Prisonner's Dillemma and economics.

[Scheavo]

For those who don't know what it is, it's a thought experiment where two people are being questioned for a crime that they committed together, but without a confession there's not enough evidence to arrest or convict on a major charge. One person confessing and betraying means the betrayer get's away, and the person who remained silent get's a 1 year prison sentence. If both confess, they both get a three month prison sentence. If neither confesses, they both get a light 1 month sentence. As a result of this, both people confess (trying to get off completely), and thus both getting a three month sentence. Whereas if both kept silent, they would've have received a lighter sentencing.

This is mostly a true phenomenon. I had a teacher perform it once in class - just after going over the dilemma - and it still proved to be rather true. You don't trust your other partner, they don't trust you, you both rat each other out. This plays into politics, economics, etc. In economics, this is important because it shows how the market and competitors can become stagnate. In regards to "green" technology and the green movement in general, for example, companies may want to invest long-term in their energy needs, pollute less, etc, but they see the short-term capital needed to do this as a hindrance to their competitiveness. However, if the companies had an open dialogue with each other, and came to terms on regulations, etc, then companies can overcome this communication gap, and all take the bullet together, at the same time. In this way, no one companies competitiveness is directly hampered, and the companies are able to make the improvements they want to make. It's the prisoner's dilemma, where a lack of communication leads to a logically worse result. The competitive nature of the market can actually hamper innovation from happening, in certain cases.

Or so that's how I see it. Where does the analogy break down for other people? Is the dilemma itself somehow flawed? IS the comparison to the market and over-reach? Hopefully I explained it all right, it's late and I'm tired.

[Orthogonal]

Or so that's how I see it. Where does the analogy break down for other people? Is the dilemma itself somehow flawed? IS the comparison to the market and over-reach? Hopefully I explained it all right, it's late and I'm tired.

I'm not sure how you think the prisoner's dilemma relates to economics at all. It presumes that both parties are under duress and significant penalties will be incurred for just about any action, with the exception of betraying someone else.

In a market transaction, both parties must mutually agree to the terms, otherwise it doesn't occur. Think of it like you going to the local convenient store. You want a Coke, they want a dollar. You value the Coke more than your dollar, they value the dollar more than the Coke, a trade is made. Both parties are now better off than they were before. There is no duress, there is no penalty for not accepting the terms, and there is certainly no incentive to betray the other person.

[bosk1]

I think he is saying there is an analogous phenomenon in economics in that there may be a certain course of action that is beneficial for everybody if everybody does it, but it is financially detrimental to one business doing it if other businesses don't do it as well, so because of "selfish" self-preservation, businesses will opt not to take that mutually beneficial step, which leads to short-term gains for that company, but loss of larger long-term gains for everyone involved.

An example from my world:  All employers in a certain industry violate a particular pay practice under the law and don't pay their people properly (maybe they classify a certain type of worker as exempt who shouldn't be, for example, so those people don't get paid overtime).  Company A wants to do the right thing, and they learn from their lawyer that they are in violation.  BUT, if they change their practices and do things the right way, they have to charge a bit more for their product or service to make up for the extra overtime they have to pay their people.  If they do that, they're product/service will be a bit more expensive than their competitors', and they believe their competitors don't care about doing the right thing and changing their practices as well.  So Company A says, "even though the right thing to do is start paying our people overtime, we lose an advantage to our competitors if we do so, so we're just going to keep doing the same thing we've always done."

[antigoon]

Or so that's how I see it. Where does the analogy break down for other people? Is the dilemma itself somehow flawed? IS the comparison to the market and over-reach? Hopefully I explained it all right, it's late and I'm tired.

I'm not sure how you think the prisoner's dilemma relates to economics at all.

FWIW, I learned about it in an international political economy course in college.

[Scheavo]

I think he is saying there is an analogous phenomenon in economics in that there may be a certain course of action that is beneficial for everybody if everybody does it, but it is financially detrimental to one business doing it if other businesses don't do it as well, so because of "selfish" self-preservation, businesses will opt not to take that mutually beneficial step, which leads to short-term gains for that company, but loss of larger long-term gains for everyone involved.

An example from my world:  All employers in a certain industry violate a particular pay practice under the law and don't pay their people properly (maybe they classify a certain type of worker as exempt who shouldn't be, for example, so those people don't get paid overtime).  Company A wants to do the right thing, and they learn from their lawyer that they are in violation.  BUT, if they change their practices and do things the right way, they have to charge a bit more for their product or service to make up for the extra overtime they have to pay their people.  If they do that, they're product/service will be a bit more expensive than their competitors', and they believe their competitors don't care about doing the right thing and changing their practices as well.  So Company A says, "even though the right thing to do is start paying our people overtime, we lose an advantage to our competitors if we do so, so we're just going to keep doing the same thing we've always done."

