Full text of Economic Commentary (Federal Reserve Bank of Cleveland) : Stock-Market Gyrations and Investment, December 1, 1987

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

finance and numerous aspects of the
tax code, including personal and corporate tax rates, investment tax credits,
and depreciation deductions. Stockmarket fluctuations influence the cost
of capital by affecting the expected cost
of equity finance. A decline in the stock
market implies an increase in the
expected cost of equity finance and,
thus, in the cost of capital.
Focusing on the cost of capital as the
link between the stock market and investment, however, ignores the advantages of the q approach. In theory,
stock- market fluctuations occur in response to new information about both future demand and future capital costs. If
true, this implies that q should be more
informative than the cost of capital.
As a practical matter, however, the
advantages of the q approach have yet
to be realized. While financial markets
seem to respond to a wide variety of
economic data, economists have not
reached agreement on how to isolate
the information contained in market
values relevant to investment decisions. If market values are not equal to
References
Elmer, Peter J., and Patrie H. Hendershott. "ReI·
ative Factor Price Changes and Equity Prices,"
National Bureau of Economic Research Working Paper No. 1449 (September 1984).
Fazzari, Steven, Glenn R. Hubbard. and Bruce C.
Petersen. "Financing Constraints and Corporate Investment." National Bureau of Economic Research Working Paper No. 2387 (September 1987).

Federal Reserve Bank of Cleveland
Research Department
P.O. Box 6387
Cleveland, OH 44101

Address Correction Requested: Please send
corrected mailing label to the Federal Reserve
Bank of Cleveland, Research Department.
P.O. Box 6387. Cleveland. OH 44101.

the present discounted value of future
returns to capital, the link between q
and the rate of investment is weakened
substantially. In addition, measuring
the replacement value of the capital
stock is complicated by continual technological change and lack of price
information for many types of unique
physical assets. Measurement of
market values of financial liabilities is
difficult because many financialliabilities are not widely traded.
Conclusion
In this Economic Commentary, we have
focused on the relations between q, stockmarket fluctuations, and investment.
The links between the three are greatest
if financial markets correctly incorporate information about future returns
from the capital stock. Even then, however, the response of investment to a
change in q is not immediate because of
costs incurred in adjusting the capital
stock and the delay inherent in the
appropriations-orders-investment
process. Another problem that persists even
Fischer. Stanley. and Robert C. Merton. "Macroeconomics and Finance: The Role of the Stock
Market," National Bureau of Economic
Research Working Paper No. 1291 (March
1984).
Keynes. John M. The General Theory of Employment, Interest, and Money, London: Macmillan
Co., 1936.
Moore. Geoffrey H. Business Cycles, Inflation,
and Forecasting, National Bureau of Economic
Research. Studies in Business Cycles, no. 24,
Cambridge. MA: Ballinger Coo, 1980.

if financial-market values are correct is
the difficulty of calculating marginal q.
The usefulness of average q, which is
more easily measured than marginal q,
is limited in analyzing short-run
changes in investment. If stock-market
values are not "correct," as may have
been the case during the recent market
rise and plunge, the link between q and
investment is weaker still. In that case,
movements in the stock market may be
expected to influence investment
through their effect on the cost of
financing investment.
However, recent research emphasizes
that some firms may fail to respond to
changes in the cost of equity finance
because of financial constraints. Of
course, a focus on the cost of capital as
the mechanism through which stockmarket fluctuations influence investment ignores the advantage of q:
q should incorporate information about
more than just the expected cost of capital. Unfortunately, as a practical matter, the advantages of the q approach
have yet to be realized.

Shapiro, Matthew D. "The Dynamic Demand for
Capital and Labor." Quarterly Journal of Economics, vol. 101 (August 1986), 512-542.
The Conference Board. "Capital Appropriations,"
First Quarter 1987.
Von Furstenberg, George M. "Corporate Investment: Does Market Valuation Matter in the
Aggregate?" Brookings Papers on Economic
Activity 2 (1977), 347·408.
Zarnowitz, Victor. Orders, Production, and
Investment-A
Cyclical and Structural Analysis, National Bureau of Economic Analysis.
New York: Columbia University Press, 1973.

BULK RATE
U.S. Postage Paid
Cleveland, OH
Permit No. 385

Material may be reprinted provided that the
source is credited. Please send copies of reprinted
materials to the editor.

