Part AIII continued Microeconomic Theory Review continued 092309

Part A-III (continued) Microeconomic Theory Review (continued) 09/23/09 Intro_F 1

Economics One-on-One continued n Cooperation/Non-cooperation n Bargaining Theory n Transaction Costs 09/23/09 Intro_F 2

Co-operation/Non-cooperation We saw that the outcome of a game of strategy can be very different if the players cooperate or don’t cooperate. Recall Mary and John: PAYOFF MATRIX Mary’s decision SELL DON’T SELL John’s decision SELL DON’T SELL 09/23/09 (-$2, -$2) ( $0, $1) Intro_F ( $1, $0) ( $0, $0) 3

We saw that if Mary and John decide to cooperative they can each do better (and society as whole will be better off in the sense that a Pareto Improvement will result - market failures – sometimes we don’t want them to co-operate? ? ) This simple game represents a very common situation. So why don’t agents always cooperate? They often do but they sometimes don’t We are very interested in why cooperation might not occur and whether or not POLICY can foster more cooperation among agents 09/23/09 Intro_F 4

Recall, if Mary and John decide that one of them will sell and one will not sell, together they will make $1 in profit. BUT, they must also decide how to split the profit ($0. 01/$0. 99 or $0. 99/$0. 01 or $0. 50/$0. 50 or whatever). Sounds simple enough but it generally isn’t. PAYOFF MATRIX Mary’s decision SELL DON’T SELL (-$2, -$2) ( $1, $0) ( $0, $1) ( $0, $0) John’s decision SELL DON’T SELL 09/23/09 Intro_F 5

Remember Mary does not know John’s profit function (his payoff’s under the various strategies) and John does not know Mary’s payoffs. Maybe a 50/50 split is ‘fair’ but maybe not. What if John tells Mary that even if she sells he can still make a profit of $0. 50 so if she wants him to cooperate she must give him $0. 75 and accept only $0. 25 for herself? 09/23/09 Intro_F 6

Or what if John is a slightly more profitable firm. Now the optimal cooperative outcome is for John to sell and Mary to not sell. (Note that the non-cooperative solution remains ($0, $0)) Together they will make $1. 50 in profit. Why should John agree to a 50/50 split? But why should Mary reveal that she only stands to make $1. 00 profit? She could tell John that if he sits idle and she sells she could make $1. 45. PAYOFF MATRIX Mary’s decision John’s decision SELL DON’T SELL 09/23/09 Intro_F SELL DON’T SELL (-$2, -$2) ( $1. 50, $0) ( $0, $1) ( $0, $0) 7

The primary reason why players find it difficult to cooperate is that each has a different information set Mary knows Mary’s decision SELL DON’T SELL John’s decision SELL DON’T SELL (? , -$2) (? , $1) (? , $0) John knows Mary’s decision SELL DON’T SELL John’s decision SELL DON’T SELL 09/23/09 (-$2, ? ) ( $1, ? ) Intro_F ($1, ? ) ($0, ? ) 8

Think of: union/employer, investor/firm, buyer/seller The first step towards cooperation is sharing information, for example: Trade association – generally questionable motives Union representation on company boards Russian military observers at NATO exercises 09/23/09 Intro_F 9

Bargaining Theory We need a framework (theory) which will allow us to understand the ‘structure of the process’ through which agents ‘negotiate’ a cooperative solution. The process itself is somewhat mysterious. Elements of Bargaining theory • Cooperative surplus (what’s at stake) • Reservation prices of each agent 09/23/09 Intro_F 10

An example Jones has a cottage for sale. She values it at $100, 000. She offers it for sale. Smith has $200, 000. He decides that he is interested in buying the Jones’ cottage. Smith decides that it is worth $110, 000 to him. What will happen? Will a sale go through? What will the price be? 09/23/09 Intro_F 11

We know: - - the cottage is worth $10, 000 more to Smith than it is to Jones the lowest price acceptable to the potential seller, Jones, is $100, 000. the highest price acceptable to the potential buyer is $110, 000. the problem is that they must agree on a price, and this agreement determines who gets what share of the increased well-being that results from the sale (the gains from the Pareto improving trade of cottage for money). 09/23/09 Intro_F 12

Cooperative Surplus - difference between the maximum buying price and the minimum selling price - the value created by reallocating resources between the agents ($10, 000 in our example) Reservation Price of the Buyer - the price above which the buyer will not buy ($110, 000 in our example) – the ‘walk-away price’ for the buyer Reservation Price of the Seller – the price below which the seller will not sell ($100, 000 in our example) – the ‘walk-away price’ for the seller Reservation Prices are also known as Threat Values 09/23/09 Intro_F 13

The transaction price – the price at which the trade takes place The transaction price determines the share of the cooperative surplus that each party will get If the agreed price is $105, 000, then each will share equally ($5, 000) in the cooperative surplus If the agreed price is $102, 000, then Jones gets 20% ($2, 000/$10, 000) of the surplus and Smith gets 80% ($8, 000/$10, 000) 09/23/09 Intro_F 14

