Wednesday, September 28, 2011


By the nature of our profession, we are naturally obsessed by what we see daily on our screens. However, in the last several years I have become fascinated by what we do NOT see there. With so much risk capital now captured by the South and the East, unfortunately for us, the opacity of the dealings will only increase. There are three, interrelated reasons for this:
1. East of Istanbul, some 75% of all deals are done in the private market
2. The experience of 2008 led to a breach of trust and much higher dependence on stocks, to the detriment of flows
3. As a result of this obsession of physical inventory control, coupled with low interest rates, contango has evolved in certain commodities, most durably in some metals

Yet, this realization is relatively new. Some three years I traveled from Charlotte NC, where I worked for a multistrat hedge fund, to New York to listen to a number of UBS strategists and analysts. I do not remember the exact date, but this must have taken place sometime between May 2008 when the commodity equities turned sharply down in London and July 2008 when Hank Paulson’s memorable speech reversed long commodity/short financials trade.

The UBS strategist was quite flippant in his dismissal of the weight represented by the emerging markets. He complained about the pressure on him to learn Mandarin and reassured everyone that the proverbial fatso from Costco would have to save the planet as he or she is ultimately irreplaceable. The argument went like this: America with 300m population consumes 11.5 tr worth of goods and services, whereas Europe with twice as many people consumes half of that figure. Japan is a distant third with $4tr worth of consumption and China, well, not in our lifetime shall we see it carrying the weight of responsibility for global growth.

But even if that UBS strategist spoke with a British accent, his view of global economy was strongly embedded in the “consumption” school of thought. In this view, consumption, or end-demand, is what propels this planet around the sun. But then, the end of the world appeared on our Bloomberg screens. Very slowly did we begin to understand that while consumption does matter, consumption on borrowed dime would eventually end in tears.

So what happened since those momentous events three years ago? Closer to home, we dusted off the Austrians. Today angry Congressmen spit out one-liners from Henry Hazlitt’s adorable paperback and Michelle Bachman carries with her a volume penned by von Mieses on a holiday break. And the rest of us? Well, we are all a notch closer to Schumpeter’s view that the way to expand the economic wealth of a nation is not to:
• use fiscal policy and liquidity injections in order to "stimulate demand"
• incur huge fiscal and monetary cost to defend holders of bad debt
• risk long-term slowdown by obstructing the restructuring of excessive debt burdens

No, from a macro perspective, living standards can be increased only in two, but diametrically different ways:
• through capital consumption, i.e. dis-saving and massive borrowing, something we avidly tried (and succeeded) during the Clinton-Bush era, or
• through capital accumulation, i.e. savings and capital investment, a model arduously pursued today by many emerging markets.

We all know that the demand for capital dropped among the developed economies, with FAI/GDP falling from 25% to 18%. At the same time, it has grown in the emerging markets. In India it stands at 33% to GDP and in China 47% and STILL growing at 25% yoy. The question remains, of course, how “productive” (from the Schumpeterian perspective) the growth in the East and the South has been. I believe that the picture there is somewhat mixed. The quality of growth in emerging markets has been uneven and with rising current account deficits and rocketing credit intensity of GDP growth (quadrupling fourthfold in China) may have even deteriorated in the last three years,. But the persistent growth differential between “us and them” is undeniable and should correspondingly affect our thinking about the commodity space across asset classes: the physical, the futures and even the equity.


Commodity professionals, we can give themselves credit that, at least since 2003-04, they have been paying a lot of attention to China. Indeed, I am sometimes amused when I hear about China being “over-dependent” on exports. It was back in 2003 that the leadership of the Chinese Communist Party (CCP) began to question a development model based on exports and on co-opting private entrepreneurs into the Party Congress, a process that had flourished in the late 1990s. 2003-04 was also the time when a decade of Zhu Rongji’s financial reforms had already borne fruit, China had become a WTO member and was in a build-up mode for the Olympic coming out party. In order to bolster the position of the state-owned enterprises, a shift from exports to fixed asset investment became a priority.

