RE: A Bull But...Page 1 5/6/2007
Executive Forum
Reserve Management
The commodity Bubble, The Metals Manipulation, The Contagion Risk To Gold
And The Threat Of The Great Hedge Fund Unwind To Spread Product
To: Global Central Bankers at the World Bank
From: Frank Veneroso
April 17, 2007
Revised as of today May 6, 2007
Introduction
Thank you for this second invitation to speak at your treasury management conference. I certainly did not expect such a second invitation.
Last year I was asked to speak only because Larry Summers, who was scheduled to speak, could not make it at the last moment and I was asked to fill in. I chose to speak on the subject of the global commodity bubble as well as the U.S. housing and housing
finance bubble and their eventual bursting. My presentation was a bit on the histrionic side, and I know it was greeted with a certain amount of amusement. But I certainly did not expect that it was of enough interest to warrant a second invitation to speak on the
same topic – particularly with the more august speakers available such as Larry Summers.
Perhaps I am reading more into this invitation than I should, but it seems to me that the events that have transpired over the last year must have intrigued some of you with the thesis that we have had a commodity and a housing bubble, that they may be in the process of bursting, and that this may have some relevance to central bankers and to reserve management.
Last year, I did not really focus on central bank reserve management, but I thought that this year I might direct my train of thought to some reserve management issues. So, in what I have to say today, I will try to work towards two assessments: first, the outlook for
gold as a reserve asset, and second, the outlook for spreads and the yield curve, which is obviously very relevant to reserve management.
The following paper is very long. So let me give a road map of where I will be going.
In the first part of this paper on the “Commodity Bubble”, I make the case that, in real terms, we have had an unprecedented commodity bubble in this decade. This bubble has occurred because of unprecedented investment and speculation in commodities, largely
by way of derivatives. The far more important engine of this bubble has been leveraged speculation by hedge funds. Over the last two years prices have climbed even though the microeconomic fundamentals of commodities have deteriorated. There lies ahead a bursting of this commodity bubble. It is now being triggered by deteriorating
Page 2 5/6/2007
fundamentals and it will be exacerbated by eventual investor revulsion which will reverse the extraordinary fund flows that have created this bubble.
In part two, “Metals: A Speculation to the Point of Manipulation”, I turn to the leading edge of this cycle’s commodity mania – metals. In base metals and to some degree in white metals hedge fund speculation has extended beyond derivatives to purchases of the physical. This has resulted in several variants of classic market “squeezes” across the
metals sector. These squeezes in the context of a runaway speculative fervor has resulted in increases in real prices for some metals that are far in excess of anything that has ever occurred before for these metals, in particular, and almost all commodities in general.
Because of the extraordinary amplitude and duration of these price moves, the microeconomic responses will be stronger than ever before, generating record surpluses that will ultimately lead to reversion to the mean or marginal cost. Investor revulsion toward this commodity sub sector should prove to be greater than for other commodities.
A consequent reversal of fund flows and the eventual liquidation of physical stocks held by hedge funds should lead to severe undershooting of marginal cost in the years to come.
In the third part of this paper, I consider gold. Gold’s fundamentals are far better than those of base and white metals. Unlike other commodities, there is zero mine supply growth in gold. Unfortunately, growth in physical demands for gold in jewelry, small bar
and official coin has been surprisingly negative over the last decade plus, especially in former gold loving Asia. The flow of gold from official stock liquidation in all its forms,
which had been depressing the gold price, has now disappeared, eliminating a former potentially bullish factor for gold. The last advance in the price of gold since mid 2005 has been driven by funds. Gold and other metals have been especially closely correlated
in this cycle, perhaps because funds have similar portfolios driven by a similar psychology toward all metals. Therefore, a revulsion by institutional holders from commodities in general and the metals sector in particular constitutes a serious contagion risk to the gold price.
Part I: The Commodity Bubble
Let us first consider the bull market in commodities in this cycle. That is a first step
toward drawing some implications regarding the outlook for gold as a reserve asset.
I will start with a recap of my thesis from last year.
Commodity prices move in cycles. Their cycle tends to be strongly correlated with the business cycle. However, since the end of the 1970s our business cycle expansions, in the United States at least, have been longer than in the past. The expansion of the 1980s lasted eight years. The expansion of the 1990s lasted ten years. Commodity cycles have not tended to be this long.
Page 3 5/6/2007
Page 4 5/6/2007
Note: The GSCI chart is very heavily weighted with energy, hence the brief spike in 1990 with the onset of the Gulf War and the rally in 2000 when most commodity prices were very weak. The CRB chart tracks most other commodities including metals.
The above charts make this clear. In the 1980s there was a recovery in commodity prices from the depths of the deep global recession. But it did not last for long. The microeconomics of the supply and demand for commodities led to a sharp correction in the mid 1980s, despite an uninterrupted global economic expansion. Around 1984 industrial metals were the first to sink back to close to the recession lows. The price of oil, held up by OPEC supply cutbacks into 1985, was the last commodity to undergo a severe correction in the middle of that decade.
Later in the decade sustained global growth put upside pressure once again on the prices of commodities, though not uniformly. When commodity prices went up again later in the expansion the rise in metals prices was particularly strong, while (except for a brief
spike during the 1990 Gulf War) the price of oil barely recovered as it bounced along its new bear market base well into the next decade.
In any case, late in that decade most commodity prices started down once again by 1989,
well before the 1980’s business expansion ended.
A similar pattern whereby the commodity price cycle was much shorter than the business cycle occurred again in the 1990s. A rolling global slowdown from the U.S. recession to the Japanese and European recessions delayed the onset of a full fledged new commodity
bull market until the middle years of the decade. A few years later peaks were reached in 1996 and 1997. Then, once again, long before the end of the U.S. economic expansion,
and for that matter the global economic expansion, commodity prices started down. The severe economic weakness of Asia in 1998 contributed to this decline. However, it was not enough to take global economic growth below its trend. Nonetheless, the price of crude oil fell to $10 a barrel in 1998 – a multi-decade low – and the prices of industrial metals fell to close to multi-decade lows in 1999 even though the U.S. economic expansion remained in full force.
I believe it is safe to say that commodity bull cycles tend to be shorter than the unusually long U.S. economic expansions that we have had since the onset of the so-called “Great Moderation” that began in the early 1980s. Why was this the case?
I think the answer is very simple. Although cycles in global aggregate demand greatly influence commodity prices, in addition to these macroeconomic determinants of commodity prices there are the microeconomic determinants of supply encouragement and demand rationing. A global economic expansion may pull commodity prices up.
But once prices are well above marginal cost, supplies are encouraged and substitution and economization lead to the rationing of demand. These microeconomic dynamics in the end drive commodity prices lower, even if a global business expansion remains intact.
Page 5 5/6/2007
That said, the commodity bull market of this decade has been unusually high in amplitude and long in duration. In fact, by some measures real (inflation adjusted) commodity prices have risen more in percentage terms than they have in any prior halfdecade bull market. It would seem that the microeconomic dynamics that have reversed
commodity prices in the past have either been absent or overwhelmed.
The New Era Thesis
In my experience, when a cyclical price trend persists for longer and to a greater degree than has happened in the past, market participants declare a New Era to not only explain this unusual persistence but to also justify the persistence of the trend forever forward.
This has certainly happened with regard to commodity prices in this cycle.
There has emerged a host of New Era advocates who share the following common arguments. The very healthy growth of the emerging economies and the super strong growth of China and India have made world economic growth different this time. We are in the midst of a new global supercycle in commodity demands that has relegated the
fairly short commodity price cycles of the past to a history that is no longer relevant.
The same New Era thinking extends to the supply of commodities. We are most familiar with such thinking as regards the oil market. The Peak Oil thesis says that all the easy to find large oil reservoirs with low extraction costs are behind us. Yes, there are smaller,
higher cost oil reservoirs to be found and exploited. But declining production from the large mature fields we largely rely upon offsets these new sources of output. For all the new drilling and development we do, we find ourselves simply running fast to stay in the
same place, as decline rates on our old reservoirs pull the ground out from underneath us.
New Era commodity bulls extend this supply side thesis as it applies to oil to many other commodities. In agricultural commodities, we are facing limits to the expansion of arable land. In industrial metals the easy to find deposits have been found and, in existing mines, ore grades are declining, and so forth.
It is my belief that these many claims for a New Era of explosive demand growth and supply constraints are largely erroneous. This is best argued by looking at individual commodity markets, for there we can document in detail if trend demand growth has accelerated or if rapid supply increases have become impossible. I will do that in what
follows. We will see that the analysis of individual commodity markets shows clearly that the New Era hypotheses are wrong. But one can also make a few generalizations that are pertinent.
Two decades ago I participated in a huge study of the copper market under the auspices of the IFC and the other partners in a project called Escondida – a project which was to become the largest copper mine in the world. We had the resources of the World Bank,
BHP, RTZ, and an array of consultants. If I go back to that copper price forecasting exercise at that point in time it becomes very clear that, despite the high economic growth
Page 6 5/6/2007
in the emerging world in this decade, we are not in a New Era, we are not in a new supercycle in commodity demands.
Making long term commodity price forecasts is largely an exercise in extrapolation. At that time, two decades ago, we were extrapolating based on the economic growth trends of the prior several decades. The emerging world was growing very fast, though the contributors to its growth were somewhat different. In those days Brazil was a 7%
grower, Mexico a 6% grower, Korea an 11% grower. Their economic growth rates today are perhaps half of what they were. This is something of an offset to the recent high 9% growth rates of China and India.
But, you will respond, the economies of the emerging world today are a much larger share of the world economy overall. And, to be sure, that is true. But offsetting this is the fact that the large developed economies were then far faster growers. Japan’s economy grew at a 9% rate in the 1960s and almost at that rate in the 1970s. It now has a
secular growth rate of perhaps 1.5%. Europe overall had a growth rate of 4%-5% in the 1960s and 1970s. It now has a secular growth rate of perhaps 2%. For the largest share of the global economy the attainment of the technological frontier, which implies lower productivity rates, plus adverse demographics, has reduced its economic growth rate
dramatically. If you take all the economies in the world, valuing GDP based on exchange rates, the overall global growth rate has not significantly changed since the mid 1970’s.
And so the demand pressures on commodities should not have significantly changed either.
And, in fact, as the chart below indicates, global economic growth and growth in commodity demands were higher in the period immediately prior to the mid 1970’s. Then much of the global economy was still recovering from a prior war ridden epoch when the disasters of wartime had set back so many economies from a greater potential
made possible by the interim advance in the technological frontier.
Page 7 5/6/2007
Mega Speculation and The Explosion In Commodity Derivatives
So if it is not a new era of supercycle demand growth and supply restraint, what has lead to such a high amplitude and long duration bull market in commodities in this cycle. My answer is speculation – nothing more. And speculation on an unimaginable scale.
The latest data provided by the U.S. Office of the Controller of the Currency and the Bank For International Settlements provides us with an ever more startling picture of
explosive growth in commodity derivatives. The data on U.S. bank positions shows a large increase into mid 2006.
Integrated Oil Update, Mike Rothman,
ISI, December 19, 2006
More striking is the data on the same derivative positions of all global banks compiled by the Bank for International Settlements (BIS).
Page 8 5/6/2007
Integrated Oil Update, Mike Rothman,
ISI, December 19, 2006
Though the six-fold increase in such positions over a few brief years is striking, it is the magnitude of these positions that is most alarming. From what we know of this data there is considerable double counting. But, offsetting this, this compilation is incomplete.
It excludes the positions of some investment banks who are extremely important intermediaries in the commodity derivative markets. And it excludes all futures and options positions on commodity exchanges, which may add another $2 billion or more to the global total. Taken all together the global total for all commodity derivatives is
probably much more than $10 trillion.
It is hard to know what to make of this data. But it is noteworthy that several years ago,
at the then prevailing lower commodity prices, the entire above ground stock of all commodity inventories was only in the hundreds of billions of dollars. If only a fraction of the increase in global commodity derivative aggregates in recent years corresponds to
a net long position of investors or speculators, it would appear that this increased demand for commodity derivative positions has overwhelmed what have been relatively small markets.
No wonder, then, that this cycle’s bull market in commodity prices has gone higher in inflation-adjusted terms and for longer than in all prior uninterrupted half-decade cycles in the past.
Page 9 5/6/2007
Investment Or Speculation?
The question arises, who are these new investors or speculators in world commodity
markets and what is their behavior?
Investment in commodities today probably refers above all to pensions and endowments
who have made long term strategic allocations to commodity futures index baskets for
diversification purposes. However, the total assets of all these baskets is somewhere
between $100 and $200 billion – up from perhaps a total of several tens of billions at the
start of this decade 7 years ago. These investment positions are quite straightforward;
there are few, if any, spread products or options or leverage that would augment this
overall position. If so, one is hard pressed to explain the increase in recent years in
commodity derivative positions in the many trillions of dollars by citing such strategic
allocations into commodity baskets.
There is another alternative: speculating hedge funds. It is estimated that the total assets
of hedge funds of all types globally now approaches $2 trillion (up from several hundred
billion 6 years ago) and many of the types of funds who speculate in commodities
employ huge leverage. The increase in gross assets of those speculators could account
for much more of the growth in commodity derivatives. But do they?
