Topic: Is deflation truly that bad for an economy? - page 8. (Read 24916 times)

This is highly unlikely since you can't effectively predict human behavior which forms the basis for taking economic decisions by individuals on the micro-scale. As long as they are free in their actions, indeed. This is where central planning would ultimately fail...

1st there is not a single nash equlibrium (not to mention that it may not even be a single point but a closed trajectory in the solution space) It is highly sensitive to the initial parameters, one could at least  set the initial parameters so (Rig the game) so that players eventually gravitate to the optimum eq.

however in central planning choice is striped from the players (not freedom) so nash equlibrium is irrelevant as there is no game and only the computational tractability of calculating the global oprimal is the problem. When that is no longer an issue, there is no point in trying to find a solution using a computationaly cheap distributed algo as in a game.

As we move to more accessible market parameters ie public transactions, public performance records, and higher computational power wtr local knowledge and low cpu power of the past. combined with the necessity of higher efficiency. I see/predict that free market with all its problems will be abandoned in the end

No i understand it.  I took a class at Yale.  We had to make those matrixes.

Then I think that you didn't understand the meaning of a Nash equilibrium.  The hypothesis set forward in game theory is that we have non-cooperative individuals who try to optimize their own gain in a rational way each independently, and the question is: what are stable configurations of strategies in such a case.  That is to say: what strategies can be applied, so that no individual has something to gain by NOT adopting it.  Those sets of strategies are then stable against the postulated desire of every individual to rationally improve his situation given his environment.  That's what a Nash equilibrium is.

Any proposed set of strategies that is hence not a Nash equilibrium, means that some individuals would have personal gain if they didn't follow it.  As such it is not stable: there's a serious probability that said individual will NOT adopt the prescribed strategy, or will change from the prescribed strategy to another one when he realises that that is in his personal advantage.

Even Jesus realized the Nash equilibrium


Indeed, the situation: "you can bring a bottle of wine, or a bottle of water, and we will drink the mixture", has a single Nash equilibrium: we all bring water and we drink it !

Although it would maybe have been better that everybody brought wine, and we drank wine, that's the UNSTABLE situation, because indeed, everybody makes the rational decision, that whatever the others bring, if you add a bottle of water instead of adding a bottle of wine, what you will drink in the end will not be very different, and what you will pay (for a bottle of wine, or a bottle of water) will be quite different.  So, whatever the others decide to do, you will be better off bringing water than bringing wine.

Most naive social engineering is based upon not realizing that non-Nash strategies are not stable.  The usual reaction to the discovery of this evident truth is then to be outraged by so much egoism

I can't follow what you are talking about.

I didn't state any hypothesis.  But you are misapplying game theory.  Game theory doesn't say "a strategy that is good for the individual is good for everybody".  It doesn't say anything is stable or not

You can't say "no social construction can be stable without a Nash Equilibrium".  That makes no sense and not what Nash equilibrium attempts to predict. NE is making a matrix of all players and possible strategies and predicting the most probable move or the Nash equilibria.

It's like saying there can be no suffering in a bell curve because 68% of income fall within one standard deviation.  This statement is true but has no common sense

I'm sorry but YOUR hypothesis was universal empathy.  If you don't have universal empathy, then any non- nash equilibrium will not be stable (that's exactly the idea of a Nash equilibrium: that you do not NEED total cooperation for it to be stable).

A Nash equilibrium is such, that, from the POV of a single individual, there is no incentive to change strategy.  That is, if a single individual changed strategy, and all the others kept their way of doing (the Nash equilibrium), the single individual would lose.  So he won't change strategy.

So in order for a situation to be stable and NOT be a Nash equilibrium, it means that some people would win by changing strategy, but they WON'T because of "empathy".  They would systematically take decisions at their own loss, for the sake of keeping the global solution.  So in order for this situation to remain stable, it would be absolutely necessary that random people suffer from universal empathy.

In other words, in order for your claim to be true (that is, there are stable optimal solutions which are not Nash equilibria), you need to make the hypothesis that random people do suffer from universal empathy.

So no social construction can be stable without being a Nash equilibrium.  However, it will be a Nash equilibrium with a very complex and sophisticated cost function, which will include all of human forms of satisfaction and dissatisfaction, including empathy.  It won't be a "bookkeepers" cost function, with just monetary gain or something.  So it may SEEM that certain situations are stable and nevertheless not a Nash equilibrium, because we don't understand the real cost function.  That's exactly what happened with the real-world prisoners dilemma.  We thought that the cost function was a naive "years of prison", while there's much more that enters it.  In fact, it is essentially unfathomable.  We're back to human action.

Anything dealing w humans are complex.  You could use math to explain stuff but easier to use sociology if you are talking about society

Being universal just means everyone has same experience.  Doesn't mean  Competition or conflict will dissapear.  Quit making such black and white arguments.

Yeah 

The point with the mathematical prisoners dilemma is that it is very abstract.  It is as if the ONLY cost/reward is years of prison.  In reality there's a lot more that plays: reputation, revenge, IOU, agreements, ...  Because, if you don't know the other person's decision, you run the risk that the other guy didn't betray you, and that you betrayed him.  If he gets free, or he has friends outside, you might be worse off outside than inside the prison !  So the REAL cost/reward structure in a real prisoners dilemma is much more complicated than in the abstract game.

Empathy is a phenomenon whereby you experience good or bad sensations as a function of what you perceive that SPECIFIC OTHER persons experience.  Empathy is always limited to a small circle of persons (family, friends), and is a phenomenon that can happen or not.

If empathy were universal, no police, no prisons, no laws would be necessary.  After all, each of us would feel much better, knowing that other people would feel better by his actions.  That's obviously not the case.  We would all sit together around the camp fire and sing Youmbala.

