Home > Philosophy > Every Model is Wrong

Every Model is Wrong

We have been discussing Austrians and their relationship to “mainstream” economics here lately and that topic raises a lot of issues.  I want to go into some of those issues in depth but I want to address them in a very broad way that isn’t really about Austrian economics.  I want to address some deep philosophical questions surrounding the nature of models and their role in economics and the mindset of many people in various “heterodox” schools including Austrians but also Marxists, post-Keynesians and so on.


In general, models serve one or both of two purposes.  They can be explanatory and exploratory.  By the former I mean that a model can be used to explain some concept that the creator has in mind in a (relatively) simple way to someone who does not necessarily fully see or understand it to begin with.  By the latter I mean that it can be used to explore a question which the creator of the model does not know the answer to.  Now clearly, a model which is originally exploratory, typically becomes an explanatory tool once it has answered the questions it is designed to answer and the function of explanation works in essentially the same way in the mind of the “student,” which is to say that it walks them through a series of logical steps which take them from a set of premises that they already “knew” to a set of conclusions which they previously did not know.  So every model is in some sense explanatory and exploratory, the distinction is in the mind of the beholder.

But the important thing to consider is the mindset of the person developing/working with the model.  Are they designing the model in order to explore some questions that they don’t know or to explain some concept that (they at least think) they know?  In my view, either one is fine.  Probably, the best models are created for the purpose of explaining a concept that is already, at least partially, understood by their creators.  However, what I think is imperative is that, when developing and working with a model, the modeler remain at all times an explorer.  By this I mean that he must be open to the possibility that the model will reveal something he had not previously understood.

This is the line between a scientist and a rhetorician.  It is also analogous to the (easier to understand) process of observation.  If a rhetorician thinks that an increase in X leads to an increase in Y, their goal is to convince you of that and they select data that reinforces their argument and try to ignore, downplay or obfuscate information that contradicts it.  (For an example of this turn on any cable news network at any time.)  If a scientist thinks that an increase in X leads to an increase in Y, they go through a careful process of gathering and analyzing data that can either confirm or refute that hypothesis and if the data refutes it, they acknowledge that outcome.

In modeling, we have a similar situation.  You can try to make a model that shows that an increase in X leads to an increase in Y but if you get in there and suddenly discover that this is not the case in your model, do you say “surprisingly, the relationship between X and Y in this model is not that which I previously speculated,” or do you say “this model isn’t working” and tweak it to get what you want or throw it out all together?  (For the record, it’s not necessarily bad to tweak it but you should notice that you have to do that because something didn’t work exactly the way you thought and this should increase your understanding.)

Science vs. Rhetoric

The same dichotomy applies when evaluating an existing model.  We all approach a new model with some set of preexisting beliefs (though the beliefs may be of varying intensities).  The question then is: how willing are we to be convinced that our beliefs are incorrect?  If someone answers “not willing at all under any circumstances,” then they are not acting as a scientist.  This may be okay in certain circumstances but it must be acknowledged.  And if you are, for instance, trying to learn economics from the internet (or from a fiery heterodox professor) you will do well to identify who you are dealing with.

Now if you encounter a model and you are willing to be persuaded that your previous belief is wrong, that doesn’t mean that you must be persuaded by the conclusions of the model.  The purpose of a mode is that it meticulously identifies which assumptions or axioms lead to which conclusions in a context in which the viewer has complete knowledge about the processes under examination.  So if the conclusions of a model are “wrong,” meaning that they are not consistent with real-world relationships, then it must be either because there is a logical flaw in the model or the assumptions of the model do not hold in the real world. (Usually with established models, it is the latter, a case of the former is “the law of diminishing marginal utility” which is why that one bothers me so much.)

But if you are looking at a model and you don’t agree with the conclusions because you don’t agree with the assumptions, there is a productive response and a non-productive response.  The productive response is to address exactly how the supposedly faulty assumption is driving the supposedly faulty result in the context of the model and show how it would be different with different assumptions.  The non-productive approach is to just say “well they assumed X, and that’s a silly assumption so we don’t need to pay attention to what they are doing.”

The former approach is how science evolves.  You try to adopt the most useful framework for thinking about issues and then you get in there and see what happens under different sets of assumptions.  If there are issues that you think are important which are difficult or impossible to address under the existing framework, you try to create an alternate framework to address them.  The latter approach is just a way of giving yourself and others a justification for ignoring a conclusion you don’t like.

And this is the problem with most (not all) “heterodox” thinkers.  They devote their energy to tearing down systems that exist because they have shown themselves to be the most useful for understanding something rather than building up a better system of understanding.  This is also why they become “heterodox” because people working within the mainstream framework see nothing of value in what they are saying and their frequently misguided attacks on the mainstream framework provide ample license to dismiss them if you truly understand the way that framework works and why it is the way it is.

The reason many (not all) heterodox thinkers go down this path is that they are rhetoricians more than scientists.  This means that they are looking for the quickest, simplest way to dismiss people who disagree with them and to inoculate themselves and their followers from any type of thinking that might lead them to these alternate views.  In short, they are trying to win an argument rather than find the truth.  I like to think of it in terms of an electrical analogy.

When confronted with a conclusion which is different from our preexisting belief, our mind is put into a situation similar to an electrical charge suspended in midair, we are in a kind of intellectual disequilibrium (my apologies to people who actually understand electricity, it’s just an analogy).  Our mind has to find a path to ground, and it prefers the quickest, easiest path.  The important distinction is in what we consider ground.  For most people “ground” is our original beliefs, so we look for the quickest path back to them which can usually be supplied by saying “well this assumption is probably wrong, so the model is probably ‘wrong,’ so I don’t need to pay attention to it, I will just keep thinking what I already though, and by the way, so should you.”

What makes you a scientist is that your “ground” is not whatever you previously thought but rather it is the truth.  So whenever this happens, you have to question whether what you thought before is the truth or whether it is something different that the model is trying to direct you to.  In order to figure this out, you have to really understand what the model is doing, not just dismiss it at the first opportunity. This path is has a lot more resistance.

Economic Models

The path of the scientist is never really in “equilibrium” because we never know the whole truth.  And every model is wrong.  At least this is true (and especially so) in economics.  This is how you can tell whether someone is a scientist or a rhetorician.  The scientist knows that his knowledge is incomplete.  He is not looking for a model that is “the correct model,” one in which all assumptions are true and nothing that is relevant in the real world is missing.  This model is not possible.  If you had that model, you would be God (at least the omniscient part, if not the omnipotence).

This fact is not nearly as clear in the harder sciences where they have the luxury of creating models which precisely predict outcomes in the real world.  A physicist can employ a model that will tell you exactly where a cannonball will land when fired from a given cannon with a given load and given wind/temperature/atmospheric pressure/etc. conditions.  If you fire the cannon, the outcome will most likely be a little off because those conditions will not be exactly what was hypothesized but they have the ability to do experiments where they control such variables and they can zero in pretty close on it.  For the same reason, a chemist can tell you (more or less) exactly what will happen if you mix certain chemicals and this can be incorporated into an industrial process the produces a certain chemical incredibly consistently.

In economics, we seldom are able to achieve this degree of precision in the real world because the starting point of our analysis is not relationships between things that are easily observable, controllable and quantifiable but is rather the subjective feelings inside the minds of countless individuals which cannot be independently observed, controlled or quantified and these interact in incredibly complex ways.  The purpose of an economic model is to put some structure to the way we think about these interactions so that specific (usually qualitative) conclusions can be reached.  But, again–and I can’t stress this enough–those conclusions are always “wrong” in some sense because the model is a highly abstract analogue for the real world.  This doesn’t mean the models are useless.

If you hire a physicist to predict where your cannonball will land and then you fire it and it lands an inch to the left, you wouldn’t say “clearly, the model is wrong, you’re ‘physics’ is all nonsense.”  And yet, that’s what many people do with economics (although, admittedly, economic models usually don’t get that close.)  The model helped you get a lot closer than you could have without it.  If you think you can make a better model that will get closer, then by all means do that, in this case you are a scientist.  But if you are trying to make a model that says the cannonball will go where you want it to go, or a model that says you can’t know where it will go–if you are just looking for a way to invalidate the model with no better alternative, then you are just anti-physics.  In that case, don’t be surprised if physicists don’t want to talk to you.

Beware the rhetoricians in long robes.

Here are some examples of attacks on economics that get us nowhere.  If you are reading criticisms like this, you may be dealing with a rhetorician and not a scientist.  Exercise caution.