We got something of a window on this world at the end of the third quarter of last year
with the collapse of the hedge fund Amaranth. Apparently, this fund lost perhaps $6
billion or more in one commodity –U.S. natural gas – in a matter of weeks. The
magnitude of this loss and the change in natural gas forward prices at the time implies a
gross position in that commodity that was at least a small whole number multiple of the
$6 billion loss. By contrast, the nominal value of all the U.S. natural gas derivative
positions of all outstanding commodity baskets was not much larger than Amaranth’s
loss.
Clearly, in natural gas one speculative hedge fund, Amaranth, was many times “larger” in
the gas forward markets than all the world’s investors in commodity baskets. And
Amaranth was only one hedge fund. According to one compilation there are now
hundreds of hedge funds dedicated solely to the commodity sector, and all kinds of more
diversified funds like global macro funds who speculate in commodities. So hedge fund
positions in commodity derivatives outweigh pension and endowment investment
positions, and by a very large margin.
So if unprecedented speculation is responsible for the amplitude and duration of this
commodity bull cycle, it can be attributed, for the most part, to hedge funds.
How Investment And Speculation In Commodities Inflates Commodity Prices
In the above discussion there is an implied assumption: hedge fund speculation in
commodity derivatives has overwhelmed commodity markets and has driven commodity
Page 10 5/6/2007
prices way beyond levels justified by fundamentals. Also implicit is the assumption that
hedge funds can employ vast leverage using such derivatives, so that limited pools of
speculative capital can create huge demands for commodities.
I believe these assumptions are valid. But there is a widely held and respectable counter
argument. Derivatives are simply contracts for future delivery between two parties. To
be sure, a speculator can take a long position in a commodity with only a small margin
commitment by going long a derivative contract. But, for his long position, there must be
a counterparty short position. The counterparty taking the short position need put up only
a small margin and can apply the same degree of leverage. So the two sides of the
contract balance out. And the leverage employed using a derivative by both the long and
the short balance out. Hence the proliferation of derivative contracts does not influence
market prices, despite their leverage potential. Because the longs and the shorts basically
have the same access to leverage and take offsetting positions.
Since the mid 1990s Fed chairman Allen Greenspan and Treasury Secretaries Robert
Rubin and Larry Summers were pushed repeatedly by Congress to bring the OTC
derivatives market under regulatory scrutiny and control. The concern of some
Congressmen was that the shadowy world of OTC derivatives could increase the leverage
of speculators and thereby the risks to markets and ultimately to economic activity.
Greenspan, Rubin, and Summers fought off these efforts by Congress with unfailing
determination. There argument was the derivatives simply reallocated risk; they did not
increase risk. And this reallocation spread risk among a greater number of market
participants, including participants who were better suited to bear it. Therefore,
derivatives were reducing- not increasing- the risks to markets.
This sounds like a solid argument with theoretical underpinnings. But, for many market
participants, it just does not seem realistic. We know of too many instances where
speculators have used derivative contracts to take on very large leveraged positions, and
have thereby exaggerated price movements which ultimately led to crashes.
There are two famous examples. The first is the 1987 stock market crash which resulted
from the widespread adoption of portfolio insurance which was based on the relatively
new introduction at the time of derivatives and synthetic derivatives on the stock markets
indices. The second example is provided by 1998. In that episode it was not just LTCM
which experienced severe difficulties; many of the large dominant macro funds at the
time took very large losses in only a matter of a few months. And Bankers Trust suffered
such losses it had to be merged into Deutsche Bank. These funds and prop desks had
pushed a whole set of markets to extremes which then all violently reversed in unison. In
many of these markets, like the fixed income and currency markets, the price distortions
were aided by the use of derivatives and the subsequent crashes were exacerbated by the
unwinding of derivative positions.
Many have argued that the explosion in fixed income and forex derivatives in this decade
– which one sees via the published reports of participating commercial banks- has not
Page 11 5/6/2007
resulted in market excesses as a result of their explosive growth since the end of the
1990s. That is probably correct. For so far, at least. So, the experience of these markets
and their derivatives can be marshaled to support the Polyanna thesis of Greenspan,
Rubin, and Summers. But one can do this only if one conveniently chooses to forget the
derivative related debacles of 1998..
How does one settle this dispute? By looking closely at the interaction of derivatives, the
“underlying”, and the price mechanism in specific markets. So let us look at all these in
the commodity space.
First, we must decide on what constitutes the “underlying”. Commodity derivatives tend
to arise through the hedging of the stock of commodity inventories and anticipated future
production. Before the commodity bubble of this cycle- when commodity prices were
close to marginal cost- the total money value of all commodity inventories world wide
was on the order of several hundred billions of dollars. Some of this was hedged, giving
rise to commodity derivatives. And a small part of future production was hedged, giving
rise to yet more commodity derivatives. Taken together all of these commodity
derivatives were comparable to the money value of the outstanding stock of commodity
inventories and a little forward production. The world’s commodity derivative
aggregates “made sense”.
As a result of the bubble in commodity prices in this decade the money value of the
outstanding stock of commodity inventories has doubled or tripled. It perhaps now
approaches a trillion dollars. But the outstanding commodity derivatives, partly visible
through the window of commercial bank books, has gone up perhaps tenfold to 10 trillion
dollars or more. Now the commodity derivative aggregates seem to be outsized relative
to the “underlying”.
Has this explosion in commodity derivatives distorted prices and increased market risk?
It is apparent from my use of the term “bubble” and the way I have framed the above that
I believe it has. Let me explain the process whereby it has.
Roughly four or five years into this decade- and two or three years into this cyclical
economic expansion- commodity markets experienced a big bull run. Without going into
the details, I think it is easy to make the case that microeconomic and macroeconomic
fundamentals were responsible for this bull move. But once it was underway the superior
performance of commodity prices attracted attention. Long-term investors like pensions
and endowments began to consider commodities as a new asset class and hedge funds,
always looking for fast action to justify their costly 2% plus 20% compensation
arrangements, began to chase commodities.
As is apparent from the above chart, the first phase of the commodity bull move occurred
without huge growth in commodity derivatives. But from the beginning of 2005 onward
there has been a vast explosion of hedge funds and pensions and endowments who have
tried to get longer and longer commodities by way of the purchase of long positions in
commodity derivatives. Looking at the pattern of expansion of commodity derivatives
Page 12 5/6/2007
one can make a prima facie case that the investment and speculative flows into
derivatives. with their huge implied leverage, pushed commodity prices further than they
otherwise would have gone after 2004.
How could this have happened? Initially, before the explosion in commodity derivatives,
the structure of commodity derivatives markets was as follows: speculators tended to be
net long, commercial consumers tended to be net short as they hedged some of their
inventory, and producers were mostly net short as they hedged some inventory and some
future production. The speculators were able to go long commodities by way of
derivatives because commercials, who were natural hedgers of some of their inventory
and some of their nearby future production, were willing to increase the volume of their
hedges.
How did this come about? Speculators wanting to go “long” had to create a price signal
that would lead to an increase in the supply of these derivatives, since for every long
there must be a short.
Buying pressure by the longs through commodity derivatives raised the price of
commodities above their marginal cost which is their long run price equilibrium.
Commercials, recognizing where marginal cost lay, were encouraged by higher prices to
hedge more of the outstanding stock of inventories. They were also encouraged to sell
more of their future production forward.
In past cycles, when commodity prices soared, spot prices tended to rise much more
rapidly than forward prices. Hedgers did not forget that the long run price equilibrium
was still marginal cost. As a consequence these markets went into steep forward
discounts or backwardations.
But, in this cycle, so great was the buying pressure of the longs in commodity derivatives
that, starting in 2005, far forward prices were pushed up along with spot and nearby
forward prices. When speculators and investors go long derivatives they are going long
the forward price. If speculators want to take on positions that are ever larger relative to
the “underlying” they have to push up far forward prices further and further in order to
induce producers to sell future production ever further forward- and thereby provide the
increasing supply of derivatives that accommodates the new speculative demands.
In keeping with the unprecedented explosion in commodity derivatives in this cycle, this
entire process has happened in spades. Most striking has been the case of crude oil. In
the past, whenever crude oil was at marginal cost, the forward market was in a small
backwardation. When the crude oil price rose sharply the backwardation became huge.
In this cycle it has been totally different. As investors and speculators bought more and
more oil by way of derivatives the far forward crude oil price was driven to a huge
premium over the spot price- something that had never, ever happened before. Why?
Because this is what it took to get commercial hedgers to generate the unprecedented
crude oil derivatives supply that rabid investors and speculators now demanded.
Page 13 5/6/2007
Let us go back to the Polyanna thesis of Greenspan, Rubin, and Summers; in derivatives,
for every long there is a counterparty short, and therefore behavior in the derivative
market is price neutral. To be sure ex post there is always a leveraged short to match
every leveraged long. But the process is not price neutral. Ex ante the rabid demands of
investors and speculators overwhelm the commodity markets and push up the forward
price. And it is only that price signal that brings forward the commodity derivatives
supply that ex post completes the identity of longs and shorts, supply and demand.
Does this price impact create a market risk? Of course. How?
If a tsunami of rabid investment and speculative commodity derivative demands hits the
commodity markets, it must drive the forward price more above marginal cost than in a
normal bull cycle. The higher the price is driven above marginal cost the more new
supply will be encouraged. These high prices will also lead to a more assiduous effort by
commodity consumers to economize and substitute, thereby rationing demand. If
unusual commodity derivative demands take prices very high and on a sustained basis,
the resulting surpluses that will eventually take down these prices will be all the larger.
But there is another facet to the increase in risk in commodity markets created by
derivative tsunamis. It is the financial risk created by the vast implied leverage of
derivatives. Speculators, by putting up only “limited margin”, can take on huge
leveraged positions. Pyramiding speculators can employ ever greater leverage as prices
soar. For those leveraged speculators it takes only a partial correction of the price rise to
wipe them out. In this way hedge funds can fail, as LTCM would have failed without its
bailout, and as Amaranth almost failed a mere few months ago.
The Pollyannas about derivative leverage always emphasized symmetry in the derivative
markets: for every long there must be a short, the leverage of the longs must be matched
by the leverage of the shorts. But in the commodity markets there is not asymmetry of
behavior.
How so? The pyramiding leveraged speculators in commodity derivatives must meet
margin calls when the price eventually goes against them. They may not be able to do so.
Then there will be failures. There is no parallel to this with regard to hedging
commercials who are short by way of derivatives. Yes they may be leveraged, but they
have the commodity in inventory on the shop floor or on the production site or in a
warehouse somewhere. Or they have future production secured in the form of extractable
reserves in the ground or the wherewithal to produce another crop. Whoever are the
dealers the hedgers they have their margin position with, the odds are that they have the
“stuff” to deliver against their forward sale when it comes due; hence, there is no margin
call or the margin call will be financed by their dealer.
It is this situational and behavioral asymmetry between the leveraged longs and the
leveraged shorts in commodity markets that creates financial risks and the potential for
crash dynamics when derivative leverage creates excessive bull moves in prices. For
the speculative longs who are leveraged, when the leverage goes against them they get
Page 14 5/6/2007
margin calls. And when the margin calls come fast and furious they must sell or be sold
out. Not so for the hedgers who are short who are “carried” by their dealers on price rises
that go against them. More importantly, when prices fall such commercial hedgers often
just sit with their shorts. They know they have the “stuff” to deliver when the delivery
due date comes, and they may simply wait until that due date and then deliver their
“stuff” against their short. So the longs are forced to liquidate. But the hedging shorts
may not be inclined to accommodate them by buying in their shorts. It is this behavioral
imbalance that creates crashes.
A Brief Note On Oil
My objective is to get in the end to our topic of reserve management – which are gold and
credit spreads. But it is probably worthwhile to touch on the single most important
commodity first, which is crude oil, before we go on to metals and gold.
The price of crude oil has appreciated almost as much in this cycle as the most bullish of
all the commodities – the base metals. So, one may ask, is it a bubble? Two years ago
the noted money manager Jeremy Grantham posed this question in an interesting way.
He presented a chart of the real inflation adjusted oil price going back to 1875.
He then noted: “Over the years we have asked over 2000 professionals for an exception
to our claim that every asset class move of 2 sigmas away from trend had broken, and not
one of the 2000 has ever offered an exception! This should be scarier than the fact that
GMO has tried so hard to find one and failed. But we always have said that intellectually
you can imagine a paradigm shift in an asset class price, even if we have been unable to
document one yet in history. Exhibit 5 shows the price of oil and 1 and 2 standard
deviation bands. If the new price averages $50 and above, it will look suspiciously like
the real McCoy. Chinese growth and supply problems could do it. It’s the best
possibility I’ve seen in my career. But the investment desert is littered with the bones of
those who bet on new paradigms.”
Page 15 5/6/2007
So for Grantham any asset class that moves more than two standard deviations from trend
is apparently a bubble and history says that such markets absolutely always mean revert,
such bubbles always burst. This happened with the bubble in oil in the 1970’s. But
Grantham opens the possibility that crude oil this time could be the first exception.