December I, 1987

Federal Reserve Bank of Cleveland

ISSN 042fl-1276

ECONOMIC
COMMENTARY
The worldwide stock-market decline on
October 19 has increased uncertainty
about future changes in employment
and output both in the United States
and abroad.
Part of the reason for this uncertainty is that changes in the level of
equity prices have been one of the best
leading indicators of economic activity
(Moore, 1980). In particular, the stockmarket decline may affect consumer
spending and business purchases of
plant and equipment.
Consumer spending may be influenced through changes in the level of
consumer confidence and changes in
consumer wealth. Business fixed investment (BFI) may be affected by changes
in the cost of financing investment and
by changes in businesses' expectations
of future demand.'
In addition, there is a presumption
that stock prices, to some extent, reflect information about future demand,
future interest rates, and a wide variety
of other factors that are related to
future economic activity. Whether stock
prices correctly reflect the best available
information is a widely debated question in the economics profession. To
the extent that they correctly reflect
new information, stock prices may be
valuable aids in forecasting future economic activity, particularly investment.
While some preliminary information
useful in predicting future consumer
spending and BFI becomes available
before actual spending data are
released, the data releases are infrequent compared to the almost instantaneous revaluations of equities
reflected in the stock market.

William P. Osterberg is an economist at the Federal
Reserve Bank of Cleveland. The author is grateful
to Mark Sniderman for helpful suggestions and to
Ted Bernard and [enifer Lee for research
assistance.
The views stated herein are those of the author
and not necessarily those of the Federal Reserve
Bank of Cleveland or of the Board of Governors of
the Federal Reserve System.

In this Economic Commentary, we
analyze the relationship between stockmarket gyrations and business fixed
investment. We focus on BFI rather
than on consumption for two reasons.
First, although BFI comprises only 10
percent of gross national product (GNP),
fluctuations in BFI are tied closely to
changes in GNP. Second, if stock prices
correctly reflect the best available
information, then stock-market fluctuations should be closely tied to BF!.
One widely used investment theory directs us to focus on the ratio of the
market value of financial liabilities to
the replacement value of physical
assets, a ratio called q. In this article,
we use the q theory to examine the relationship between the stock market and
BF!. We find that, even if stock-market
values are "correct," stock-market fluetuations only indirectly influence BF!.
Investment and q
A relationship between stock-market
fluctuations and investment was predicted by Keynes (1936, p. 151):
...daily revaluations of the Stock
Exchange ... inevitably exert a decisive influence on the rate of current investment. For there is no
sense in building up a new enterprise at a cost greater than that at
which a similar existing enterprise can be purchased ....
This passage, and subsequent developments in investment theory, implies
that there should be a close relationship between q (the ratio of the market

1. Business fixed investment (BFI) refers to the
nonresidential business fixed investment component of the GNP accounts. Expenditures on plant
and equipment constitute about 90 percent of BFI.

Stock-Market
Gyrations and
Investment
by William P. Osterberg

value of financial liabilities to the
replacement value of physical assets)
and the rate of physical investment."
The theory relating q to investment is
the "preferred theoretical description of
investment" because it links investment to expectations about the future
(Fischer and Merton, 1984).
The link between q and investment
is strongest if the financial markets'
valuations of debt plus equity correctly
reflect relevant economic information
and if investment decisions are made so
as to maximize the market value of a
firm's liabilities (debt plus equity)."
In theory, the correct value of debt
plus equity reflects all information
about the returns to be received by the
owners of a firm's physical capital. In
fact, the market value of debt plus
equity should equal the present discounted value of the future after-tax
returns to be received by the bondholders and stockholders. Thus, if the
financial markets' valuations are correct, they reflect two types of information: 1) information about the future
returns to be received after firm
revenues are used to pay wages and
taxes, and 2) information about rates of
return available on alternative investments (these are used to "discount" the
future returns, that is, to translate
them into present values).
If the value of debt plus equity
reflects the value of the returns to be
generated by the capital stock, and if
our second assumption is met, changes
in the value of debt plus equity should
be related to decisions to add to the capital stock. Because investment is an

2. While this article focuses on BFI. it is difficult
to separate the market value of physical assets
such as plant and equipment from the value of
other assets such as land. inventories, and
intangibles such as future investment opportunities or patents.