A decision was taken that the "pillar" industries, including energy, mining, steel manufacturing, automotive industry and telecom would be dominated by companies under government control. Granted, private enterprise was there to stay, but tolerated only as providers of jobs, foreign exchange and technology. In December 2003, PBOC established Central SAFE investments (Huijin), which later helped recapitalize state-owned banks using foreign exchange reserves, thus entrenching a growth model based on state-owned banks offering loans to SOEs and the same banks coming back to the market every couple of years to recapitalize themselves. The ensuing lending binge unleashed a tsunami of construction activity which durably transformed many commodity markets. In response, Sir Bob Wilson, upon his retirement as Executive Chairman of Rio Tinto, cautioned against commodity markets’ excessive dependence on China.

But dependent we became. The observation of overdependence has been particularly glaring in the commodities that China is net short: metallurgical coal, copper, iron ore, oil, platinum group of metals, soybeans.

We are often victims of our own wishful thinking, expecting the world to become somewhat more Anglo-Saxon, more free-market-ish and more democratic in general. Meanwhile, despite the trappings of (our) modernity, the Chinese Communist Party’s model still has a lot more in common with the Soviet system than with ours. Unlike the pre-Gorbachev Soviet system, which tried hard (and eventually failed) to control the flows and volumes of products, components, subcomponents and commodities throughout the value chain, CCP devised a system in which it controls not the widgets, but the supply of economic factors, and consequently their price. As most of us still remember from the economics class, there are three basic economic factors: land, capital and labor. This is how CCP achieves this in a state-controlled economy with Chinese characteristics:
• land supply control, through state-ownership and land registry at municipal level (I will return to this later),
• capital control, which remains underpriced through capital account controls and distributed through lending quotas,
• labor supply control, with the two-tier population registry (hukou) system: the locals and the migrants and the central guidelines for provincial decisions regarding the minimum wage.

This system may be bursting at its seams now, but it is still firmly under CCP’s control, with the possible exception of underground lending.

Now, if you control all the factors, then regardless of your personal make-up (which may be highly technocratic and pragmatic, rather than ideological), you and your organization will develop a Leninist control culture around it and will be pretty upset if there are ANY inputs whose supply you do NOT control.

The iron ore market is a case in point. China Iron Ore and Steel Association (CISA) is a bureaucratic entity mandated to protect the interests of the country’s steelmakers. If you meet CISA’s officials in Beijing and overcome their initial “we won’t tell you ‘cos you’re a foreigner” (我不告诉你,因为你是外国人) you will soon learn that Western governments, (yes, “governments”) have devised a perverse plot to deprive China of access to iron ore resources and thus to strangle its birthright growth rate.

Much has been said about iron ore market’s high Herfindahl index of supply concentration. However, Sir Bob Wilson was not entirely wrong. Today as much as 25% of global seaborne iron ore of about $1bnt p.a. is destined for just one market: long steel products used in Chinese construction. And when asked about how sustainable this is, most analysts point to the continued “urbanization”.

Contrary to the claims of such reductionist commentators, urbanization itself is NOT a market phenomenon, but a process driven by a confluence of institutional decisions in China, whose sustainability depends largely on the durability of the system based on three pillars:
1. the Constitution of PRC, which turned all land into the property of the State
2. the lopsided structure of provincial budgets, responsible for 77% of expenditure but entitled barely to 46% of federal tax transfers, which turns these budgets dependent on land transfers. As much as 70% of the provincial budget relies on land transfers (leases below cost, sales tax, spec sales).
3. the capacity of local officials to single-handedly transform the value of land by re-allocating land from “rural” (whereby rural land can only accrue value from 30-year leases) to “industrial” or “residential” (where 70-year ‘lease’ is possible).