Why? As alluded to above, it is a popular view (the peak oil thesis) that, in this cycle and
maybe forever forward, the supply of oil will be severely constrained. And that this will
be even more so if there is an adverse geopolitical development in the Middle East that
disrupts global oil supplies. If oil is the first exception in history to the bursting of all
bubbles, this will be why!
Let us look at oil’s fundamentals. When one looks at the supply/demand data for crude
oil over this cycle it falls within ranges that I would regard as reasonable relative to
history. Demand growth, which averaged perhaps 1.6% per annum in the prior decade,
was elevated in 2004 to a 4% rate by a synchronous global expansion led by the emerging
world and by a power generation shortage in China that led to a transitory need to use
large quantities of diesel to supplement the coal fired power grid until new power plants
came on line. That rate of global oil demand growth has since been tempered. This has
been due in part through conservation and substitution (price rationing), as one might
expect. It has also been due to less reliance in China on diesel for power as the coal fired
power grid expanded. So based on the global demand data, the emergence of China as an
economic power has not changed the global demand trend by that much. On the other
hand, so far at least, we have not seen any of the severe demand rationing that occurred
after the 1970 bull market in oil which led to a very large outright decline in oil demand
in the early 1980s.
On the supply side, the depletion of the large mature oil fields found decades ago that the
world now relies upon and the absence of comparable sized new discoveries has surely
constrained supply. Global capacity growth has been positive, but it has only risen from
a 1% annual rate to only a 3% plus rate this year, despite the high prices that have
prevailed in this cycle. So the fundamentals of oil supply suggest more supply restraint
than in the past. But so far there is no “peak oil”; high prices still encourage some
supply.
So if one looks at the global oil market, demand growth exceeded capacity growth early
in the decade by a small margin (2%-3%). Now, in response to the very high oil price,
demand growth has come down by two or three percentage points, capacity growth has
gone up by perhaps 2 percentage points, and a buffer of unutilized capacity has slowly
emerged and is gradually on the rise.
In the case of the oil market we are looking at modest changes and differences in the
supply and demand growth rates. This is typical of most commodity markets. There is
nothing really shocking about any of the fundamental developments in this sector. There
is none of the huge increases in primary supply that we are now seeing in base metals, as
I will illustrate later.
Page 16 5/6/2007
That said, I believe that much of the explosion in overall commodity derivatives over the
last half-decade must be attributable in part to investor and speculative long positions in
crude oil. This investor and speculation pressure is most apparent in the emergence of a
supercontango – a premium in the futures and forwards curves that recently has
sometimes been several times the cost of carry. Theory says such forward premiums or
supercontangos should not occur because arbitragers should buy the physical and hedge
with a forward, locking in an enormous riskless arbitrage profit. I believe that such a
super contango can exist only if investment and speculative demands in the forward
market are so large that virtually all available storage is filled by arbitrageurs, making it
impossible for them to arb away any further such supercontango.
For this reason I believe that investment and speculation in crude oil derivatives has
materially inflated the price of oil, even though the supply and demand fundamentals are
not seriously “out of whack”. The tsunami of hedge fund speculation in commodities is
responsible for much of the amplitude and duration of the bull move in most commodities
in this cycle. This no doubt applies to crude oil to some degree. But not to an absolutely
overwhelming degree, as I believe is the case in the metals sector.
Part II Metals: A Speculation to the Point of Manipulation without Precedent in the
History of Commodities
Preface
Before proceeding with this section, which is replete with extreme statements about the
current state of the metals markets, I believe a reminder to the reader is in order. A
decade ago a famous manipulation of the copper market was revealed. This led to a big
bear market in copper and many class action lawsuits against the perpetrators and deal
dealers and banks who were “peripherally” involved. That these things happen in these
markets seems to be forgotten in today’s “euphoric” market environment. But, at this
very time, after many prior settlements of similar claims with large awards to the
“injured”, the last such case is now being taken to court. I think it is worth while to quote
a recent description of this coming court case and its claims to remind the reader that
what follows does not lie outside the realm of plausibility in metals markets.
“Last legal action from 1996 copper scandal to go to trial
Metals Insider - 3 May 2007
A US federal judge in Wisconsin has ordered a trial next month in a long-running
antitrust lawsuit claiming J.P. Morgan & Co. Inc. conspired with Japanese trading house
Sumitomo Corp to manipulate the copper market in the mid-1990s. It is the last legal
hang-over from the 1996 copper crisis with most of the other claims long settled out of
court. This action is a consolidated law-suit brought on behalf of around 20 US copper
consumers seeking an estimated $1 billion in damages from J.P. Morgan Chase & Co.,
the successor to J.P. Morgan. "We're going to show a jury of good Wisconsin citizens the
chicanery and manipulation that went on and how manufacturers of copper wire and
copper rod suffered," Atlanta lawyer James Bratton, who represents Southwire Co. and
Page 17 5/6/2007
Gaston Copper Recycling Corp, told local media. "We're looking forward to getting a
big verdict." The centrality of the allegation against the bank is that it provided
financing to Sumitomo to artificially inflate the copper price. The bank argued in court
documents that its transactions with the Japanese company were proper and did not have
an impact on copper prices. However, its motion to dismiss the case was rejected by US
District judge Barbara Crabb and the trial is scheduled to begin on May 29.””
Introduction
In every market bubble there is some cutting edge where the greatest extremes are to be
found. In retrospect, the extent of the speculation has always seemed almost impossible,
though, amidst the fury of the bubble, very few recognized it for what it was. The
chronicles of bubbles in the past like Charles Kindleberger’s, “Manias, Panics, and
Crashes”, and Edwards Chancellor’s, “Devil Take the Hindmost” show that, in almost all
such bubbles, at such a cutting edge speculation is not only unimaginable but involves
some measure of fraud and manipulation.
I alluded to some such behavior in the metals sector in my presentation to you last year,
though with little specificity. This year there will be no allusions. I think we know
enough to say that speculation in metals markets in this cycle has gone further than in any
other cycle in history. What we have been undergoing is a speculation to the point of
manipulation, perhaps involving collusion, across a whole array of related metals
markets.1 I argue that it is as though the famous episode of the Hunts in silver decades
ago has now been taken to a power.
I understand that these are very strong words. To back them up I will discuss a few base
metals where it has become quite apparent that something like this is happening. I will
then touch briefly on a few others as well as the white metals – before we get to the
subject of gold.
My reasons for making such an unusually strong claim are many. Some come from
“inside” reports about the activities of hedge funds and others operating clandestinely in
these markets. Some come from analysis that points to large anomalies in these markets
over recent years. And some come from claims and analyses about extreme speculation
to the point of manipulation that has fallen into the public domain.
In fact, since I last spoke to you a year ago, a great deal of commentary has surfaced in
the public domain. We have put together a compilation of this commentary which we are
attaching as an appendix. In what follows I will avoid all “inside information” and
confine myself to analysis and a small subset of this commentary that we have culled
from the public domain.
1 Am I going too far with these words? The head of the enforcement at the CFTC just issued a warning:
“Regulators need more funds to guard against fraud and manipulation”. Why this choice of words? See
the above preface. And see more on this later in this text.
Page 18 5/6/2007
But let me just say, we have many, many reports from market participants that are more
specific than the statements we have culled from the public domain. Also, many of these
reports are from investment firms who distribute bullish metal market assessments
supposedly based on facts to their clients when, at the same time, they privately report
manipulation and collusion as the dominant force in these markets.
In analyzing base metals one has to bring up an all important and rather unsavory feature
of these markets: the emergence again and again in most past bull cycles of squeezes or
attempted corners. (In this regard, see the quotes from Paul Krugman in a later section
entitled, “The Risk of Revulsion II, Revulsion From A Hamanaka on a Massive Scale.”)
In the past metal merchants, faced with a bull market environment, often hoarded
physical metals. By taking metal out of circulation prevailing shortages were
exacerbated. This amplified panics by consumers caught short of physical, which in turn
fueled further the bull market move. This happened in most cycles in base metals since
the late 1960s. But in these past cycles such merchants usually sold when the markets
had not yet gone into significant surplus. In effect, they tended to get out when the going
was good.
A squeeze operation like this, by taking metal out of circulation, always distorted the
supply, demand and stock data. As hidden stocks were built, visible stocks declined
faster and fell further. Market conditions appeared tighter than they really were.
Not only was the stock data distorted, so was the data on demand. We define the demand
for a metal commodity not as the demand for that metal in all the products that
incorporate it which consumers buy – that measure of demand is far too hard to calculate.
To simplify, statisticians in commodity markets define demand as simply the absorption
of metal by first stage processors. Take for example copper. The primary product – a
copper cathode – is put into a furnace by a wirerod mill. The mill produces wirerod
which is then turned into wire. And that wire is incorporated in many products that
consumers buy. Copper demand is defined as the absorption of the metal by a wirerod
maker rather than a wire manufacturer or the maker of final products that embody wire.
But even this concept of demand is hard to calculate for almost all economies. It turns
out it is simply too taxing to monitor the purchases of metals by first stage processors.
To simplify yet further statisticians, for the most part, use a concept of apparent demand.
Demand is defined for a given country as domestic production plus net imports (in other
words overall supply) adjusted for the change in visible stocks in that economy. The
residual produces the estimate of apparent demand.
It follows that, if there are builds of unreported stocks (due to commercial inventory
building or merchant squeeze operations), total supply from production and imports will
have to be higher and the residual, which is apparent demand, will have to be
correspondingly higher. So squeeze operations with their typical hidden stock builds not
only portray less inventory than actually exists; they inflate apparent demand relative to
the true level of demand.
Page 19 5/6/2007
Earlier in this paper I described an almost unimaginable explosion in commodity
derivatives due to massive investment and speculation by pensions and endowments and
hedge funds. It would not be out of keeping that some of these investment and
speculative demands would have spilled over into the accumulation of physical stuff
rather than just derivatives. But, remember, all the physical commodities in the world at
the prices of several years ago only had a total value of several hundred billions of
dollars. At today’s prices the value of these above ground stocks is much higher. But it
is still a very small fraction of the total mountain of commodity derivatives and probably
the total investment and speculative claims on commodities by way of derivatives. If this
is so, even a moderate spillover of such investment and speculative demands into
physical stuff could involve an accumulation of physical inventories which is large
relative to total inventories and the flows of supply and demand.
Has such a spillover occurred? Yes. There are now commodity oriented hedge funds
that make it clear in their documents that they purchase physical commodities as well as
commodity derivatives. It has also become fairly clear from detail provided on the
ownership of metal on warrant with the commodity exchanges that hedge funds hold very
large amounts of these physical stocks – a point I amplify on below. So if hedge fund
documents declare that such funds may own physical, if exchange reports suggest they
sometimes own very large amounts of visible physical stocks, it is certainly possible they
own “off warrant” material as well. So the issue is not one of whether we have
investment and speculative holdings of physical stuff in the cycle; it is rather a question
of how much and in what form.
Unless the holdings of physical commodities by institutional investors and speculators
take the form of exchange warrant claims on exchange stocks, there will be no record of
such physical holdings. These stocks will be “hidden”. Increases in such hidden stocks
are recorded by statisticians as increases in apparent demands that exceed the increase in
real demands. In effect, just as with our case of the merchant squeezer, the spillover of
fund speculation into physical commodities may result in an understatement of the true
level of above ground stocks and an overstatement of the level and growth rate of
demand. Where markets are balanced, they will appear to be in deficit; where markets
are in surplus, the surpluses will be understated. Given the possible magnitude of such
investment and speculation in physical commodities indicated by the new Mount Everest
of commodity derivatives, these distortions in the stock data and the supply and demand
data could be very large.
Now, let us put together the historical behavior of merchant squeezers and the more
recent behavior of institutions speculating in commodities – including physical. A
question arises, have hedge fund forays into physical commodities, and particularly into
metals, occurred simply as passive investments or have they occurred with the objective
of conducting squeezes and corners as merchants and speculators have done in the past.
The answer is clear – in many cases the motive must have been to squeeze or corner.
Why is it so clear? Because there is considerable evidence that, in the metals markets,
hedge funds have taken positions in exchange warrants – physical holdings – even though
Page 20 5/6/2007
these markets have been in significant backwardations. And, as our so many of the
commentators in our appendix indicate, these funds have held off warrant material as
well. In a backwardated market holding a forward (which is at a discount to the spot)
provides a positive return as one rolls one’s future or forward into a successor contract.
In the last two years, there have been periods when this “roll yield” into a significant
backwardation has been extremely high – perhaps as much as 20% -30% annualized. By
contrast, holding exchange warrants and off warrant physical provides no positive roll
yield under such circumstances; rather, one must pay the cost of financing plus storage
and insurance. Why would any fund hold physical at a considerable “carrying” cost
rather than a future or forward with no such cost but with a handsome positive roll yield
instead? The only reason I know of is to conduct a squeeze or corner operation that
yields other returns if the squeeze succeeds. (Again, see Paul Krugman’s comments later
in this paper.)
If hedge funds decide to conduct squeeze plays and attempted corners as merchants
traditionally have – but with their far vaster financial resources – we face the possibility
of builds of hidden stocks and distortions in our measures of inventory, apparent demand,
and market balances that could be far greater today than they were at any point in the
past.
