections fail this test because there is not an allowance for diverse opinion as voters are typically limited to a choice of a few candidates. Democratic elections also fail because publicity and advertising skew opinion and lessen the independence of thought.

    Financial markets allow a greater diversity of choice but they still can face the risk of those choices being dependent, or otherwise the “madness” of the crowd piling into the same side of the trade because that is where “everyone” is making money.

    The “wisdom of crowds” and the “madness of crowds”. There is evidence both are always at work. How does know which crowd is making the decision?

    “What happens if this so-called wisdom, while theoretically true, is, practically speaking, not so easy? A study just published in PNAS suggests that all it takes is a whiff of social influence (the knowledge of how others are acting) for the wisdom to evaporate – and for crowds to become even less wise than individual decision makers.”

    http://bigthink.com/artful-choice/the-wisdom-of-crowds-revisited-when-the-crowd-goes-from-wise-to-wrong

  18. Gravatar of W. Peden W. Peden
    28. May 2016 at 04:35

    On Bayesianism-

    “An ideal Bayesian should include *all* information when doing the calculation.”

    That’s true, but it’s not a distinctively Bayesian principle. It’s sometimes called the “Principle of Total Evidence”. The name comes from Rudolf Carnap, who could reasonably be called an early Bayesian (albeit with a different view of probability from most modern Bayesians) but the basic idea is there in Jacob Bernoulli’s work in the 18th century, and there’s also an interesting discussion in Keynes’s “A Treatise on Probability”.

    On futarchy-

    There’s actually an element of betting on values in Hanson’s 2013 article: bets are to be pairs of bets on (a) a current national welfare index and (b) a future national welfare index.

    Dan W.,

    A crowd, like an individual, is wise only under the right circumstances. One of the problems with democracy is that individual voters have insignificant power to affect change, yet informed voting requires a very significant personal sacrifice of time/energy. So you wouldn’t expect voters to be particularly wise, individually or in the aggregate. In contrast, investors can literally make or break their living by being well-informed, and as James Surowiecki has argued, markets are particularly good (relative to other groups) at avoiding groupthink. And groupthink is a sense of “bubble” that I think even Scott can tolerate.

  19. Gravatar of ssumner ssumner
    28. May 2016 at 05:26

    Dtoh, I agree, but individuals suck even worse than crowds at theory.

    Dan, Still don’t get it, eh?

  20. Gravatar of Dan W. Dan W.
    28. May 2016 at 05:51

    Is not the mass investment in index funds “group think” behavior?

    What percentage of money buying stocks on any given day is “smart” money, based on real effort to predict the future value of each company that makes up the index? What does it matter what the smart investor thinks if the crowd sets the market price and persists in its delusion of pricing?

    Thing is we only know the right price when there is a way to actually prove the right price. Consider the Preakness race. The crowd favored the wrong horse. The crowd was wrong. But we only know this because there was a race to determine the fastest horse. Until this race occurred the crowd would have persisted in claiming Nyquist was faster than Exaggerator.

  21. Gravatar of AN AN
    28. May 2016 at 07:43

    Scott,

    Could you give an example of these two levels of belief in action? The reason I ask is because it seems that your posts have been eliding the problem of measurement. Many (in fact, the vast majority of) beliefs are not directly measured by markets. The main problem, then, is the interpretation of market results into a belief system.

    I think that Bryan Caplan hit the nail on the head in his blog post on Econlog. You said that he took “The view a betting market would take, if one existed.” In his critique of you when he said: “But in any case, this is a circular explanation for why I win, because it doesn’t explain how I ballpark a betting market position when there is no betting market.”

    The transaction costs associated to setting up traditional betting markets for most belief questions are prohibitive (as a side note, I’m very interested in figuring out ways to reduce these transaction costs or inventing “nontraditional” betting markets using technology). Thus, I think the main problem lies in interpreting markets.