Urban Development Investment Corporations or UDICs (城市发展投资公司) and other many similar entities (6600 of them nationwide) have been structured to bypass the inability of municipalities to sell bonds directly. The existing products often showcase a glaring mismatch between maturity and revenue generation. However, UDIC bonds (and infrastructure loans) do not have principal payments until years after the communist officials in charge of the province are long gone. Meanwhile, the interest payments are being satisfied with land sale proceeds. Today, the system is saddled with an estimated $RMB11tr (or nearly $2tr) of debt, representing some 42% of China’s GDP. The system’s cheerleaders tend to profess high level of comfort with this number (given the value of assets potentially offsetting this burden), conveniently forgetting that under conditions of uncertainty, the duration of a financial institution’s liabilities shortens and the duration of its assets lengthens.

This process of forced urbanization, embedded in this institutional framework does not affect commodity markets in which China functions as a significant primary producer, e.g. zinc with its global surplus of 256kt this year. But it certainly does matter for iron ore, where the problem is compounded by the dropping exports from India, as illustrated by the recent political and legal wrangle in Karnataka and Orissa. China’s iron ore reserves are of low quality and despite $1bn annual exploration spending, the aggregate reserve depletion is among the fastest on the planet. Looking at the data from US Geological Survey, the 60mt per month importer may run out of domestic sources of iron ore within 9 years.

This is a linear, finite view of the emerging market commodity phenomenon, but cyclicality and seasonality are of equal importance.


Since most humans evolve in a climate characterized by regular changes – four seasons in temperate climates, dry and wet seasons in the tropics, near permanent darkness and midnight sun in the Arctic – most of us also come to expect some form of recurrent patterns. Some traditions have even injected such hopes into religious thinking, thus avoiding the eschatological destiny of much of the Western heritage. By the virtue of climate, tradition AND the related credit subcycles, seasonality imposed by the emerging markets has begun to trump such well-respected recurrent references as the US driving season, European winter heating season, US hurricane season or sometimes even northern hemisphere corn planting season. In fact, as we could observe in the last several years, the industrial metal demand dances around the Chinese credit cycle.

The Chinese credit market is directed with an annual, rather than Japanese fiscal or Australian calendar. As the state owned banks have, by government fiat, guaranteed 3% spread loan/savings rate ratio on some $2 trillion worth of savings, they are keen to open their loan books as early as possible after the beginning of the calendar year and earn maximum interest within the official quotas permitted by the regulators for that particular year. Much of that credit goes to builders, contractors and manufacturers. But a lot of this lands with legal, quasi-legal and illegal underground lending system, starting with the commonly tolerated Minjian Jiedai 民间借贷 – or “civil borrowing”, through mutual assistance societies Huzhuhui (互助会), all the way to subterranean loan sharks Gaolidai (高利贷), who would later use this liquidity throughout the year at interest rates we can only remember from Vito Corleone movies.

As a result, the builders are in a position to contract new projects and the commodity import machine is set in motion, with the concomitant impact on the seasonality in China’s current account. Every year, depending on when exactly the Spring Festival falls, sometime between February and April it becomes fashionable for a wave of Western analysts (and some politicians) to express a collective sigh of relief that “Chinese surplus is shrinking and the problem of undervalued renminbi will soon go away”. Consequently, it would be highly improper, impertinent and uncivil of us to dub the Chinese government as a “currency manipulator”. And so it goes. China continues to intervene in forex markets at $1.4bn a day, and Chinese trade surplus is indeed shrinking (from $295bn in 2008 to $183bn last year), but not necessarily in terms of bilateral exchanges with the US as our soybeans exports (even coupled this year with our corn exports) prove insufficient to quench Chinese thirst for inputs, much of which remains very seasonal.


The law of structural dynamics means that every self-reinforcing loop will eventually encounter sufficient constraints to slow down the process. Such constraints are already present in the Chinese economy. Robert Mundell was right. If you do not want to realign your prices in relative terms through exchange rate, you will sooner or later pay for this with real price adjustment. This is exactly what happened in China with 87% increase in M2 over the last three years. It is now frequently quoted that this is a country with a third of US GDP and a monetary mass 30% higher than the US. I find it intriguing to compare China of today not with the US of today, but with the Nixon era. The 1970s show how long the lag could be between the M2 avalanche and the onset of inflation.