It is my position based on much information that this is in fact what has occurred over the
last two or three years. As I have said above, I am not alone in this regard by any means.
Just scan the appendix with our compilation of commentaries on such speculation and
manipulation of metals markets from many market observers. And, as an example, look
at his recent commentary on this very subject by Morgan Bank.
“This week we examine the tension in the metals market between (visible signs of
weakness in) industrial trends and the impact of the now super larger and super
leveraged commodity funds. We test the water to ask if there is manipulation or collusion
of if the sector has inadvertently become a price support and inventory sterilization
mechanism without consciously planning to be.”
The author then cites two historical examples where price support was done through the
accumulation of physical commodities: De Beers in the diamond market and a producer
cartel in tin during the early 1980’s.
“Excess inventory gives consumers pricing power and low inventory gives producers
pricing power. This is something that the tin cartel realized and for years that industry
thought that it may have found the holy grail of production, inventory and price control
but the eventual and inevitable demise of that cartel was a history lesion showing that
such systems cannot last forever, even if they can last for quite some time.
In the current base metal markets it is possible that such a system is operating either
unintentionally, by default, or intentionally. Whenever there is a possibility that
commodity funds with huge cash flow decide to make their investments not only via
Page 21 5/6/2007
futures and options but by holding metals then there is the risk that a crude De Beers type
inventory limitation is in place.”
The Nickel Market
On to examples.
Today the price of nickel is close to $50,000 a tonne. Its historical mean is more like
$7,000 a tonne. The price of nickel in this cycle has now risen almost 900% from its
prior cycle low. Above I showed Jeremy Grantham’s chart of the “real” oil price over
130 years. The move in the real nickel price is an even greater deviation from the mean
than was oil in the 1970’s. That is really amazing since, with oil in the 1970’s, there was
the very inflationary psychology of the era which fostered hoarding, there was the loss of
roughly 10% of global output with the revolution in Iran, and there was the price support
provided by the OPEC cartel.
If Jeremy Grantham’s observation that all such two sigma plus departures from the mean
eventually end in reversion to the mean, if this “rule” proves right once again in regards
to nickel, we will see in the years to come that nickel will be $7,000 a tonne again.
How did the nickel price get to such a lofty level?
The Wall Street analysts will tell you that it has done so because demand is super strong
and because of the booming economies of China and India. They will tell you that
supply growth has been restrained for many reasons. As a result, the market has been in
an acute deficit and prices have soared.
On the face of it this argument seems to have some merit. There have been many
primary nickel projects that have been delayed. Demand for stainless steel, which is the
principal market for nickel, soared an amazing 16.7% last year.
But people close to the nickel market argue otherwise.
About two years ago there was a meeting of nickel producers and consumers in Portugal.
That meeting ended in a dispute between nickel consumers and producers that was
disorderly to a point approaching pandemonium. At the time nickel consumers claimed
there was no shortage of nickel and that an artificial shortage had been engineered by
hedge funds. They claimed that speculation by funds on the LME had divorced the price
of nickel from reality and that the LME had ceased to function as a “mechanism for price
discovery”.
These claims have never ceased. You will see in our appendix references by the press to
hedge funds that own all the physical nickel.
One of the most interesting complaints has been made by China’s largest nickel
producing company, the Jinchuan Group – a company that is largely owned by the
Page 22 5/6/2007
Chinese state. Jinchuan says more than once on its website that the “LME nickel market
is excessive of speculations and with a suspicion of manipulations”, “the LME is no
longer a place for fair dealing of metals but a paradise of speculations”. It warns
customers “not to be puzzled by deceptive information of nickel stock, conditions of
supply and demand, as well as price released irresponsibly by a few foreign agencies.”
Part of its ire seems to be directed at the LME which allows such shenanigans to run
amok. “Lots of doubts were aroused in a survey about LME’s fairness, justice and
openness, and also about its role in price discovery.”
As I have said, the Jinchuan group is not alone. Speaking about the last run up in the
nickel price to its peak the Russian nickel producer Norilsk made the following comment:
“David Humphries, chief economist for Norilsk Nickel, said hedge funds had moved in
for the kill, triggering a violent “short squeeze” on the futures markets.” Implying that
physical stock is being hidden, ABN Amro notes, “The task of the nickel longs is to keep
inventories at these critical levels. They will then be rewarded with acute pricing
tension.”
Since that conference in Portugal two years ago, those who have complained about an
unrelenting short squeeze by hedge funds and merchants, often suggested to be operating
in collusion, argue that the market is ceasing to function as a market. This may be true.
Recently, the nickel price made a new high above $50,000 a tonne. An LME trader
reports it was done on no volume, with the usual comments about the lunacy of the
market.
“Nickel's three-month price broke the $50,00 mark at 9am London time on Thursday
morning when 1 lot traded at this level in a market traders have dubbed "lunatic".
The price of the alloying metal has not traded below $49,200 since the start of the
trading day, after it closed at $49,400 on Wednesday, but volumes have been extremely
thin, with only 28 lots having traded by 9:29am on Thursday morning.
"This is lunacy," said a physical trader. "I better close shop and come back when we are
at $40,000 again. The phone has not rung once this morning. But we're heading for a
fall at these levels. Material is coming from consumers now offering to us. Not small
little consumers. The big [stainless] producers."
He said the market lacked natural sellers, and the price surges are on speculative trading
only.
"It's completely artificial. There is no connection whatsoever between the LME and the
physical market now. The LME has its own life."”
The alleged nickel squeezers which encompass hedge funds may have pressured so many
short side participants out of this market that it has been reduced to the trading of a mere
handful of lots on what would seem to be a price pivotal day.
Page 23 5/6/2007
Let us assume that much of this is correct. Then hedge funds and perhaps merchants as
well have built hidden stocks. The metal is not as scarce as it seems. More importantly,
apparent demand has been inflated by the build of hidden stocks. Most statisticians and
analysts assume the nickel market was in a small deficit last year. But, instead, the
market may have been in a surplus rather than a deficit.
Jinchuan’s website goes far beyond complaints about speculation in, and manipulation of,
the nickel market. They provide documentation of a very rapid move towards lower
nickel bearing stainless grades and non-nickel bearing types of stainless steel. They cite
a proliferation of primary nickel projects down the road which they believe will not be
needed. They fret that today’s price distortions will lead to permanent demand
destruction for nickel producers and a painful glut when the speculation is over.
And that point may be not far off. Stainless steel demand was up 16.7% last year. Part of
that was a rebound off a prior year where there was a global stock liquidation. But even
so, it is a growth rate more than four times the past trend. It must have reflected very
substantial stock building in stainless. This is what is now being reported by the stainless
producers and the service centers that distribute stainless. The very pronounced
inventory cycle of stainless may have now gone so far that a reversal is probably
underway. MEPS, the “industry statistician”, reflecting this, is predicting a 4% decline in
stainless steel demand this year. Add to this the economization and substitution in nickel
use that Jinchuan and many others talk about and a significant decline in nickel demand
should be at hand. For example, the steel giant Posco has just announced that it will use
14% less nickel in stainless production this year as it turns to non-nickel bearing
stainless.
Is this happening? Apparently it is. In Europe there are excess inventories. The
European stainless producers report falling orders. The price of stainless has fallen from
1900 euros a tonne to 1300 euros a tonne in six weeks. Major producers like Outokumpo
have announced 10% production cuts.
But the turning point dynamics do not end there. Last year China, looking to bypass high
nickel prices and the alleged shortage of nickel, began to access already mined laterite
ores in Australia and the Philippines. They used idle blast furnaces to process these ores.
The investment involved was minimal, as the ores have been mined and there was ample
blast furnace capacity everywhere in China. Last year nickel output from such laterites
rose from almost nothing to 30,000 tonnes. This year they predict an increase to 60,000
tonnes, representing a surprising 5% increase to supply in a mere two years. Some
analysts who have looked at this issue think that production from nickel laterites by
China and Japan could be much higher than 100,000 tonnes this year. In addition, to this
the world will finally see a significant ramp up in the primary production of nickel from
more traditional sources in 2007 – perhaps on the order of 7%. This implies an overall
increase in primary supply of 10% and possibly significantly more.
When I look at all these impacts on the balance – a reversal in the stainless inventory
cycle, substitution and economization, increases in conventional and non-conventional
Page 24 5/6/2007
supplies – it would seem that the market balance in nickel could swing this year by 15%
of demand/supply or more. Such a swing is unheard of. At the same time, adjusting for
distortions in the apparent demand data, the market may have already been in surplus last
year despite a likely large stainless inventory build associated with that 16.7% rise in
stainless demand. In effect, the nickel market may already be moving into a vast
unprecedented surplus.
And yet, owing to the “LME paradises of speculation and manipulation” the nickel price
is still soaring further and further beyond its two standard deviation “bubble” status
threshold.
Copper
The price of nickel in this cycle has gone up almost nine times from it prior cycle low.
The rise in the price of copper has not been quite so extreme. It rose more than 6.5 times
above its cycle trough at last May’s $4.07 a pound high. It rose well in excess of five
times in real terms, and is within 10% of last May’s price peak a whole year later. This
deviation of the real copper price from trend surpasses that of crude oil in the late 1970’s
despite the generalized inflation psychosis of the 1970’s, the Iranian revolution that shut
down one of the world’s biggest oil producers, and the support of OPEC, the most
famous commodity cartel in history. By Jeremy Grantham’s criteria, copper in this cycle
is a two standard deviation bona fide bubble event.
You can divide copper cycles in different ways depending on your choice of endpoints.
If one’s choice of end points is half decade cycles in inflation adjusted terms, the increase
of the copper price in this cycle was already greater than in any cycle since 1900 once we
got to $1.90 a pound in 2005. By this measure no rise in the inflation adjusted copper
price has ever approached what has happened in this cycle.
Page 25 5/6/2007
The above refers to cyclical moves in the copper price in real or inflation adjusted terms
from 1900 forward. I have found a similar analysis of copper cycles in nominal dollars
since the year 1860. This analysis considers longer duration cycles, out to as much as ten
years. This period, starting in 1860, encompasses the Civil War, the First World War, the
Second World War, the Korean War, and the Vietnamese War. Wars tend to consume a
lot of copper. Wars also spawn generalized inflations, and this period since 1860
encompassed the inflations of the Civil War, the First World War, the Second World
War, the Korean War, and the persistently inflationary period from the mid 1960’s to
well into the 1980’s. This period also encompassed the advent of electricity and
telecommunications in the late 19th century. This was the most important technological
revolution and economic engine of those decades. Copper was the material essential to
that New Era industrial boom.
Even though this history since 1860 encompassed all these wars and inflations and a
copper critical industrial New Era the largest percentage increase in the dollar copper
price in any of these past cycles was only 246%.
IN THIS CYCLE IN A MERE FOUR AND A HALF YEARS INTO MAY 2006 THE
COPPER PRICE ROSE 575% AMIDST THE LOWEST INFLATION IN THE U.S.
GENERAL PRICE LEVEL IN ANY ECONOMIC EXPANSION IN ALMOST A HALF
CENTURY.
Page 26 5/6/2007
Some will say, this must surely be because of a new era of super cycle copper
consumption led by China, India and the emerging world. In fact there is absolutely
ZERO evidence for this in the official data on global copper consumption.
Global Copper Consumption
Average Annual Growth
1980 – 2005 2.3%
1990 – 2005 2.6%
2000 – 2005 2.6%
Page 27 5/6/2007
2006 2.3%
Source: ICSG
In general, when confronted with this data, people simply refuse to believe. It is obvious,
they say, that copper consumption in China has been booming; therefore, such data must
be wrong.
But what has been happening is something that has been going on for decades. As I
described above, base metals consumption is defined as the absorption of metal by
primary processors. Since the 1960s the primary processing of copper cathodes has been
migrating from the first world to the emerging world because such fairly low tech
manufacturing makes more economic sense in emerging economies than in higher cost
developed economies. Now that China, with its super high investment ratio, is displacing
more and more manufacturing that heretofore had been done in the U.S. and other
advanced economies, its booming copper fabrication industry is displacing at a dramatic
rate this industry in the advanced countries of Europe, the United States and Japan.
World Consumption
2000 – 2006
Average Growth Per Annum
W. Europe -1.5%
USA -5.5%
Japan -0.6%
China 12.6%
In the end, with all this furious migration of the primary processing of copper cathodes
from the first world to China, the trend in the overall global consumption of copper has
remained basically the same.
If this is so, why has the price of copper soared way, way beyond all historical precedents
in this cycle? Is it a constraint on primary supply unlike the world has ever seen? In fact,
that has not been the case either. Yes, a reduction in new investment in primary
production and a closure of mines followed the period of depressed prices at the end of
the 1990s. This resulted in a lagged production response when this global economic
Page 28 5/6/2007
cycle took off, thereby creating a deficit in 2003 and 2004. But, despite that lagged
supply response, the trends in copper supply growth look, if anything, more positive than
in the past.
Primary copper production comes in two forms: the production of ore concentrates which
must go to smelters and direct on site production of refined cathode using solvent
extraction (SXEW).
