Understanding the Late 1990s Large-Cap Stock Market Bubble

1
Understanding the Late 1990s Large-CapStock
Market Bubble

• David L. Debertin
• Professor of Agricultural Economics
• University of Kentucky
• DLDebertin_at_aol.com

2
Foreword

• This slide show is designed to help better
  understand the conditions that led to the bubble
  in large-cap stock prices in the late 1990s, and
  what happened as and after the bubble burst

3
The following graph provides a comparison of two
actively-managed mutual funds that are
representative of the performance of actively
managed funds that can be characterized as
Large-Cap Growth versus Small and Mid-Cap Value
for a 15-year period beginning in 1990. The
Large-Cap Growth fund is Fidelity Growth Company
(FDGRX) and the Small and Mid-Cap Value fund is
Fidelity Low-Priced (FLPSX). Share prices for
both have been normalized to 1 a share for both
funds at the start of the period.
4
(No Transcript)
5
Note that in the mid 1990s, FDGRX starts to
increase much more rapidly than FLPSX, and there
are instances in 1998 and 1999 where FLPSX is
nearly flat. During these periods, investors are
underinvesting in small and mid-cap value stocks
to take advantage of the boom in large cap growth
stocks
6
In March of 2000, The NASDAQ and FDGRX peaked,
followed by a steep sell off. Notice that FLPSX
continued to rise throughout this period as
investors began to rotate some of the proceeds
from the sale of Large-Cap Growth stocks into
purchases of Small- and Mid-Cap Value stocks.
There was a secondary large cap growth peak in
August of 2000, but after that large cap growth
stock prices deteriorated rapidly. However, some
of the proceeds from the sale of large-cap
growth stocks continued to be reinvested in
small- and mid-cap value stocks, and FLPSX
continued to rise through most of the period.
Even in the steep overall market sell off of 2002
(22 loss in the SP 500), FLPSX lost very little
(about 6 ). Large-cap funds focusing primarily
on stocks with above market average P/E ratios
fared even worse than the SP 500, with FDGRX
Being off over 33 percent that year. This
performance is typical of Large-Cap Growth funds
with portfolios containing a lot of above market
average P/E stocks in 2002.
7
Consider the period from 1995 through 1999, when
even the SP 500 was going up at 20 per year,
and funds with larger NASDAQ 100 exposure were
going up even faster. Given the market
capitalization represented by these
large companies, it took huge amounts of new
injections of cash each year during the period to
sustain these annual percentage gains. In
retrospect, its surprising that a large cap
rally of this magnitude was sustainable for as
long as it was. At some point investors run out
of new cash to invest.
8
In contrast, a 20 rally in small and mid-cap
value stocks is much easier to sustain because
the the total market cap of the entire group of
stocks is much smaller. Far smaller annual cash
inflows are needed to sustain 20 annual
appreciation rate in Small- and Mid-Cap
Value than in Large-Cap Growth. To the extent
that lots of cash flows into funds
holding primarily small and mid-cap value stocks,
the major problem fund managers face is
identifying stocks that are cheap on a P/E basis
yet represent favorable opportunities for
further gains.
9
So we see lots of small and mid cap value funds
closing to new investors (FLPSX, RYLPX) as soon
as these inflows start to get unmanageable. A
fund manager for a large-cap growth fund facing
inflows will not likely run into the same
problem to near the same degree. Fortunately for
investors with new money to invest, there are
now index funds concentrating on small and mid
cap value (i.e. VISVX) as well as Exchange
Traded Funds (ETFs) that allow investors to
purchase SP/Barra and Russell Value indices
directly. Still, the underlying value stocks need
to be able to absorb the cash inflows without too
much increase in the P/Es .
10
Since early 2000, we have yet to have a sustained
rally in large-cap growth stocks, although
investors in small and mid-cap value stocks (and
funds with investment objectives similar to
FLPSX) have continued to fare well through most
of the period. We have had some short periods of
1-3 months in outperformance of Large-Cap Growth
over Small-and Mid-Cap Value but not longer time
periods. This suggests that having gotten burned
by holding large cap growth stocks after April
2000, investors are very wary of investing
in large-cap stocks that do not appear to offer
good value. Investors in 2004 appear to have
shown increasing interest in large-cap value
stocks as the year progressed, however (GARP).
11
In the following diagrams, assume that the large
green circle represents the total market
capitalization of all large-cap growth stocks,
and the red circle represents the total market
capitalization of the universe of small- and
mid-cap value stocks at the beginning of the
run-up in large cap growth stocks in the mid
1990s. The red circle is much smaller because the
total market capitalization of all small- and
mid-cap value stocks (those with below- market
P/E ratios) is much smaller than for large-cap
growth.
12