However, too many of us get carried away by this obsession with inflation and commodities. Not even gold, a financial product par excellence, is perfectly correlated with inflation. Others argue that the value of gold is simply a mirror image of the trade-weighted dollar, but this conveniently dismisses the fact that most other currencies are also losing their purchasing power, not in terms of CPI-related indexes, but in terms of their capacity to acquire assets. As real buying of gold occurs also outside of the USD currency zones, gold represents a useful yardstick of value for all of those currencies, not just the dollar.

No, where gold truly reacts against the extremes, it is in terms of how we connect future and present prices. And we do it via interest rates, in real terms. Gold perfoms well when the expectations become entrenched that real yields will crawl in the gutter for a while. The main exception is the period 1994-95, marked by a Greenspan interest rate hike that panicked the bond market.


Let us move now to the impact on the futures market. Here too, the institutional framework in the emerging markets, and particularly in China is actively shaping the global marketplace.

A critical juncture came in 2008, when the ultra-capital intensive system designed by the Chinese Communist Party and described above was severely tested and then further enhanced. Half a decade into the fixed asset investment binge that made China a linchpin for all the commodity markets with the exception of oil, the 800 pound gorilla trader was still sitting on a 19th trading infrastructure. As China slowly integrated into the world economy, the letter of credit (LOC) system remained key element in its dealings with overseas trade partners. The growth was extraordinary. By 2007, almost 70% of China’s exports were financed with LOCs. But on the import side, so vital for supplying China with raw materials, letters of credit were practically the only avenue for trade financing. It is understandable (at least for sinologists) that pre-payment schemes are not very “Chinese” culturally speaking, but Beijing failed to develop alternatives to the LOC system, which are necessary to keep the trade flowing during periods of credit stress. Factoring could have been one way to deal with the issue. It is astounding that in an economy of 1.3 billion people, the Chinese government issued only two import factoring licenses over the 30 years of reforms.

As many remember, the sudden collapse of China’s LOC system in late 2008 led to a catastrophic slump in most commodity prices. One day I could be sitting with a CFO of a mid-size Australian iron ore company, who professed his confidence in the future, and the next day his panamax vessels were floating idly in Southwest Pacific in search of a willing customer. What did this lesson mean for the buyers? Do not trust the flows, trust only the stocks. So a quarter of a century after the Just in Time system thinned out the value chains, upstream commodity business went in the opposite direction as if we were all in perpetual preparation for a war or, at the very least, for a collision with a large asteroid. China’s Strategic Reserve Bureau began its frenetic buying spree: becoming the world's top importer of copper, soybeans, iron ore, cotton and natural rubber and among the largest buyers of coal, vegetable oil, sugar and potash. Importantly, this hoarding behavior coincided with record low interest rates globally, and it was no different in China.

Two years into the great credit-lubricated party, inflation expectations had become so pervasive that multiple increases in reserve requirement ratios proved insufficient to cool the economy. In October 2010 China officially entered the “tightening mode”. On that day, commodities blipped, but nobody remembers that. By the second hike in December 2010, the commodity market shrugged it off completely. Although some argue that any increase in domestic interest rate exposes its central bank to sterilization losses, the PBOC is not really an independent policy bank. Rather, it is a ministry subjected to the decisions of the State Council, where so-called “stability” is paramount and the commercial considerations secondary. Yet this is a ‘stability’ where food price inflation is in double digits and rents in Beijing shoot up 100% yoy. Someone got scared and the screw had to be tightened.

For the rest of us, the story that has unfolded since then is one of copper.

In January this year a friend announced to me: “this year I am going to short gold and go long copper”. I asked ‘why would you do that?’. Well, “gold is a bubble, but copper has real demand, Chinese really buy it”. I asked: do you know what they do with this when they buy it? He was not interested: “supply and demand”.