That's a good way of putting it, and a great example.

In a market transaction, both parties must mutually agree to the terms, otherwise it doesn't occur.

You always draw too small of a scope. The economy is more than just parties agreeing to terms, its the sum total of all parties agreeing to terms, and more. It's like you don't even acknowledge such a thing as macroeconomics.*

The economy is more than "the market," and any theory which tries to define the economy as the market place is putting the part before the whole.

FWIW, I learned about it in an international political economy course in college.

Hehe.

[Orthogonal]

I think he is saying there is an analogous phenomenon in economics in that there may be a certain course of action that is beneficial for everybody if everybody does it, but it is financially detrimental to one business doing it if other businesses don't do it as well, so because of "selfish" self-preservation, businesses will opt not to take that mutually beneficial step, which leads to short-term gains for that company, but loss of larger long-term gains for everyone involved.

An example from my world:  All employers in a certain industry violate a particular pay practice under the law and don't pay their people properly (maybe they classify a certain type of worker as exempt who shouldn't be, for example, so those people don't get paid overtime).  Company A wants to do the right thing, and they learn from their lawyer that they are in violation.  BUT, if they change their practices and do things the right way, they have to charge a bit more for their product or service to make up for the extra overtime they have to pay their people.  If they do that, they're product/service will be a bit more expensive than their competitors', and they believe their competitors don't care about doing the right thing and changing their practices as well.  So Company A says, "even though the right thing to do is start paying our people overtime, we lose an advantage to our competitors if we do so, so we're just going to keep doing the same thing we've always done."

I see what you are saying now, not necessarily a prisoner's dilemma, but a similar phenomena. In the specific example you gave, this would be more a case of contract theory and enforcement. Worker pay or their classification as exempt/non-exempt would be spelled out in their employment contract. As long as the company pays them according to agreement, there is no harm done. If they are not paying people for overtime, regardless of the motivation, they are in breach and susceptible to lawsuit. Further, any short term gain (whether this is the company that wants to do the right thing, or the one who doesn't give a shit) would be compromised as worker morale takes a hit and they suffer a loss of productivity. Perhaps people even quit outright.

I can think of a real world example that's going on right now that illustrates this phenomena.In the semiconductor industry, manufacturer's really want to push moving to larger wafer sizes (currently at 300mm, want to go to 450mm) to increase die output at a fraction of the wafer cost increase, the problem is that it is incredibly expensive to drive this change and no company alone has the money to do it. The problem is further compounded because the semiconductor manufacturer's must rely on the tool manufacturer's to develop tools for the larger wafer size and they must get all of them to work in concert, otherwise no one could build out an entire production line. Clearly, everyone would benefit by going to a larger wafer size, but no one can afford to do it alone, so if your theory is hypothesis is correct, no one has the incentive to make the push due to the cost being so prohibitive. What we see instead is that several manufacturers came forward (Intel, Samsung and TSMC) and agreed to collaborate and pool their capital for the development of 450mm manufacturing technology. Later IBM expressed interest and is getting involved too. All the companies are working together to bring about the change as well as establish industry standards.

https://www.theinquirer.net/inquirer/news/2112790/ibm-partners-intel-samsung-tsmc-fab-research

[Scheavo]

Yes, that's a wonderful example of when the market can come to agreement on such issues. That doesn't prove that this is the rule, or that this has to always be the case. I could find examples of slaver owners freeing their slaves out of conscious guilt in the antebellum south, but that doesn't mean the South was overall a society which felt guilty over it's ownership of black men.

I also found this interesting in the article:

One big name missing from the deal is Globalfoundries, the primary manufacturing partner of AMD. It will be interesting to see if Globalfoundries enters a similar research deal with other chipmakers, although after this deal there aren't many more left.

Just showing you a real life example (and not one I think is a problem) of where private practices can force market-wide decisions. If there was a government law enacted, at this moment, that set a goal of 450mm for wafer size, the results would be almost identical.

[rumborak]

Scheavo, I'm pretty sure modern Economics (i.e. the science) is fully aware of these effects. To my understanding, more and more economic research relies on "agent-based" simulations, because the simplistic theories simply can't explain the sometimes erratic behavior of markets.

rumborak

[Super Dude]

tl;dr except for the OP, but the illegality of cartels must surely come into play here. For better or worse, I can see how cartels or even just of corporate collusion might be a tremendous help in dealing with such problems.