1.  Economists assume people have complete knowledge but they don’t.

We all know that.  The question is what are you going to do about it?  This doesn’t mean that a model showing what would happen if they did have perfect knowledge is useless.  It is actually a very useful jumping-off point for considering what might happen if they don’t.  Many “mainstream” economists have tried to create models in which people don’t have perfect knowledge but you have to somehow explicitly define what they know and don’t know and what they believe about what they don’t know in order to get any useful implications out of it.  If someone just say “people don’t know everything so we can’t do anything” then they are not being productive, they are just trying to tear down the existing framework.

2.  You can’t use probability to represent peoples’ beliefs.

This follows on from number 1 and I may do a more complete post on probability in the future but, as I said above, if you are going to have a model in which people don’t know everything, you have to somehow define what they believe.  If someone just says “they have no idea” then there is no way to logically infer what makes sense for them to do.  In other words, they are just saying that it is impossible to make any sense of anything so we shouldn’t bother, we’ll just keep believing whatever we believe.  If you can think of a better way of defining what people believe, there’s probably a Nobel prize in it for you.

3.  Markets aren’t always in equilibrium.

Again, there are lots of “mainstream” models in which markets are not in equilibrium.  The key point is that such markets always try to give a specific reason why they are not in equilibrium.  Instances of this abound even in introductory econ and include price controls, taxes, money illusion, sticky wages, adverse selection and moral hazard.  But these models all have some concept of “equilibrium” in the model which may or may not always correspond to the competitive, Walrasian market equilibrium because without such a concept it is impossible to reach any kind of conclusion about what would happen.  If you have no equilibrium concept, then you are just making arbitrary conjectures or else you are just saying “anything could happen” so I will presume that the thing I already thought would happen is the thing that will.  This approach is not productive scientifically.

4.  You can’t aggregate individual preferences/beliefs/behaviors.

Yes you can.  The question is how best to aggregate them.  If their criticism amounts to “people aren’t all the same,” then yes that’s true but nobody is saying they are, they are probably just imagining that they are in order to illuminate some other principle.  There are legitimate “fallacies of composition” but in those cases, it is necessary to show how what is being done specifically is actually a fallacy and how specifically it is driving the supposedly erroneous result.  If they are just saying “we can’t aggregate anything so we can’t do macroeconomics” then they are not being productive.

5.  Money is endogenous.

Have been over this ad nauseam, so won’t say too much more about it but if this is their argument (and to some extent it is my argument), then it is incumbent upon them to explain why the process of money determination gives a meaningfully different result than a model which takes it as exogenous, and not to just say “it’s not exogenous so your model is wrong.”  So far I haven’t seen that model.  I also haven’t created it yet but am working on it (it’s not all that easy).

6.  There’s too much math.

If in other sciences we should arrive at certainty without doubt and truth without error, it behooves us to place the foundations of knowledge in mathematics.

-Francis Bacon

I happen to agree, generally, that there is too much focus on mathematical rigor in economics and not enough on understanding the meaning of what we are doing.  But there is a good reason to place the foundations of our knowledge in mathematics.  Mathematics is pure logic.  The more you can reduce your logic to mathematics, the harder it is to argue with it.  This highlights the assumptions and the conclusions of the model as the important things rather than having to wonder if each logical step we made along the way from one to the other was really entirely logical (see the law of diminishing marginal utility).  If you are only trying to convince someone who already agrees with your conclusion, this may not be necessary but if you hope to change anyone’s mind about anything, it helps to work within an agreed upon logical framework.  And again, if they are just saying “it’s too mathematical, it can’t be right,” they are probably just looking for an excuse to ignore it without really understanding it.

This criticism is general, it is not Austrian-specific.  I just tend to focus on Austrians (or at least a certain type of person typically considers themself “Austrian”) because I have more experience with them and because I am ideologically close to them.  This is my general perception of this type of person (call them “Rothbardians,” or “pop-Austrians,” or “internet Austrians,” or “Mises.org acolytes,” or whatever).

Their primary motivation, the reason they got into economics, is that they are suspicious of government and think most forms of government intervention are bad.  They see a lot of “mainstream” economists making economic arguments for the types of interventions that they think are bad.  They find these arguments outrageous.  They gravitate toward ways of thinking that offer the easiest path to dismissal of all of those arguments which they find outrageous.  They discover Mises.org and other blogs or commentators that offer them a plausible way to believe that they understand what the “mainstream” is doing without actually understanding it and dismiss it all with a few waves of hand and go on believing what they already believe.

Now please keep in mind that I agree entirely with those beliefs.  I just think that they are throwing the proverbial baby out with the bathwater and that this is not a scientifically productive approach, even though it is having some rhetorical success.  And what’s more, you don’t need to do this!  The neoclassical framework offers plenty of room to support free markets.  The same phenomenon can be observed in more leftists heterodox ideologies, post-Keynesians for example, but we have to clean up our own house first.  Rhetorically, It’s easy to convince people who already agree with you that you are right.  To convince people who don’t, we have to be scientists.


P.S. Some disclaimers:

1.  There are, apparently, different breeds of “Austrian” out there.  This does not necessarily apply to all “Austrians.”  If you fancy yourself an Austrian and you aren’t doing these things, I have no beef with you.  In fact, I would love to have a few beers with you and discuss how great free markets and entrepreneurs are and how to save more souls from Mises.org.

2.  An argument could be made that many like Noah Smith are doing the same thing to the reasonable Austrians by pointing to crazy things that some self-proclaimed Austrians say and using it to convince people to dismiss anything any Austrian says.  That is not my goal here.


  1. Chris
    July 8, 2014 at 6:41 am

    Good stuff.

    So I want to offer a way to think about how all the things that you’ve been talking about (and a little about what I’ve been telling you).

    Mises adopted the use of praxeology from earlier Austrians (who got it from other philosophers who got it from the Greeks) to offer an alternative to the historicism being presented by the German School. Mises saw the emergence of logical positivism as just another form of historicism so that is the beginning of that whole issue.

    With logical positivism losing its luster due to the work of Kuhn, Popper, Quine, etc., the main conflict came between Mises’ Kantian a priorism and Popper’s falsification. There are Austrians who believe that Mises’ praxeology is compatible with Popper’s falsifiability, but they are few and far between. For a more academic division between Mises and the positivists and Popper, see “Was Mises Right?” Two other things should be noted here: Mises’ use of a priorism in economics was not uncommon at the time (Lionel Robbins and Frank Knight were also a priorists.), and Hayek arguably never accepted Mises’ a priorism and instead was more influenced by Popper and Quine. However, Hayek wasn’t completely comfortable with falsifiability in economics, which may go back to Quine’s influence (as explained in the Boettke piece I just mentioned.).

    Anyway, after the end of the great influence of logical positivism, modern Austrians became more focused on the issues with formalism. As you stated, there is an issue with the over use of formal mathematics. This isn’t just a position held by people like you or Austrians, It is also a fairly large deal to philosophers of science. The nuanced way to put this is that formalism closes off certain kinds of communication that could be used when formal mathematics fails to express specific issues intelligibly. To use one of your examples, the issues that Austrians have with the knowledge assumptions and the work done with them (and this is a concept of the highest importance to Austrians) isn’t so much that perfect knowledge is divorced from reality (while that is of some importance), the real issue is that even when that assumption is relaxed, formal mathematics lacks some of the ability to adequately discuss the nuances of the transmission mechanisms and subjectivity involved with knowledge. So there is a desire to combine the formal models with what are called “analytical narratives” that were once the standard of discourse due to their explanatory power and relative rigor. So, at least from the Austrian side, the critique is never a simple “hurr people don’t have perfect knowledge dummy.” It’s a relatively substantive critique.

    The same thing goes for the other points of which you worry. For instance, Austrians the issue of how equilibrium models fit into Austrian work itself is still up to debate. There is some concern as to how equilibrium models have affected the actual thought of the profession. However, the majority of Austrian work is heavily influenced by disequilibrium economics and monetary disequilibrium theory. Robert Clower and Axel Leijonhufvud are required reading. Their work on the importance of disequilibrium to Keynes is relatable to the project that Hayek was attempting to show about the error-detection capacity of the non-equilibrium prices and thus the adaptive properties of the price system. Once again, the concern that these Austrians have is that this type of work doesn’t necessarily map well to extreme formalism and they worry about what is lost when that is the only language being used.

    Of the rest of your concerns, I don’t really find much to comment on except to mention that there are a bunch of people now doing empirical tests on Cantillon Effects.