    There’s one more part of your post I would like to address: the idea that the academic economic consensus is equivalent to the “wisdom of crowds.” There is no clear market mechanism forcing academic economic consensus to mirror reality. If anything it’s the opposite: if academic economists had incredible insight into the economy, they should be able to parlay it into personal wealth. Rather, academic economists are very good at appealing to other economists and the politically powerful, not at making money.

  22. Gravatar of bill bill
    28. May 2016 at 08:14

    What book by Rorty do you recommend to start with?

    BTW, the Harris book I recommend is The Moral Landscape.

  23. Gravatar of Ray Lopez Ray Lopez
    28. May 2016 at 08:49

    A remarkably humble mea culpa by Sumner, which shows his pendulum is in its center now. Only for now.

    OT–Sumner’s view that FDR’s New Deal prolonged the Great Depression (which was largely over by 1934/5) is shared by the proto-libertarian, Isabel Paterson. https://en.wikipedia.org/wiki/Isabel_Paterson

  24. Gravatar of Fed Up Fed Up
    28. May 2016 at 14:00

    Does the price inflation target apply to the lender of last resort function of the central bank to the commercial banks?

  25. Gravatar of dtoh dtoh
    28. May 2016 at 14:05

    @Scott
    99% of individuals suck worse than crowds at theory.

  26. Gravatar of Major.Freedom Major.Freedom
    28. May 2016 at 15:36

    bill:

    “What book by Rorty do you recommend to start with?”

    FYI, Richard Rorty was a philosopher who reintroduced the ancient times of nihilism and skepticism. If his own philosophy were applied to his own pronouncements, then his own pronouncements would have to be considered as ultimately nothing but hot air.

    In his book “Philosophy and the Mirror of Nature”, Rorty wrote that Rationalism:

    “…is a desire for constraint—a desire to find “foundations” to which one might cling, frameworks beyond which one must not stray, objects which impose themselves, representations, which cannot be gainsaid…” (p. 315)

    “The dominating notion of epistemology is that to be rational, to be fully human, to do what we ought, we need to be able to find agreement with other human beings. To construct an epistemology is to find the maximum amount of common ground with others. The assumption that an epistemology can be constructed is the assumption that such common ground exists” (p. 326)

    However, Rorty claims that no such common ground exists: hence the false idol of rationalism must fall and a “relativist” position termed hermeneuticsmust be adopted:

    “Hermeneutics sees the relations between various discourses as those of strands in a possible conversation, a conversation which presupposes no disciplinary matrix which unites the speakers, but where the hope of agreement is never lost so long as the conversation lasts. This hope is not a hope for the discovery and of antecedently existing common ground, but simply hope for agreement, or,at least, exciting and fruitful disagreement. Epistemology sees the hope of agreement as a token of the existence of common ground which, perhaps unbeknown to the speakers, unites them in common rationality. For hermeneutics, to be rational is to be willing to refrain from epistemology—from thinking that there is a special set of terms in which all contributions to the conversation should be put—and to be willing to pick up the jargon of the interlocutor rather than translating it into one’s own. For epistemology, to be rational is to find the proper set of terms into which all contributions should be translated if agreement is to become possible. For epistemology, conversation is implicit inquiry.For hermeneutics, inquiry is routine conversation” (p. 318)

    After reading Rorty’s whole book, I thought to myself it would be appropriate to ask “What, then, about Rorty’s own pronouncements?” If there is nothing like truth based on common, objective ground, then all of Rorty’s talk can’t claim to say anything true. In fact, it would be self-defeating to do what he seemed to be doing: denying that an objective case can be made for any statement, while at the same time claiming this to be the case for his own views. In so doing, one would falsify the content of one’s own statement.

    This then is the self-refuting nature of Rorty’s worldview. In order to understand Rorty correctly, one must first realize that he cannot truly be saying what he seemed to be saying.