Large-Cap Growth
Small-and Mid-Cap Value
13
As we move into 1998 and 1999, Large-Cap
Growth Substantially outperformed Small and
Mid-Cap Value. Fund managers complained that no
one was interested in buying below-market P/E
stocks, especially in the small- and mid-cap size
range. Funds emphasizing these stocks fared
poorly, even relative to the SP 500 (which was
increasingly dominated by small, high market cap
technology firms with few employees or earnings,
but nosebleed stock prices). Often these were
technology or Internet-related firms as investors
showed little interest in stocks of firms in
sectors thought to be slower growing at any
market capitalization level large or small.
14
This was a period of extreme euphoria in the US.
The Cold War had been won, the federal budget was
showing an increasing surplussomething that had
not happened in ages. Business schools were
talking about new theories of stock valuation
that no longer relied as much on current
earnings, and companies were stressing their pro
forma profits that excluded many expense items.
Pro forma profits as reported to Wall Street
could be nearly whatever the company officers
wanted them to be, ignoring costs whenever
convenient. Sadly, many accountants went along
with this new accounting math. Meanwhile you
had near startup Internet-based companies with
billion dollar market caps, and officers with
net worths of 10s and even 100s of millions of
dollars based on the price of the stock in a
recent public offering.
15
You had companies like AOL becoming an SP
500-sized firm based on market capitalization
and having enough value to enable it to gobble up
older, established but slower-growing
traditional media firms such as Time Warner.
By early 2000 an increasing number of people
including investors started to believe that
share prices could not grow to the sky without
real (and not just pro forma) earnings, and the
bubble burst, first in April of 2000 with a
steep sell-off in high-growth, high P/E
technology firms, followed by a short rally into
August 2000 which was quickly followed by
a sharper and still deeper downturn. The air was
starting to go out of the large-cap Growth
balloon!
16
Large-Cap Growth
Small-and Mid-Cap Value
17
Whenever investors sell stocks, the proceeds
from The sale must be parked somewhere. We know
now that in the sell off starting in 2000, many
investors wanted to believe that the downturn
was very temporary. Therefore they parked large
amounts of cash in money market funds and other
accounts where their principal would at least be
safe. All this cash caused interest rates on
money market funds to fall to historically low
levels. Option 1 Hold Cash with the idea of
moving back into the market soon
18
Large-Cap Growth
Small-and Mid-Cap Value
Cash, Money- Market Funds
19
A second option was to park cash from the
proceeds in US Government or high-grade
corporate bonds. Investors who did this fared
quite well for a period of time, as the Federal
Reserve was increasingly worried about the threat
of a recession (in part brought on by the
puncturing of the Large-Cap balloon), rather
than inflation. As interest rates begun their
slow decline, bondholders who had gotten out
early saw the value of bonds they purchased in
2000 and 2001 appreciate in value. Option 2
Invest in Government Debt or High-Grade
Corporate bonds
20
Large-Cap Growth
Small-and Mid-Cap Value
Cash, Money- Market Funds
Government and Corporate Bonds
21
A third major option involved rotating proceeds
from the sale of Large-Cap Growth stocks and
mutual funds into another segment of the market.
Here the relative size of the large-cap growth
side of the market versus small- and mid-cap
value side is important. If only a small portion
of proceeds from the sale of large-cap growth
stocks get reinvested into small- and mid-cap
value stocks and funds, because the total market
cap is so much smaller in that segment, that will
be enough to fuel a strong up-movement in small-
and mid cap stocks and funds containing those
stocks. Option 3 Invest in Small-and Mid-Cap
Value stocks
22
Large-Cap Growth
Small-and Mid-Cap Value
Cash, Money- Market Funds
Government and Corporate Bonds
23
The period of relative outperformance of small-
and mid-cap value stocks and funds relative to
large cap growth funds continues through 2004,
with only brief largely unsustainable rallies on
the large-cap growth side. It is important to
note that because of the comparative size of the
total market capitalization of stocks labeled
Large-Cap Growth versus Small- and Mid- Cap
Value, it takes far less newly invested cash to
sustain this relative overperformance on the
Small- and Mid-Cap Value side of the market. In
short, the outperformance of Small and Mid-cap
Value relative to Large-Cap Growth could continue
for some time to come.
24
Meanwhile we still have lots of investors
with proceeds from stock sales when the large-cap
growth bubble burst still earning low rates of
interest on money market funds. Life is about to
become far more difficult for those holding cash
in bonds, as the value of more recently purchased
bonds decrease as interest rates begin a slow
but steady rise. The bond market quickly becomes
