Scribd
Upload
152 views6 pages

ECON 440/640 Problem Set 2 Answers

This document contains solutions to problems from an economics class. Problem 2.3 considers a two-stage game and finds the subgame perfect equilibrium payoffs. Problem 2.5 analyzes a two-period game between a firm and worker over investment and promotion. It finds the equilibrium wages and profits. Problem 2.17 considers an infinitely repeated game where workers choose effort and gets the condition for cooperation to be sustainable.

Uploaded by

Bharavi Kothapalli
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
152 views6 pages

ECON 440/640 Problem Set 2 Answers

This document contains solutions to problems from an economics class. Problem 2.3 considers a two-stage game and finds the subgame perfect equilibrium payoffs. Problem 2.5 analyzes a two-period game between a firm and worker over investment and promotion. It finds the equilibrium wages and profits. Problem 2.17 considers an infinitely repeated game where workers choose effort and gets the condition for cooperation to be sustainable.

Uploaded by

Bharavi Kothapalli
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

ECON 440/640 Problem Set 2 Answers

Problem 2.3
Stage 2. Player 1 accepts player 2s offer if S2 1 s. Offering that minimum accepted offer leads to payoffs (1 s, 1 1 s) and making a lower offer leads to (1 s, 2 (1 s)). Hence progressing to stage 2 leads to payoffs (1 s, 1 1 s).

Stage 1. Player 2 accepts player 1s offer if 1 S1 2 (1 1 s) or S1 < 1 2 (1 1 s). If player 1 makes the minimum acceptable offer, the players payoffs are (1 2 (1 1 s), 2 (1 1 s)). If she makes a lower offer, player 1s payoff is 1 s.

Innite repetition. Following the books argument, s is dened by s = 1 2 (1 s)s(1 + 2 1) = 1 2 s = 1 2 . 1 + 2 1

Worker

Invest

Not Invest

Firm

Firm

Promote

Not Promote

Promote

Not Promote

Wd - C Yds - Wd

We - C Yes - We

Wd Yd0 - Wd

We Ye0 - We

Figure 1: Game Tree for Problem 2.5

Problem 2.5
Period 2. because The rm will not promote the worker without investment on the workers part (0, E ) = YE 0 WE > YD0 WD = (0, D) as long as WD WE , since YD0 < YE 0 . The rm will promote the worker if the worker invests as long as (S, E ) = YDS WD <YES WE = (S, D) YDS YES >WD WE .

(1)

Period 1.

Assuming (1) holds, the worker will invest if WD C > WE , or WD WE > C .

Wage setting. Lower wages are good for the rm, so it makes sense to consider two possibilities 1. WD = WE = 0 = the workers reservation utiity, and 2. WD = C , WE = 0, the minimum wages that induce investment. Equilibrium prots under option i are (0, E ) = YE 0 and under the second option are (S, D) = YSD C . By assumption investment is efcient (YSD YE 0 > c, so the rm makes the second set of offers.

Problem 2.13
Collusion. Colluding rms will charge the same price pC > c. Each rms prot from the arrangement is a pC = D ( p C c) = (pC c) 2 The prot-maximizing price level is pC = (a + c)/2. Hence the current and discounted future value of this arrangement is ( a c) 2 (a c)2 (a c)2 (a c)2 + + 2 + = . 4 4 4 4(1 )

Deviation. A rm can deviate and charge pD = pC , providing a one-time payoff of (a c)2 /2. Comparing this payoff to the value of maintaining the collusive agreement, collusion is supportable if (a c)2 (a c)2 > 4(1 ) 2 2 > 4(1 ) 1 > . 2

Problem 2.14
As in the previous problem, pH = aH + c 2 aL + c pL = . 2

Colluding in period t leads to expected prots in t + 1 to of discounted value V = = (aH c)2 (aL c)2 (aL c)2 (aH c)2 + (1 ) + (1 ) + 2 + ... 4 4 4 4 (aL c)2 (aH c)2 + (1 ) . 1 4 4