World Concentrate Production
Growth Rates Per Annum
2000 - 2006 2.4%
2007 - 2010 4.1%
2007 - 2015 4.3%
World SxEw Production
Growth Rates Per Annum
2000 - 2006 3.9%
2007 - 2010 8.9%
2007 - 2015 4.1%
Last year was a year marked by an unusual number of work disruptions in copper mining.
The global mine capacity utilization rate was very low. Nonetheless, copper mine and
refined production grew at or above the long term trend. Based on a recent return to a
higher capacity utilization rate and a round of mine start ups and expansions early this
year we should expect in 2007 much higher mine output growth and continued well
above trend growth in refined output. Looking further ahead, almost all compilations of
future total primary production (both SXEW and concentrates) based on what has been
Page 29 5/6/2007
announced to date results in a supply trend of close to 5% for years to come – a rate that
is almost two times the historical average rate.
It should be noted that such projections are based on what has been announced. When
prices are high exploration budgets rise. Exploration produces results. Existing deposits
become larger, additional investments are therefore made, and surprise expansions tend
to proliferate. Also, new low cost projects are found., In the past, at least, these
unforeseeable successes have proved to be the most important engine of supply
expansion over time.
So what then has caused the copper price to soar? Once again, unprecedented
speculation to the point of manipulation. In the case of nickel, I gave you accounts of the
role of speculation according to China’s leading nickel producer Jinchuan and a few
analysts from investment banks. In this case of copper I will provide you with an account
based on two recent presentations of Nexans, the largest copper fabricator in the world.
Nexans operates in 33 countries and accounts for about 7% of all copper wire
manufactured globally. Nexans has laid out its analysis of the copper market in two
power points used in two recent public presentations
Nexans starts their market analysis by noting the difference between official statistics
based on “apparent” consumption and real consumption – that is, what is going into
furnaces. Nexans believes the copper market has been in an increasing surplus, possibly
since autumn of 2004. Where has this surplus gone, since there has only been roughly a
200,000 tonne increase in visible exchange stocks over this period?
February 8, 2007
Who hoovered away so much physical copper?
February 8, 2007
Page 30 5/6/2007
Again, in a second presentation:
April 20, 2007
In our appendix you can find many references from the public domain that correspond to
Nexan’s claims. But, we should ask, is there any analytical basis for such a claim that the
copper market, which ICSG official statistics2 admit was in surplus last year at a
significant rate of 350,000 tonnes, has in fact been in a larger surplus and for years and
which has been masked by a “hoovering” away into hidden locations of more than a
million tonnes by the end of last year?
My answer is yes.
If such a build of hidden stocks by speculators has occurred, it might be discernable in
trade data which reflects the flow of metal across borders. We have looked into this. In
the case of the United States and Canada combined we compared total supply from
imports and domestic refined production with fairly good proxies for domestic
consumption. Provisionally, these respective data series embody a positive residual of
supply over consumption in 2006. This residual is considerably greater than the rise of
visible stocks in the United States during the period. The implication is that there may
have been a hidden stock build of perhaps 100,000 – 200,000 tonnes in North America
2 I realize one must be careful about this ICSG estimate. Since early 2005 ICSG produced initial estimates
of a surplus only to revise them away. Well, well after the fact the initial deficit for 2004 has been revised
up, thereby shifting subsequent year balances in the same direction. Another statistical agency, WBMS,
was carrying a surplus for 2006 through November that projected perhaps a 500,000 tonne surplus for the
year. Suddenly there are revisions and, based on the last report I have received, the estimated surplus for
2006 including December is now 377,000 tonnes. More striking is the WBMS estimate of a 115,000 tonne
surplus for the first two months of this year. Demand is probably a bit above average in January and
February; the seasonally adjusted surplus for these two months combined might be 150,000 tonnes. That
annualizes out to a 900,000 tonne surplus for 2007, which is huge relative to supply/demand by historical
standards. Obviously, one cannot reconcile such a large surplus with a copper price of close to $4. I think
in time we will see efforts being made to tweak and fudge the data toward something that is more
consistent with the prevailing sky-high price. As I say elsewhere in this report, the copper bulls including
the mining companies who fund them will look at the statisticians and say, how can you mere bureaucrats
estimate a large surplus in the market when the market says otherwise. Hence the tweaking and fudging.
Page 31 5/6/2007
over the last year. I may add that the same calculations hint toward something similar at
the end of 2005 and the very beginning of 2006.
Last year the import based ICSG apparent demand data on the copper market in Europe
showed an anomalously high rate of copper consumption. They estimated an 8% year
over year increase. If you look back at the first table in this copper section you will see
that, owing to the migration of copper manufacturing out of Europe, this European
demand growth rate is usually negative. The analyst Simon Hunt has taken the ICSG
data on European copper consumption and has compared it to a measure of true
consumption based on reports from Europe’s fabricators. These two measures of
consumption correspond over the early years of this decade. Then, suddenly last year,
apparent demand based on imports soars relative to the above demand data. Again,
implying a build of hidden stocks by someone.
The situation with China is less clear. Last year Chinese imports of refined copper fell
very sharply. As a result, ICSG’s estimate of Chinese copper demand showed great
weakness. It is widely argued that a liquidation of hidden stocks by commercials and the
Chinese Strategic Reserve Bureau was responsible for this. And there is probably some
truth to this.
On the other hand, if one looks at the two years prior, the ICSG demand data, again based
on imports, showed double digit growth in apparent demand. At the same time, all other
demand indicators suggested that demand growth was much lower. China’s National
Development and Research Commission (NDRC), its government affiliate Minmetals,
and the data we have on copper semi-fabrication in China all suggest that actual copper
consumption growth was in the single digits. The difference of course is an implied build
in hidden stocks. If one takes the several years combined it would appear that there have
been hidden stock builds in China over the last several years. It appears that analysts and
the ICSG have inflated numbers on Chinese copper consumption in this most recent cycle
and that China may be one of the places where Nexan’s alleged surplus has been
“hoovered up”.
We have done some tentative, albeit so far weaker, analyses along the same lines for
several other Asian countries. Again these analyses point to imports above and beyond
domestic fabrication, implying builds of hidden stocks in these locations.
That is one avenue of analysis that supports Nexans thesis of hedge fund and investment
bank hoovering up of a recent large copper surplus into hidden locations. There are
others.
If there has been a build of hidden stocks in recent years it should have inflated ICSGs
consumption data. But could that be? Since, for the last two years combined, 2005 and
2006, ICSG shows a rise in global consumption of less than 2%. Could that feeble a rise
still be inflated?
Page 32 5/6/2007
The answer is yes. As the first table in this section makes clear, on trend global copper
consumption has been a mere 2.5%. But, with the price of copper now far higher relative
to the price of substitutes like aluminum than has ever prevailed before, one might expect
aggressive substitution and a growth rate well below trend. Remember, this has
happened before with other commodities in spades. When the oil price rose by a
comparable percentage amount in real terms in the 1970s, global crude oil production
actually fell by a whopping 15% when the global economy was experiencing trend
economic growth.
From Nexan’s we get a window on this potential.
February 8, 2007
Nexan’s tells us that its copper consumption has been flat over the past two years. In the
wire sector substitution is more difficult, at least initially, than in the brass mill sector
where plastic tubing, thinner tube walls, etc. are options. So if Nexans has been flat,
overall industry consumption may have been down. Why has Nexan’s consumption been
flat? Because much cheaper aluminum is making inroads in wire, as their explosive
aluminum consumption testifies.
For last year ICSG reported a surplus in the copper market of 350,000 tonnes. If
consumption of copper over the last two years was a little negative rather than a little
positive, one gets a surplus of perhaps three quarters of a million tonnes. ICSG can
account for most, but not all, of its estimated surplus for 2006, implying a possible hidden
stock build on the order of 100,000 tonnes. If the real surplus was higher, so was the
implied hidden stock build.
Let us try another avenue. ICSG has estimates on refined production and mine
production. Commodity statisticians time and again make errors at turning points. And
one of those errors is a systematic edging of the supply and the demand data to generate a
balance that is consistent with prevailing prices. ICSGs significant 350,000 tonne surplus
in 2006 is already wildly inconsistent with today’s prices which are unlike anything ever
seen before. Could it be that they have been edging down their production data estimate
– because otherwise they would show an even larger surplus, which would be even more
out of whack with today’s sky high prices.
There are many compilations of global mine and refined production from statistical
agencies, metal consultants, etc. We have looked at many of these. In particular, we
have looked at one where, given the business situation of the analyst, there would be a
bias, if anything, towards a lower, not a higher, number. We have looked at this analyst’s
Page 33 5/6/2007
strengths and weaknesses to make a judgment on any likely error. We have found that
this compilation stands up to detailed scrutiny. It generates a production number that is
many hundreds of thousands of tonnes higher than that of ICSG.
If we plugged this number into the ICSG framework, their 350,000 tonne surplus
becomes a three quarter of a million tonne surplus or higher. If we make both the above
adjustments to ICSGs refined production estimate and their apparent demand estimate, all
of a sudden we have a surplus of over a million tonnes. Of course, the larger the actual
surplus relative to the now standing ICSG estimated surplus, the larger the possible
implied hidden stock build. So you can see that there may have been a hidden stock build
in 2006 well in excess of a half a million tonnes.
Such possible errors in the official data have further implications for the possible
accumulation of hidden copper stocks in this cycle. Whatever the underestimate by the
official statisticians of the surplus last year, it most likely carries back into the prior year.
If ICSG has the 2006 percentage changes in demand and supply more or less right but the
level of both are wrong, a much larger surplus in 2006 implies a smaller but still
significant surplus in 2005 – rather than their now prevailing estimate of a deficit. The
difference between such a surplus and the official estimate of a deficit corresponds to a
possible further hidden stock build.
Therefore several avenues of analysis can provide support to Nexan’s claim that the
market has been in surplus, that the surpluses have become quite large, and that someone
has hoovered away a million tonnes of surpluses into hidden stock locations.
What can we say going forward? U.S. housing still absorbs 10% or more of copper
embodied in final products produced both in the United States and elsewhere in the
world. The trend in copper consumption is being affected adversely for the first time in
this cycle by cyclical factors with the bust in U.S. housing. If substitution was significant
in 2004 and 2005, with a much, much higher price signal in 2006 substitution should now
intensify. There is a general agreement that global mine capacity will rise healthily this
year, and so far this year the outlook is for a higher capacity utilization rate. It is not hard
to make the case that the copper surplus will increase by several percent of demand this
year. If all of the above happens, the copper surplus could increase by 3% or 4% or 5%
of global supply/demand of roughly 17 million tonnes plus. If so, we may be looking at a
surplus of perhaps 1.5 million tonnes this year. That approaches 8% of supply and
demand.
The copper market has never experienced such a large relative surplus. When it has been
approached it has always been at deep price troughs. But the copper price is not low. It
remains monstrously high. So projects are now underway that cannot be stopped, and
that ensure yet further net gains in supply going forward. And production processes that
result in economization and substitution are also underway that cannot be stopped. For
example, wire makers have found a way to clad much cheaper aluminum with copper and
achieve the same functionality with a small fraction of copper consumption and
massively lower overall costs. Initially, manufacturing aluminum wire with a thin copper
Page 34 5/6/2007
clad proved to be difficult, but this production process has finally been perfected in
China. Production lines to do this started to expand last year. The practice is now
disseminating globally. We are still in a very early stage of diffusion of this practice.
This is typical of the demand rationing process. Such lagged demand responses to the
crazy copper price increase over the last year and a half, coupled with unstoppable
increases in supply, ensure yet higher surpluses beyond 2007.
Should we be surprised? No! Speculation in copper in this cycle may have resulted in a
build of hidden stocks that dwarfs anything that merchant squeezers did in the past. The
result has been a price distortion that has no historical comparison – a six fold rise in the
dollar copper price in this cycle versus rises of 246% or much less in all prior cycles
since 1860. Microeconomic logic tells us that such an unprecedented price distortion
should lead to an unprecedented surplus.
What happens if the market goes into a surplus of perhaps 8% of supply/demand and this
grows and grows because new projects with sunk costs cannot be stopped and
investments that destroy demand are just beginning to take hold? In the end, the
cumulative surpluses will become unmanageable for any speculators, no matter how well
heeled they are.
It is not clear what will happen to the hidden stocks that have been hoovered away in this
cycle. I know that in prior cycles, when squeezes failed or when prolonged bear markets
with full contangos took hold, merchants accumulated and then held on to large hidden
stocks of copper and other base metals which they could finance almost costlessly in a
full contango market. Something like this might happen, at least initially, to the hidden
stockpiles that have been built in this cycle. But, in the end, even if the speculators
somehow hold on to their hidden stocks, a mega surplus must be dealt with. At some
point, it will not be possible for the speculators to simply keep on buying up such a
surplus. There will have to be an eventual closing of mines to the point where the market
is, at a minimum, in balance. Aside from depressions, I do not know that we have ever
seen an episode where so much production capacity had to be shut down to achieve that
end. And to do that will require a copper price below the cash costs of many copper
producers, not just the most marginal ones. Look at the current copper cost curve; that
will be a very low copper price, indeed.
