This
is our 82nd report on the consequences of the stock market mania
since
we first published this website on January 15, 1999.
Our readership
spans 185 countries.
Our paid
subscription stock market newsletter has only two rationales for its existence;
Powerful
Commentary and Unique Perspectives.
Please
check out the testimonials on our Kudos
page.
Printable
files of this report accompanied by our forecast are available only to
paid subscribers.
If you
are interested, please CLICK
HERE. Don't
feel like committing to a longer term?
Special
3-issue subscription offer available: CLICK
HERE
The articles below are
reprinted from previous issues of Crosscurrents
and are chosen for their timeliness
and relevance.
Every Reason To Expect A Worst
Case Scenario
REPRINTED FROM THE JUNE 29th
ISSUE OF CROSSCURRENTS
Every few months, we open the latest report (see
http://1.usa.gov/1GsVmj4)
on derivatives published by the Office of the Comptroller of the Currency
(OCC), hoping to find some reason to allay our fears that another
major derivative nuclear blowup and fallout lies just around the next corner.
The OCC’s data only goes back to 1991, so we are dealing with a very short
history of 24 years. However, in a span of 28 years, we have seen
three major derivative events, the impact of the second far exceeding the
impact of the first and the impact of the third far exceeding the impact
of the second. Each of these events pushed the financial world to
the brink of collapse and should be a frightening and constant reminder
of the fragility of our financial markets.
At the end of the fourth quarter of 2014, the total
notional value of derivatives was $220.4 trillion, down 7.9% from the third
quarter total of $239.3 trillion. This sum still represents 12.6
times the country’s GDP and nine times total stock market capitalization.
Notional values are so vast that the potential for disaster is always present.
Each of the three major derivative events were catalyzed by an entirely
different category of derivatives and each proved that reliance on derivatives
and the assumption of correct pricing models to justify those derivatives
simply invites disaster.
By May 1987, stocks had become grossly overvalued
accompanied by margin debt running to over 2% of total market capitalization
in the late summer, the first time such lofty levels had been achieved
since the Roaring Twenties. However, portfolio managers were convinced
their holdings were protected by a brand new strategy known as portfolio
insurance. This scheme was made possible by stock index futures,
a derivative product developed only a few years earlier. The theory
was if prices declined by 3%, you could sell stock index futures to protect
your portfolio against further decline. However, when everyone began
using the strategy, stocks would sell off 3%, triggering more sales of
stock index futures, which triggered still more sales until a cascade was
in effect. The selling of stock futures became a self fulfilling
prophesy and the crash was enabled. Around 12:30 pm on October 20,
1987, the Dow Industrials traded briefly below the crash low of October
19th. After opening 200 points higher, the Dow had quickly faded
and was dropping like a stone. One could easily make the case that
we were no more than minutes away from a total financial collapse.
It didn’t take long to recover. By early
December, the Dow had a huge three-day breakout. The bad times were
behind and a brave new world was once again possible. One decade
later, derivatives were far more widely utilized for trading and offered
huge potential for gains. Notional values had grown to 2.6 times
total U.S. stock market cap and 3.7 times GDP. Long Term Capital
Management (LTCM) seemed to be the savviest player of all, utilizing extreme
leverage to return over 21% after fees in its first year, 43% in the second
year and 41% in the third year. The Board of Directors included Myron
S. Scholes and Robert C. Merton, who shared the 1997 Nobel Memorial Prize
in Economic Sciences for a "new method to determine the value of derivatives".
What could possibly go wrong?
With leverage and derivatives, everything can
go wrong. In 1998, LTCM lost $4.6 billion in less than four months
following the 1997 Asian financial crisis and 1998 Russian financial crisis.
At one point, it was posited that the LTCM derivatives portfolio which
included very risky options on the S&P 500 index, threatened a daisy
chain of $2 trillion in risk, placing the stability of seversl of our largest
financial institutions in doubt. Not only was the Federal Reserve
forced to intervene, the Fed had to coerce banks and brokers to cooperate
in the liquidation and dissolution of the mess LTCM left behind.