The incentives for each rm to deviate are going to be greatest when a = aH and the benets of cheating are large. The payoff to defecting and pricing at pH is (aH c)2 /2. Firms are only better off colluding if (aH + c)2 (aH + c)2 (aH + c)2 (aL + c)2 < + + (1 ) 2 4 1 4 4 2 2 2 (aH + c) (aH + c) (aL + c) (1 ) < + (1 ) 4 4 4 (aH + c)2 < ( + 1)(aH + c)2 + (1 )(aL + c)2

so (if Ri prots from collusion when a = ai ) = (aH c)2 ( + 1)(aH c)2 + (1 )(aL c)2 RH . = RH + RL + (RH RL ) 3

As a rudimentary check of whether this answer is correct, note that when = 1, = 0.5 as it should. Reducing pH will reduce the incentive to deviate, so when 0.5 < < , setting some intermediate price between pL and pH will allow credible collusion. Let p( ) be the maximum price that can be charged in aH periods if collusion is going to be successful. Let R( ) prots(p( )) = (aH p( ))(p( ) c)/4. Then p( ) is dened by the equation 2R( ) = R( ) + Solving for R( ), p2 R( ) = R( ) + (1 ) L 4 2 p (1 )R( ) = R( ) + (1 ) L 4 2 p (1 )R( ) = (1 ) L 4 (1 ) p2 L . R( ) = 1 4 1 Substituting in the denition of R( ), (aH p( ))(p( ) c) = 4 p( )2 + p( )(aH c) aH c = (1 ) 1 (1 ) 1 p2 L 4 p2 L 1 R( ) + (1 ) p2 L . 4

and solving the quadratic equation gives p( ).

Problem 2.17
Let et {0, c} be the tth workers effort choice and wt be the wage the rm pays to that worker. Consider the following strategies: wt = e1 = c et = 0 if et1 = c and wt1 < c c otherwise 4 0 if et = 0 c if et = c

Conditional on the rm following this wage-setting strategy, workers who put in effort get c + c > 0 so et = c is optimal. For the rm, offering wt = c gives 1 1 ( y c)

and paying 0 gives y . Hence the rm will prefer to not deviate if 1 1 (y c) > y y c > (1 )y y > c c > . y

Problem 3.3
When bi = bL , i (pi ) is i (pi ) = (a pi bL E [pj ])pi so di = a bL pj 2pi dpi a bL E [pj ] . p i = pL = 2 Likewise when bi = bH ,
p i = pH =

a bH E [ p j ] . 2

The expected price the other rm sets is


E [pj ] = E [p] = p H + (1 )pL

so E [p] = a bL E [ p ] + (1 ) 2 a E [p] = . 2 bL (1 )bH a bH E [p] 2

It is possible to use E [p] to solve for p H and pL , but as far as I can tell, doing so does not lead to anything interesting.

Problem 3.6
The problem asks you to show that in there are n bidders, then the strategy of bidding bi = (n 1)vi /n is a symmetric BNE. Given that everyone plays this strategy, the probability of bi being the winning bid is Pr(bi > max{bj |j = i}) = Pr(bi > b1 ) Pr(bi > b2 ) . . . = Pr(bi > (n 1)v1 /n) Pr(bi > (n 1)v2 /n) . . . = Pr(v1 < nbi /(n 1)) Pr(v2 < nbi /(n 1)) . . . = nbi n1
n1

The rst step follows because for bi to be the maximum bid, it must be greater than each other bid. Using this result, we can rewrite i as i = Pr(bi > max{bj |j = i})(vi bi ) nbi (vi bi ) = n1 n1 nn1 bi vi nn1 bn i = (n 1)n1 and use the rst order condition to nd b i (vi ):
2 n1 i (n 1)nn1 bn vi nn bi i = =0 bi (n 1)n1 n1

so (n 1)vi nbi = 0 (n 1)vi b . i (vi ) = n

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy