model. With probability (1 - α) the project is ”bad”: it yields a zero payoff
with some positive probability x. The stochastic structure of the payoffs for a
bad project is summarized in the following matrix:
RH e2 - ∆ - x
RL 1 - e2 + ∆
0x
where e2 is effort exerted by the second entrepreneur.
At the contracting stage neither the venture capitalist nor firm 1 knows
firm 2’s type.
Timing
The timing of events is as follows:
t=1 (Contracting stage) The investor offers a contract to Firm 1.Firm1
accepts or rejects. At this stage neither the firm nor the investor knows Firm
2’s type.
t=2 (Start-up stage) In the process of starting Firm 1’s project, the en-
trepreneur and the venture capitalist observe a signal about Firm 2’s prof-
itability: as a result, they find out whether Firm 2 is good or bad. The signal
is verifiable.28 At the end of this stage both parties have an exit option: they
can abandon the project if their continuation payoff is negative.
t=3 (Production stage) Firm 1 picks a level of effort e ∈ [∆, 1]. This
decision is not observed by the investor.
t=4 The investor decides whether to fund Firm 2 or not; if he decides to
do so, he offers a contract to Firm 2, which then accepts or rejects. Firm 2
observes whether Firm 1 has been funded and knows its type when considering
the investor’s offer. If Firm 2 accepts the contract, it then picks an effort level
e2 ∈ [∆ + x, 1].
t=5 Payoffs are realized according to each manager’s level of effort and
whether the investor has funded one or both firms. The first entrepreneur and
the venture capitalist share returns according to the contract signed at t=1.
If Firm 2 was funded at t=4, then the second entrepreneur and the venture
capitalist share Firm 2’s profits according to the contract they signed then.
Firm 2’s credit-worthiness
We define:
V2 = Maχe2,Rl2,Rh2 (e2 - δ - x)(R - RH) + (1 - e2 + ^(RL - RL2) - I2
s.t. :(e2 - ∆ - x)RbH2 +(1 - e2 + ∆)RbL2 - Ψ(e2) ≥ 0
RH - RL2 = ψ0(e2)
28If the signal is not verifiable, the analysis would proceed in much the same way. There
would be an additional incentive compatibility constraint on the investor’s program (given
below) to ensure that the investor indeed converts his debt when he would otherwise face
the Coase problem.
24