Page 35 5/6/2007
Of course, as prices tumble along the way, speculators may have to disgorge their hidden
stocks. I do not believe it is possible to sell hidden stocks of a magnitude that may now
exist and will surely exist in the near future. In prior cycles stock liquidation on a much
lesser scale created powerful undershooting of the mean. This time such stock
liquidation dynamics might dwarf those of the past. This means a deeper undershooting
of marginal cost than in the past. This means greater pressure to reduce production well
below the level of consumption in order to not just bring the market into balance but to
go further and absorb the stock overhang.
I think we cannot imagine how severe the industry strains will be when this comes about.
Aluminum
The nickel price relative to the cost of production and past trends is a crazy, crazy price.
The copper price, the same. But the aluminum price is not a crazy price. Almost two
decades ago in 1988 it soared to a brief peak above $1.80. So today’s price of $1.30
looks high, but not necessarily artificially so.
Nonetheless, there is a lot of evidence that the same is happening in aluminum as has
happened in copper and nickel.
In the case of nickel I cited the Chinese nickel producer Jinchuan calling the LME a
paradise of speculations. In the case of copper I cited Nexans. In the case of aluminum, I
find it useful to return to Chinese spokesmen.
Formerly China’s planning commission, the National Development and Research
Commission (NDRC) may be China’s most important body responsible for economic
matters. In recent years it has been very focused on reducing the exposure of China’s
manufacturers to high commodity costs. It has also been very concerned about over
investment by China in certain economic sectors, including steel and aluminum. So it has
thought a lot about metals prices and their prospect.
Page 36 5/6/2007
Given that perspective, I find the following statement by the NDRC pertinent: The NDRC
said: "Possibilities of a steep fall in aluminum prices could not be ruled out if
international hedge funds pull out of the aluminum futures market next year." It stressed
that hedge funds' massive buying into aluminum futures was another cause of the bullish
prices this year."
China Daily, December 28, 2006
So, for the NDRC, the aluminum price is high today because of massive hedge fund
buying.
Has this hedge fund buying been the passive actions of many investors, or has it been the
result of a concerted effort by a few that mimics the merchant squeezes of the past, but on
a larger scale. In recent months the press – Financial Times, Reuters, Bloomberg, and
elsewhere – have provided so much commentary that the latter clearly seems to be the
case. First a note from the prestigious Financial Times.
A battle raged in the aluminium market on Monday between one investor holding a
$1.7bn long position – betting that prices will rise – against a number of shorts who are
equally determined that prices will fall.
Those holding short positions have made no effort to decrease their exposure in the
market and they potentially face significant losses if prices move against them.
Market talk suggests the holders of the short positions have substantial amounts of
aluminium available for delivery to market.
This could wreck the strategy of the investor, thought to be a hedge fund, holding the long
position, which is understood to have maintained its position since mid-December.
Analysts say the position is equivalent to nearly 645,000 tonnes, almost equal to the size
of the LME’s aluminium stockpiles of 695,000 tonnes. At current market prices, it would
cost $1.7bn to buy 645,000 tonnes of aluminium.
Copyright The Financial Times Limited 2007
Again, you hear the same from Reuters.
One party has maneuvered itself in a position where it could take delivery of almost
650,000 tonnes of the metal, or about 93 per cent of total LME aluminium stockpiles. At
current prices, it would cost $1.7bn to buy 650,000 tonnes.
If the demand for aluminium was booming, there would be a clear-cut reason for the
confidence to hold such a position. But demand from aluminium users is not robust, with
Page 37 5/6/2007
US orders falling and forecasts that production is set to outstrip supply in China this
year.
Still, cash aluminium prices were priced at more than a $80 premium to the benchmark
three month forward contract yesterday. The premium has narrowed from more than
$100 a tonne on Monday. The aluminium cash price tends to trade at a premium to the
forward price, when demand is strong and inventories are low. The last time aluminium
cash prices were trading at such a high premium to the three-month price, the London
Metal Exchange launched a probe into the aluminium market about “possible collusion
between market participants”. The investigation was launched in August 2003 but was
abandoned the following year after regulators failed to find any evidence of collusion.
However, this time the LME said it could see nothing untoward in the aluminium market.
The LME has the power to inspect the trading books of all its members, which are mainly
financial institutions, and their clients. But Robin Bhar, base metals strategist at UBS,
said: “Aluminium cannot be described as a tight market and based on current supply and
demand trends there is no need for cash prices to be trading at a premium.
“This suggests that the current premium has been financially engineered by a large fund
player, who maybe is acting alone or with other players.”
All eyes focus on aluminium (Reuters)
By Kevin Morrison
If you check our appendix, you will see that this kind of commentary goes on and on in
the press and in the statements of traders and analysts.
I described above the typical dynamics of a merchant squeeze play. I suggested that
today’s hedge funds in commodities may not be just investors, like pensions and
endowments, but players engaging in squeeze tactics on a more massive scale than in the
past. In the above quotes and others in our appendix we see a clear cut indication that
hedge funds are “financially engineering” and may be acting with others. Commentary
regarding the latter suggests another “possible collusion between market participants”
that was investigated years ago.
Such massive squeeze tactics would typically involve accumulations of physical metal
held off the market in order to make the squeeze in the futures market work. If such
hidden stock builds in aluminum are occurring, how big might they be?
Just look at the comments above of the visible positions of one or a few hedge funds.
There is roughly 800,000 tonnes of aluminum in the LME. That is no small amount. For
one or a few hedge funds to own almost all of the aluminum in the LME is to hold one
visible physical position worth billions of dollars. There are similar holdings in nearby
futures. Why not then assume that there are comparable holdings of invisible off warrant
metal to make the squeeze work?
Let us now look at aluminum’s fundamentals. Among base metals, aluminum has had a
higher secular growth rate over the last decade and a half – closer to 4% for aluminum as
Page 38 5/6/2007
opposed to 2.5% for copper. In addition, recently the price of aluminum has been one
third of copper. In the past it has been almost equal to the copper price. There has been a
two and a half fold rise in the price of copper versus aluminum. Changes in relative
prices induce substitution. As demonstrated above in the Nexans case, aluminum
production over the last two years may be higher than the past trend because of
substitution out of copper wire into aluminum wire.
Nonetheless, if hedge funds and merchants have built hidden stocks in aluminum, along
with other base metals, our apparent demand data for aluminum should also be inflated.
The statisticians tell us that aluminum demand rose perhaps 7% to 8% last year. Maybe
in reality it was 6%. The statisticians tell us that, based on apparent demand data, the
aluminum market was roughly in balance or in a small deficit in 2006. But adjusted for a
hidden stock build maybe it was in a small surplus.
What about this year?
Minmetals is a Chinese state entity with responsibilities in the metals sector. Star Futures
is an affiliate. Star Futures produced a report in which it projected an 11% increase this
year in global aluminum production. Minmetals Star Futures is not without some
knowledge in the sector, as half of the increment of production this year was estimated to
come from China’s own smelters.
Western agencies and most market participants have a somewhat more conservative view
of the outlook for aluminum production growth this year. Many are looking for a
production gain of 7% or 8%, with the market in a balance or a slight surplus. According
to the official statisticians, global aluminum production was up 10.5% year over year in
January, with Chinese production up 35%. According to most recent Chinese data,
Chinese aluminum production rose by 40% in February and 37% in March. So far the
data is saying Minmetals Star Futures may be more right than the Western consensus.
Whatever, the case of aluminum makes it clear that in some metals industries there are no
serious restraints on supply. This is not a market where supply is growing a little more or
a little less in its past trend, as it is in oil. It is an industry where, already, supply may be
growing at almost three times its past trend. Some new era of supply restraint!
The outlook for rapid aluminum production goes beyond this year, above all because of
investment in the sector by China.
Alumina is the primary cost component of aluminum. Last year, heavy investment in
alumina resulted in Chinese production growth of alumina of 50% or more year on year
starting at the middle of last year. This alone added an eight-percentage point increment
in global alumina production in 2006. Not surprisingly, the alumina price fell in 2006
from a peak of $640 a tonne to a low of $225 with only a partial recovery to $400 so far.
I have seen a compilation of alumina projects in China that are slated to go ahead. It
takes about two years to build an alumina plant and bring it on stream in China. If all
Page 39 5/6/2007
these projects were to go ahead, China would add 40% to global alumina supplies – a tenyear
increment at past global trend rates of growth, all in just a few years. I should add
that China is investing in large alumina facilities outside China as well, such as Vietnam.
And there are many new plants and expansions coming on stream elsewhere in the
world. The outlook is for a renewed large decline in the alumina price.
But at such an alumina price the spread between today’s high price of aluminum and the
total cost of its production, which includes alumina and energy, is very wide. This makes
investments in aluminum production very profitable. So there is a price signal for
Chinese provincial governments to build aluminum plants. According to the NDRC
investment in aluminum smelting was up 124% in China in the first few months of 2007
versus a year ago. Today’s leaps and bounds in Chinese aluminum production reflect a
combination of restarts and an investment in smelters a year or two or three ago when
such investment was only a fraction of its early 2007 level.
Given the signal of today’s aluminum price that has been inflated by hedge fund
speculation and manipulation, it would seem that Chinese aluminum production will keep
racing ahead at recent past rates, creating ever wider surpluses in the years to come. The
NDRC is alarmed. It sees the hedge fund basis for the current inflated price and the
inevitability of rapidly growing gluts based on China’s own investment actions.
Zinc and Lead
I will not go into great detail anymore. Let me be brief.
The zinc market in this cycle, with a six-fold increase in the zinc price, comparable to
that of copper, is a bona fide bubble by Jeremy Grantham’s criteria.
Are super cycle demands from China or New Era supply scarcities pushing the zinc price
to an unprecedented new high plateau?
From what I can tell zinc, lead and silver – all of which come from the same ores – are
about as scarce as dirt. Production of lead and zinc globally is now exploding. The great
irony is that China is the biggest source of the great supply surge.
China produces roughly 30% of the world’s lead and zinc. It has very favorable geology
for lead and zinc production. However, the industry has been very backward. Much of
the mine production has been done by small miners in inaccessible areas using quite
primitive procedures. In many cases, there has been no infrastructure, hence no vehicles.
Ore has been carried in baskets on the backs of mules. Processing plants have been
rudimentary; therefore, recoveries of metals from ores have been low. Exploration has
been minimal. One specialist who tours the lead and zinc mining regions of China
extensively has reported that, for the first time ever, one has encountered drilling rigs on
a Chinese lead and zinc mining site this past January. For someone from the mining
culture of the West this is incredible: to mine without advanced drilling is to walk blind.
Page 40 5/6/2007
All of a sudden, this is all changing. Small operations are being merged and rationalized.
Funds for investment are available. Improved recovery techniques are being employed.
Infrastructure is penetrating the far hinterlands, replacing mules with vehicles. I have
been told of one instance where the introduction of roads alone is resulting in a small
mine going from a negligible 1,000 tonnes a year to 45,000 tonnes a year – which is a
medium sized mine in the world of zinc – all in a mere year.
Therefore, it should not be a surprise that growth in China’s production of lead and zinc
are now far outpacing growth in their consumption. Last year, according to Chinese
statistical sources, zinc mine production rose by 16% year over year versus an average
10% growth rate in years past.
Therefore, it should not be surprising that China is suddenly becoming a major exporter
of zinc. The recent numbers on Chinese gross zinc exports are dramatic, as displayed in
the bar chart below.
CHR Metals, March 23, 2006
The same is happening to Chinese lead production, though it has yet to surface in
booming exports.
It looks like there is nothing to stop such rapid progress in this aspect of the Chinese
domestic mining industry. The Chinese authorities, burned by the explosion in the prices
of natural resources, are committed to investing huge amounts of money both at home
and abroad to develop and secure sources of supply of all commodities. The odds are
that Chinese production of lead and zinc will continue to rapidly outpace domestic
consumption, adding continuingly to global supplies.
Page 41 5/6/2007
According to the best supply work I have seen similar zinc/lead mine expansions are
occurring all around the globe. Serious analysts intent on creating accurate supply
schedules cannot keep up with the many smaller expansions and new mines in many
other places in the world. Zinc production may be on course for more than 10% growth
this year. If demand grows on trend, which is around 3%, a market balance or small
surplus (apparent demand and hidden stocks again?) last year would become a huge
surplus this year.
Based on the work of the analyst I believe is the world’s best on zinc, one cannot square
Chinese domestic zinc production growth with many indications of much lower domestic
Chinese consumption growth. There is an implied build of hidden stocks. According to
recent press reports there has been a rapid build of hidden stocks in Shanghai very
recently. Hedge fund shenanigans again, anyone?
The same can be said to some degree about silver, because more than half of all silver
production comes from lead and zinc ores. Over the decade prior to 2005 silver exhibited
one of the poorest demand trends among metals, with virtually no growth in demand over
the decade. At the same time, despite low metal prices in the late 1990s and early in this
decade, primary supply rose at 3% per annum. A large market deficit in the early 1990s
turned into a small surplus. Now, with the huge expansion of lead and zinc production,
along with other projects globally with silver bearing ores, that small surplus has
probably become a huge surplus.