    Rorty’s mistake is the same mistake philosophers have made since Plato. Language is not some ethereal medium disconnected from reality, but is itself a form of action. It is an offshoot of practical cooperation and as such, via action, is inseparably connected with an objective world. Talk, whether fact or fiction, is a form of cooperation and thus presupposes a common ground of objectively defined and applied terms. Not in the sense that one would always have to agree on the content of what was said or that one would even have to understand everything said. But rather, in the sense that as long as one claimed to express anything meaningful at all, one would have to assume the existence of some common standards, if only to be able to agree on whether or not and in what respect one was in fact in agree-ment with others, and whether or not and to what extent one in fact understood what had been said.

    Sumner presupposes this common objective ground in all of his blog posts. So do all the commenters here. And doing so is not even a choice. It is required. It is objective because it is action.

  27. Gravatar of E. Harding E. Harding
    28. May 2016 at 16:00

    “I’m not familiar with deBoer, so I have no idea whether he’s being dishonest or if he’s simply unaware of global realities.”

    -He’s basically a democratic Communist who believes Democrats are indistinguishable from libertarians. Berniebro. Whiner.

    “Does he not know what it means to see per capita income rise from say $1000 to $2000, or $5000, or $10,000?”

    -Of course he doesn’t; he’s a Berniebro.

    Also, as I have repeatedly pointed out, manufacturing productivity in the U.S. has flatlined in the past five years. Not just growth in manufacturing productivity -manufacturing productivity.

    https://research.stlouisfed.org/fred2/graph/?g=Vzx

    And, yes, you’re right about Tyler being wrong. I suspect Cowen has an IQ of 115. Just the fact he knows how to use turgid language and is well-read doesn’t mean he’s smart.

  28. Gravatar of Major.Freedom Major.Freedom
    28. May 2016 at 16:01

    Sumner is hypostatizing the concept of “crowds”.

    Crowds are in fact composed of separate individuals. If a crowd of 10 people have 10 different expectations of the future, or if a crowd of 10 people have contradictory beliefs (say 5 people believe X is true while the other 5 people believe ~X is true), then where is the “wisdom” here?

    Hermeneutics would have the talk itself as wise, regardless of the future and regardless of whether X is in fact true.

    Yet only those particular individuals whose expectations, and only those particular individuals whose belief concerning X, are “wise” to the extent they have fidelity with a standard external to the beliefs themselves. Otherwise, how could a group of people who have contradictory beliefs concerning the future, as well as X, be as a group “wise”? Does wisdom require incompatiable ideas to be regarded as both true as the same time? Is an individual wise if they believe that next year there will be a recession AND that next year there will not be a recession?

    Whatever wisdom is, it isn’t what Sumner says it is.

  29. Gravatar of ssumner ssumner
    28. May 2016 at 17:44

    Bill, I’m not sure, as I have not read very much Rorty. I believe Philosophy and the Mirror of Nature is the book that popularized his ideas, but I never read it. I read Irony, Contingency and Solidarity, which I thought was excellent. I also read a short book where he debated another philosopher, but I don’t recall the name of it. It’s only about 60 pages. I’m not very knowledgable about philosophy.

    AN, You said:

    “There is no clear market mechanism forcing academic economic consensus to mirror reality. If anything it’s the opposite: if academic economists had incredible insight into the economy, they should be able to parlay it into personal wealth. Rather, academic economists are very good at appealing to other economists and the politically powerful, not at making money.”

    This is not true, due to the EMH. It’s almost impossible to beat markets, except through luck.

    I realize that betting markets don’t exist for many topics, but we can be reasonably confident that most highly qualified economists do not believe that monetary offset is true, at least at the zero bound.

  30. Gravatar of Art Deco Art Deco
    28. May 2016 at 19:18

    I suspect Cowen has an IQ of 115.

    That would be the median for an ordinary BA holder. Fairly unusual for a man with a signed dissertation in a math-intensive subject.

  31. Gravatar of AN AN
    28. May 2016 at 20:29

    Scott,

    Thanks for the response and for pointing out the fallacy. My point was that academic economists’ incentives are derived more from politics than from prediction accuracy. Indeed, with sufficient political skill, it’s easy to get by while making false or even contradictory predictions.