BTW.  There was a prisoners dilemma experiment w real prisoners and the results show majoriry of prisoners dont snitch each other. IOW not NE

"We should all help each other" is called human empathy.  Its neither moralising nor a game.  Happens all the time because thats how humans are

That's correct, but the fundamental problem with laws that forbid INTENTIONS and not acts, are that intentions are unfathomable.

If your solution is not a Nash equilibrium, it will simply not be adopted in a non-cooperative game.  So even though it might be Pareto-optimal, it won't work.  As such, the ONLY solutions you should consider, are Nash equilibria, simply because the others are unstable and will not be adopted, or remain adopted.

The prisoners dilemma is indeed the best illustration of that.  The point is that the optimal solution will not be adopted, simply because each individual can improve his situation by changing strategy.  As such, the only Nash equilibrium is not the optimal solution, but it is the only stable solution.

It is the same reason why cartels can't last.

Its not that easy to corner the market.  Look up Silver Thursday.  BTW its illegal to hoard if your intention is to corner the market

That is not what game theory says.  Nash equilibrium does not mean "good for everybody".

Game theory 101.  Prisoners dilemma, Nash equilibrium is not good for both prisoners.  Its the move each expected to make knowing all possible outcomes

I would say: hoarding can be 2 things: a form of saving, or a form of cornering a market.

Because otherwise they wouldn't remain rich ?

Poverty has the tendency to propagate.  It breeds, until it dies of poverty.  Malthus.


Nash equilibrium.  If your "global optimal" solution isn't a Nash equilibrium, it won't last.
You reach a Nash equilibrium when everybody applies a strategy, such that no-one can gain anything by changing only his own strategy.

This is another argument why "a strategy that is good for an individual" is also a strategy that is good for everybody.  Because otherwise it wouldn't be a Nash equilibrium.

It is the problem with all "moralising" utopia.  "We should all help one another".  It simply isn't a Nash equilibrium: someone can win in that game, by having all others help each other (him included) and not helping others himself (change in just his own strategy).
 

Hoarding is an one way transfer from the economy to your hoard, Saving is saving for the bad days, Cornering the market is part of the game. 3 diffrent things

Congratulations, you have now linked genetic diversity with wealth, and you are on your way to devolution. And before you tell me thats not the case, take a look at all the noble families of europe in middle ages, or modern day Luxembourg.
But anyway in reality rich people breed less than poor people why do you think that happens?

Greedy algorithm may be the best distributed algo and most efficient computationaly, but will not give you the global optimal basic game theory

I think the misunderstanding between both of you comes from two different views of hoarding.

The first view on hoarding is that you store stuff when it is cheap (that is, when there is a lot of offer, and not much demand), and you sell when it is expensive (that is, when there isn't much offer, and there's a lot of demand).

This is the "good" kind of speculation, that softens scarcity.  It is the chipmunk that puts nuts aside in summer for the hard season.  When nuts are abundantly available, they don't consume everything, but they put aside some.  When there are no more nuts available, they live off their hoarded savings.

Hoarding that way increases prices when things are cheap, and decreases prices when things are expensive, as such stabilising price.  The benefit you make on hoarding is exactly equal to the price difference.  Hoarding stops being interesting when price is totally stabilised and there is no "cheap" and "expensive" season any more.

But the other view of hoarding is "pump and dump" market manipulation.  In an a priori constant price market without temporary preferences ("summers" and "winters" for nuts), you might try to manipulate the market by storing a lot of stuff, making buying contracts at "normal price", by your hoarding, you increase the market price, and when the stuff that became scarce because of you having bought half of the available production, you make sales contracts at higher prices, and then dump everything on the market.

The point is of course that in order to do so, you need to be able to buy a significant fraction of the market.

So "pump and dump" can only happen with very significant sole players on the market, or by a cartel.  However, the normal hoarding behaviour can be done by many many small independent hoarders.

Many small, individual hoarders cannot do a pump and dump, because the whole trick is to be perfectly synchronized in buying "just before" the scarcity, and selling "just before" everything floods the market.  If you are slightly out of sync in a pump and dump, you loose instead of gaining.

This is the fundamental flaw in collectivist thinking.  If a strategy is good for an individual, it is good when applied to every one because every one is an individual.  If it isn't good for every one, then it wasn't good for an individual either.

Typical example: having a lot of children.  Is that a good or a bad thing for an individual ?  Is it good or bad that I have 10 children ?  If my aim is Darwinistic, and I want to get as much of my genetic material in the next generation, what should I do ?  Breed like rabbits, have one kid, have no kids ?

You may say: if YOU have 10 children, that might be Darwinistically good for YOU, but if EVERYONE has 10 children, the world will go down in over population.

No.  In fact, I should have as many children as I can economically support efficiently.  If I have enough economic means to educate well, 5 children, that's what I should do.  If I make less children, I will diminish my genetic presence in the next generation, and if I make more of them, they will be economically weaker and may not survive the next round.  If I optimize the number of children, and my economic value so that each of them reaches a good level to start in live, then that's my best solution.

And now, it turns out that if everybody would do that, that is, optimise its number of kids as a function of his own economic power, then that that would also be globally the best solution.  Poor people wouldn't make kids because they would starve any ways.  Rich people would have a lot of children.  In the end, you would have an economically better balanced next generation.

But that doesn't happen, because there are obliged economic transfers from the rich to the poor, who allow poor to breed more poverty. 

So even in the most "collectivist" a priori questions, you see that if everybody really optimises for one-self, everybody optimizes for one-self.  In fact, that's nothing more than game theory.

I don't want to play.