    So as you can see, there is a whole lot going on here with the methodology, philosophy, and sociology of the profession…most of which I tire of pretty quickly. For what it’s work Ronald Coase made similar points to what the Austrians made in an essay entitled “How Should Economists Choose?”. It’s a good piece and discusses his rather deep disagreements with Friedman’s “Methodology of Positive Economics” which center around the importance of intelligibility from models as their central purpose instead of predictive power. By the way, one of Hayek’s regrets was not engaging Friedman on that paper.

    • Free Radical
      July 8, 2014 at 5:45 pm


      Good stuff here. As I tried to put into a disclaimer, I don’t mind people doing things outside the mainstream (like trying to use a more narrative approach) as long as they aren’t also leveling baseless attacks on mainstream models. I do think that having to express something mathematically imposes a degree of discipline (it’s harder to fool yourself with math than with narrative, and of course, as Feynman said, you are the easiest one to fool). But if you think something can’t be properly explained with math, then feel free to try something else. I’m not convinced that this is the case with uncertainty but I’m not trying to shout anyone down. But I don’t think we should look at what other people are doing and criticize it for being too mathematical. Probably (hopefully) this kind of thing isn’t too prevalent in the halls of George Mason but it is all over Mises.org.

      For the love of God, don’t mention Cantillon effects, you might start a world war! (In case you missed it, this was sort of a watershed moment for me where I lost all faith in Austrians–yes I mean ‘internet Austrians’–listening to reason.)

      Will have to look at the Coase paper. I love Coase. =)

      • Chris
        July 8, 2014 at 8:04 pm

        I’ve never seen anyone in an academic setting be criticized for being too mathematical. The problem with formalism isn’t the use of math. It’s when there is a lack of economic intuition in the models. And note that when I say formalism here I don’t mean all formalism. I’m referring to the section of which philosophers of science have a problem

        I have, however, seen people be shouted down for “not doing economics” when there models were deemed not mathematical enough (this occurred at a conference and between professors).

        Haha. I did miss that. I don’t usually frequent blogs to seek out these types of things. But I don’t usually stray too far into monetary or business cycle stuff so I promise not to start any wars over that.

        We constantly here a Friedman quote in class: “there is only good economics and bad economics.” I hope that I’m conforming to the former.

      • Free Radical
        July 8, 2014 at 11:36 pm

        I think you’re on the right track.

    • Free Radical
      July 8, 2014 at 6:30 pm

      One more thing. Regarding equilibrium, there seem to be two things that people mean by equilibrium and I am often not sure which one people mean (and I suspect that often they are not sure either). There is what you might call “market equilibrium” or “Walrasian equilibrium” meaning, the state in which all margins are equal and everything is Pareto efficient. This is a special case of equilibrium. But essentially every model has some sort of equilibrium concept, even when they don’t result in the Walrasian equilibrium. It is difficult to imagine an economic model without one. Perhaps the stuff that Steve Keen is doing (that I have been talking about here) qualifies where he just puts arbitrary values into an arbitrary system of differential equations and then watches it evolve over time. But there is no economics in that, meaning that there is no treatment whatsoever of thinking individuals making decisions between scarce alternatives. Maybe a hard-core, anti-equilibrium Austrian would say “yes and that’s why we shouldn’t use mathematical models” but that strikes me as awfully extreme.

      • Chris
        July 8, 2014 at 8:11 pm

        Well right that’s the same criticism that is put towards Representative Agent Models. And it’s true. But the response has been positive (even from Austrians) to the work being done (in New Mexico or Arizona) where the’re trying to give the representative agents something that represents a form of agency.

  2. John S
    July 8, 2014 at 6:36 pm

    Thoughtful post as always. I agree, too much of the blogosphere is going for ideological inoculation rather than scientific analysis.

    Just to change gears a bit, I had a few (very) rough and random thoughts while looking at one of JP Koning’s old posts. I’d appreciate any input you could give (please don’t feel obligated to respond point by point).


    I didn’t really know what “front running” was before, but I get the gist now. In a nutshell, my question is: by announcing pretty much exactly what and how much it’s going to buy (in terms of monthly targets), isn’t the Fed almost certainly carrying out QE in a wrong/exploitable fashion according to basic game theory?

    Let me back up a bit. Generally, Sumner seems quite scornful of the idea of asset price “distortions” (i.e. bubbles) caused by Fed policy. He admits that asset prices swing up and down quite a bit, but that’s just the EMH in action, and anyway the Fed always buys assets at market prices, so what’s the problem?

    I have two quibbles with this.

    1) Every time the Fed buys assets, it’s adding demand to that market that wouldn’t otherwise exist and it thus pushes the price of that asset up by a few pips. Also, I believe that Bernanke and the FOMC have directly stated that their goal is to boost asset prices and stimulate consumption via the wealth effect. So I’m thinking, A) Why not believe the Fed’s own statements that it’s trying to move asset markets? and B) Isn’t it almost meaningless to call the resulting prices “market prices” since the Fed isn’t constrained by profit/loss and liquidity/solvency issues in the same way as other financial institutions? (although I realize there’s a debate about whether CB solvency matters)

    2) By pre-announcing its purchasing “strategy,” isn’t some level of front running (and overpaying for assets) inevitable? As JP notes, the Fed is a massively large and dumb trader, and under QE3 it’s committing itself to acting that way for the indefinite future. How can this not: A) attract capital and investment from other sectors by traders looking to exploit this? and B) make it that much harder for investors to disentangle the “real” info that asset prices are conveying about future prospects from the “signal fuzzing” being introduced by Fed actions?

    I don’t know that much about game theory (I’m guessing you do, since your focus is micro), but I do know that a basic principle of optimal strategy is information hiding. So in a game like poker, revealing your cards would obv be a dumb move, but so would adopting any easily predictable pattern such as betting small with your bluffs but large with your value hands. To bring it back to the Fed, it would be bad enough if it engaged in an easily detectable pattern of purchases every month, but by stating its targets, it’s effectively laying its hand face-up, leaving itself ripe for exploitation.

    Another game theory concept, at least in poker, is balance. To stretch the poker analogy (perhaps to the breaking point), for any given bet size there is a mathematically “correct” mix of strong, medium, and weak value hands and bluffs which yields the maximum expected value against the full range of counter-strategies. Any deviation from this proper mix is sub-optimal relative to a strategy that more closely approximates the correct balance, regardless of how well the player conceals information about his hand strength (in other words, the more balanced player would have a higher EV even if he wasn’t paying attention to the unbalanced player’s strategy at all).

    I think Fed asset purchases can lead to “unbalanced” outcomes in two senses. First, given the “true” conditions underlying the productive capacity of the economy, there must be some maximally efficient (but unoberservable) mix of investment across all asset classes. Fed asset purchases could in theory be moving the investment mix either toward or away from this portfolio, but I think it’s reasonable to suspect that capital being diverted to front running is not being deployed in an optimal manner. Second, the Fed is unbalanced since it is constraining its range of options. To avoid exploitation by front running, the Fed should be capable of performing any action[1] — buy, sell, or be inactive — at any time, in any asset market (not just Treasuries and agency MBS). In a free banking system, all financial institutions would act like this, since they would be extremely careful to avoid exploitation by overpaying, which I think would lead assets to being closer to their true values. But giving the Fed this much discretion seems potentially dangerous.

    So the Fed is in a catch-22: act transparently (but encourage exploitation and possible distortions in asset markets) or be secretive (and open up possible avenues of corruption). In a free banking system, financial institutions would want to take advantage of all profitable investment opportunities, but they would also be constrained from overextension by profit/loss & liquidity/solvency issues, which would lead to asset prices that provided more accurate information on the true productive capacity of the economy.

    I have a couple other thoughts on this, but this comment is already salad, soup, and spaghetti mixed in a pot, so I’ll stop here for now.

    [1] Just as a good poker player should be able to credibly represent having a balanced mix of strong, medium, and weak hands in any betting sequence.

    • Chris
      July 8, 2014 at 8:15 pm

      I can’t find the links right now, but John Cochrane has some posts (and links to his publications) about these issues over at grumpy economist.

    • Free Radical
      July 8, 2014 at 8:19 pm


      I will try to go through this but I might take a break in the middle.

      First, announcing purchases ahead of time is not necessarily “exploitable” because it is announced to everyone simultaneously. This means that everyone has an equal chance to evaluate the effects of the purchases and adjust their behavior accordingly. It may (I don’t think this is so clear) mean that the Fed faces a less favorable price than if they tried to do it covertly but that is not their goal so I don’t think we can say that it is “wrong.” If you are trying to front-run another trader, the benefit to you comes from obtaining a unique bit of knowledge about their future behavior that the rest of the market doesn’t have. Since the Fed gives it to the whole market, it doesn’t give anyone in particular any kind of unfair advantage. Of course, if you can process the implications faster than everyone else, you might be able to get the quicker and benefit from it but that is basically how markets work.