    I think it’s important to consider the amount of distortion (i.e. difference from a pure betting market) involved in any consensus view. For example, it would be nuts not to acknowledge the “consensus” price for a stock (i.e. try buying a stock for a different price). On the other hand, the “consensus” view of Republican voters is that Trump is the best candidate for President, which is just as crazy but on the opposite side of the equation.

    Obviously, most macroeconomic questions lie in between these extremes. But the value of the fiscal multiplier is a heavily political question, so I would be inclined to place more importance in your independent derivation of monetary offset.

  32. Gravatar of Philo Philo
    28. May 2016 at 21:16

    Let’s try out your idea about two levels of belief on an example. Suppose that, while engaging in some fundamental securities analysis (but without having access to any inside information), I learn something that suggests to me that Chevron is undervalued, compared with Exxon Mobil. But I accept the EMH. (So why was I doing the securities analysis? Just for fun: I was bored. Let me add that these are the only two companies I have investigated.) Should we describe me as having beliefs on two different “levels,” on one level believing Chevron stock is a better buy than Exxon Mobil stock, on another believing they are both appropriately valued? (By the way, since these “levels” are incommensurable, the “level”-metaphor is infelicitous.)

    The answer is to be found in my behavior. If I buy Chevron and short Exxon Mobil, or something more or less equivalent–employing a significant amount of my net worth–then I really believe in my analysis, and, although I may have professed to believe in the EMH, I really do not. But if I do not do this, though the opportunity is readily available, then I really do believe in the EMH and I do not really believe in the conclusion suggested by my own analysis of the two companies.

    Beliefs are guides to action, and it is from my actions that my beliefs are to be inferred. I cannot at the same time believe incompatible propositions. One may have other psychological relations besides belief to various propositions: for example, one may have an *inclination to believe* a proposition–a relation that can hold even if belief itself does not; but it should be easy to distinguish such psychological relations from belief proper.

  33. Gravatar of W. Peden W. Peden
    29. May 2016 at 06:59

    Art Deco,

    Unusual, but not as rare as you might think. Willpower outpredicts IQ for academic success by a large margin, according to every study I’ve seen on the subject, although it’s also true that it’s very unlikely that someone with a sub-90 IQ will do well academically.

  34. Gravatar of Dan W. Dan W.
    29. May 2016 at 08:14

    Scott,

    You write: “I think the EMH is basically what Rorty meant when he said truth is what my peers let me get away with.”

    I agree. In terms of prices EMH means the “crowd” sets the price and it is impossible for an individual to predict with any advantage what price the crowd will set in the future.

    But why should it follow that EMH means markets have predictive capability? Is this a provable thesis? Or a narrative you wish to believe? How many examples of markets failing to predict outcomes would it take to nullify the hypothesis that “crowds” have predictive capability?

    Consider this research on the commodities futures markets that asserts this observation: “we document a broad decline in the predictive content of commodity futures prices since the early 2000s.” http://www.ssc.wisc.edu/~mchinn/commodityfutures.pdf

  35. Gravatar of Art Deco Art Deco
    29. May 2016 at 08:42

    Unusual, but not as rare as you might think. Willpower outpredicts IQ for academic success by a large margin

    Per the Digest of Education Statistics, there were 97,000 STEM doctorates awarded during the 2013/14 academic year. Let’s offer that 1 in 6 were to the foreign born. That leave 80,000 awarded to natives. You’ll collar a few thousand more from the natural science wing of psychology and from economics, demography, urban geography &c. That’s not much more than 2% of a typical birth cohort.

  36. Gravatar of ssumner ssumner
    29. May 2016 at 09:59

    AN, You said:

    “But the value of the fiscal multiplier is a heavily political question, so I would be inclined to place more importance in your independent derivation of monetary offset.”

    I’m not saying politics never plays a role. For instance, political correctness may bias results on some questions. But I think most economists could pass a lie detector test denying monetary offset.

