Divison of Labor??

I am currently reading a really great paper by Dr. Richard Thaler and Dr. Cade Massey entitled "The Loser's Curse: Overconfidence vs. Market Efficiency in the National Football League Draft."  I am not done with it yet, but it is very insightful so far.  (If you're interested, it can be found on here.) 

Similarly, I have not spent much time thinking about this quote (I will post my own thoughts soon), but I wanted to change things up a bit and ask for your opinion first.  Here is the quote from Dr. (presumably) Gary Becker as noted in an article by Dr. (again, presumably) S. Stewart:

"Division of labor strongly attenuates if not eliminates any effects caused by bounded rationality. ... it doesn't matter if 90 percent of people can't do the complex analysis required to calculate probabilities.  The 10 percent of people who can will end up in the jobs where it's required."

So, a proverbial penny for your thoughts??

The Protestant Work Ethic and Capitalism

Here is an excerpt from a paper I wrote as an undergrad about the consequences of Puritanism in America (please pardon the flat and overall poor writing style):

America is the great capitalist society in history.  According to the highly acclaimed sociologist Max Weber, capitalism is a result of Protestantism in Europe.  With belief in the doctrine of predestination, many believers sought methods that would indicate if they were chosen by God to go to heaven or not.  One method devised by the leaders of the Protestant movement was economic success.  They said that if a good Protestant believer worked hard and was successful, this could be a sign that they were to be brought to heaven.  They effectively turned work and economic prosperity into religious practice and signs from God.  This initial Protestant movement urged frugality.  While economic success may be a sign from God as to one’s status in the afterlife, spending all of these monetary gains on things other than the church was frowned upon.  In fact, in John Winthrop’s A Model of Christian Charity, he quotes Proverbs 3.9 “honor the Lord with thy riches” to say that wealth should be accorded to the church or attained with God’s purpose in mind.  Thus, the initial impetus for capitalism can be traced to the Protestant reformation and their belief in the doctrine of predestination.  Since Puritanism is a branch of Protestantism, the Protestant work ethic, as this has come to be called, was also transplanted from Europe to the New World along with the settlers.  The Puritans sought what they called ‘competency’, the possession of enough property to ensure the family’s economic independence.  While seeking familial economic independence, surpluses were traded and acquired at fair market rates and services were traded.  Thus, it can be claimed that the genesis of American capitalism is the Puritan colonists that settled parts of America over 350 years ago.

Just a little background for you:  The Protestant schism was started by Martin Luther posting his 95 Thesis, essentially 95 talking points about how the Church's then-current practice of selling indulgences (no matter what the Catholic Church's official stance on indulgences is at this point in time, at that point in time, indulgences were essentially pardons for a sin that you bought from the church) was, for lack of a better word, bad.

Anyway, this eventually lead to the establishment of Lutheranism (which is the oldest and, hence, original branch of Protestantism).  Puritanism is another branch of Protestantism (or at least was originally...  Things change you know.)

…Don’t I know you?

Okay, so there is a slight problem with my previous post… it only takes into consideration a single interaction and completely ignores repeated interaction. What I mean is that if you offer John $4.99 and he turns you down, what will happen if it is then his turn to make the offer and he offers you $4.99? We can go through that entire analysis again, but we would also need to take the history of transactions into consideration. In other words, don't you think you would be more likely to turn down the offer if he turned it down? You may say that that is illogical because you would be better off accepting $0.01 in the second iteration if you got nothing the first time around because at least you are getting something out of it, but this logic depends on 1) an absolute valuation of the transactions and 2) that you remove your emotion from your decision. Revenge is a powerful motive…

So essentially I wanted to briefly discuss repeated market interactions because it is a concept that is extremely important in economics. But instead of just leaping into it, let's gradually build up to it by starting with a classic problem from economics commonly known as 'prisoner's dilemma.'

Prisoner's Dilemma

The classic question referred to as the 'prisoner's dilemma' goes something like this: Bob and Joe are apprehended by the police. The police put Bob in one room and Joe in another. The police lay out the options to Bob and Joe. Both are given the option to remain silent or confess. The payoffs for each option are given below in the table. (A note on the question is that you are to keep the two criminals separate in order to, essentially, make their fears that their partner will rat them out push them to confession. By putting them in the same room, however, you're more likely to elicit a confession from at least one of them because if Bob confesses in front of Joe, Joe should confess as well. If Bob denies in front of Joe, Joe should confess. You might argue that whomever answers first has that aforementioned fear, but the second player just acts rationally and with perfect information instead of acting upon imperfect information. While this does produce the desired outcome as well, it does not illustrate the economic concept of the Nash Equilibrium.)

