Investors are worried over the prospects that the long-term
momentum behind the stock market recovery of 2009-2015 may be in danger of
complete dissipation this year. That
would mean a certain date with an extended bear market and, potentially, an
economic recession perhaps sometime later this year.
Normally, within the context of an established bull market,
worry would be a good thing given that the bull tends to proceed along a “wall
of worry.” In view of recent actions
undertaken by central banks, however, those worries are legitimate as I’ll
explain in this commentary.
There are two established ways of killing forward momentum
and induce economic recession. One is to
sharply reverse monetary policy or margin maintenance policy from very loose to
very tight. An example of this is what
happened in the months leading up to the 1929 crash; yet another example was
the margin requirement tightening in the gold, silver and copper markets in
2011.
The other way of reversing forward momentum is to slowly
suffocate the financial market through subtle, incremental policy shifts which
favor holding cash over equities. The
central banks of Europe and the U.S. have opted for the latter course. The ultimate outcome of this policy are being
felt even now in Europe and elsewhere, but likely won’t become abundantly clear
in the U.S. until later this year.
Tight money policy, especially at a time when the financial
market is vulnerable to overseas weakness, is nothing short of a recipe for
disaster. Timothy Cogley, an economist
with the San Francisco Federal Reserve Bank, admitted in a 1999 research paper
that the Fed’s tight monetary policy in 1928-29 likely contributed to the stock
market crash of 1929. Cogley observed:
“In 1928 there was a synchronized, global contraction of
monetary policy, which occurred primarily because the Fed was concerned about
stock prices. These actions had
predictable effects on economic activity.
By the second quarter of 1929 it was apparent that economic activity was
slowing. The U.S. economy peaked in
August and fell into a recession in September.”
[“Monetary Policy
and the Great Crash of 1929: A Bursting Bubble or Collapsing Fundamentals?”]
The Fed’s mistake in those days was in trying to prevent a
speculative bubble in the equity market.
In so doing, however, the Fed inadvertently contributed to an even
greater problem: the implosion of a speculative bubble. Moreover, the speculative bubble was fueled
in part by a loose monetary policy in the years leading up to the 1928-29 run
up in stock prices.
Cogley concluded: “In retrospect, it seems that the lesson of
the Great Crash is more about the difficulty of identifying speculative bubbles
and the risks associated with aggressive actions conditioned on noisy
observations. In the critical years 1928
to 1930, the Fed did not stand on the sidelines and allow asset prices to soar
unabated. On the contrary, its policy
represented a striking example of The
Economist’s recommendation: a deliberate, preemptive strike against an
(apparent) bubble. The Fed succeeded in
putting a halt to the rapid increase in share prices, but in doing so it may
have contributed one of the main impulses for the Great Depression.”
While the Fed’s recent quarter-point interest rate increase
may seem insignificant at face value, the magnitude of the move can only be
appreciated by realizing the rate of change involved. The following graph provides some idea of
just how huge in percentage terms the Fed’s policy tightening is.
It’s important to put this chart into its proper long-term
context. The bull market which began in
2009 was largely fueled by a loose monetary policy, courtesy of then Fed
Chairman Ben Bernanke. His successor,
Janet Yellen, has reversed Bernanke’s accommodative measures and seems intent
on tightening the noose around the economy’s throat.
Although many observers denigrate Bernanke’s stimulus
measures as being excessive, there can be no denying that they were successful
in not only reviving the stock market and housing market, but also the overall
economy to some degree. Many economists
consistently underestimate the extent to which the U.S. economy is tied to the
financial market. Yet the saying has
never been more apropos than it is today: “As goes the stock market, so goes
the economy.”
Fed Chair Yellen evidently doesn’t understand that
truism. Her restrictive monetary policy,
if pursued further, will eventually choke the last remnants of forward momentum
in the economy, particularly in the manufacturing sector. Truly, now is the time the Fed should pursue
an aggressively looser policy. As one
observer put it, “With as much headway as the economy has made since 2009, why
not open up the monetary floodgates and gun for prosperity?” Why not, indeed!
One of the biggest criticisms the Fed faced when it initiated
quantitative easing (QE) and Operation Twist is that its loose policy would
inevitably lead to runaway inflation.
Yet here we are some seven years later and inflation is nowhere to be
seen. Nay, deflation actually threatens
the global economy and, by extension, aspects of the U.S. financial
system. Why not then throw all caution
to the wind and open up the monetary spigots full throttle? What have we possibly got to lose?
I’ve long maintained that the single biggest threat to the
financial system is not an aggressively loose money policy, but an
unjustifiably tight one. Everyone fears
financial bubbles these days, but bubbles wouldn’t necessarily lead to
catastrophe if central banks and governments didn’t consistently pop them by
tightening money. While it’s true that
every bubble has its natural limit, they need not always end in catastrophic
implosion. Indeed, bubbles in an economy
as dynamic as ours are welcome events which bring quantum leaps forward in
technological progress. They also tend
to raise living standards for nearly everyone.
One could argue that without the many bubbles of the last 30 or so
years, America wouldn’t enjoy her current high standard of living.
Let’s
now briefly turn our attention to the stock market outlook. While several reasons were given by analysts
for the latest stock market rally attempt, the most common one was the hope for
additional stimulus from the Bank of Japan and the European Central Bank
(ECB). On Jan. 21, ECB President Mario Draghi indicated the bank
would consider additional stimulus at its next meeting in
March. Investors were elated by this statement, though it’s
surprising that the ECB is sufficiently unconcerned by the global financial
crisis to wait until March before even considering action. If
anything, Draghi’s statement is a further testament to the complacency that’s
still rife despite the trillions of dollars in damage already inflicted by the
crisis.
That
word “complacency” seems to capture the prevailing sentiment among both retail
investors and policymakers right now.
Remember that in a bull market the “wall of worry” is what helps to
establish the upward trend in stock prices.
In a bear market it’s the “slope of hope” that predominates. There seems to be a lot of “holding and
hoping” going on right now, and that’s not promising from a contrarian’s
perspective. We need to see a much
bigger manifestation of fear, doom and gloom among mainstream investors before
we completely cast our concerns about the bear aside.
Another
aspect in great need of improvement is the market’s internal momentum
picture. Since last spring, the number
of stocks making new 52-week lows on the NYSE has consistently been above
40. That’s a sign that internal weakness
is still present in the broad market.
Moreover, the important 200-day rate of change in the new highs-new lows
has been declining now for over a year.
The last time this happened was heading into the 2008 bear market. The following graph shows the enormity of the
decline in this indicator since last year.
No comments:
Post a Comment