      Now “assets” could mean a lot of other things, and indeed, their stated goal is to get inflation up to 2% which means raising the value of “assets” in general, but that is different from the price of the financial assets which the Fed actually buys.

      Regarding your first two quibbles:

      1) First of all, you have committed a cardinal Sumner-sin. With nearly every other asset, an increase in demand causes an unambiguous increase in price but if the asset is a bond, and the demand is from the Fed, it may either increase or decrease in value because, in addition to adding demand in the present, they are changing expectations about monetary conditions in the future and this may actually lower the price (raise the yield). For instance, imagine that the announced increase in demand causes people to think that inflation will spike. In this case, they will want to get rid of the bonds (You can think of it as an increase in supply but be careful with supply and demand for bonds, the line between them is pretty ambiguous) and the yield may rise to account for the increased expected inflation.

      I don’t think it is right to say it isn’t a market because the Fed isn’t motivated by profit. The market is still there. The Fed does what it does and then “the market” which is made up by many firms and individuals acting through mutually voluntary exchange motivated mainly by the pursuit of profit, reacts. Certainly, the market outcomes are different than if the Fed were operating from a profit motive, or if the provision of money were left to “the market.” But we can draw the boundary of “the market(s)” such that the Fed falls outside of it without denying the existence of a market all together.

      2. The concept of “attracting capital” is actually problematic. For one thing, typically in economics we take “capital” to mean physical means of production, which of course, the Fed does not attract. But even if you take this to mean “financial capital,” it doesn’t really attract it either. There is a tendency to think about money as being “tied up” in various assets because from an individual perspective that is how we think about it. But in the aggregate, money doesn’t change into stocks or bonds, etc., it just goes from one person to another. The prices of these things are determined by the willingness to trade them for money on the margin, not how much money is “in” them. These prices can vary greatly with very little “flow” of funds from one asset to another.

      Now a case can be made, at least in the short run, that by distorting the marginal conditions, monetary policy can “misallocate” capital. In essence, that is the whole point of monetary policy in the Keynesian tradition, except that they say it is already misallocated in one direction and they are just trying to nudge it back in the right direction with monetary policy. However, I think the “pulls capital away from other sectors” argument is, at least, not as clear as you suggest.

      As far as fuzzying up the signals in the market, I think there is some potential for that (especially since one of the main components is future inflation expectations which depend highly on future Fed policy which we must glean from their attempts to telegraph it) but remember that traders still have every incentive to try to extract the appropriate signals as much as possible and the market still functions in what I would consider a pretty efficient manner to aggregate their estimates into market prices which reflect those judgments. This doesn’t mean that the market is always “right” about the future of course, but it does mean that if you can do better, there’s a lot of money in it for you and that is a powerful method of estimation.

      Okay, lunch time. Will be back for more later (I do like a good game theory question).

      • John S
        July 9, 2014 at 12:32 am

        When I say exploitable, I’m not talking about traders being able to exploit one another, I’m talking about the Fed possibly overpaying for assets by making its intentions transparent. This might lead to a misallocation of financial capital by banks and other institutions looking to profit from this (although as you say, the idea of “pulling capital away from other sectors” is problematic; this has been a confusing sticking point for me).

        their stated goal is to get inflation up to 2% which means raising the value of “assets” in general, but that is different from the price of the financial assets which the Fed actually buys.

        As I mention in comment below, I’m still not exactly sure about the chain of events from increased financial asset prices to increased inflation, esp. in today’s environment. I really do want to understand this, so any help would be appreciated.

        if the asset is a bond, and the demand is from the Fed, it may either increase or decrease in value because, in addition to adding demand in the present, they are changing expectations about monetary conditions in the future and this may actually lower the price (raise the yield).

        This is really interesting (and confusing! 🙂 I don’t read Sumner everyday, so I don’t remember any specific posts about this. But I’m re-reading something from Miles Kimball where he said:

        Whenever the Fed buys any asset, its price goes up. You can think of it as the Fed adding to the demand for the asset or subtracting from the supply of the asset in private hands. Either way, when the Fed buys an asset, its price goes up.


        As long as any asset has a nominal interest rate above zero, buying that asset will raise its price and lower its interest rate.


        So I’m genuinely confused and will have to think about this a lot. I do get what you’re saying, it’s just that I have very bad intuition about bond math, aside from “price up, yield down” and vice versa.

      • Free Radical
        July 9, 2014 at 1:01 am

        Here is the Sumner post you want.

        Regarding the “exploitable” part, like I said, the Fed is not hurt by this (since maximizing profit is not their goal), so I don’t think it is appropriate to call it exploitation. It’s true that resources are spent competing over the profits that can be had by figuring out the implications of Fed policy fastest. In some cases, it may seem ridiculous and wasteful (like when they build their servers across the street from where the announcements are made so their algorithms can trade 1/1000 of a second faster than another guy) but I see this as the maintenance cost for the market mechanism. You can say the same thing about the market for pork-belly futures. If you have any other mechanism (like helicopter drops) you will have people using a lot of resources trying to predict what the Fed will do next and what the effects of it will be. Of course, the more you can make it predetermined and less arbitrary (like an NGDP futures market and level target) the less there will be to predict. But, again, it’s about the overall policy regime, not the way that individual transactions are carried out.

      • Free Radical
        July 9, 2014 at 1:26 am

        Almost forgot about your transmission mechanism question.

        Sumner (although he isn’t thrilled about the concept of transmission mechanisms in general) would say it’s a hot-potato mechanism. This means basically the following: When the Fed buys assets it increases the money base. This means that given the previous prices and interest rates, people end up holding more money than they want to hold so they try to get rid of it by spending it. Since people are trying to get rid of money by trading it for other goods, the price of those other goods in terms of money increase (and short-term interest rates decrease, but long-term interest rates may rise or fall depending on what it means about the stance of future monetary policy) until the quantity of money people desire to hold becomes equal to the no larger quantity of money.

        I don’t have a problem with that treatment but I might say it a different way. Perhaps something like this: When interest rates are lower, people borrow more. They take the money which they borrow and use it to buy stuff (both investment and consumption). This pushes prices up. This seems like overly simplified treatment and it is but I think you can dig into it and find some interesting stuff. I won’t do that here. I will say this is sort of an intermediate step in the hot-potato effect. When banks are reserve constrained and interest rates are significantly positive, it happens automatically when you increase the base. If they have a lot of excess reserves, then the connection between increasing the base and increased borrowing/lending is not so direct.

      • John S
        July 9, 2014 at 1:03 am

        we can draw the boundary of “the market(s)” such that the Fed falls outside of it without denying the existence of a market all together.

        I shouldn’t have used the word “meaningless,” but what I was getting at is that, yes, the prices are determined in “a market,” but those prices are different from what would occur in “the free (banking) market.” So in a free banking world, there’d be some supply and demand for different maturity Treasuries and (if we were unlucky enough to still have GSEs) agency MBS, but those prices would be different from what we have now, and they’d convey more straightforward information about the true underlying value of those assets than they currently do (since these assets now have the extra role of being the means of carrying out QE).

        JP did a post asking if it matters how the CB carries out monetary policy (“Toying with the monetary transmission mechanism,” Oct. 9, 2013). He said, in terms of setting the price level, no. But what if the Fed carried out its monetary policy by purchasing, say, crude oil futures? It might shake out to the same level of inflation, but in the meantime wouldn’t it also massively screw with the heads of oil traders and people buying gas? As GMU Austrian Steve Horwitz says, it adds to the “epistemic burden” of all actors in the market (please don’t be put off by “Cantillon” — he’s not making a “run to the ATM” type argument).


        (Also — and I’m talking way out of my depth here, of course — I’m vaguely aware that Treasuries and agency MBS are widely used as collateral for repo in shadow banking, so I just have a gut feeling that “fuzzing up the signal” around these prices could have pretty wide repercussions.)

      • Free Radical
        July 9, 2014 at 1:37 am

        This I certainly agree with.

        “those prices are different from what would occur in “the free (banking) market.” So in a free banking world, there’d be some supply and demand for different maturity Treasuries and (if we were unlucky enough to still have GSEs) agency MBS, but those prices would be different from what we have now”

        This part is a very tricky road to go down.

        “and they’d convey more straightforward information about the true underlying value of those assets than they currently do (since these assets now have the extra role of being the means of carrying out QE).”