At the worst point, we might only have been hours from a financial collapse.
Somehow, in less than an additional decade, banks
and brokers created a bubble in housing prices via incredibly aggressive
lending policies which enabled another derivative fiasco. Zero down
mortgages were offered to impossibly risky borrowers and the low quality
loans were repackaged and touted as higher quality. It was all an
illusion and as it became apparent that too many low quality borrowers
had become overextended, it all rapidly fell apart. By the time Lehman
Bros. was forced to close its doors, we were once again perhaps only hours
from a collapse.
Our featured charts tells you all you need to know.
Virtually all notional values are tied up by nine large banks, and 95%
of the total belongs to just the five largest. It is no wonder that
banks were bailed out when the housing bubble burst. There was literally
zero room for error and the best policy at the time was to ensure stability
by any means possible. The ensuing recession was the worst in decades
but at least a collapse was avoided and the economy was enable to recover,
albeit weakly.
Unfortunately, nothing much has changed.
The banks that were too big to fail in the last blowup remain too big to
fail. If a replay occurs, there will be no choice for the Federal
Reserve. Any major players poised on the brink of extinction will
be bailed out. We have created a perfect scenario for bad players,
who not only may take on undue risk, why would they not take on undue risk?
There is no price to pay for foolishness. Goldman Sachs CEO Lloyd
C. Blankfein owns 1.9 million shares of GS worth more than $400 million.
Mr. Blankfein need not worry one whit at whatever course his company takes,
however foolish or even illegal. As the previous bubble fallout clearly
showed, there is an elite class of individuals in this country immune to
the problems and pitfalls suffered by the other 99%.
Credit default swaps account for 96.1% of all notional
values in credit derivatives, for a total of more than $9 trillion.
See http://bit.ly/11zlIjW for a primer,
which tells us “A buyer of a CDS might be speculating on the possibility
that the third party will indeed default.” Therein lies the rub.
Buyers may have a vested interest in a default and could conceivably act
with precisely that objective in mind and actually contribute to circumstances
that lead to the default. Given the impossibility the housing bubble
could play out without disaster, we believe there were considerable and
concerted efforts to accelerate and worsen the outcome. There were
billions of dollars to be made. The same situation prevails today.
There are billions of dollars
to be made via defaults.
As the charts above imply,
there is simply so much risk
built into the financial system that a disaster seems inevitable.
Couple these pictures with
those illustrating record leverage in stocks
and you have all the ingredients
required for the inevitable disaster.
We’re still looking for Dow
14,719 on the downside.
"Fascinating, Insane or Telling?"
REPRINTED FROM THE MARCH 30th
ISSUE OF CROSSCURRENTS
- MARGIN
CHART UPDATED THROUGH JUNE 2015 -
CNBC recently proved for the umpteenth time why
the station has lost so many viewers over recent years and why they are
relegated to lose a lot more when the next bear market finally arrives.
The question is not how many more viewers will watch the soaps instead,
but may be whether the financial channel can even survive. Given
the staying power of a six year bull market in which stock prices have
tripled, the circumstance of Neilsen ratings plunging to 21-year lows last
year strongly suggests the product CNBC sells is one consumers can do very
well without.
Our rant has always been CNBC’s style over the
years, consistently the cheerleader and all smiles in the quest to keep
viewers interested at all times. Unfortunately, bear markets are
a fact of life and they necessarily result in a contraction of viewers
as investors move to the sidelines and wait. This is a completely
normal situation but is not acceptable when advertisers pay your bills.
Viewers must be retained, even if it means that fantastic and unsustainable
bubbles such as the tech mania in 2000 or the housing mania in 2007 receive
less recognition as manias than they deserve. Thus, the tack has
been to remain positive as often as possible and of course, the game played
out with ratings slipping anyway as viewers fled to self preservation mode,
uninterested in any best case scenarios offered on air.
