Caveat: You move pretty quickly from talking about "high-risk"
projects to talking about the company beta, and it isn't obvious
whether the "higher cost of capital" on your first line refers to the
standard finance theory concept or to an internal company capital
charge. For the purposes of this answer I am assuming that the
Equipment Manufacturing Division is investing in high-beta projects
rather than "risky" projects where the risk is uncorrelated with the
general market.
First, note that the effect on the firm's beta depends partly on how
the investment was financed. In an all-equity firm, the decision to
invest in higher-beta projects would tend to drive up the firm beta.
In a firm with debt finance, if the Equipment Manufacturing Division
was using internal cashflow rather than debt to finance its risky
projects, this would have the effect of reducing the firm's financial
leverage, and this effect could in principle swamp the effect of a
higher asset beta on the equity beta -- I would regard this as an
unlikely case, however.
A higher beta would mean a higher cost of capital abstracting from tax
effects. If, however, there were favourable tax effects arising from
the nature of the projects that the Equipment Manufacturing Division
was engaged in (if they could be financed out of bonds with a
favourable tax treatment, for example, as is the case for some
European industrial projects), then there would be an offsetting tax
effect on the weighted average cost of capital. Again, in principle,
this could offset the effect of the higher beta.
But abstracting from these unusual cases, we would normally assume
that the effect of investing in higher beta projects would be to raise
the intrinsic beta of the firm. Further, we might as well assume that
an efficient market would note this and would value the stock
accordingly with immediate effect (I can't see any reason to start
complicating the issue by assuming anything weaker than semi-strong
form efficiency here).
Most brokers and investment advisory services tend to measure beta
using some variant on a backward looking Ordinary Least Squares
estimation of the CAPM, estimated with a rolling five years' worth of
daily data. If we assume that the market knows about the new beta
immediately, one would expect to see the change in intrinsic beta
being incorporated into the estimated beta over a period of five
years, more or less evenly over time, and subject to the normal
standard error of the estimation process.
Note finally that in so far as the tax and leverage effects noted
above were material, brokers and investment advisory services would be
able to incorporate them into their estimated equity betas from the
date of the next published financial statements. |