        I think “adds to the epistemic burden” might be an appropriate way of saying it but I am not very familiar with that line of research off the top of my head. It would certainly impact the price of crude oil (though it would depend what they did with it) but that is why they don’t do it with crude oil and instead do it with treasury securities which are not connected to any real good, only money–which of course is the thing they are trying to manipulate the value of in the first place.

        Your last comment is out of my depth too, I don’t know much about “shadow banking.” I wouldn’t be surprised if your general feeling turns out to be correct at some point but I can’t say much specifically about it.

        Tread carefully with that…(haha)

    • Free Radical
      July 8, 2014 at 11:35 pm

      Okay, regarding game theory and poker: This is not quite right.

      “for any given bet size there is a mathematically “correct” mix of strong, medium, and weak value hands and bluffs which yields the maximum expected value against the full range of counter-strategies. Any deviation from this proper mix is sub-optimal relative to a strategy that more closely approximates the correct balance, regardless of how well the player conceals information about his hand strength (in other words, the more balanced player would have a higher EV even if he wasn’t paying attention to the unbalanced player’s strategy at all).”

      If you take a game theoretic approach to poker (and assume that players don’t know each other’s hand) you would find a mixed-strategy Nash equilibrium (at least I suspect you would, I haven’t seen a careful model but it is something I have thought about as a big fan of bot poker and game theory). But this doesn’t mean that the strategies in that equilibrium are optimal against the full range of counter strategies. Let player A’s strategy in equilibrium be SA. What it actually means that the full range of counter strategies by player B are all equally effective against strategy SA. But for any particular strategy by B which is off the equilibrium path, the best response by player A may be something different. In a mixed strategy equilibrium, (where all players play a mixed strategy) all players are indifferent between all their possible strategies. This is one of the things about Game theory that is difficult for many to wrap their heads around.

      Now what actually goes on in a poker game is different from what you would get in a NE anyway because players are not really randomizing. They are trying to appear unpredictable but they are also trying to predict what the other players are doing and act in a purposeful way to counter it.

      Now, I think what the Fed does is not analogous anyway because I think you are mischaracterizing what happens when you call it “exploitation” and things like that. It’s not. It’s what they want to happen. They are trying to move markets. They expect to be “front-run” and have no particular interest in avoiding it. As long as the information that moves them is provided to the entire public simultaneously, it’s not exploitation or corruption, it’s just markets doing what they do. Of course, there may still be problems with centralizing control of the financial system in this way and I’m not advocating it. But I don’t think the real problems are related to the way their orders are executed. They are buried much deeper in the structure of money and debt and are not so much about prices or interest rates varying from the “right” values at any given point in time but more about them varying from what had been previously expected.

      • John S
        July 9, 2014 at 1:21 am

        This is a really interesting comment, and I will have to dig into it tomorrow. Much obliged for the all answers and links.

      • Free Radical
        July 9, 2014 at 1:26 am

        No problem, thanks for reading!

      • John S
        July 10, 2014 at 12:25 am

        What it actually means that the full range of counter strategies by player B are all equally effective against strategy SA.

        I’m not sure what “full range” means here — literally any counter strategy? For example, say player A is playing an equilibrium strategy (I’ll call it “optimal,” though that may be the wrong terminology) in heads-up no limit holdem. What if player B decides to fold every hand? That can’t be equally as effective as the strategy of a good human player, it’d be dramatically worse. I must be misunderstanding something here.

        My only formal exposure to this topic is The Mathematics of Poker by Bill Chen and Jerod Ankenman. You should def check it out; it’s a serious math book, but with comprehensible summaries for people like me.

        Now what actually goes on in a poker game is different from what you would get in a NE anyway because players are not really randomizing. They are trying to appear unpredictable but they are also trying to predict what the other players are doing and act in a purposeful way to counter it.

        Not necessarily. Chen & Ankenman describe two approaches to poker. The first they call “optimal.” It assumes you are playing against “the nemesis,” a player who can perfectly exploit any mistake you make. The way to do this is to imagine how someone might exploit your strategy if you openly described it, and then take steps to fix it. Keep doing this and you’ll have a strategy that is both maximally protected vs. exploitation but also performs very well against non-optimal strategies.

        The other approach to poker is what you mentioned, “exploitative,” i.e. predicting opponent mistakes and taking advantage of them (“for any particular strategy by B which is off the equilibrium path, the best response by player A may be something different”). But there’s a danger in taking the exploitative path — since it represents a deviation from optimal play, it opens oneself up to counter-exploitation.

        Jerod Ankenman sums things up here:

        “Playing game theory optimal/equilibrium strategies is “playing defense.” Estimating your opponents’ strategies and trying to exploit them is “playing offense.” The first important fact here is that playing defense ignores your opponents. Defensive strategies are exactly the following: they maximize against an opponent who exploits you maximally.”


      • Free Radical
        July 10, 2014 at 3:49 am


        On game theory:

        Yes! You are right, I’m the one who is missing something, I tried to generalize from a much simpler game. It’s still not true that the NE strategy is the best strategy against any strategy by the other player (otherwise “offense” would be the same as “defense.”) What I said is only true if the NE strategy is randomizing over all possible strategies in all possible situations. This won’t be the case. I’m shooting from the hip here, as I said, I haven’t thought it through carefully. But if you have the “nuts” you shouldn’t randomize, and there are probably other situations where you also shouldn’t. This makes it more complicated. You only have to be indifferent in situations in which you are randomizing and with regard to the strategies over which you are doing so (so there may be situations in which you would never fold but you randomize between betting small and betting large, and you would have to be indifferent between those in equilibrium). But, like I said, it’s still not necessarily the best strategy against whatever your opponent does. To see this, imagine that your opponent folds to any bet, then you would always want to bet, which isn’t the NE strategy (I presume).

        Now, the first strategy you described is not really “randomizing” as in a mixed strategy NE. Playing a mixed strategy would be like having a list of probabilities of each action you could take in every situation (which would be a very long list) and you would have to follow it not only in aggregate but at each point. So imagine you had a two ten-sided dice and every time the action was on you, you rolled them and acted according to the roll you got and the list of probabilities. This would take any element of strategy out of actually playing the game (obviously deriving the probabilities is a “strategy” but you would follow it mechanically.

        This is a big ball of wax and it’s difficult to jump into game theory with a problem this complicated so I will be impressed (with both of us) if this makes any sense to you. =)

      • John S
        July 10, 2014 at 12:48 am

        Now, should the Fed worry about being “exploited?” For now, I’m not talking about whether it matters if it suffers capital losses. But even if it doesn’t matter, even if the Fed wants to get front-run so that asset prices rise, the fact that it is allowing itself to get front run is going to “attract capital” (tricky wording, I know) from where it would have otherwise gone, perhaps leading to malinvestment and/or fuzzing up of asset price signals. Is this enough to worry about? In normal times, probably not, but when Sumner talks about “buying up Planet Earth, if needed,” it makes me think that it might be worth trying helicopter drops first before engaging in large scale asset prices in special situations like now, when there’s a lack of borrowers and the HPE isn’t working so smoothly.

      • John S
        July 10, 2014 at 1:36 am

        [The real problem is] more about them [prices and interest rates] varying from what had been previously expected.

        But… who can predict when QE3 (or Abenomics) will end? Seems like there could be big discrepancy from what the market expects and what the Fed actually does. And when QE does end, will the bottom drop out of asset prices again?

      • Free Radical
        July 10, 2014 at 4:06 am

        Actually, I think the answer to this is yes, but getting there is a longer (intellectual) road than most people think. I happen to think that in order to keep the bottom from falling out of assets, the Fed (or the government) will eventually have to “buy up the entire world.” That is what I think the problem is with central banks. Helicopter drops would help with that problem. Of course, I think it would be better to just keep the government out it.

        I don’t think NGDP fluctuations are a good reason to centralize control over the macroeconomy but that’s just me. I do have trouble seeing how you could target NGDP with free banking, but I think Selgin sort of has different levels of “free” banking, which include some forms of centralized policy. I need to read up on it more.

      • Free Radical
        July 10, 2014 at 4:08 am

        P.S. They told us when QE3 will end today (October). Sure, they could change their minds, but these things are estimable. Who could predict how much gold will come out of the ground over the next hundred years? And yet the value of gold today depends on that estimate.

      • John S
        July 10, 2014 at 11:59 pm

        Re-reading JP’s post again, I see that I’ve really missed his point, and that “QE via frontrunning” need not lead to distorted asset prices (in fact, Fed openness about its purchases helps reduce signal fuzzing and front running actually speeds up that process). Since no one wants to hold overvalued assets, arbitrage ensures that asset prices rise across the board.