[Scheavo]

Scheavo, I'm pretty sure modern Economics (i.e. the science) is fully aware of these effects. To my understanding, more and more economic research relies on "agent-based" simulations, because the simplistic theories simply can't explain the sometimes erratic behavior of markets.

rumborak

Well ya... but I'm trying logic, since empirics has failed.

[Riceball]

Holy-economic-theory-spooge Batman, how did I miss this?

The Prisoner's Dilemma is taught in undergrad microeconomics at uni; mainly because its both the simplest way to teach someone who doesn't really want to be doing economics something useful and invites nerds like you and I to come up with all kinds of situations where it does and doesn't apply - which seems to have been done here.

IMO, theoretically the Prisoner's Dilemma is sound; I mean its a simple optimisation equation. In practise, however, its only applicable under reasonably specific circumstances:

 - Both parties are aware of the full game (decisions of the other party, payoffs etc).
 - Both parties can make their decisions unilaterally and sequentially
 - Both parties must make a decision that is specified in the payoff matrix

I mean, think about it. When does that ever really happen ever? Particularly in a macro setting.

[Orthogonal]

I also found this interesting in the article:

One big name missing from the deal is Globalfoundries, the primary manufacturing partner of AMD. It will be interesting to see if Globalfoundries enters a similar research deal with other chipmakers, although after this deal there aren't many more left.

It's not terribly surprising that AMD/Global Foundries isn't involved because they've never done their own R&D for manufacturing. Instead they are part of a consortium with Sony and Toshiba who all pay IBM for their process technology.

Just showing you a real life example (and not one I think is a problem) of where private practices can force market-wide decisions. If there was a government law enacted, at this moment, that set a goal of 450mm for wafer size, the results would be almost identical.

So, all it takes is for some politicians to say "make it so", and it's as good as done? If that were true, all we'd have to do is get them to pass the US Utopian Act and every problem would be solved, forever. Sorry, you can't legislate results.

[Scheavo]

Just showing you a real life example (and not one I think is a problem) of where private practices can force market-wide decisions. If there was a government law enacted, at this moment, that set a goal of 450mm for wafer size, the results would be almost identical.

So, all it takes is for some politicians to say "make it so", and it's as good as done? If that were true, all we'd have to do is get them to pass the US Utopian Act and every problem would be solved, forever. Sorry, you can't legislate results.

I never said that, now did I? I specifically said that if such a law were passed right now, it would be of hte same effect. Obviously, it's practical, so you can't make the argument that the government is demanding some utopian ideal (as you are); additionally, I said "set a goal," not "mandate this change tomorrow."

And no, I don't think government should be regulating this, I'm showing you the inherent similarities between regulations and normal economic activity.

Cant legislate results? Lol, that's just comical. We have and we do.

[Scheavo]

Holy-economic-theory-spooge Batman, how did I miss this?

The Prisoner's Dilemma is taught in undergrad microeconomics at uni; mainly because its both the simplest way to teach someone who doesn't really want to be doing economics something useful and invites nerds like you and I to come up with all kinds of situations where it does and doesn't apply - which seems to have been done here.

IMO, theoretically the Prisoner's Dilemma is sound; I mean its a simple optimisation equation. In practise, however, its only applicable under reasonably specific circumstances:

 - Both parties are aware of the full game (decisions of the other party, payoffs etc).
 - Both parties can make their decisions unilaterally and sequentially
 - Both parties must make a decision that is specified in the payoff matrix

I mean, think about it. When does that ever really happen ever? Particularly in a macro setting.

If you wnat to read a book, full of statistics and studies, which partly deals with this, I'd suggest Making Democracy Work, by Putnam. It's really where I'm drawing most of my idea's right now from. He did a 20 year case study of Italy, and he specifically talks about the industrial sector, their comments, and the gains made.

Also, I'd argue that such is happening right now with green technology and green investments. It seems to me to hold some water, for while companies are doing minimal things to reduce their inefficiency, I'm sure many of them would like to do more, just are afraid of investing the capital due to competition with rivals who may not invest in greener technology. Whereas if there was a massive agreement, or possibly enforced regulations regarding the issues, these worries fade away.

[Orthogonal]

Cant legislate results? Lol, that's just comical. We have and we do.

Then why can't they legislate prosperity? We're still in a recession despite numerous spending bills. Results require actions, bills are just words on paper.

[Scheavo]

Cant legislate results? Lol, that's just comical. We have and we do.

Then why can't they legislate prosperity? We're still in a recession despite numerous spending bills. Results require actions, bills are just words on paper.

I never said the government is omnipotent, or that it can create a utopia. I said that government can legislate results, which it can. You make this a hyperbolic, "well it cant 'do whateve it wants," which is an absurd reduction.

Also, they did legislate prosperity. It's the 50's, 60's, and 70's.