We have seen our fair share of bull over the years
but this particular segment (see http://cnb.cx/1G1Ju5F)
in which the WSJ’s Jonathan Clements and Premier Financial Advisors’ Mark
Martiak trumpet the advantages of effectively taking the proceeds of a
seven year car loan to buy stocks, pretty much takes the grand prize.
Last we looked, autos were a rapidly wasting asset. In the same four
years an auto depreciates by half, one could also be risking a hefty loss
on the amount “invested” in stocks. Whether or not the car is held
for seven years is irrelevant. Using an auto loan as leverage to
buy stocks is far worse than buying stocks on margin. This sure sounds
like imprudent advice to us. To her credit, reporter Erin Burnett
questioned the safety of such actions but this is the most highly leveraged
society in history and without exception, levering up in any way, shape
or form is typically accepted with open arms. Ms. Burnett was a bit
nonplussed, admitting she didn’t know if the idea was “fascinating, insane
or telling.” We’ll go with insane.
The U.S. equity market has morphed into a Frankenstein
monster driven by high frequency trading; millions of trades, many of which
unwind in mere seconds. For the most part, time horizons have shortened
so significantly that there is no long term to buy for anymore and as our
featured chart clearly illustrates, seven year holding periods for stocks
went out of fashion over 50 years ago. As a consequence of the modern
era and high frequency trading, traditional valuation measures have become
relatively meaningless and mutual funds now hold as little cash as possible
in reserve. They are near fully invested and will lose more in a
market decline and be less able to take advantage of lower prices when
a bear market eventually provides compelling values, as typically occurs.
Shocking as it may seem, current average holding periods are shorter than
they were in 1929. Dollar trading volume (DTV) versus gross domestic
product (GDP) is at its fourth highest annual level ever, while the last
nine years represent all of the nine highest ever. Interestingly,
the tenth highest was in 2000, when the tech mania peaked. There
are so many measures that scream “bubble” and “mania” that any suggestion
for investors to increase leverage must be treated as sheer lunacy.
At center, we featured a similar table from Zero
Hedge in the March 2013 issue but have altered it somewhat to better represent
what we view as striking comparisons to the October 2007 peak, from which
stocks collapsed by 50%. While we do not expect another halving in
prices, the comparisons are quite unfavorable and strongly imply things
are far worse, not better than in 2007. Consider that the Fed’s policy
has indeed pulled off its primary objective; to inflate asset values.
More than $5.8 trillion in stock wealth has been created from the October
2007 peak and as we know in retrospect, even that peak was unsustainably
inflated. The additional wealth works out to over $18,000 for every
U.S. citizen, yet GDP growth is lower, we have twice as many unemployed
for longer than 27 weeks and still have a near record number of people
receiving food stamps. What is wrong with this picture?!
At bottom left, combined margin debt as measured
by the NYSE plus NASD, now totals $485 billion. An expansion in margin
debt is actually bullish to a point, since it signifies a growing interest
in stocks. However, not only do we now have a situation of excessive
leverage, margin debt is no longer growing and this strongly suggests prices
are too high. For all intents and purposes, total combined margin
debt topped more than a year ago in February 2014. The draw down
of leverage in both the 2000-2002 and 2007-2009 bear markets was over 50%.
A similar unwind today would equate to more than $250 billion of excess
supply, easily enough for prices to suffer a 20% downside this year.
While forecasting the bursting of a bubble is precarious work, we have
no choice when viewing so many charts at seemingly pivotal extremes.
Clearly, the gap between margin
and mutual fund cash ratios
is enormous in comparison
to the two prior mania peaks.
Fascinating, yes.
Telling, yes.
However, insane wins hands
down.
The Case For A 25% Decline
REPRINTED FROM THE JUNE 29th
ISSUE OF CROSSCURRENTS
At least twice a year, we show a chart of the Dow
Industrials dating back over a century accompanied by a 5% regression line.