        Still, this discussion has really helped me understand things better and provided me with lots of food for though (and it indirectly introduced me to a great new blog in “Macroresilience”) so I really appreciate you being patient with my questions.

        To get back to poker:

        I had to read both of your paragraphs a couple of times, but yeah, I think I do get what you’re saying! Actually, if I understand correctly, it’s not that different from what I said originally (“mix of strong, medium, and weak value hands and bluffs”) b/c I’m including taking different actions (at randomized frequencies) with the same hand in similar situations when I say “mix.”

        So this would apply to threshold hands, e.g. a weak made hand on the river that’s just slightly too weak to check-call. You might fold it 80% of the time (and all worse hands 100%), but the remaining 20% (determined by some randomizing device, like dice) you turn it into a bluff and check-raise (actually you’d prob choose a hand with card removal effects w.r.t. the nuts, like the bare Ace on a 3-flush board, but same idea).

        if you have the “nuts” you shouldn’t randomize

        Interestingly, prior to the river you should def randomize how you play the nuts so that you can credibly threaten to have it in any betting sequence, both to 1) prevent your opponent from wantonly overbetting the pot; and 2) add credibility to your bluffs in unlikely sequences, like when you check-raise the river. Normally, with the nuts you’d be putting in a check-raise somewhere on the flop or turn, but to play GTO you’d need to show up with nuts at least some % of the time when you check-raise the river to make your bluffs here credible.

        (I think there is an interesting game theory angle to QE in that the “buy Earth” threat [the nuts, so to speak 🙂 ] might not be seen as credible, but I won’t inflict further GT questions on you here).

      • Free Radical
        July 11, 2014 at 2:10 am

        Regarding this “if you have the nuts you shouldn’t randomize.”

        That’s (another) misspeak on my part. I meant that you should never fold (there’s a lot of dimensions to keep track of here). So yeah, you don’t want to do the same thing every time you have the nuts, sometimes you should slowplay it and sometimes you should bet big and try to look like you’re bluffing etc. This is probably the case in all or almost all cases where you would play a hand, there are probably few if any cases in which a NE would prescribe that you make a certain bet with probability 1, but there are cases where you would fold with probability 0.

      • John S
        July 11, 2014 at 12:05 am

        It’ll be really interesting to see if the Fed keeps its promise to end QE3 and how the market reacts to all of this. There certainly won’t be “silence” on Sumner’s blog from now until October, that’s for sure.

      • Free Radical
        July 11, 2014 at 2:12 am

        To my ear, it sounds like they are ready to use interest on reserves in lieu of more QE should the need arise (which I suspect it will). I think the market monetarists will consider that a (minor) victory.

  3. John S
    July 8, 2014 at 11:57 pm

    Just to give you an idea of where I’m headed with this, I’m wondering: Should Sumner and other MM’s start thinking more about direct helicopter drops to households (or Miles Kimball’s Federal Lines of Credit [FLOC] idea, which would be like repayable heli-drops)? Beckworth seems quite open to heli-drops, Sumner more hostile; he often says they were 1) tried and didn’t work in Japan and 2) might cause hyperinflation.

    I might be wrong about “asset market distortions,” but if I’m not, I think heli-drops or FLOC might cause less distortion in asset markets since they would more directly stimulate nominal spending, which is what I believe NGDPLT calls for. Starting with asset purchases first seems like a circuitous route to get to higher NGDP, esp. now that banks are completely flush with reserves.

    (I must confess at this point a gap in my understanding. I’m starting from JP’s view [“The fed funds rate was never the Fed’s actual policy lever,” Aug. 23, 2013] that pre-2008, the Fed controlled the price level by keeping reserves artificially tight and altering the marginal convenience yield on reserves. So for example, open market purchases increase the supply of reserves, decreasing their convenience yield, and induce banks to shed unwanted reserves and buy up other assets.

    I’m not sure of the exact chain of steps leading from increased asset prices to general increases in stickier prices in the real economy such as wages, goods, and services. I’m guessing that, since yields on financial assets are lower, it’s now attractive on the margin for banks to increase lending for projects that wouldn’t have offered a sufficient return prior to the overall increase in asset prices. Does this sound right? Also, I’m doubly confused about how the mechanism is supposed to work now, when banks are flush with reserves, and why it didn’t increase lending when excess reserves exploded after the crisis, though I’m sure interest on reserves didn’t help).

    • Free Radical
      July 9, 2014 at 12:46 am

      I actually think you are making essentially a new-Keynesian argument here, just substitute “fiscal policy” for “helicopter drop.” Keep in mind that, if you don’t like a lot of government spending, you can pursue “fiscal policy” through tax cuts as well. During the Bush administration, they did a “tax cut” that was really a tax credit. In other words, they gave everyone $200 (or something like that). No change in marginal tax rates, no change in incentives whatsoever=helicopter drop. The policy was immortalized in this Futurama episode.

      I don’t think that such a “helicopter drop” would necessarily be any less distorting than open market operations but it would change everything if it became the predominant policy tool because it would disconnect the quantity of money from any form of debt. This really would open the door to hyperinflation and all of the stuff my internet-Austrian friends get all worked up about. Of course, it would make it a lot easier to raise inflation and interest rates and avoid a deflationary recession. In my mind it’s a kind of Scylla and Charybdis situation. Of course Sumner would say you don’t need to do that to avoid a “liquidity trap,” that you can make monetary policy “looser” in other ways (obviously through an NGDP futures market/target, but also QE or negative interest on reserves).

      Regarding bank lending. Yes, I think you are thinking about it the right way: lower yields on other financial assets increase their willingness to lend. I would say that now they are “flush with reserves” because they can’t find enough borrowers who (when applying the appropriate risk premium) are willing to borrow at a high enough rate relative to the rate of interest on reserves. I did a couple posts about this here and here (you may have already read them).

      • John S
        July 10, 2014 at 1:22 am

        I did a little Googling last night on the Fed and game theory, and I came across the great blog “Macroresilience.” I have to say his posts and subject matter clicks with about 90% of the stuff I’ve felt in my gut the last year or so but couldn’t express clearly. If you get time, I’d be interested in what you of his critique of QE:


        As you said, helicopter drops would also lead to a lot of analysis about how they would affect spending patterns and asset prices. But since individuals are going to be determining how to spend this money, this type of research seems more useful b/c it’s aimed at trying to satisfy individual preferences. It seems more “bottom up” rather than “top down” emergency monetary policy.

        On a side note, a common criticism of free banking (even at Sumner’s blog) is that, taken to the extreme (I only say this b/c Selgin always emphasizes that free banking and NGDPLT are not mutually exclusive), base money and NGDP might swing around a lot. Other critics say, how do you address recessions? I think helicopter drops (maybe a salabe tax rebate), used sparingly, could be the answer. But money creation would generally be left to the private sector, eliminating the need for a central bank.

      • John S
        July 10, 2014 at 1:25 am

        By “salable tax rebate,” I mean a kind of tax credit that substitutes for a certain dollar amount of taxes. So if the receiver didn’t owe any taxes, he could sell it off.

      • John S
        July 10, 2014 at 2:00 am

        As far as labels, go it does seem Keynesian/NK. But Nick Rowe said somewhere (can’t find link) that we need to think of helicopter drops as a hybrid fiscal-monetary policy, not either or.


        Maybe helicopter drops could be combined with a negative income tax policy (I’ve seen Noah Smith, Yglesias toy with the idea). I mean, what is the govt’s responsibility w.r.t. recessions? I’d say, ultimately it’s not to maintain NGDP or some other target, but to mitigate human suffering from biz cycle downturns. So I don’t have much of a problem with fiscal policies of this type. I can be an Austro-Keynesian to some degree.

      • Free Radical
        July 10, 2014 at 4:15 am

        There is a distinction between a tax credit and a true helicopter drop which is that the former, theoretically increases the deficit. So whether or not that matters depends on whether or not you think the deficit matters (particularly the part of it that is held by the CB). For instance, if your model has a no-Ponzi-game constraint, then it matters. If you tend to think about money the government owes the Fed as money it owes to itself, then I think they are equivalent.

        I would say the government’s responsibility w.r.t recessions should be nonexistent (except for their responsibility in every other situation which should be to delineate and enforce property rights). Once you start accepting the premise that the government has a “responsibility” to manage the economy, then you open the door for all kinds of debates about exactly how they should manage it and I think we would be much better off without all that.