    Philo, I agree that beliefs influence action, but I’d say that each of these two levels can influence different kinds of actions. In some cases my “within my skin” perspective influences my actions, as when I blog on market monetarism. In other cases, I’d rely on the consensus, the 20,000 miles up view, for instance if I was made dictator of the Fed.

    For a market participant, they might invest their own funds based on their research, but tell their mother to buy an indexed fund.

  37. Gravatar of AN AN
    29. May 2016 at 11:47

    Scott, You said:

    “I’m not saying politics never plays a role. For instance, political correctness may bias results on some questions. But I think most economists could pass a lie detector test denying monetary offset.”

    I don’t think outright lying is the biggest cause of market distortion here (in general, it’s hard to motivate people to lie consistently). I think the most likely problem is selection bias: you don’t become a macroeconomist unless you think some combination of monetary and fiscal policy is effective at changing real outcomes. Monetary policy is implicitly fixed at 2% inflation targeting, so apparently the only independent variable is fiscal policy.

    If the only variable economists can affect right now is fiscal policy, they are naturally going to be disinclined to say it’s worthless from the macro perspective (again, probably not from lying outright, but from selection bias and the need to get funding and write relevant papers).

    People in general have a bias towards believing their work is worthwhile. There’s a reinforcing mechanism here: people are attracted to purposeful work, people convince themselves that their work is purposeful in order to keep working, and the most convincing, purposeful workers advance in status. This is necessary for society to function — worker motivation is a function of sense of purpose (as well as mastery and autonomy).

    Without market structure to ground people in reality, tribalism and memes overwhelm evidence. It’s not because of intentional dishonesty. It’s just the way our minds work.

    The dual positive and normative nature of macroecon is incredibly interesting from a philosophical point of view, but the current level of dialog is disappointing. For the above reasons, I don’t trust the consensus. But I have a solution: only pay attention to people who make falsifiable predictions, and grade prediction-makers based on their accuracy.

  38. Gravatar of Matthew Waters Matthew Waters
    29. May 2016 at 15:50

    On the EMH, “The Myth of the Rational Voter” made the argument that markets are efficient “wisdom of the crowds” while political contests will not be efficient. That argument never really made sense to me.

    Markets may be correct, or wrong. Voters may be correct, or wrong. There’s irrationality on both counts.

    My epistemological beliefs are summed up in “It’s a pretty messy world, compared to relative neatness of hard sciences. You take different empirical data, some logical arguments and cobble together a Bayesian posterior.”

    My current Bayesian posterior is the EMH in the “weighing machine” sense. As Ben Graham said, “in the short term, the market is voting machine. In the long run, it is a weighing machine.” Unlike voting, longer-term investments approach a “true” underlying value.

    On shorter timelines, the true value of investments has a less clear link with the actions of intermediaries and end investors. The issues with this link even leads to repeatable violations of the EMH. “Value vs. Growth stocks” has been repeated in many different markets, the “Equity premium puzzle,” etc.

    On another note in Trump Derangement Syndrome, the three recent Will Wilkinson posts are very, very good. The newest one is very frightening.

    http://www.willwilkinson.net/

  39. Gravatar of Major.Freedom Major.Freedom
    29. May 2016 at 21:23

    Matthew Waters:

    ““The Myth of the Rational Voter” made the argument that markets are efficient “wisdom of the crowds” while political contests will not be efficient. That argument never really made sense to me. Markets may be correct, or wrong. Voters may be correct, or wrong. There’s irrationality on both counts.”

    Did you read the book? Caplan does not dispute the notion that there is irrationality in both markets and politics. What he argues is that ‘since delusional political beliefs are free, the voter consumes until he reaches his ‘satiation point,’ believing whatever makes him feel best. When a person puts on his voting hat, he does not have to give up practical efficacy in exchange for self-image because he has no practical efficacy to give up in the first place.’

    With markets on the other hand, delusional beliefs incur direct costs on the individual, and practical efficacy does need to be sacrificed.

  40. Gravatar of Postkey Postkey
    30. May 2016 at 02:25

    “It’s almost impossible to beat markets, except through luck.”

