170 M. Baker et al.
in the firm and fiduciary duty. This leads to a similar setup to the market timing objec-
tive in Section2.1, except that an optimistic manager believes there is never a good time
to issue equity. In particular, since the capital market is efficient and values the firm at
its true fundamental value off−K, the manager believes that the firm is undervalued
byγf, and thus in selling a fraction of the firmehe perceives that existing, long-run
shareholders will lose
eγf (K,·).
Putting the two concerns together, the optimistic manager chooses new investment
and financing to solve
max
K,e( 1 +γ)f(K,·)−K−eγf (K,·).We do not explicitly include a budget constraint. Instead, again to keep the notation
simple, we consider its reduced-form impact onf.
Differentiating with respect toKandegives the optimal investment and financial
policy of an optimistic manager operating in efficient capital markets:
fK(K,·)=1
1 +( 1 −e)γ, and( 1 +γ)fe(K,·)=γ(
f(K,·)+efe(K,·))
.
Put into words, the first condition is about investment policy. Instead of setting the
marginal value created from investment equal to the true cost of capital, normalized to
be one here, managers overinvest, to the point where the marginal value creation is less
than one. The more optimistic (γ) is the manager and the less equity (e) he is forced to
raise in financing investment, the greater the problem. The second is about financing.
The marginal value lost from shifting the firm’s current capital structure away from
equity is weighed against the perceived market timing losses. As in the analysis of
irrational investors, we consider some special cases.
Investment policy. If there is no optimal capital structure, so thatfeis equal to zero,
the manager will not issue equity, settingeto zero, and there is no interaction among fi-
nancing, internal funds, and investment. In this case, the optimistic manager will clearly
overinvest:fKis less than unity. InHeaton (2002)andMalmendier and Tate (2005),
there is an optimal capital structure, or more precisely an upper bound on debt. If the
manager needs equity to invest (fegreater than zero, here), the degree of overinvestment
falls.
Needing equity is akin to having little cash or cash flow available for investment.
Thus in this setup, investment can be strongly related to current cash flow and profits,
controlling for investment opportunities. This leads to a behavioral foundation for the
Jensen (1986)agency costs of free cash flow. But instead of receiving private benefits
of control, managers are simply overconfident and overinvest from current resources
as a result. Leverage reduces the degree of overinvestment by increasingfe, thereby
increasing equity issueseand reducingK.