Purpose of bargaining To allow the seller to try to determine the buyer’s true reservation price, while trying to convince the buyer that her reservation price is quite high. To allow the buyer to try to determine the seller’s true reservation price while trying to convince the seller that he has a very low reservation price. The process of bargaining is intended to help each side develop a notion of the other side’s reservation price. At the start, the buyer and/or seller might not even know if a trade is possible. The cooperative surplus might be zero. 09/23/09 Intro_F 15

Information revealed in the course of bargaining must be viewed as ‘credible’ by the other side - prices of other cottages in the area - the physical state of the cottage - the fact that the buyer’s spouse would rather move to Florida - the fact that the buyer is having trouble arranging a mortgage - the fact that the cottage represents a lifetime of memories to the seller and she is having second thoughts about selling - etc. , etc. 09/23/09 Intro_F 16

‘Bargaining’ makes it a ‘cooperative game’, as opposed to a ‘non-cooperative game’ No guarantee that the bargainers will be able to agree on a price (a sharing of the surplus). Negotiations might break down and no trade result. (The Pareto improving transaction might not take place. ) Negotiations might be impossible 09/23/09 Intro_F 17

Returning to Jones and Smith and the cottage Lets look at some possible outcomes and see how far bargaining theory can actually get us. Can we predict the selling price of the cottage? Cab we predict that a trade will take place? 09/23/09 Intro_F 18

Jones dominates the bargaining process Player Non-cooperative solution payoff (Threat Values) Cooperative solution payoff (Jones dominates) Cooperative surplus Jones $100, 000 cottage $109, 999 money $ 9, 999 Smith $110, 000 money $110, 000 cottage for cottage $ 90, 000 money $ 90, 001 money $ Total wellbeing in society $10, 000 09/23/09 $300, 000 $310, 000 Intro_F 1 19

Smith dominates the bargaining process Player Non-cooperative solution payoff (Threat Values) Cooperative solution payoff (Smith dominates) Cooperative surplus Jones $100, 000 cottage $100, 001 money $ 1 Smith $110, 000 money $110, 000 cottage for cottage $ 90, 000 money $ 99, 999 money $ 9, 999 Total wellbeing in society $10, 000 09/23/09 $300, 000 $310, 000 Intro_F 20

‘Reasonable Solution’ bargaining outcome 50/50 split – warning this is a purely theoretical concept Player Non-cooperative solution payoff (Threat Values) Cooperative solution payoff (Jones dominates) Cooperative surplus Jones $100, 000 cottage $105, 000 money $ 5, 000 Smith $110, 000 money $110, 000 cottage for cottage $ 90, 000 money $ 95, 000 money $ 5, 000 Total wellbeing in society $10, 000 09/23/09 $300, 000 $310, 000 Intro_F 21

If a cooperative solution results (they agree on a price and a voluntary trade takes place) the total value of the cooperative solution will always be $310, 000 Bargaining is simply a process to determine if a cooperative solution is possible and who will get what share of the surplus If bargaining is successful (a trade is made) then an ‘efficient’ Pareto improving trade occurs - who gets what is a matter of ‘distribution’ of wellbeing 09/23/09 Intro_F 22

This bargaining game can tell us the value of the cooperative surplus - it cannot tell us how the surplus will actually be divided between the two parties, or even if a trade will actually occur. Rational self-interested behaviour on the part of each agent is not enough to determine what the price might end up being in such a model Often we cannot even predict whether or not bargaining will be successful 09/23/09 Intro_F 23

When we face such situations in economic modelling. We generally assume that a ‘reasonable’ as opposed to a ‘rational’ price results – if bargaining is successful. The ‘reasonable’ solution will be the one for which the buyer and seller divide the surplus evenly. Economic theory provides no basis for assuming that the ‘reasonable’ distribution rule will actually apply. The ‘reasonable solution’ has nothing to do with fairness (or rationality). It is just a convenient assumption. 09/23/09 Intro_F 24

Some variations on the cottage sale QUESTION: Suppose Jones’ neighbour offers an identical cottage for sale at a price of $103, 000, while Jones is still bargaining with Smith. How would this change threat values, the surplus from cooperation and the reasonable solution? ANSWER: Jones’ threat value remains $100, 000 Smith’s threat value falls to $103, 000 since he now has an alternative ‘offer’ of sale at that price Cooperative surplus (from a trade taking place between Jones and Smith) now falls to $3, 000, since $103, 000 is the most the Jones cottage is worth to Smith ‘Reasonable solution’ is now a price of $101, 500. 09/23/09 Intro_F 25