    Is this ‘luck’?

    http://www.techinvest.ie/downloads/sample.pdf

    P 12.

  41. Gravatar of Craig Craig
    30. May 2016 at 03:34

    Scott,

    I think I grok most of what you said, but I wish to try to construct a paradox. Your two hats that you like to wear are (1) Scott Sumner personal views hat, and (2) wisdom of the crowd hat (ideally a prediction market, but if not then consensus / textbook views).

    This does strike me as reasonable as when asked a question, your answer can be of the form, “I believe X, but you should know that the majority of the field disagrees and believes Y. And there are plenty of smart or smarter people in the Y camp.”

    My constructed paradox: What if hat (2) WOTC doesn’t believe in either WOTC or a prediction market? What if the economic consensus switched and stopped believing in EMH? What if the consensus view is that you shouldn’t subscribe to the consensus view? Where would that leave you?

  42. Gravatar of bill bill
    30. May 2016 at 05:32

    Today’s (5/30/16) Dilbert seems related to this topic. There are smart people on both sides of every issue. Which makes sense. With respect to all the zillions of things that all the people agree on, then, by definition, they aren’t issues any more.

  43. Gravatar of Mike Rulle Mike Rulle
    30. May 2016 at 07:14

    Dan on “wisdom of crowds”

    I always thought of this as an EMH type concept. Its the best guess of the future. In markets, e.g., todays prices are the best guess of the PV of future cash flows.

    In politics, the right analogy on Trump is the probablistic chance he will be president—-not that its “wise” he should be president. One could set up a prediction market on specific outcomes of a hypothetical Trump presidency. That would be a wisdom of crowds concept—again, not that those outcomes would be wise per se.

    I am having trouble closing the loop of this thought as it relates to Scott’s “outside” perspective on MM. It seems he says markets prefer outcomes, rather than pricing outcomes in the sense that they differ from his inside views. I am obviously missing something.

  44. Gravatar of Matthew Waters Matthew Waters
    30. May 2016 at 17:04

    MF,

    I said later that the markets often have forcing mechanisms which voting mechanisms do not. The general sense that markets are better than, say, Soviet central planning has been taken too dogmatically. Real-world evidence shows market frictions and failures can be far more non-trivial than the real-world evidence would indicate.

    Postkey,

    Yes, it could be luck. Mutual Fund companies have been known to start many mutual funds with small bits of capital and then only have the top performers survive. The lesser-performing mutual funds get folded into the top performers, and then the top performers are further folded in.

    The Page 12 of that newsletter seems particularly problematic. They look like portfolios from the newsletter rather than their actual open-ended fund. The portfolios look like small-cap stocks which will have higher transaction costs. The illiquidity and variability means part of the portfolio’s return is liquidity and risk premium. Their actual mutual fund, which has transaction costs and has to buy stocks big enough for the fund, has not done as well.

    The portfolio on page 1 is particularly misleading. It says an average return of 37%, which is an awful way to give returns. The actual compounded return is 24%, which is still good. But again, you have pure luck + no transaction costs + liquidity/risk premiums. It may be honestly above-average, but the case is far from mixed.

  45. Gravatar of mpowell mpowell
    30. May 2016 at 18:00

    I’m not going to say you’re entirely wrong, but you are overlooking something very critical – it is objectively much harder to become an expert at financial predictions than some other topics (chess, for example). Read Kahneman’s Thinking Fast and Slow to find out what people are good at vs not. So there are some subjects where I would trust an expert over crowds. And in the field I am an expert… I usually know who to trust.

  46. Gravatar of Bob Murphy Bob Murphy
    30. May 2016 at 20:17

    Scott,

    I can tell people are getting confused by your statement: “My two types of beliefs do not have a rank order; they are incommensurable concepts. Both are essential, and one is not more or less important than the other. There’s no answer to “What do I really believe about monetary offset?” I believe different things, at different levels of belief.”