Worse yet, some silver uses like silverware are very price elastic. In the past, owing to
huge above ground stocks of silver in various forms, silver scrap supply has been very
price elastic. We will not see again the very high scarp supply elasticity that helped bury
the Hunt’s almost three decades ago, but we should still see a considerable scrap surge in
response to price. These are two more reason why silver should be moving into a large
surplus.
If all these markets are going into surplus, where are those surpluses going? The odds
are, to the same or similar hedge funds who are “operating” in the base metals we delved
into in detail above.
We see this also in some minor metals. A case in point is palladium. This market is
apparently in surplus, as exchange stocks have risen from negligible levels to over a
million ounces in a few years (which is big for little palladium). Yet the palladium price
has more than doubled, reportedly because hedge funds have been willing to buy all these
visible stocks and no doubt invisible stocks as well.
Another case in point is uranium, whose price has gone up 15 fold in this cycle. Because
it is a strategic material individuals cannot directly own uranium. It must be owned via a
special facility. As a result, it is public information that a large share of all spot
transactions are absorbed by hedge funds, and it has been this incessant buying that has
skyed the uranium price.
Page 42 5/6/2007
And on and on. When it comes to metals, we see hedge fund speculation, hoarding and
squeezing everywhere. Not only have some metals markets been driven far, far higher
in this cycle compared to all past cycles; we see the same phenomenon across all
metals. It is the combination of both the amplitude and breadth of the metals bubble
that probably makes it the biggest speculation to the point of manipulation in the
history of commodities.
On Costs and Long Term Prices
Metals bulls often justify today’s heretofore unimaginably high prices by pointing to
recent higher production costs. As regards prevailing prices the thesis is absurd: though
capital costs and cash costs, both average and marginal, have risen a lot in this cycle,
prices are still massively above the total costs of the most marginal producer.
But the question arises, how great is the current metals bubble? Is it less than the
historical price experience would suggest because costs have risen by a great deal?
The answer is, short run costs have risen, but not long run costs.
In every metal cycle there are demands on capital goods and current cost items because,
over the short run, the supply of capital goods and inputs lag. Capital goods in mining
are not off the shelf items; for a while, during mining booms, demands for such
specialized equipment simply exceed the ability of capital goods producers to deliver.
The same is true of spare parts, key chemicals for processing, and the like. The same is
true of skilled labor. Typically, the demise of the prior boom results in the layoff of so
many geologists and engineers that people leave the business and the professional schools
wither. But as profits of suppliers and wages of skilled workers rise, new supply comes
on with a lag. When the metals cycle turns down, supply to the industry keeps roaring
ahead. Scarcity gives way to glut. The profits of capital goods and engineering firms
collapse. Equipment costs and consulting services fall in price. Skilled workers in the
industry are laid off again in droves. Those that hang on accept lower wages once again.
This pattern is perennial. It has happened in every cycle over the past decades. Most
striking, even though the period from the early 1970’s to the late 1990’s saw much
inflation in the general price indices, the nominal costs of the production of metals barely
rose. In effect, we saw a pattern that is centuries old, whereby real metals costs and
therefore prices fall in real terms on average by perhaps two percent a year.
We should not be surprised. Excluding the cost of energy, which might rise secularly in
real terms in this cycle, the cost of capital goods and the current inputs in mining are the
result of nominal wages plus productivity gains in mining. The latter have, on average,
been significant – perhaps somewhat higher than in manufacturing industries. This
combination of nominal labor costs and strong sector productivity has, through each
cycle, resulted in almost flat nominal and declining real costs. The rate of decline has not
been high like in tech goods where productivity gains are very large on average, but such
Page 43 5/6/2007
productivity gains have been higher than in some manufacturing industries where
productivity gains have been fairly hard to come by.
Today’s metal bulls present another argument for a higher cost base for metals in this
cycle – declining ore grades. There is no doubt that in many of today’s mature mines ore
grades are declining, and new deposits do not tend to have the high ore grades of the past.
But the latter is precisely the point. Ore grades have been declining since the days of the
Romans. Over the period from the early 1970’s to the end of the 1990’s, when base
metals prices barely rose in nominal terms and fell hugely in real terms, average ore
grades fell fairly dramatically. Yet, typical incremental productivity gains obviously
more than offset such ore grade declines.
One can argue that, in this cycle, we are seeing a secular rise in energy costs, which are
major for mining, and that will not be reversed this time. One can argue that productivity
gains will be less. On the other hand, there is less inflation in general prices indices in
the U.S. and elsewhere versus prior cycles. The above considerations argue that, as in
past cycles, the current rise in costs all up and down the cost curve presented below will
be reversed to some degree.
But that is not the real reason why costs in metals will fall far more than anyone now
expects. The real reason does not have to do with the level of the cost curve; it has to
do with the degree to which the cost curve shifts outward. And the degree to which the
demand curve fails to shift outward.
Throughout the history of the industrial world at any point in time there is a cost curve
for metals. That cost curve is rather flat in its infra marginal portion and then rises very
steeply as in the above chart. In other words, most mines have quite low cash operating
costs. But there are a few standby mines with very high cash costs which can be brought
into production when demand is strong and supplies are strained.
Page 44 5/6/2007
But even though marginal cost under conditions of strong demand is always way above
average cost owing to the steep nature of the far out or marginal end of the cost curve, the
real price of base metals has always fallen over time. The answer to why that happens,
despite the steeply rising far end of the cost curve, is simple: new deposits with low cash
costs are continually being found and developed. Their inclusion in the cost curve is in
the low inframarginal part of the curve. This pushes the entire cost curve outward. In
effect, over history such additional capacity has been pushing the cost curve out more
rapidly than the demand curve shifts out. The intersection of the two is therefore never in
the steep marginal part of the curve – except for brief periods – but is rather further into
the inframarginal portion where new mines keep cash costs low. As for the highest cost
marginal mines, when the cost curve shifts outward, they are simply dropped from the
cost curve, forgotten, consigned to the dust bin of history. Some of today’s inframarginal
mines then become the new marginal mines that constitute the steep high marginal end of
the ever shifting cost curve.
So the relevant cost of production that determines where metals prices will gravitate is
determined by the speed to which the global cost curve shifts outward.
What determines that rate of outward shift? Two things. First, the success of new
exploration, which finds hitherto unknown deposits with low cash operating costs.
Second, output prices and the availability of finance, which bring high capital cost/low
cash cost projects which are “on the shelf” into production, thereby pushing out the cost
curve.
When people consider technological change and its role in reducing mining costs, they
focus on the impact of such change on production processes. But it is important as well
to focus on technological change in the exploration process. Modern exploration
techniques are able to see “beneath the earth’s surface” in a way miners could not do in
the past. Technology has resulted in a high pace of new exploration success whenever
money has been spent. This has added to the availability of low cash cost projects, even
though the globe has been girdled and the “easy to see” surface deposits have been
picked over.
When metals prices are low little is spent on exploration. Little new is found. This is
what happened when metals prices collapsed late in the 1990’s. When metals prices are
sky-high exploration budgets tend to explode. Which is what is happening now.
Exploration finds new low cash cost projects. This eventually shifts outward the supply
schedule over long periods of time. This happened repeatedly over the period from the
early 1970’s to late 1990’s. Exploration success had a lot to do with the containment of
nominal cash operating costs and nominal marginal cost despite significant inflation in
general price indices over those decades.
But there is my second point – which is more relevant now. There is always a large
inventory of base metals projects that have been discovered and delineated. Their
implementation is simply a matter of economics. Typically, in metals, projects have very
Page 45 5/6/2007
high capital costs and very low cash operating costs. The hurdle to their development is
coming up with the capital needed to get them into production. Once the capital costs are
sunk you do not close down such mines until they lose money on a cash operating cost
basis. Therefore, very high metals prices, high profits and available finance can rapidly
add to the low inframarginal portion of the cost curve and thereby shift outward the curve
and marginal cost.
I have said earlier, we have never seen such a huge increase in base metals prices relative
to past historical trends and such a persistence of super high prices. Under these
conditions projects which have been on the shelf because capital costs were too high
suddenly become super economic. Many projects which heretofore could never yield an
adequate return to the initial capital outlay now have paybacks of one year or less. The
totally anomalous high amplitude of this cycle’s rise in metals prices is giving the go
ahead to many high capital cost/low cash operating cost projects. This is why most
compilations of future increments to mine production show a growth rate in capacity that
is far higher than the growth rates in the past. As these projects come on stream the cost
curve exclusive of capital costs and comprised solely of cash operating costs gets shifted
outward faster than in the past. If the demand curve shifts at a lesser pace that is more
consistent with history, the intersection of the two curves or marginal cost can actually
fall even if there is inflation in cash operating costs.
Worse yet if sustained super high prices also lead to demand destruction. Then the
demand curve shifts out more slowly relative to the past, while the supply schedule is
shifting out more rapidly. Again, the implication is a lower intersection of the two
curves; that is lower marginal costs.
Investor “Revulsion” From Metals I:
Strategic Diversification Into Commodities Overall Is Self Destructing
Part of the investment and speculation in commodities in this cycle comes from strategic
investments in commodity derivatives by pensions, endowments, etc. and part comes
from speculation “and more” by hedge funds. Will the former, which have helped inflate
metals prices, persist?
There are several tenets to the recent strategic allocations to commodities in long term
institutional portfolios.
1.) You cannot time markets, so you should make a strategic allocation
independently of whether an asset class looks cheap or dear.
2.) Your objective is to maximize the risk adjusted return where risk is measured
in terms of volatility.
3.) You can reduce your volatility and raise your risk adjusted returns by
investing in commodities future baskets because commodity futures baskets
yield almost as much as bonds and stocks but are inversely correlated, thereby
reducing overall volatility.
Page 46 5/6/2007
There is an important point that must be stressed. Strategic allocations to commodities
only increase risk-adjusted returns if the return to commodities is reasonably high. One
can accept the lower return to commodities relative to the returns to stocks and bonds, but
it cannot be that much lower, since the reduction in volatility from inclusion of this
inversely correlated asset is limited. In effect, the foregoing of return must be less than
the gain from reduced volatility.
Why, traditionally, have pension and endowments not included commodities in portfolios
as diversifiers? Because traditionally they thought about diversifying with the physical
“stuff” itself. And if you diversify with “stuff” you give up too much return to make the
diversification work.
What is the return to “stuff”? Historically commodity prices have fallen in real terms. In
other words they have lagged inflation in the general price indices. Exactly how much
they have fallen in real terms and exactly what their nominal return has been depends
upon the time period and the commodity basket you choose. Over the last several
decades, but prior to the last two years of this “super boom” in some commodity prices, a
good set of assumptions is that the general price level has inflated by about 3% per
annum and commodity prices have risen by about 1.5% per annum.
In fact, for large portfolio investors the return to “stuff” has been a little less. Unlike
stocks and bonds, holding commodities involves certain non-negligible storage and
insurance costs. More importantly, physical commodity markets are surprisingly small.
The total value of all inventories of commodities globally has typically been only a few
percent of the value of the total outstandings of bonds or stocks. And most of these
physical commodity stocks have been working stocks that cannot be readily purchased.
This means that buying and selling large amounts of commodities (appropriate to the
needs of global pension and endowment portfolios) will disturb prices and involve
significant transaction costs.
Given the above considerations it is likely that the net nominal return to portfolios from
investing in physical “stuff” has not been more than 1% per annum. By contrast, in a 3%
inflation environment, bonds have yielded somewhere between 5% and 9% and equities
have yielded somewhere between 8% and 11%. In effect, you gave up an immense
amount of yield if you diversified out of bonds and stocks into commodities. You did
gain by reducing overall portfolio volatility, but that gain was not large enough to offset
the loss in yield. Diversifying with “stuff” did not enhance risk-adjusted returns.
Enter the innovation of the commodity futures basket.
In the early 1990’s Frank Russell Associates and Goldman Sachs pioneered the idea of
using commodities futures baskets rather than “stuff” as a strategic diversifier.
Commodity futures baskets exhibited the same inverse correlation as the physical
commodities themselves. But the yield or nominal return to commodity futures baskets
was perhaps 6 percentage points higher in a 3% inflation environment.
Page 47 5/6/2007
Here is how it works. Instead of investing a certain amount – say 100 – in “stuff” you
purchase commodity futures with a nominal or face value of 100. To do this you only
need put up a margin of 10 in a non-interest bearing account. The remainder of the funds
you hold on deposit and earn the prevailing rate of interest, which, in a 3% inflation
environment, might be 5%. In doing so, this interest income adds 450 basis points to
your overall nominal return.
In addition, from the perspective of the early 1990’s looking backward, you earned an
additional return from rolling your commodity futures forward. The annualized
backwardation on a commodity basket like the GSCI was perhaps 1% or 2%. Rolling
your commodity futures forward to the next liquid contract as first notice day approached
provided an additional nominal roll yield of perhaps 1.5%.
