Ayn RAnd and the Elephants

[From Bill Williams 19 January 2004 12:30 PM CST]

  [From Rick Marken (2004.01.19.1000)]

  Bill Williams (19 January 2004 6:10 AM CST) --

  > Hey, Bill Powers thinks it is just friendly conversation to call me,
  > "bent."

  I don't know that that's true.

  You are right, of course, I was being sarcastic. Bill Powers wasn't telling the truth.

  But it still seems like there are better ways to handle insults than by insulting back.

  But, it isn't an insult. Bill Powers really is a crank when it comes to economic issue. Runs in the family.

  > But, in regard to economics Bill really is a crank

  > Somehow you managed to publish the H. Econimcus paper, and then
  > given it at two CSG conferences, and then have it subjected to
  > a Very through criticism, a negative and destructive criticism,
  > by Bill Powers, and the result is that you still think it was a
  > good piece of work.

  Why should I take to heart the criticism of a "crank"? Why would you, the
  person who says Bill's a crank, think such criticism is worth considering?

  There is such a thing as what's called a "short certificate." If someone finds a needle in a haystack after a very long search,
  you can satisfy yourself that it is in fact a needle, and do so very quickly. So, I could read through Bill Powers critique
  without much effort-- looked like he had it right to me. It was a _very_ through job.

  Actually, I do take Bill's criticism very seriously. Many of his criticisms
  were very reasonable. But I do still think the paper is worthwhile.

  Time for you to talk to Bill again.

  The model described in the paper is a working implementation of TCP's circular
  flow model of the macro economy.

  This is a contradiction in terms--- working model, TCP's model.

  By making the model work as a dynamic
  computer program (using difference rather than TCP's differential equations)
  I had to make some assumptions that were not obvious in the TCPs analytical
  model. For example, because it is a closed loop model I had to implement it
  using two control systems, one a system controlling consumption and the
  other controlling for a balance of income and expenditures.

  Income and expenditures are equal by definition.

  The implementation of the model showed clearly that the effect of leakage on
  GDP growth rate was not part of the model. Leakage affects the ratio of real
  to potential output (Q/Qo) but not the rate of growth in Q (dQ/dt). So TCPs
  conclusion that leakage affects growth rate was put in as an assumption of
  his analysis, not derived from the behavior of the model. Apparently WTP
  came to the same conclusion without the benefit of simulation but the
  simulation helped me see it.

  I'm doubtful that this is in fact what happened.

  The model also illustrated what I think is a useful way of looking at the
  big picture of macro economics.

  Useful as compared to what? And, useful for what?

  It shows that the macro economy can be viewed as an aggregate control system, producing goods and services for it's own
  consumption and using money as the basis of purchasing the specialized outputs of the production process. It shows, I believe,
  that an economy can be viewed as human nature writ large, as a collective of control of input systems.

  I will say, that controlling an identity so that the two terms are equal is quite sophisticated.

  I believe that the H. economicus model was a faithful computer
  implementation of TCP's circular flow model of the economy.

  In the sense that both are fundamentally mistaken, I would agree with you.

  I think Bill's criticisms, which were, indeed, harsh, relate to TCP's approach as much as to mine since H. economicus is an
  implementation of TCP's model. The article on H. economicus shows data that I used to make sure the the program produces the same
  output as TCP's equations.

  You may be onto something here. I wondered why Bill Powers waited until after you'd published the H. Economicus, given it at
  Boston, Given it at St. Louis, before really doing a nice job critiquing it. Although, the bit about "A giant leap in the wrong
  direction may have been somewhat gratuituous. Acutely perceptive, but it was perhaps unneccesary.

  As I recall, Bill's criticism's had to do with the fact that my model of the "economic environment" was too simple.

  Among other things your "recall" is in error.

  You could easily read again what Bill Powers said. My mistake, "I" can easily read it. I can see that for you the situation is somewhat different.

  An excess of simplicity wasn't the problem that concerned Bill. The question that really got Bill's attention was the possiblity
  that you would make the same mistakes in the future.

   Money, for example, simply appears when it is needed.

  How wonderful!

  There is no mechanism built into the model that represents the mechanisms (banks, international loans, etc) that constrain
  economic activities in the real world. I accept all these criticisms. The model was just meant to be a very high level
  illustration of how a closed loop economy works according to TCPs analysis.

  Again "Works" and TCP's analysis amounts to a contradiction in terms.