      • John S
        July 11, 2014 at 12:18 am

        I would say the government’s responsibility w.r.t recessions should be nonexistent (except for their responsibility in every other situation which should be to delineate and enforce property rights).

        This was sort of surprising to read. I definitely can’t dismiss it out of hand on principle, but given the history we’ve had with the Great Depression and Recession, I think this position would receive almost zero support from anyone in blogosphere (certainly not Sumner).

        Do you feel there’s any place for welfare, neg. income tax, or redistribution policies in an ideal libertarian state?

      • Free Radical
        July 11, 2014 at 2:28 am

        No to the last question. If you crack the door for the government to redistribute wealth and manage the economy, you will always end up in a bad place. If you want to talk about misallocating resources for rent-seeking activities, just consider how much effort is spent lobbying governments to make laws to protect someone from competition or funnel tax money into their pocket. State, local, and federal registers are filled with more stuff like that than you could read in your lifetime. And all of it was done under the guise of improving the “welfare of society.” There isn’t a lot of support for a truly limited government right now because it’s not out of control enough yet that the center cannot hold it, but I think it’s a good idea to start thinking about such a thing.

        I think you are right about not finding much support for that but I don’t think you can use the great depression to justify that position since we certainly didn’t have a hand-off approach to the economy (or anything even remotely close to that) during or leading up to it.

      • Free Radical
        July 11, 2014 at 2:35 am

        By the way regarding income taxes, I don’t think there is any place for them at all (positive or negative). The constitution originally specifically prohibited them, they had to pass an amendment under the worst president of all time in order to get it through. They promised it would only be a few percent and only on the very rich. Something like a decade later (I don’t really remember the timeframe but it was not very long), the top bracket was up in the nineties. Today, it’s obvious, if you are paying attention, that the “powers that be” are using the IRS to squash their political opposition which should have everyone seriously considering whether we really need a giant bureaucracy with extra-constitutional abilities to seize the wealth of individuals without due process of law and a government that can use the tax code to engineer society and/or target individuals in any way they see fit.

        BTW this is where the “free radical” comes from 😉

      • John S
        July 11, 2014 at 1:33 pm

        Last post from me in this thread.

        I don’t think you can use the great depression to justify that position since we certainly didn’t have a hand-off approach to the economy

        Any thoughts about 1890s Australia? It seems to be the one major depression under a free banking gold standard.


      • Free Radical
        July 11, 2014 at 6:50 pm

        Unfortunately, I don’t know anything about Australian banking in the 1890s so I can’t say anything to refute this but I never find these type of “see a bad thing happened once when there as a system that they called free. For one thing, it’s not my position that if you have truly free banking and finance, that you will never have a recession. For another, I am suspicious about what “free banking” actually meant in 1890s Australia. For instance, in America around the same time we had what we typically call “free banking” and it was relatively free compared to what we have now but there were still requirements about what banks could use for reserves and it was the inability to get these things fast enough that led directly to the “bank runs” that are always used as a reason to be afraid of too much freedom in banking.

  4. Tom Brown
    July 9, 2014 at 5:51 am

    Mike, nice post! You make a good point about exploration vs explanation, and I liked the bit about rhetoricians. Good word. Good list of things to watch out for too… actually I just noticed that I missed part of the article, shoot. I need to read all your points there.

    As you know, I’m a fan of Jason Smith’s blog. I think he has that spirit of exploration that you talk about here, yet also some humility and an admirable desire for his results to dovetail with existing mainstream economic principles. Plus, he’s always checking himself against the empirical data, which appeals to me. I know he’s am amateur, so perhaps he’s not the best source for learning econ, but it’s fun to watch someone approach the subject fresh, who’s obviously had experience using the scientific method on other complex systems…. and not have their mind made up about what they’ll find.

    Why am I telling you all this? Because I asked Jason today about his priors when he first starting developing his theory. I thought he had an interesting answer, that ties into your post here.

    You might also want to check out his thread with Vincent Cate the hyperinflationist (right above the comment I link to above). Vincent commented here the other day too I notice.

    Also, there was this recent post on “What does ‘model’ mean?” that you might find interesting.

    Anyway, good post Mike! Now I’ve got to go back and cover the things I missed.

    • Free Radical
      July 9, 2014 at 5:56 pm

      Thanks Tom, I’ll check out these links in a bit.

  5. July 9, 2014 at 8:15 am

    Nice post.

    Unfortunately, I would have to agree that some (many?) heterodox economists reject mainstream approaches out of hand, rather than looking to see what may or may not be useful. But, I think the position taken by some (not all) in the mainstream doesn’t help. As you say, economic models can never prove anything – they just help us understand the way things may or may not work, and some are better at that than others. But some mainstream economists give the impression that they actually think, because there is some mathematical proof involved, that these models therefore prove something about the real world.

    The other aspect of this is the apparent insistence of the mainstream that economic argument can only be carried in a certain way, specifically by deduction from microfoundations. I consider myself to be heterodox, but that does not mean I am averse to using things like representative agents or rational expectations when appropriate. I just don’t want to be bound by it. Often, it’s just not the best way of looking at an issue. All models have to sacrifice a lot in order to serve their purposes of informing and illuminating. By refusing to sacrifice certain principles, mainstream models can end up having to sacrifice something much more important.

    This is not to say that I think that microfounded models cannot be used to examine the sort of things that heterodox economists are interested in. On the whole, I think they can. The problem is that they are not always the best tool for the job. To achieve tractability whilst retaining the microfoundations, they are often forced into making arbitrary and unrealistic assumptions. As a consequence, many of the most important things about the way modern economies work are dealt with as marginal issues rather than being seen as core.

    To look at the things I think are important, I believe I need to use all sorts of approaches, including non-microfounded models and not just limit myself to the rules set down by the mainstream. I think that is what makes me heterodox.

    • Free Radical
      July 9, 2014 at 5:52 pm


      Thanks. Yours is, I think, a very sensible “heterodox” approach which I condone entirely.

      “This is not to say that I think that microfounded models cannot be used to examine the sort of things that heterodox economists are interested in. On the whole, I think they can. The problem is that they are not always the best tool for the job. To achieve tractability whilst retaining the microfoundations, they are often forced into making arbitrary and unrealistic assumptions. As a consequence, many of the most important things about the way modern economies work are dealt with as marginal issues rather than being seen as core.”

      No complaints with that. I do think coming up with alternative approaches that are also tractable is a pretty ambitious task but I would be the last person to say you shouldn’t try. I’m just worked up lately because I have been listening to Steve Keen too much. Hopefully I put enough qualifiers and disclaimers in to leave room for people to work outside of the established framework. I just want them to do it thoughtfully and not be trolls.

      • July 9, 2014 at 6:31 pm

        Yeah – listening to Steve Keen will do that to you.

      • Tom Brown
        July 9, 2014 at 7:52 pm

        Ha!… I was going to mention Steve Keen as someone with an approach and attitude which I think is in sharp relief to either Nick Edmonds or Jason (and that’s not meant to be a favorable comment about Keen).:D

      • Free Radical
        July 9, 2014 at 8:30 pm

        Yeah, a Steve Keen lecture is basically one of these issues after another: “Neoclassicals think markets are always in equilibrium, that’s garbage, neoclassicals think consumers calculate their utility over millions of different bundles, that’s garbage (!!!), neoclassicals think people act rationally, that’s garbage (I actually forgot that one, that’s basically the mother of all unproductive criticisms). The only reasonable approach is to make a model that doesn’t say anything about how people make decisions and just make everything completely arbitrary. Obviously Marx understood this but neoclassicals refuse to appreciate his great wisdom because they are driven by politics.”

        I was trying to take them on one at a time for a while but the barrage was so overwhelming I decided it wasn’t the best use of my time.

      • Tom Brown
        July 9, 2014 at 9:16 pm

        Mike, you write, in what I assume is your paraphrased quote from Keen:

        “The only reasonable approach is to make a model that doesn’t say anything about how people make decisions and just make everything completely arbitrary.”

        From what I understand, Jason’s approach isn’t all that different from that… Lol … **however** he also said this, when I asked him about his priors:

        “Tom, regarding your philosophical question, my only theoretical prior was: any new model should reduce to known economics models in some limit (like the quantity theory, AD/AS model or ISLM model for example). I seriously doubted that economics as conducted today was based on wholly wrong-headed views.”

        Not exactly something Keen is likely to say.

        To my understanding, Jason’s approach might be described as assuming “maximum ignorance” regarding individual human behavior. I’m reading this paper he gave as a reference to understand the general approach better.