		Joe	Joe
	Bob/Joe	Deny	Confess
Bob	Deny	1 Year/1 Year in Prison	5 Years/0 Years in Prison
Bob	Confess	0 Years/5 Years in Prison	3 Years/3 Years in Prison

Okay, so now that we have the payoffs, what should each person do? Both should deny the charges? Well, there is a problem with that… If I am Joe and I think Bob is going to deny the charges, wouldn't I be better off by confessing? (Sure, you can argue that I should also deny to save my partner in crime some jail time, but, as the saying goes, "there is no honor among thieves.") So then they should both confess? If both confess, a total of 6 years is spent in jail. True, each person spends fewer years in jail than if they deny and their partner confesses, but again, it is rational for each partner to assume that, so the apparent answer is to confess. In fact, confessing is considered to be a dominant option. Let's take a quick look at things from Bob's point of view (the first payoff in each cell). If he thinks that Joe will deny the charges, Bob is best off confessing, right? Okay, now what if he assumes that Joe will confess? Again, it is in Bob's best interest to confess. Hence, it is always in Bob's favor to confess.

Let me guess, you are thinking, "Okay great, so now I know how I should act if I commit a crime with a partner, we both get caught, and these are our options. How exactly does this help me with repeated interactions?!?"

Let's now assume that you go to a store, any store. Here are the possible situations that will occur:

		Store	Store
	You/Store	Give good	Keep good
You	Pay	Pay/Give	Pay/Keep
You	Don’t Pay	Don’t Pay/Give	Don’t Pay/Keep

Admittedly, it seems like the store really only has one option (which is to give you the good you pay for), but that isn't because of the market… that is due to an external force: the law. So from a pure economics point of view, the store really does have two options. If you will only shop at the store once, the optimal strategy for the store is to keep the good whether you pay for it or not. That way, they might get your money and they still have the good available for sale. Of course, your optimal strategy (morals aside) would be to steal the good. But here is where the concept of repeated interactions comes into play. Let's assume you pay, but don't receive the good. From now on, you will be wary of shopping at the store. If you stole the good (we are assuming the store knows you stole it), the store will be wary of your presence. Here is the catch though: this is a micro view of the situation. The real consequences of this interaction are played out on a much larger scale.

Think of it this way, you pay for the good, but the store does not give you the good. What is the first thing you do? (Remember, we are talking economics, not law, so 'call the cops' is not a relevant option) You will probably tell your friends about what happened. So based on your single experience with a store, you are affecting the store's future interactions with your friends. It is very possible that, if this happened to you, you would never shop at that store again and neither would your friends. This is the market effect we are really concerned with. Unless you are a large enough buyer to have actual buyer power in the Porter's Five Forces term of the phrase, the store will not really care about losing you as a customer, but they will care about losing a lot of yous.

In the case of customers shopping at a store, reputation and information sharing are built-in mechanisms that ensure fair treatment and instantly turn your single interaction into repeated interactions by multiplying 'you'. For some perspective, the purpose of 'law' from an economic perspective is to incentivize reputable action.

Clearly, repeated interaction is a powerful force in the market, but how can it affect you outside of shopping experiences? Repeated interaction is the basis for statistical analysis of many things ranging from financial markets to sports. In baseball, for example, statistics are kept on how a batter does against a right-handed pitcher (or the specific pitcher he is facing) with a full count, with the bases loaded, with the wind blowing in, at night. Seriously. What is the point of such a statistic? Well, presumably, by measuring how he did with this specific set of circumstances in the past, you can predict what will happen in the current at bat (okay, 'predict' might not be the best word… but it will help you to statistically forecast what will happen). Unfortunately, I cannot find a free source online to show you, but if you ever watch a baseball game on ESPN/ABC, they get their statistics from the Elias Sports Bureau who publishes a yearly report of all of their baseball statistics. Similarly, all of the statistics kept on how stocks perform are used to help forecast how a stock will perform in a market with similar conditions (since 'identical conditions' would almost certainly never happen).

So yes, I am saying that statistics is essentially the quantification of the economic concept of repeated interaction.

Just keep in mind that whenever something happens in which the history of that event is taken into consideration, you are looking at repeated interactions.

