Texts: played by a worker (W) and a firm (F).
10.10
M
6.10
F
N
WH
F
(r)
q
4.4
c
dyH
c
0.4
10.0
M
Low 1-p
M
3.0
i(1-r) N' WL
q)i
E'
-3.4
The worker has private information about her level of ability. With probability p, she is a high-ability type (H) and with probability 1-p, she is a low-ability type (L). After observing her own type, the worker decides whether to obtain a costly education (E) or not (N); think of E as getting a degree. The firm observes the worker's education, but the firm does not observe the worker's quality type. The firm then decides whether to employ the worker in an important managerial job (M) or in a much less important job (C). The payoffs are represented in the above extensive form.
a) Is there a separating perfect Bayesian equilibrium in which the high-ability type worker obtains an education and the low-ability type does not? If yes, fully describe such an equilibrium. If no, prove there is not such an equilibrium.
b) Are there any pooling perfect Bayesian equilibria in which both types of workers take the same actions (pure strategies regarding education)? If yes, fully describe all such equilibria. If these depend on p, describe how. If no, prove there is not such an equilibrium.
c) If there are equilibria in both a and b, which does the firm prefer? Explain.
d) If there are equilibria in both a and b, which does the low-ability worker prefer? Explain.
e) If there are equilibria in both a and b, which does the high-ability worker prefer? Explain.
f) In light of your answers to c, d, and e, is it possible to rank the pooling and signaling/separating equilibria from a social welfare perspective? Explain.