Interpretation Parts of this response might appear counter-intuitive. We started with Smith valuing a cottage such as that offered for sale by Jones at $110, 000. Presumably he still values such a cottage it at $110, 000. But Jones’ neighbour offered such a cottage for sale at $103, 000 and thereby established a maximum price that Smith would be required to pay Jones. The action of Jones’ neighbour granted Smith $7, 000 in guaranteed ‘consumers surplus’, something he no longer has to bargain over or share with anyone. What remains for him is to bargain (with either Jones or her neighbour) over the remaining $3, 000 which represents the new lower cooperative surplus – although we do not know the neighbour’s reservation price. 09/23/09 Intro_F 26

QUESTION: Suppose Jones’ neighbour does not offer his cottage for sale but Jones receives an offer to purchase at a price of $105, 000 from Wilson, while still negotiating with Smith. How would this change threat values, the surplus from cooperation and the reasonable solution? QUESTION: Suppose that Smith begins to feel that purchasing a mobile home might also be enjoyable and as a result, lowers his valuation of the cottage to $95, 000. Jones’ still believes that the cottage is worth a minimum of $100, 000. How would this change threat values, the surplus from cooperation and the reasonable solution? QUESTION: What if the government imposes a 10% buyer’s fee on all cottage purchase? How would this change the original threat values, the surplus from cooperation and the reasonable solution? (Land Transfer Tax) 09/23/09 Intro_F 27

Bargaining model - summary The bargaining model reduces the bargaining process to three basic elements: 1 The threat values (reservation prices) of the buyer and seller. 2 The cooperative surplus. 3 Agreement on terms for distributing the surplus between the agents. But why does actual bargaining sometimes fail? Transaction Costs 09/23/09 Intro_F 28

Transaction Costs Is bargaining free (does it have zero costs)? No, in the simplest setting the agents must take the time to meet and haggle. In more complicated settings, they might need to spend time carefully assessing each other’s threat value (hire accountant’s, engineers, etc. ) hire bargaining agents (lawyers), etc. Bargaining can at times be very costly. These costs are referred to as transaction costs 09/23/09 Intro_F 29

Transaction costs are the costs of carrying out trade – the resources required to bring about Pareto improving voluntary exchange. In well organized markets, these costs can be very low – e. g. trading securities on the internet. At times they can be high relative to the value of what is being trade – fine art and collectible auctions 30% or more. Sales and other trading taxes are also part of transaction costs. 09/23/09 Intro_F 30

Whatever they are, transaction costs ‘cut into’ (detract from) the cooperative surplus. Consider Jones and Smith, if there is a 15% land transfer tax on the sale of cottages – can a trade go through under any conditions? Individuals trade net of transactions costs and the general rule is: The greater are the transaction costs the less trading will occur. (transaction costs are not a gain to either the buyer of seller) We need to understand the nature of transaction costs 09/23/09 Intro_F 31

The elements of transaction costs Three components of transaction costs: 09/23/09 1 search costs 2 bargaining costs 3 enforcement costs Intro_F 32

Search costs: Costs associated with locating a buyer or seller. These costs tend to be relatively high when trading in unique goods or services (a spare part for an vintage automobile) and low for standardized goods (most consumer goods). 09/23/09 Intro_F 33

Bargaining costs: Costs associated with carrying out the negotiations. If each party’s threat value and the cooperative surplus is known to all parties then bargaining costs will be relatively low. In general the more information concerning all aspects of the trade that is public to the parties, the easier it will be to reach a bargain - otherwise the bargaining process will have to make that information known. Bargaining will be more difficult (more costly): the greater is the number of parties; the greater is the amount of ‘primary uncertainty’ associated with the transaction (requiring agreement on contingencies); the greater is the extent of hostility between the bargainers; the extent to which there is ‘overreaching’, on the part of one or both parties. 09/23/09 Intro_F 34

Enforcement costs: Costs associated with monitoring the behaviour of the parties and punishing failure to fulfil agreements. If there is no time lag between the bargain being reached and being fulfilled, then there will be no enforcement costs. If there is a lapse of time between bargaining and fulfilment of the bargain conditions, then monitoring and enforcement costs will arise. Note that many apparently instantaneous transactions carry warranties or guarantees of one sort or another and therefore monitoring and enforcement costs exist. 09/23/09 Intro_F 35

Consider the purchase of illicit drugs versus alcohol All three components of transaction costs are likely to be higher for illicit drugs: search costs, bargaining costs and enforcement costs. This results mostly from the fact that recreational drugs are illegal. Would legalizing the sale of recreational drugs increase their use? It would certainly make the transaction costs fall, implying more successful bargaining. 09/23/09 Intro_F 36

To summarize At times the cooperative solution to a potential conflict of interest will yield a Pareto improvement over the non cooperative solution – there is a cooperative surplus, or gain from trade) However, the agents must cooperate successfully – agree to bargain and eventually come to some agreement on the division of gains Since neither agent generally has perfect information this bargaining is costly (the information must be developed) This gives rise to transaction costs Bargaining will breakdown (fail) if the transaction costs exceed the cooperative surplus We are interested in situations in which the law can facilitate successful cooperation. 09/23/09 Intro_F 37