  I view it as kind of an illustrative demonstration of principle --
  a "toy model" of the macro economy -- like my hierarchical control

  In that case it is a broken toy-- and it always has been.

  Now that demonstration is a fait accompli.

  The demonstration may be a _fait accompli_ but the question is of what?

  I agree that Bill didn't like it much

  Yes. "Bill [really] didn't like much" what he said was it is a "a giant leap in the wrong direction." You do have a way with

  but it helped me see the big picture (just as TCP's work did).

  We've all been impressed that you are a "big picuture" guy.

  So I still consider it worthwhile and that's why I put it in my book.

  I'm sure your readers will find it extremely helpful.

  Perhaps someday, someone will see the value in it that I see.

  I really do think there is hope, as Mr. Barnum said, "There's one born every minute."

  But by then it may be beside the point since we will have more detailed models, like the Test Bed model, when it includes active
  control agents, of the general, circular flow model I describe in the H. economicus paper.

  I can't say how pleased I am that you've decided to help Bill Powers out.

  > ( Just to keep the record straight, Bill
  > Powers thought it was a giant leap in the wrong direction. )

  Again, why listen to a crank? But I don't know if he thought it was quite
  that bad.

  Yes, Bill has his own way with words. The subtle nuances contained in his expression "a giant leap in the wrong direction" fill me
  with envy.

  By-the-way would you believe it, my _New College Standard Dictionary_ published in 1950 and purchased by me recently for 50 cents
  doesn't have that nifty German word that is spelled sort of like "Schudeunfreud. "

  I agree that he was pretty negative about it.

  You really did understand at least part of what Bill was suggesting.

  I agreed with many of his criticisms and disagreed with others. But I'm happy to discuss it again.

  Rick, we are well aware that you are always happy to discuss Rick. But
  I think Bill Powers would just as soon not. For the moment all of us may have had about all the fun with economics that we can

  I was just trying to make a start a macro economic modeling.

  The most obvious element in this effort was as you say, that it
  was a "start."

  I think the Test Bed approach, with control agents, will basically be an improved, though far more detailed, version of what I was
  doing with H. economicus.

  I have this strange impluse to say profound things in something that sounds like German. Which is strange, because I know almost
  no German. And, I don't even look German.

  > You think you understand "a lot" about economics. I'd wait to
  > say that untill you have your first economic model working.
  > All you've done so far is get in the way.

  H. economicus did work.

  You really do need to talk this over with Powers.

  I have also looked at a lot of macro economic data
  in terms of the relationship between things like discount rate, inflation
  rate, tax rate, growth rate and so on. So I think I know a bit about macro
  economics in terms of the data as well.

  Having mastered economic theory, I didn't expect that your next trick would be to memorize train schedules.

   I discovered the strong, positive
  correlation between Fed rates and inflation by plotting the raw data myself.

  All by yourself? I had once had an idea that sounds similar to the relationship you discovered. You might like to check this
  out. There seems to be a very strong relationship between increasing prices and inflation. And this seems like a pretty reliable

  The exact same relationship was independently reported by E. Ray Canterbery
  (a real economist) in _Wall Street Capitalism_, which also refers favorably
  to TCP's work.

  I wasn't aware of this.

  I also discovered an error made by Canterbury in that book.
  He shows a plot of Fed rate and growth rate (dQ/dt) and, based on it's
  appearance, concludes that there is a negative relationship between Fed rate
  and growth rate. I got the raw data and found that the graphical appearance
  is an _illusion_.

  Good work. Wonder how he made such a mistake. Have you pointed it out to him?

  The actual correlation between Fed rate and growth rate is
  ~0.0, which is what is qualitatively predicted by the H. economicus model if
  Fed rate is assumed to be a component of leakage.

  If I remember, Powers attempted to explain to you how it was that your model generated an output that matched the value you
  inserted into it. I have to admit I had been puzzled-- puzzled about where the model was. Maybe it was where ever the leakages

  I'm a bit puzzled here, about how the Fed rate-- which I thought I understood, could be a part of leakage-- which well anyway.

  Maybe for the second edition of your book, you could include an explaination of how when you are worried about the elephants going
  extinct, the thing to do is to kill more elephants. With the way I'm sure the books are flying off the shelf, it isn't at all too
  early to begin thinking about such matters.

  I guess you really aren't interested in how capital costs may be mistaken for capital expenditures.

  I can't tell you how much I've enjoyed this, and how much I've learned too.

  Bill Williams