    Let’s consider a standard Bayesian model. There are 30 kids in the classroom who get to bid on a jar of nickels that the teacher has brought in. Each kid goes up to the teacher’s desk and can visually inspect (but not touch) the transparent jar, which is filled to the brim with nickels.

    Then each kid privately writes down a bid. It is a second-price auction.

    So, if we model this using standard auction theory with rational expectations blah blah blah, then the kids take into account the “winner’s curse” and realize that if they have the winning bid, then it means 29 other kids wrote down a lower number. Since the kid knows “he is not special” (he reads Scott Sumner after all), he doesn’t actually write down what he thinks is in the jar. Instead he adjusts for the winner’s curse and writes down a lower number.

    If we then switched settings to an auditorium freshman lecture with 500 students, then the students downwardly adjust their bids even more significantly.

    So would you say in this setting that the distinction between the private estimate and the “20,000 foot view” estimate is impossible to make?

    Suppose a student thinks there is $30 of change in the jar, based on his own calculations, but then learns that the median guess of everyone else is $22, with a maximum guess of $26. Wouldn’t the student adjust downward his OWN subjective estimate?

  47. Gravatar of Bob Murphy Bob Murphy
    30. May 2016 at 20:25

    Shoot I realize I was a bit unclear myself in the previous comment, Scott. What I’m saying is that your remarks about the two types of estimates being “incommensurable” seems weird to me. If you have a well-specified model of imperfect information etc. then it’s crystal clear why someone might say, “Yes, I am the high bidder on this oil lease because I think there is more oil here than anyone else in the industry. We all are relying on the same sources of information, and I am getting a more optimistic ‘signal’ than anybody else. This *has* to be true of the person who ends up winning the auction. We have all adjusted for this phenomenon and are bidding rationally, and I have no reason to suspect that I am overbidding for this property.”

    Is your understanding of the economics profession and its opinion of monetary offset analogous to Bayesians bidding on a federal lease for oil development, Scott, or are you thinking of something qualitatively different?

  48. Gravatar of Postkey Postkey
    31. May 2016 at 04:19

    Thanks for that.

  49. Gravatar of Postkey Postkey
    31. May 2016 at 04:29

    “The portfolios look like small-cap stocks which will have higher transaction costs.”

    “All transactions take full account of prevailing bid-offer spreads. Commission is charged at a rate of 1.95%
    on deals up to £10,000, less thereafter. No credit is taken for dividends paid by companies nor for interest on
    cash balances. Current holdings are valued using mid-market prices.”

  50. Gravatar of Matthew Waters Matthew Waters
    31. May 2016 at 10:35

    Thanks for that Postkey. Sorry I didn’t see it.

    I just found it particularly interesting that their own mutual fund invested broadly in larger tech stocks, instead of the newsletter portfolio. Although it uses the current bid/ask spread of these small-caps, large purchases would also move the bid/offer up significantly. Then sales move the bid/offer spread down. The Portfolio also had a lot of variability, so excess returns include risk premium.

  51. Gravatar of Tom Brown Tom Brown
    31. May 2016 at 13:51

    If back in 1290, we wanted to know whether it was witches or hobgoblins that caused disease and crop failure, the optimal guess would not have been derived by asking a single physician, but rather setting up a prediction market.

  52. Gravatar of Tom Brown Tom Brown
    31. May 2016 at 14:05

    “Justified true belief” is one way to describe knowledge. I don’t see a problem with it.

    It’s justified, in that the preponderance of evidence supports the belief rather than disbelief.

    True, in that the belief is true (independent of the available evidence or who believes it)

    Belief in that someone (the one with the “knowledge”) believes it.

    This guy debunks the arguments of the so-called “presuppostional apologists,” but in this part (~1 minute) he covers the concept of “justified true belief” with a simple analogy (starting here).

  53. Gravatar of ssumner ssumner
    31. May 2016 at 17:29

    AN, You said:

    “Monetary policy is implicitly fixed at 2% inflation targeting, so apparently the only independent variable is fiscal policy.”

