01F Mid1mu
01F Mid1mu
01F Mid1mu
1\2 L M R
T 3,2 4,0 1,1
M 2,0 3,3 0,0
B 1,1 0,2 2,3
T
B
2 2
L R L R
(3,1) (0,0) (5,0) (0,1)
3. Consider two agents {1, 2} owning one dollar which they can use only after they divide
it. Each player’s utility of getting x dollar at t is δ t x for δ ∈ (0, 1). Given any n > 0,
consider the following n-period symmetric, random bargaining model. Given any date
t ∈ {0, 1, . . . , n − 1}, we toss a fair coin; if it comes Head (which comes with probability
1/2), we select player 1; if it comes Tail, we select player 2. The selected player makes
an offer (x, y) ∈ [0, 1]2 such that x + y ≤ 1. Knowing what has been offered, the
other player accepts or¡ rejects ¢the offer. If the offer (x, y) is accepted, the game ends,
yielding payoff vector δ t x, δ t y . If the offer is rejected, we proceed to the next date,
1
when the same procedure is repeated, except for t = n − 1, after which the game ends,
yielding (0,0). The coin tosses at different dates are stochastically independent. And
everything described up to here is common knowledge.
(a) Compute the subgame perfect equilibrium for n = 1. What is the value of playing
this game for a player? (That is, compute the expected utility of each player before
the coin-toss, given that they will play the subgame-perfect equilibrium.)
(b) Compute the subgame perfect equilibrium for n = 2. Compute the expected utility
of each player before the first coin-toss, given that they will play the subgame-
perfect equilibrium.
(c) What is the subgame perfect equilibrium for n ≥ 3.