a Chinese water torture in which assets are
depleted a drop at a time, and investors will
increasingly want to find other opportunities
for the money invested in bonds as the Fed does
its thing with respect to slowly raising rates,
and the full impacts of the huge federal
deficits start to hit the credit markets.
25
Over the next several years, barring major
terrorist attacks or another major economic
downturn, more and more of the money now parked
in money market accounts and shorter-term governme
nt bonds will likely come back into the equities
market in some form. An interesting question is
what side of the market will that money tend to
favor. Investors still clearly remember what
happened to their investments when the balloon
burst in 2000, and those that simply held on
still are a long ways from having fully
recovered. Meanwhile they also observe the recent
data showing outperformance of Small- and
Mid-Cap Value. Even a small portion of new cash
gets invested on this side a rally in these
stocks may very well continue.
26
Because of its size, the large-cap side of the
stock market has a far greater capacity to absorb
huge cash inflows than does the small and mid-cap
side. Further, if we are looking at the universe
of small- and mid-cap stocks that can be bought
at below market average P/E ratios, unless
earnings rise in line with the purchases using
net cash inflows, P/E ratios could quickly rise,
making the stocks no longer a Value play. Both
the indexes and actively-managed funds no doubt
have definite sell rules when a once value stock
is to be sold or dropped from an index.
Otherwise a Value fund would not remain a Value
fund for long!
27
Meanwhile you have the effort toward privatizing
a portion of social security. The Social
Security Trust fund is running a current surplus
in that current collections exceed current
payouts. Currently, that surplus is used to
finance the national debt in that it is used to
purchase government bonds. The federal
government could fairly readily privatize a
portion of social security by permitting working
people to place a share of that current surplus
in a privately held account with their own
individual names on iti.e. a vested versus the
current non-vested (pay benefits as you go)
plan. One option in a vested plan would be for
the individual to invest that money in US
government bonds. That is what happens to the
trust fund surplus right now, except that
those bonds are not tied to a particular worker.
28
But the most interesting possibility would be
allowing at least a portion of the privatized
dollars to be invested in the equities markets.
Even a small portion privatized would represent
a huge sum of cash for the equities markets to
absorb, and quickly would go beyond the amounts
the small-and mid-cap side of the market could
absorb. Further, the Federal government at least
initially would likely believe that the large-cap
equity markets represent less long-term risk
than the small- and mid-cap markets, and workers
may not have alternatives that go beyond funds
dominated by large-cap stocks. So social
security privatization might lead to a
mini-boom in large-cap stocks and funds as the
markets absorb the new cash (cash currently in
government debt instruments).
29
If workers are permitted to move privatized
social security accounts into investments other
than US Government debt, in essence they will
each be selling some of the Government bonds now
represented by the current surplus in the
Social Security Trust fund. The Government will
still have to finance the debt once financed by
trust fund surpluses, and in order to do this,
interest rates on government debt will likely
rise. Meanwhile investments in equities by
workers paying into social security should lead
to growth in the private sector. Tax revenues
from this growth might to a degree offset
increased expenditures by the federal government
for interest on the national debt.
30
The Corporation
One way of viewing a corporation is that it is
simply a mechanism for deploying capital From a
capital budgeting perspective, corporations seek
to deploy capital in such a way that will
permit them to earn above market average internal
rates of return Were this not true the investor
would do better by simply investing in the
market index
31
Large corporations need large capital projects
for them to have any meaningful impact on rates
of return and earnings Many large-cap
corporations appear to be running out of ideas
for deploying capital at high rates of
return Microsoft gives investors back a large
dividend rather than deploying capital in high
payoff projects within the firm ATT deploys
capital in projects earning below average rates
of return? GE In the 90s an exception, but now?
32
Mid- and small-cap firms need to implement
smaller projects than large firms do in order to
positively affect overall rates of return and
earnings Smaller firms have more opportunities
to increase Economic Value Added (EVA) Smaller
not larger firms have the potential for faster
year over year growth rates Projects with high
potential payoffs can be riskier than projects
with lower payoffs
33