Our chart data goes back to 1890 and since then, the Dow has averaged annual
gains of roughly 5% ex-dividends, thus we believe the line is a fair representation
of normal expectations. Surprisingly, the Dow has been below the
line 74% of the time. This seemingly weird circumstance simply mirrors
what we know about markets. Bear markets tend to greatly overdo the
downside and this factor tends to keep the Dow under the regression line
longer than one might expect. Below left, we illustrate just how
overbought stocks were in 2000 and 2007 by their percentage above the regression
line.
There have been only four occasions in which the
Dow has greatly exceeded the regression line on the upside. The first
was during the Roaring Twenties, arguably the biggest mania of all time.
As stocks peaked in 1929, our indicator soared to 86% above the line.
The 1929 high was not exceeded until 1955 and the super bull market did
not begin until August 1982, thus the Dow remained below the line for decades,
finally rising above in 1996 as the tech mania approached.
How amazing was the tech mania? Our indicator ran to almost exactly
the same level as in 1929. The third peak arrived in July 2007, when
both housing and stocks were in manic mode. And now, the present
overbought condition has taken us almost 37% above the line. Bear
in mind, this is only the fourth time we have seen this circumstance and
each of the prior circumstances resulted in a horrific bear market.
The dénouement need not be an exact replay of any of the cited prior
bear markets. The Roaring Twenties resulted in an 89% decline in
the Dow. Stock prices fell 50% from the tech mania peak and 50% again
following the busted housing bubble. Although our present downside
bear target is Dow 14,719, we would caution that a scenario in which prices
briefly dip to the zero percent line on our chart at lower left would take
us to under Dow 13,500, down roughly 25% from today.
History shows us that a 25%
decline occurs on average every 5.8 years.
A bear market of this magnitude
would actually be quite ordinary.
Standard Deviations
REPRINTED FROM THE JUNE 29th
ISSUE OF CROSSCURRENTS
After a brief timeout, Investment Advisors are
again flowing into the bull camp. In our quest to show the persistence
and permanence of the bullish camp. the chart below right is a 104-week
moving average of the ratio of advisory bulls to bears (per Investor’s
Intelligence). Using a baseline of December 2002 through December
2013, we can see that the average two-year ratio has climbed well above
three standard deviations from the norm (average 1.76, median 1.86).
Statistically speaking, two
standard deviations cover 95.45% of all values.
Three standard deviations cover
99.72% of all values.
The odds of us witnessing this
kind of one-sided sentiment is clear off the charts,
an explicit indication of
a mania.
OUR EVENTUAL
TARGET REMAINS DOW 14,719
DOWN 19.8% FROM THE
DOW PRINT HIGH OF MAY 19, 2015
WE STILL
HAVE AN OPEN BEAR MARKET TARGET OF DOW 12,471
DOWN 32% FROM THE DOW
PRINT HIGH
We also
believe the March 2009 low of Dow 6469 will likely never be seen again.
THE CONTENTS
OF THE ENTIRE WEBSITE ARE COPYRIGHT 2015 CROSSCURRENTS PUBLICATIONS, LLC
I hope you have enjoyed your visit. Please
return again and feel free to invite your friends to visit as well.
Alan M. Newman, July 15, 2015
The entire Crosscurrents website has logged
nearly 4 million visits.
All information on this website is
prepared from data obtained from sources believed reliable, but not guaranteed
by us, and is not considered to be all inclusive. Any stocks, sectors
or indexes mentioned on this page are not to be construed as buy, sell,
hold or short recommendations. This report is for informational and
entertainment purposes only. Persons affiliated with Crosscurrents
Publications, LLC may be long or short the securities or related options
or other derivative securities mentioned in this report. Our perspectives
are subject to change without notice. We assume no responsibility
or liability for the information contained in this report. No investment
or trading advice whatsoever is implied by our commentary, coverage or
charts. |