The combination of the interest on your cash balance and the “roll yield” increased the
total nominal return to holding a commodity futures basket by 600 basis points above the
return to mere “stuff”. If “stuff” yielded 1% your commodity futures basket might yield
7%. That total yield is almost as much as the return to bonds and not far below the return
to stocks. With this new commodity futures strategy the strategic investor was giving up
very little in the way of yield to achieve reduced volatility. The diversification with
commodities worked: the risk adjusted return to the overall portfolio would be higher.
The problem with this strategy is simple. The commodity markets are too small relative
to pension and endowment and other portfolios to accommodate it. As commodity
futures baskets became popular as a strategic diversifier, the purchase of these baskets
and similar purchases of commodity derivatives by hedge funds pushed up futures prices
relative to spot prices. First, these commodities futures baskets went from their historical
backwardation to a small contango. Then the contango became very large. In energy, we
went into supercontangos, which became far greater than the cost of carry. Very large
contangos began to emerge in grains and other soft commodities. Large backwardations,
which prevailed in industrial metals, began to disappear. Depending on your basket,
these contangos in some cases are 15% annually or more.
With such high contangos the roll yield is now disastrously negative. Let us go back to
the above example. We must now replace a positive roll yield of 150 basis points with
perhaps a negative roll yield of 1500 basis points. Our previously calculated positive
nominal return to a commodity basket of 7% now becomes a negative return of 9.5%. If
diversification using “stuff” with a positive nominal return of 1% did not work,
diversification with a commodity’s futures basket with a deeply negative return will
surely not work.
Investment bankers have tried to change the weights in their commodity baskets away
from supercontango markets to backwardation markets in order to reduce this negative
roll yield. But most of the few commodities that remain that do not have large contangos
are small markets. Therefore, not much can be done in this regard, since any significant
such reallocation will throw these few remaining commodity futures into full contangos
or supercontangos.
Page 48 5/6/2007
But it is actually worse than this. The tsunami of buying by the “pinstripe investor” in
commodities has raised commodity prices above their marginal costs (which is their long
run equilibrium). Over time, it is inevitable that demand will be rationed and supply will
be encouraged. Surpluses in commodity markets tend to increase contangos in their
futures markets. That means that, over time, the presence of the “pinstripe investor” will
ensure full contangos or supercontangos everywhere. There is no getting around this: the
buying pressure of the “pinstripe investor” has destroyed and will continue to destroy the
return assumptions that led him to commodity futures baskets in the first place.
“Pinstripe investors” of all kinds and their investment bankers who sell them commodity
futures baskets are backward looking. The “pinstripe investors” are backward looking
because that is their orientation. Investment bankers may not be backward looking; they
may recognize that the current supercontango and its implied negative roll yield is here to
stay. But they are salespeople. However, in time, it will be realized by everyone
involved that the buying pressure of the “pinstripe investor” in commodities has changed
dramatically the returns to commodity futures baskets. And, in doing so, has destroyed
the diversification case for such strategic allocations.
Some commodity baskets now have deeply negative returns. At today’s high commodity
prices this can only get worse. No strategic investors with a basic menu of high yielding
stocks and bonds can continue to hold commodity baskets with negative returns.
Therefore, over time, much of the strategic investments that “Pinstripe Investors” have
made in commodity baskets will be liquidated. I am hearing that some strategic investors
are now liquidating, though new commitments may be offsetting. Such flows may now
be close to zero on a net basis. Eventually they will be negative on a net basis.
Only when this liquidation is largely over will the prohibitive contangos go away and the
diversification rationale for investing in commodities become valid again.
Once this process is well under way “pinstripe investors” in commodities may feel
foolish. They may feel they have been persuaded by investment bankers who sold them
deeply negative return instruments even though they realized their inherent return had
turned very negative. This could create a climate of “revulsion” which could accelerate
such liquidation. Such revulsion will apply to all the commodities included in these
baskets. That will include all the metals. This is the one of two “revulsions” that the
metals sector will experience in time.
The Risk of “Revulsion” II:
Revulsion From A Hamanaka on a Massive Scale.
The metals bubble is far worse than the commodity bubble. What I have described so far
is a base metal sector that is out of all touch with reality. We see price distortions relative
to underlying fundamentals the likes of which have never happened before.
Page 49 5/6/2007
The reason is investment and speculative demands. Strategic investment demands have
been a contributor, but a lesser one. Speculative hedge fund demands have been
overwhelming, and across the entire metals complex.
And in this cycle such speculation has gone to the point of manipulation.
This is an old story. It should be a familiar one. It happened a decade ago with the
Hamanaka incident in only one metal – copper. As illustrated in the preface to this
metals section, just recently a new class action lawsuit is scheduled to go to trial against
Morgan Bank for financing Sumitomo (Hamanaka) and the fraud, manipulation, and
collusion involved. You would think the world would have remembered. Yet, despite
the published commentary we have found about these metals being out of touch with
reality because the LME is a paradise of speculation and manipulation there is little such
commentary among mainstream analysts and consultants.
After the Hamanaka incident unfolded Paul Krugman discussed this age old process of
speculation to the point of manipulation, which I have described above.
“So far so good. But a long time ago somebody--I wouldn't be surprised if it were a
Phoenician tin merchant in the first millennium B.C.--realized that a clever man with
sufficiently deep pockets could basically hold such a market up for ransom. The details
are often mind-numbingly complex, but the principle is simple. Buy up a large part of the
supply of whatever commodity you are trying to corner--it doesn't really matter whether
you actually take claim to the stuff itself or buy up "futures," which are nothing but
promises to deliver the stuff on a specified date--then deliberately keep some--not all--of
what you have bought off the market, to sell later. What you have now done, if you have
pulled it off, is created an artificial shortage that sends prices soaring, allowing you to
make big profits on the stuff you do sell. You may be obliged to take some loss on the
supplies you have withheld from the market, selling them later at lower prices, but if you
do it right, this loss will be far smaller than your gain from higher current prices.
It's nice work if you can get it; there are only three important hitches. First, you must be
able to operate on a sufficiently large scale. Second, the strategy only works if not too
many people realize what is going on--otherwise nobody will sell to you in the first place
unless you offer a price so high that the game no longer pays. Third, this kind of thing is,
for obvious reasons, quite illegal. (The first Phoenician who tried it probably got very
rich; the second got sacrificed to Moloch.)”
Why is this incident not in the forefront of everyone’s minds? Why, after the Hamanaka
incident, which sent some to jail, some fleeing outside the reach of the law, and a new
court case now at hand, have the supposedly self-policing exchanges and their regulators
not been active?
Without going into details, I can assure you the exchanges know what is going on. They
know who the dominant longs are that everyone talks about. They know of exchange
metal movements that involve non-commercials. They probably know all that Jinchuan
or Nexans or the NDRC know. They probably know what the traders and analysts cited
Page 50 5/6/2007
above know and what is behind their comments about the markets’ speculations and
squeezes.
And so do their regulators. The exchanges pass information to the regulators. Others
have complained to the regulators and with evidence.
Paul Krugman says that such activities are quite illegal. I am not sure. I have been told
by the U.S. regulators that squeezes are illegal in the U.S., but hard to prove. I am told
collusion is illegal, but hard to prove. I am not sure in London, under the rules of the
FSA and the LME, such operations are illegal. But I believe that the FSA recognizes
there can be undesirable market “abuses”.
So, we may ask, as the exchanges and the regulators know what is going on, why is
nothing done? The answer is the Anglo-Saxon approach to regulation. The principal
metals markets are in the U.K. and U.S. In one conversation I have had with U.S.
regulators we both agreed that the current policy of regulation in Anglo-Saxon
commodity markets is to let “the fox guard the hen house”.
Paul Krugman made this point with respect to the Hamanaka incident a decade ago. “The
funny thing about the Sumitomo affair is that if you ignore the exotic trimmings--the
Japanese names, the Chinese connection--it's a story right out of the robber-baron era,
the days of Jay Gould and Jim Fisk. There has been a worldwide rush to deregulate
financial markets, to bring back the good old days of the 19th century when investors
were free to make money however they saw fit. Maybe the Sumitomo affair will remind
us that not all the profitable things unfettered investors can do with their money are
socially productive; maybe it will even remind us why we regulated financial markets in
the first place.”
It is possible that the regulators in the United States may finally be concerned about what
the fox guarding the hen house is doing to the hens. In a recent FT article entitled,
Futures enforcer issues warning, we are told Regulators need more funds to guard
against fraud and manipulation. April 20, 2007. This article suggests that such concerns
of the CFTC have to do with energy and perhaps not with metals.
This is understandable. Prior to the Amaranth incident consumer groups got wind of a
possible hedge fund manipulation of the price of natural gas. They took their complaints
to the attorney general offices of several states, as well as to the U.S. Congress. They got
a sympathetic hearing, at least at the state level, but the states realized they had neither
the financial resources nor the expertise to follow through. And they also found they got
no support from the CFTC.
But after the Amaranth debacle these allegations about hedge fund manipulation of
natural gas, which is politically sensitive because it is the home heating fuel of American
consumers, came before the Congress again. Hearings are now being held to see if there
was manipulation and why was the CFTC asleep at the switch. Hence, the emergence of
concern by the CFTC’s chief enforcer.
Page 51 5/6/2007
But it is not clear whether his concerns about fraud and manipulation extend to the metals
markets. Prior to the Amaranth debacle, there was very little comment in the public
domain about manipulation in natural gas. From my perspective, there was little smoke
that might have indicated a “fire”. By contrast, in the public domain there is now huge
talk about possible manipulation in metals. So it would be surprising to see so much
publicity over CFTC enforcement’s concerns about fraud and manipulation but no
attention to metals at all.
As for Britain’s FSA and the LME, the assessment of Jinchuan and Nexans of the
behavior of these UK agencies suggests we may never see the modicum of concern than
may now be surfacing in the U.S. – at least until the inevitable unwind of the current
shenanigans in LME metals occurs and brings with it its consequences and revelations.
Perhaps there is now so much “smoke” the U.S. and U.K. regulators are now scared of
looking for fire. Look at it this way. In the Hamanaka incident, only one metal – copper
– was involved. There were only two colluding parties, Sumitomo and another. And the
price of copper was only pushed to $1.46, close to, but not even, at the top of its prior 3
decade trading range.
In this case, there may be no manipulation and collusion. Or maybe there is but it cannot
be proven. But if the many commentaries in the public domain have merit, the scope of
the manipulation and collusion could be vastly greater than in the Hamanaka incident.
Many metals are involved. Many more than two players may be working together. And
the price distortions are hugely greater. Metals like copper, nickel and zinc are not at
multi decade highs – they are several times multi decade highs.
If there is manipulation and collusion on such a scale, what would be the eventual
consequences in terms of legal actions. They could vastly surpass in aggregate claims
those that followed the Hamanaka incident.
Would the U.S. or U.K. regulators, having let things get to such a point, have the resolve
to investigate and act. What would be the impact on metals prices if they uncovered what
the many public commentaries suggest? What would be the impact on the exchanges if
there were such revelations?
It is not hard to conclude that, if the many publicly voiced claims and suspicions have
merit, that the regulators are now afraid to act because of the consequences they might
provoke.
Does this mean that, if the regulators leave things alone, all will be well? Absolutely not.
Such absence of regulation allows excesses that lead to outcomes that become socially
intolerable. In the U.S., cheered on by the likes of Allen Greenspan, mortgage lenders
lent to people who could not pay on terms that ensured massive future defaults. Two
years ago the regulators began to see what would happen and called for stricter standards.
But, typical of the laissez faire Anglo-Saxon New Financial Architecture, private lobby
Page 52 5/6/2007
groups in finance and real estate objected and the regulators did nothing. The result is a
housing bubble that is now bursting and a mortgage finance crisis which has become the
focus of the ire of the public and the congress. And the mess may be only in its early
stages.
Conclusion On Metals Revulsion
In the end, no matter what are the games of speculators in base metals, the fundamentals
will prevail. Surpluses will become too great to hoard and hide. Prices will fall to and
below a marginal cost that will prove to be little different than in past cycles.
Institutional investors that have got caught up in the bubble in various ways will suffer
huge losses. As with the tech bubble, all the Wall Street cheerleaders of the bubble, with
their crazy New Era fundamentals, will be discredited. There may be Amaranth type
crises which disgust institutional investors in the speculating hedge fund sector. And, as
has happened again and again in the past, the regulators who sat idly by will be accused
and attacked.
When bubbles burst the asset class always becomes profoundly discredited. There is
always revulsion. But when it turns out that there is the skullduggery typical of so many
of the big bubbles in the past, when the regulators in charge of maintaining the rules of
fair play have been seen to fail in their role, the revulsion is far more profound.
In my judgment the odds are very, very high that this will all transpire. Investment funds
will eventually be taken out of the metals sector with savagery. The reversal of the fund
flows responsible for the commodity bubble described in the prior section will be more
complete in the metals sector. This reversal of the tsunami of fund flows will deepen
further any price declines to marginal cost or below that will be made inevitable by the
fundamentals.
So today’s record price distortions will generate a fundamental response: record
surpluses that will cause mean reversion, overshooting the mean to the downside, and a
very long period of low prices to force mine closure and absorb the inventory overhang
of record surpluses. And this “nuclear winter” will be deepened by revulsion that causes
reversal of the speculative fund flows that have fed, in myriad ways, the metals bubble.