(One final note… While the concept of repeated interactions is exemplified by looking at the history of an event or by tracking statistics of an event, that does not mean that 'repeated interactions' in the economic sense is what is being analyzed… but it could be.)

$10

Today a friend of mine was telling me about an experiment she was a part of. Imagine that you are given $10 and told to split it between you and another person, John. The catch is: you make an offer to John, if he rejects it, neither of you get any money. If he accepts it, you both get the agreed upon amount. What is the optimal strategy? Split it 50/50, right? ...Right?!? Well, maybe...

Perhaps the simplest way to look at this is to say that splitting the money in half between you and John is the optimal solution.  The problem is that that assumes that John knows 1) that you have money that you are supposed to split between the two of you and 2) how much money that is.  If John doesn't know about the money, you could get away with offering him $0.01 (assuming you have to offer something) and still get to keep the $9.99.  Assuming John knows about the money but not the amount of money, you could still get away with $0.01... at least presumably.  But let's throw these assumptions out the window and say that John knows both that you have money to split with him and how much money you have to split with him.  So 50/50, right?

Well, before we answer that question, let's look at a very simple and a very simplistic evaluation of the situation.  It could be argued that if you offer $4.99 or less to John, you are being greedy and he should reject.  Alternatively, it could be argued that if you offer John $5.01 or more, you are being unnecessarily generous.  So, 50/50 then is viewed as a 'fair' amount.  .....But we still have a couple problems with this line of reasoning.  

The first big problem here is that we are assuming that John has as much right to the money as you do.  Only under this assumption would we say that anything less than $5 exemplifies you being greedy and anything more than $5 exemplifies you being generous.  Of course you could be told that you are acting as the CEO of a publicly traded company in which case the earnings you are to distribute really do belong to John.  Even still, you can probably come up with a legitimate way to spend the entire $10, thus justifying not giving anything to John for expected higher stock prices.  But what if you aren't the CEO of a publicly traded company?  What if you are just a person randomly picked off the street to participate in this experiment?  Under these circumstances, does John really have any right to the money?  It is in your possession to begin with... And anything you offer will make him better off than he previously was.  So if we operate under the assumption that John has no legitimate claim to the money, you can get away with offering him $0.01, right?  Maybe...

This greatly depends on how John and you are valuing the transaction.  In many economic theories, wealth is valued in the absolute.  So according to these theories, John should accept any offer of $0.01 or greater because it increases his wealth.  So the answer is $0.01.  Similarly, you should be comfortable giving John $9.99 because you would increase your wealth by $0.01 with 100% certainty.  So are either of these answers the right answer???

Well, we still can't say that with any certainty.  The problem with this method of transaction valuation is that it doesn't take into consideration one of the most fundamental human characteristics: emotion.  Okay, before we go any farther, let us just remember the situation: 1) you have been given $10 to split between yourself and John, 2) John knows that you have $10 to split between yourself and him, and 3) if John rejects your proposal, you both get nothing.

Okay, so you are probably thinking that the penultimate sentence in the paragraph before last is kind of fishy, but that requires you actually having possession of the money before divvying it up...  If you should split the money in the form of two checks, one for you and one for John, things may very well be valued differently because there are externality costs associated with a check as opposed to cash.  For instance, John now needs to go to a bank to either cash the check or to deposit it into his account; this money is not immediately available to him.  This simple fact may cause John to value the potential transaction differently than if it was cash.  If it is cash, John may prefer $9.95 to $9.99 because it would be fewer coins to carry and no pennies (presumably). 

As I said, the amount you offer depends greatly on how both you and John are valuing the transaction.  Many theories claim that people value transactions in a relative sense, not an absolute one.  If this is true, then John should accept $5 or higher only, right?

We still can’t say that, unfortunately, because of emotion.  At the most basic level, John would be indifferent on an offer of $5, so theoretically he would reject it 50% of the time.  By increasing the offer by a single penny, the probability that John will accept increases greatly, but not to 100%. 

So if we assume that you want to maximize the amount of money you would get from this transaction, there are a couple of factors to take into consideration: how John is valuing the transaction, how entitled John feels to the money, and how much information about the transaction does John really have? 

In general, my advice would be that if you are dividing cash, offer half plus a nickel; if you are dividing a check, offer half plus a penny.

100+% APR ?!?

Another great article from Tim Harford discussing microfinance and microcredit:

http://timharford.com/2008/12/are-loans-at-100-per-cent-apr-good-for-the-poor/