addition to the capital stock, however,
firms need to compare the market
value of the returns to be generated by
additions to the capital stock and the
cost of adding to the capital stock. If
the ratio between the market value and
the cost for additions (marginal q)
exceeds one, net investment (BFI
minus depreciation) will be positive. If
marginal q falls below one, net investment will be negative.
The act of building up or reducing the
capital stock alters average q (the market value of all existing liabilities divided
by the replacement value of the existing
capital stock) by increasing the total
replacement value of the capital stock.
Average q will eventually equal one, as
indicated in the quotation from Keynes.'
Under the assumptions of the q theory, the value of current q contains
most, if not all, of the information
needed to predict investment. An alternative approach would focus on individual variables affecting the market
value of the firm's liabilities, because
maximization of that total is the objective for investment decisions.
Measuring expectations of such variables is difficult, and the information
required is not readily available. The
advantage of the q approach is greatest
if financial markets and, thus, q correctly reflect information about expectations. In this case, no individual variable relevant to investment decisions
should help q predict investment, since
such variables influence the market
value of the firm's liabilities (the
numerator of q).
In figure 1 we have plotted average q
and the rate of investment for nonfinancial corporations." The likely
impact of the stock-market decline on
average q can be approximated by noting that equity has comprised 60 percent of the total market value of liabilities. The average price of stocks, as
measured by the Standard and Poor's
500 index, declined 22 percent from
June 30, 1987, to the end of November
1987, so we would expect the numerator of q to fall by 13 percent. According
to q theory, the rate of investment following this decline will be lower than it
otherwise would have been.s

3. The studies cited in the text differ as to the
assumed objective.
4. A slightly more complicated version of the q
story implies that average q may not equal one in
the long run. An investment· tax credit, for example, alters the replacement value of the capital
stock and, thus, q.

Figure 1 Average q and the Rate of Investment for
Nonfinancial Corporations
1.5r------------------------~.020

O~~~~~~~~~~~~~~~~~~~~~~~~~.OOO

1950

1960

1970

1980

1990

SOURCE: See footnote 5, below.

Lags in the Response of
Investment to a Change in q
Even if a change in q reflects a change
in the correct valuation of the future
returns generated by the capital stock,
the response of investment to a change
in marginal q will be delayed and
drawn out. If the stock market rises,
increasing q, firms first have to decide
that a larger capital stock is desired
before appropriating funds and placing
orders for future delivery. If q falls,
firms could reduce investment by canceling orders, or could decide to reduce
appropriations or orders.
Since few orders are actually canceled, investment expenditures depend
on previously placed orders. To further
complicate matters, expenditures do
not necessarily coincide with the delivery and installation of new capital
goods. In fact, there is some evidence
that payment is spread out between the
time that orders are placed and the
time of delivery.
Because it is costly to change the size
of the capital stock rapidly, the response
of investment will be spread out over
time. Investment could be costly if
firms have to pull resources away from
production to adapt to new plant and
equipment. The cost, in terms of lost
output, of increasing the capital stock
may vary with the rate at which the
capital stock is increased. This implies
that the net return from investing
decreases with the rate of investment.

At first, if q has increased, a relatively high and costly rate of investment can be justified. After the initial
response, the net benefit from additional investment falls (since the marginal product of capital declines with a
larger capital stock), and subsequent
rates of investment will be lower.
Several empirical studies have examined the lag lengths involved in the
investment process. Von Furstenberg
(1977) found that, given an average
ratio of unfilled orders to shipments, 20
percent of new orders for plant and
equipment received one quarter ago
were shipped in the current quarter. In
the second, third, and fourth quarters,
44 percent, 28 percent, and 7 percent of
new orders were shipped, respectively.
The response of investment to a
decline in q, however, should be somewhat different from the response to an
increase in q. At the extreme, investment can be reduced to the point at
which the capital stock is allowed to
wear down or be slowly scrapped.
Cancellation of some orders or
appropriations is a more likely response
to a decline in q. In fact, the ratio of
orders canceled to new orders increases
in downturns (see Zarnowitz, 1973). On
average, the rate of cancellation of
orders or appropriations is low. Only 5
percent of appropriations by manufacturing firms were canceled in 1986 (see
The Conference Board, 1987).

5. The calculation of q follows that of Von Furstenberg (1977). For years since 1976, I assume
that the ratio between preferred and common
dividends is still at its end·of·1976 value. Real
capital stock series are calculated separately for
structures and equipment, then added together.
For each series I utilize the perpetual inventory

formula with the starting values set at the endof·1950 net-constant-dollar
stocks for nonfinancial corporations published in the Survey of Current Business. The expenditure series are the
real expenditures on plant and equipment from
the National Income Accounts. I utilized .014 and
.032 as the quarterly depreciation rates in the
perpetual inventory formula. The investment
rate was then calculated as [K(I+ 1)·K(t)]/K(I),