• Larger firms have some advantages
• Less competition
• Dominance over markets
• Access to capital
• Name recognition

34
Markowitz-Sharpe Portfolio Theory
35
Nobel-prize winning economists Harry
Markowitz and Robert Sharpe developed the
underlying theory that forms the basis for the
mutual fund industry. They argued that a
properly diversified portfolio of volatile
stocks would be substantially less volatile than
each individual stock in the portfolio taken
alone Their solution to diversification
suggested picking individual stocks for a
portfolio that all have a good potential for
high returns over a long period of time, but
combine stocks whose returns patterns often do
not move together.
36
Markowitz and Sharpe would likely have suggested
that an investor diversify by putting half the
original investment in FLPSX and the other half
in FDGRX. The returns would have looked like the
red line on the following graph
37
(No Transcript)
38
The Diversified (Markowitz-Sharpe) strategy
which puts half the initial investment in each
fund (FLPSX and FDGRX) appears to be better than
the all-Growth (all FDGRX) strategy but more
volatile and risky than the all-Value (all
FLPSX) strategy. Losses in 2000-2002 for the
Diversified strategy while a bit muted are still
substantial, and month to month variation in
returns remains high.
39
The Holy Grail of Investing
40
The Holy Grail of investing is a model that
would tell the investor that relative gains from
one fund (or stock) are starting to increase in
comparison with another fund or stock, and then
also signal when the relative gains are
beginning to narrow. Such a model would provide
the necessary signals so that the investor
could ride up a market bubble such as that which
occurred in Large-cap growth stocks and
funds, and them move into something else (perhaps
a Value fund) just after the bubble begins to
burst.
41
I have developed such a model and that model
appears to accurately signal when an investor
should be in Large-Cap Growth (FDGRX) versus
Small and Mid-Cap Value (FLPSX). Graphing the
model results, we get the following graph from a
one-time 1,000 investment in April, 1990 to
December, 2004 All Growth 1,000 FDGRX bought
and held All Value 1,000 FLPSX bought and
Held Switching 1,000 invested according to
signals from the Debertin
spreadsheet Switching Model
42
(No Transcript)
43
The Debertin Switching model generates an ending
value about twice as great as the buy-and-hold
Value (FLPSX) strategy, and about four times as
great as the buy-and-hold Growth strategy
6,135 All Growth 1,000 FDGRX bought and
held 12,055 All Value 1,000 FLPSX bought and
Held 9,095 Markowitz-Sharpe Diversified
(500 in FLPSX, 500 in FDGRX)
25,359 Switching 1,000 invested according
to signals from the Debertin
spreadsheet Switching Model
44
The Switching model requires only a limited
number of rotations between FDGRX and FLPSX over
the entire period, and signals when they
should occur with a high degree of
accuracy Further details about implementing the
Debertin Holy Grail Switching model can be found
at http//www.dldebertin.com/finance/grail.htm
----David
L. Debertin