In fact, all we usually know about demand in China is the so-called “apparent demand”, i.e. production plus net imports plus/minus changes in Shanghai exchange stocks. When in 2009, Chinese net imports rocketed from 1.360 million to 3.112 million tonnes, it was hard not to see the hidden hand of hoarders: legitimate commercial restocking, strategic moves by SRB, and speculative stockpiling. Conversely, strong end-use and low apparent demand in 2010 pointed to destocking.

By comparison, what has happened since the tightening of interest rates at the end of 2010? Since then, Chinese copper importers have become… bankers. Unfortunately for my buddy the investor, they did not import copper to consume it. And the difference matters. They exploited the system to obtain – and provide - renminbi.

With new official loans in the first five months of this year falling 12% to 3.55 trillion renminbi ($549 billion), it was only a matter of time till the loophole was filled with typically Chinese ingenuity. Already in 2010, letters of credit issued by Hong Kong-listed Chinese banks jumped 70 percent, faster than the nation’s overall trade growth of about 25 percent.

It is useful to review the cat and mouse game between the importers and the regulators and to see what the impact on copper market has been, especially in Asia.

Companies could apply for a 90-day or a six-month dollar-denominated letter of credit to import refined copper, but not really to consume it. The only price was a requirement to put down 20 percent of the value of the imports as a deposit to the bank for the LOCs.

First, copper was imported into China by speculative investors for resale in China. They would sell the metal domestically once it arrived and lend out the earned (renminbi) cash at higher (unofficial) rates before repaying the letter of credit at maturity.

However, because Chinese copper prices have maintained discounts (up to nearly $200/t) to the London Metal Exchange prices since 2010, it forced the speculators to put contracted imports into bonded warehouses, most of which are located in Shanghai area. By March 2011 this shadow stockpile reached a reported 1mt. From here on, there were two ways to obtain renminbi. Investors could either:
1. use the bonded inventory to borrow collateralized loans at 90% or even 100% of the stored material and most frequently re-lend this capital at higher interest rates, or
2. re-export the copper.

Why would re-exports be of interest to investors? First, remember that in an economy with the closed capital account, exporters have to exchange their dollar inflows into renminbi. Secondly, following some successful lobbying by importers, the regulators scrapped the rule whereby these stocks had to pay 17% VAT in order to be re-exported.

However, last April, the Chinese authorities tightened rules on repatriating foreign currency from re-exports. Now, the Chinese firms were required to leave such foreign currency earnings in pending accounts and were not allowed to convert them into renminbi until they received receipts of import payments and re-export incomes. Financial intermediaries also had to cut their advance payments from foreign importers and delayed payments to exporters to 20 percent of the total foreign exchange they sold or bought over the past year. That naturally affected all those who carried several outstanding letters of credit at the same time.
But the demand for renminbi credit continued to outstrip the official lending quotas and the cat-and-mouse game continued. Enter offshore renminbi. To much fanfare, some 20 months ago Beijing had opened up Hong Kong as an offshore venue for investors willing to gain access to renminbi-denominated assets, allowing foreign companies and banks to raise funds in Chinese currency for cross-border trade and investment. Some 67,000 Chinese companies were allowed to participate in the offshore renminbi business. Now, investors trying to resell their bonded copper overseas began to ask foreign buyers to settle trades in the “offshore” renminbi (CNH), remitted from offshore banks in Hong Kong. This summer, however, PBOC moved to tighten the screws on offshore trading and notice No. 145 stressed that the onshore market had not been liberalized and the capital account remained closed. Widespread CNH selling has intensified since and continues to date (late September 2011).

More importantly for the copper market, it soon became possible to deliver the material from the bonded warehouses against Shanghai Futures Exchange contracts. Previously, imports were subject to VAT payments in advance of the delivery against exchange contracts. There are two consequences of this change. First, it removed one hurdle to trading the arbitrage between the Shanghai bourse and the London Metal Exchange. Secondly, in case backwardation appeared in Shanghai, you could now deliver physical against it. For futures traders it could be a way for the shorts to cover their exposure, instead of buying back shorts. The outcome would be reduced volatility.