    You’ve got it exactly backwards. If monetary policy is fixed at 2% inflation, then fiscal policy is powerless (to affect demand).

    Matthew, Thanks for the links. On the voting question, it also depends how the political system is designed. The better the design (i.e. Switzerland) the better that voting works.

    Craig, I suppose it depends on the question. If it’s an economic question then it might be the case that the economist consensus and the market consensus disagree. In that case I go with the market consensus, which I view as superior.

    Thus suppose that in late 2008, economists thought monetary policy was appropriate, but markets thought it was too tight. I would have gone with the markets. I only rely on experts when markets don’t exist. I’d even prefer markets over physicists on some fundamental physics question.

    Bill, Yup, if it’s an issue then people disagree.

    mpowell, I agree, but in those cases the crowd of experts will usually beat an individual expert.

    Bob, When money is involved people often switch to the 20,000 mile up view, although in the case of auctions I think people usually tell themselves “This is risky, so my bid better allow for a big profit margin” not “there is a winners curse, so my bid better allow for a big profit margin.”

    I’m not sure that applies to things like my belief in market monetarism, which does not expose me to financial risk. I have an incentive to report exactly what I believe, whereas the bidder on the oil lease will bid substantially below what he think’s it’s worth.

    But it may have something to do with why people often don’t want to bet with someone like Brian Caplan. The 20,000 miles up view is looking over their shoulder, and whispering in their ear not to bet. (Although my reluctance to bet is different, I prefer to bet with abstract markets, and not have to worry about collecting on a bet.)

    Tom, You said:

    “True, in that the belief is true (independent of the available evidence or who believes it)”

    Rorty and I have trouble understanding what that means, except perhaps as a prediction of future belief.

  54. Gravatar of Tom Brown Tom Brown
    1. June 2016 at 08:13

    Scott, I’m assuming there’s a reality independent of the existence of any humans or other beings capable of belief in the reality.

  55. Gravatar of Tom Brown Tom Brown
    1. June 2016 at 08:15

    QBism doesn’t make that assumption AFAIK.

  56. Gravatar of ssumner ssumner
    1. June 2016 at 17:05

    Tom, I’d encourage you to read Rorty. The question is not whether there is some sort of reality, it’s whether we know things to be true, and what that statement means. How do these statements differ?

    1. I believe X.

    2. I believe X is true.

    3. X is true.

    I claim they are all the same statement.

  57. Gravatar of Gene Callahan Gene Callahan
    1. June 2016 at 19:30

    “If my philosophy is wrong then my market monetarism is equally wrong. Either the wisdom of the crowds is true, or it isn’t.”

    Sorry, another example of bad philosophy. Because Joe is good at diagnosing plumbing problems doesn’t mean he is good at coaching an NBA team. Because crowds are good at picking stocks doesn’t mean they are good at determining the nature of being!

    Another example is the last trio of statements. 1 and 2 are about your beliefs. 3 is not. It is almost trivially to see that 3 is about something different than 1 and 2. E.g., “Last year, I believed the square root of two is a rational number. But even back then, it was true that it isn’t.”

    If 1, 2 and 3 are the same statements, then I just babbled nonsense. But clearly that wasn’t nonsense, so 1, 2 and 3 can’t be the same statement.

    I encourage you to read Green, Bosanquet, Collingwood, Whitehead, Blanchard, Croce, McTaggart, Ryn…

  58. Gravatar of ssumner ssumner
    3. June 2016 at 12:42

    Gene, You said:

    “Another example is the last trio of statements. 1 and 2 are about your beliefs. 3 is not.”

    Of course it is. If I say “X is true” then I am expressing my believe that X is true? Why is that so hard to see?

    I encourage you to read Rorty.

  59. Gravatar of Those Silly Austrians Keep Making Econ 201 Errors Those Silly Austrians Keep Making Econ 201 Errors
    25. June 2016 at 20:37

    […] the ultimate last point I’ll make: In other posts (e.g. here), Scott said that you shouldn’t trust his personal views on monetary theory, but instead you […]

Leave a Reply