Marginal and Average q
Because investment concerns additions
to the capital stock, the relevant measure of q is marginal q. The stock market, however, reflects not just the value
of the returns from the new capital
stock, but the value of the returns from
the existing capital stock. Because of
the difficulty of separating the change
in the value of new capital stock from
the changed value of old capital, economists focus on average q, calculated as
the ratio of the market value of a firm's
existing debt plus equity to the replacement cost of its existing capital. The
link between average q and investment
is weakened by divergences between
average and marginal q.
An example of such a divergence is
when new information leads to an
upward revision of expected future
energy prices. The market's value of
the existing, less-energy-efficient capital would decrease relative to the value
of returns anticipated from new capital
designed to economize on expensive
energy. In such a case, average q may
fall in spite of increased incentives to
invest. More generally, changes in expectations of relative factor prices may
stimulate investment while decreasing
average q.
Changes in expectations about the tax
code can also cause marginal and average q to move in opposite directions.
The investment tax credit affected the
replacement cost of new capital and,
thus, marginal q. Tax-code provisions
regarding deductions for depreciation
affect the replacement value of the
entire capital stock.
There is some evidence that changes
in expectations can explain some of the
historical relation between average q
and investment. Elmer and Hendershott
(1984) demonstrated that changes in expected prices of capital, energy, labor,
and materials can explain the decline in
average q during the late 1960s and
1970s.
Therefore, when expectations about
future relative input prices are being
revised, the link between average q and
investment is weakened and the stock
market is a poorer indicator of future
investment. However, if expectations
are not systematically high or low,

where K(I) is the stock of structures and equipment centered on quarter I, to correspond to the
centering performed in the calculation of q.

average q would be a better guide to
investment in the long run than in the
immediate future.
Stock Market "Bubbles" and q
So far, we have assumed that market
values accurately reflect information
about future returns from the capital
stock. If actual market values do not
accurately reflect such information,
how useful is q? Many analysts have
suggested that the stock-market decline
in October did not reflect a revaluation
of future returns. Does this imply that
BFI will be unaffected?
Economists refer to the excess of the
actual stock-market value over the
value calculated as a present discounted value of future returns as a
"bubble." It can be argued that firms
will respond to a change in q even if
firms' managers view the market as
being "irrational" (too high or too low).
Firms may respond to stock-price
changes simply because stock prices
affect the cost of investment. When
stock prices rise, the cost of raising
funds via stock issues falls. The total
cost to a firm of financing via equity
equals the rate at which it pays out
dividends, plus the rate at which it
must reinvest retained earnings to generate capital gains for the shareholders.
Since dividend policy is relatively stable, when the price of a share rises, the
cost of equity falls. Further investment
then becomes profitable, since investment decisions involve comparing the
rate of return on the investment with
the cost of financing investment.
In other words, even if the firm's management has not revised its view of the
future, it may respond to the stockmarket rise. If the price of equity falls,
by the same reasoning, the cost of financing investment rises, and previously profitable investment opportunities
may be cut off. Rather than issue
shares, firms may now use funds previously designated for investment to
repurchase their shares at lower prices.
It is not clear, however, that investment will immediately respond to the
issue or repurchase of equity. Firms'
decisions to change their capital stock

6. There are other ways that a firm can adjust
output and affect the return to its physical capital. Firms can adjust hours and employment, for
example, and thus avoid potentially high costs
associated with changing the capital stock (see
Shapiro [1986]).

will be based on their own internal
assessment of future demand for their
output. In addition, many investment
decisions involve large and irreversible
commitments. However, market fluctuations give firms the opportunity to
lower the cost at which they obtain
funding, either by issuing stock when
the price is high or by buying back
equity when the price is low.
Financial Constraints and the Link
Between q and Investment
Recent work (for example, Fazzari,
Hubbard, and Petersen [1987]) has
explicitly shown how constraints on
firms' ability to raise funds for investment can weaken the link between q
and investment. Not all firms will issue
shares if their share price rises or buy
back equity if their share price falls.
So, fluctuations in q will not affect the
rate of investment for all firms.
Small and growing firms are likely to
face constraints on their ability to
finance investment. Small firms may
not be as well-known, so investors may
require a higher rate of return, or lower
price, in order to hold their shares. A
growing firm needs to utilize external
sources of funding, since funds required
to finance investment will tend to exceed
available internal funds, or cash flow.
In order for such firms to respond to
an increase in equity values by issuing
stock, the stock price must rise enough
to compensate for the additional return
demanded by investors and to make the
cost of equity issue lower than the cost
of using internal funds. Fluctuations in
stock prices that do not succeed in reducing the cost of share issue below the
cost of retained earnings will not result
in equity issue. Because such firms tend
to be constrained by cash flow, investment would tend to respond more to
fluctuations in cash flow than to q.
The Cost of Capital, q,
and Investment
An alternative to focusing on q as the
link between stock-market fluctuations
and investment is to focus on the cost
of capital. The cost of capital is the cost
of financing investment, taking into
account the costs of various methods of