Ch. 6: Security Offerings 335
are low compared to the stock returns of similar non-issuing firms (also shown in
Section5.3below). However, they argue that this observation follows naturally when
projects are viewed as options on the cash flow potentially generated by the project.
When project execution is flexible in time, a project becomes a real option. Managers
can time the starting time of the project to maximize the value of the firm. An option
to grow the company through execution of the project is a levered claim. The required
return on a levered claim is higher than the required return on an unlevered claim on the
same assets. Exercising the real option, i.e., making the investment necessary to start
the projects, unlevers the claim. Thus, when firms grow they convert real options into
assets in place. The assets may be risky, but an option on these assets is even riskier.
Thus, when projects are financed using seasoned equity, the model predicts that realized
returns on average should be lower after a SEO. This does not happen because the SEO
is timed, but rather because there has been a fundamental shift in the riskiness of the
firm’s assets. Since growth options only are exercised when they move sufficiently in-
the-money, the model also explains the pre-issue stock price runup.
In the model ofCarlson, Fisher, and Giammarino (2005, 2006)the required return
is endogenous and depends (among other things) on the optimally timed investment
decisions made by the firm. If the expected return is assumed to be time varying but
exogenous, more projects will become profitable as the discount rate drops. This will
increase investments and lead some firms to raise capital. Thus, time varying expected
returns predict that stock prices will rise prior to equity issues and that returns will be
lower after the issue.Pastor and Veronesi (2005)develop a model of IPO waves along
these lines. Their model predicts that IPOs should cluster and that such IPO waves
should be preceded by high market return and followed by low market return.
The relationship between investments and stock return was first formalized by
Cochrane (1991). In a production based asset pricing model, Cochrane shows that a
firm’s investment return (the rate of return obtained on the marginal real investment)
should be equal to the stock return. Thus, when the real investment level is high, the
marginal return on invested capital is low, and stock returns should be correspondingly
low.Cochrane (2005)interprets this argument as a first-differenced version ofQ-theory
of investment.Zhang (2005)develops theQ-theoretical argument further. Zhang fo-
cuses on time varying expected return and shows howQ-theory, among other things,
implies that firms conducting a SEO should have lower post-issue returns than oth-
erwise similar firms.Lyandres, Sun, and Zhang (2005)explore the investment based
explanation for the low long-run stock returns of SEO firms. They find the investment
to asset ratios of SEO firms are about twice as large as the investment to asset ratios of
non-issuing firms. Thus, under theQ-theory of investment, the expected return of SEO
firms should be lower than the expected return for non-issuing firms.
In sum, the investment based theories predict that subsequent to an SEO, a firm will
have lower market risk and thus, lower expected rates of return. This offers a potential
explanation for the finding, discussed in detail in Section5.3below, that stock returns
are relatively low—but not necessarilyabnormallylow—following SEOs or IPOs. It
also suggests that matching an equity-issuing firm with a non-issuing firm based on size