What does this opening mean for the copper curve? A copper-collateralized loan costs an importer 6 to 8 percent, but the renminbi obtained this way is then re-lent in the “underground” market at rates 3x that much! What is the significance of this? It’s an alchemist’s dream. We turn copper into gold.

One of the reasons why the cost of borrowing base metals, such as copper were generally higher than prevailing interest rates (i.e. µ<λ) was that unlike gold, they usually responded strongly to physical supply-and demand fundamentals. When demand was strong, inventories were depleted and the metal came at premium, putting an even greater upward pressure on prices. As the metal prices rose, the lease rates rose as well in response to the scarcity of the metal. Similarly, when the demand was weak, inventories would build up, making the metal more abundant and pushing prices lower. Not surprisingly, high lease rates were associated with high spot prices and low lease rates with low prices.

However, as long as µ>λ there is an implied rate of return (interest rate less lending rate) on holding the metal. Gold usually provided such an implied rate of return, but base metals did not. Global gold inventory does not matter because the metal is plentiful. In fact, this is how Indian jewelers finance their gold inventory, through gold lease rates, thus avoiding currency risk. And even though gold lease rates generally have a negative correlation with spot prices, the calculated base rates do not ultimately matter.

Gold may have been in contango since Lucy left Olduvai Gorge, but in the case of Chinese copper, if 民间借贷 or 互助会 interest rates remain above the LOC interest rates, then we end up with gold-like characteristics:

Not surprisingly, contango appeared in Shanghai despite all we know about the tightness of copper concentrate, labor activism in Chile, winters in Atacama desert, strikes at Grasberg, low TC/RCs and high merchant premia. This contango would be here to stay, if the futures market in Shanghai was not so jittery about the maze of potential regulatory changes concerning:
• LOCs deposit, interest rates and associated currency exchanges
• offshore renminbi repatriation
• bonded warehouse taxing

Indeed, State Administration of Foreign Exchange has introduced rules to make it harder to use the metal as collateral and in late August People’s Bank of China required banks to place a part of the original collateral held against LOC in low yielding reserve accounts, instead of using it to make further loans. That means that it is going to become more expensive to issue the LOC.

Interestingly enough, the dearth of renminbi credit led not only Chinese investors and merchants to use dollar-denominated LOCs. Even Chinese producers of semi-finished copper products have begun to use letters of credit to purchase USD-priced bonded copper in Shanghai, rather than in the CNY-denominated spot market, for which they do not have cash. Bonded metal in Shanghai can be delivered to buyers in two days after stock owners pay the VAT. Banks in Jiangsu and Zhejiang have been advised not to issue LOCs to purchase bonded copper from Shanghai warehouses, arguing that these only applied to imports and not to the metal already stored in the country. But local banks would continue to provide credits if their bonded copper purchases are resold in the domestic market, rather than re-exported.

Assuming the symmetry in terms of CNY/USD preference, the above pyramid build on copper (or soybeans, or any other) collateral could potentially crumble the moment capital flees towards USD, as it is commonly the case during the episodes of global liquidity stress. This could explain some of the vicious Chinese selling of metals (including copper) throughout most of last week. But more ominous signs could be just around the corner. There are reports of letters of credit refused for imports. The sensitivity of Chinese trade to global credit woes has not diminished.


It is time we abandoned the dream that global markets, including financial markets, will somehow lead to homogenization of the world exchanges. Capital and goods move freely precisely BECAUSE of differences between localities, their institutions and their cultures. It’s Asia’s fears that drive its demand for gold, its quest for liquidity and fresh credit or the taste for the control of the physical inventory.

At a conference in Oxford 10 days ago someone was stunned that a decade of global growth delivered a zero return in the US stock market. I find this tunnel vision baffling. It is irrational to expect that Asians would put their wealth in pension funds which, in turn, will efficiently allocate capital to underpriced US stocks. What do we know about their economic behavior, obsession with land ownership, real asset control, inter-generational wealth transfers, seasonal consumption patterns and underground lending structures indicates that commodities, thanks to their fungibility, still remain the best bet we have in the public markets to participate somehow in the waves triggered by China’s and India’s fears.