      • Free Radical
        July 10, 2014 at 2:39 am


        Yes, that is my sense of what Jason is doing but you’re right, I haven’t seen him level any baseless attacks on mainstream economics so it doesn’t bother, even though it isn’t the approach I would take. In both his case and Keen’s case I think it is partly the result of being trained more in the harder sciences. In other sciences, you just try to notice relationships and then measure them. There is no need when considering a falling body to wonder “does it make sense for the falling body to act this way?” This is sort of the fundamental question in economics. If you have a model where you just say “people do X” but it is not rational for them to do X, most economists would say you have a serious problem. You can’ say “well people don’t act rationally” if you want but once you say that, you can have them do anything, so it is very easy to get your model to say whatever you want it to say. Of course, if you are a rhetorician, this is a big plus!

        Note that “rational” doesn’t necessarily mean that they don’t make mistakes, but it means you have to have some rationale explaining how and why they make mistakes.

        Also, for the record, yes that Keen “quote” was a made-up quote, not his words, just my playful attempt at paraphrasing what seems to me to be his general mindset. Hopefully, nobody takes it otherwise.

      • Tom Brown
        July 9, 2014 at 9:22 pm

        Actually, the abstract on this version of Fielitz & Borchardt’s paper is a little better (notice they added econ to their list of possible applications).

      • Tom Brown
        July 10, 2014 at 7:15 pm

        Mike, it just so happens that Jason responded to a question of mine yesterday about the nature of the underlying theory he built his theory from, and I really liked his write up! I think it gives a good sense of the intended “scope” of his theory, and of what sorts of things the general IT theory can be applied to:


        Jason often makes an analogy with how the ideal gas law was developed in the 19th century. I’m going to do a little interpolation here, rather than find you a quote, but as I understand it, it went something like this:

        Observations were made regarding pressure, temperature, volume, etc, and how one changed when another was held constant, and a third was varied, and from that an “ad hoc” type “law” could be inferred (w/o any explanation of “why?” like you mention in your comment). Later on, in an attempt to answer “why?” the field of statistical mechanics was developed. At the time, the physicists had very little idea of what a molecule was or its physical properties. They just knew there were lots of them, and that they could move, rotate and some of them could probably vibrate internally (e.g. molecules composed of two or more atoms). Does statistical mechanics attempt to model a “representative molecule” and how it might move and vibrate? No, not at all… the only micro-physical details that “survive” at the macro level is a count of the degrees of freedom (“modes”) available to each molecule. Amazingly enough that was good enough to explain the macro-scale ideal gas law, and turn it from an ad-hoc explanation into a theoretical explanation. It’s possible that “molecules” with very different micro scale physical properties could produce exactly the same gas law.

        I think that’s akin to what Jason is attempting, except that he’s starting with the theory rather than the observations: he’s postulating that he can capture certain limited dynamics of the economy by simply “counting degrees of freedom” w/o having to build a model of individual agent scale motivations, expectations, etc. So he really does have a theory: not just ad-hoc formulas to try to capture observed macro scale dynamics. That’s my impression anyway!… and please forgive me if I got any of the above wrong: I’m kind of shooting from the hip here. My point is he’s been looking at the “why?” from the beginning, only he’s looking at it from a different perspective. And that’s not to say you can’t apply his theory to microecon… actually, at it’s core it really does use concepts of supply and demand that are better suited to a microeconomic view… but he’s relying on the market being large. Similar to how an individual molecule doesn’t have a pressure, a single purchase in a market doesn’t produce supply and demand curves. “Pressure” is an emergent property of a huge ensemble of molecules. I think he’d make a similar argument about microecnomic supply and demand curves.

        What I also think is cool is his theory lets him skip right to the good part of comparing his results with empirical data … without having to first build up a lot of abstract layers of groundwork about utility optimization, indifference curves, expectations, game theory, equilibrium conditions, etc. Granted he has his own underlying abstractions (in the generalized information theory), but overall it is an attempt to find a “shortcut.” Ideally there will be lots of correspondences with mainstream econ. …just as we’d all expect that if a modern physicist were to construct a high fidelity computer simulation of gazillions of carefully modeled individual gas molecules, we’d expect that he’d confirm the results of 19th century statistical mechanics on the macro scale.

        Well anyway, I’m interested because I think it’s great fun watching someone with a physical science background delve into something like this, and what could be more challenging than econ? And yet he’s very aware of how badly that kind of thing has gone in the past. Here he quotes Samuelson:

        “I also found this quote by Samuelson:

        There is really nothing more pathetic than to have an economist or a retired engineer try to force analogies between the concepts of physics and the concepts of economics. How many dreary papers have I had to referee in which the author is looking for something that corresponds to entropy or to one or another form of energy. Nonsensical laws, such as the law of conservation of purchasing power, represent spurious social science imitations of the important physical law of the conservation of energy; and when an economist makes reference to a Heisenberg Principle of indeterminacy in the social world, at best this must be regarded as a figure of speech or a play on words, rather than a valid application of the relations of quantum mechanics.


        Also, he did a post on how “Economics is neither physics nor computer science” that you might enjoy:


        BTW, yesterday I shared your post here on pragcap. Commentator John D. (who’s also a physicist I think) gave it a favorable review 😀


        I really like it too… I think I’ll draw Cullen’s attention to it as well. I think he might like it.

      • Free Radical
        July 10, 2014 at 8:06 pm


        That’s great, thanks for sharing. I am getting a little behind on following up these links (this blog is starting to take up a lot of time. That’s a good thing, but I have to stand back once in a while and try to get some actual work done) but I will check them out soon. I like the Samuelson quote. I especially like that he mentions entropy, which I took Keen to task for (that was one of my favorite posts). I still don’t understand what Jason is doing entirely, so it’s hard to comment on it confidently but I will say this. Part of the reason that economics relies so heavily on theories of rational behavior is that empirical laws are almost impossible to determine. This is because it is usually impossible to observe important variables like preferences and it is usually impossible to carry out controlled experiments. So if you build a sort of top-down empirical model, you can calibrate it to fit the data but it is very hard to know all the variables and all the relationships that went into generating that data and this means that it is hard to be certain that whatever relationship you are capturing isn’t dependent on some other set of circumstances you are missing that could change at any time. So, for instance, you might make a model that perfectly fits the data from the U.S. over the last 100 years without having any idea why things work that way. But if you have a question like “what would happen if we changed the monetary regime to something totally different?” you would have no way of thinking about this because you can’t just do an experiment where you change the monetary regime and see what happens empirically. Most economists will feel better if they can come up with some explanation for why those relationships hold. So finding persistent empirical relationships does contribute to our knowledge but mostly because people will look at it and try to figure out some way to explain why it works that way.

  6. Tom Brown
    July 16, 2014 at 12:25 am

    Mike, that’s a great reply, thanks. You write:

    “I still don’t understand what Jason is doing entirely…”

    Nor do I! However, just to be clear, I wouldn’t call what Jason is doing a “top down empirical model.” I’d just say it’s a non-traditional theoretical model, that if true, allows one to “take a short cut” when analyzing a limited set of macro dynamics.

    You write:

    “But if you have a question like “what would happen if we changed the monetary regime to something totally different?” you would have no way of thinking about this because you can’t just do an experiment where you change the monetary regime and see what happens empirically.”

    This is a great point, and even though Jason starts with theory (rather than empirics), there’s only a limited set of “what would happen if” scenarios he can address. I think he might agree that is a weakness of his approach: most of the time his theory is not going to facilitate letting you work our the macro implications of proposed policy changes. There might be some exceptions: for example, if the central bank were to target M to the exclusion of all else (and not even look at any other data), I believe Jason’s model says you leave one solution to his equations (the “endogenous” solution) and enter another solution (the “exogenous” solution). This could result in hyperinflation, or accelerating inflation, or perhaps deflation (not sure on that last one). But does his model tell you what happens if the CB targets an inflation rate vs NGDP levels? No… that’s “in the noise” as far as he’s concerned (which is why I think of his model as a “very macro” — though he might not agree with that description). Likewise his model probably won’t be helpful in working out the macro implications of changes in tax law, changes in government debt levels, or a host of other things that economists are probably very interested in (again, he might not agree with that specific list, but I think I’m on the right page with the general concept).

  7. Tom Brown
    July 16, 2014 at 2:16 am

    Mike, since I do a considerable amount of “interpolation” w/o fully understanding Jason’s theory (or anyone else’s for that matter… but hey, I’m working on it!), keep this story (first few paragraphs) in mind about what it means to be a “modern jackass” while reading anything I write. 😀

  1. No trackbacks yet.

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: