Copper: Enough already or not enough?

The title of this article is intended to address two questions related to the copper market — questions that reflect two alternative views: Have prices spiked to unsustainable levels? Or are stocks so low that prices shall still be forced higher?

This time last year, we provided our outlook for copper, and as such, we thought it might be interesting to look back at our forecast to see how things have developed.

In October 2003, prices had risen by 17% during the year to US$1,883 per tonne as at Oct. 10, and the major question for the market was whether the upward trend could be sustained. Our answer, and frankly that of most market participants at the time, was a definite yes. Low interest rates and a weaker U.S. dollar were encouraging interest from macro-oriented investors who saw copper as a “reflation trade.” Improving copper order books and falling inventories were encouraging speculative interest. Chinese copper imports were surging higher, and an already tight concentrate market was suddenly having to contend with news that Grasberg, the world’s second-largest mine, had suffered a major landslide and would lose several months of production.

Putting all this together, we forecast that 2004 would see a copper market deficit of less than 100,000 tonnes and that copper prices would peak at US$1,950 per tonne, which we quickly revised higher to US$2,400. At the time, that was one of the more bullish forecasts in the market. A year on, it has proved far too conservative, with the copper market now looking at a 2004 deficit of at least 600,000 tonnes and the 3-month price averaging US$2,730 per tonne to date. Perhaps this is why we commodity analysts try not to review our past projections.

In retrospect, commodity analysts had little chance of giving accurate price projections last October. In fact, LME Week 2003 can now been seen as the pivotal period for base metals prices. Three-month copper prices surged from US$1,883 per tonne on the Friday close before the week to US$1,955 per tonne by the next Friday, US$2,300 per tonne by the end of the year, and continuing to its 9-year peak of US$3,055 per tonne in March.

Prices hovered at these heights until April 2004, when signs of rising U.S. interest rates, credit restrictions in China, and a firmer U.S. dollar encouraged mass long liquidation from macro-oriented investors. By May, copper prices had corrected back to US$2,500 per tonne.

It was back in April, most notably during Cesco week in Santiago, that we began to outline our view that copper prices were likely to experience a “twin peak.” The view was that the retracement in prices from the first peak was driven largely by the investment community — macro funds, commodity-specific hedge funds, and CTAs. However, we thought we would see a second price peak, driven primarily by strong physical fundamentals, following the end of the summer slowdown. After some grave concerns during the summer, this has proved one of our better calls. Prices are now in the midst of a second price peak, and prices are challenging the earlier peaks. The question now is: Where will the second peak be?

A year on, the environment is much more complex. Demand is still positive but slowing. The concentrates market is still tight, but mine output is rising. And inventories are at record lows, meaning that even modest differences, in view in supply and demand, can have a huge impact on one’s opinion on price and even on price direction.

As a result, this year’s LME Week (Oct. 11-15) will most likely include many heated discussions on the key challenges facing the copper market in the year ahead. In this article, we’d like to set the tone for those discussions and provide our views on what we think will be the key issues for copper.

The following are the key concerns regarding copper in the year ahead:

— Low inventories

— What about prices when growth is slowing?

— Can production meet the deficit?

Low inventories

Copper inventories are approaching historical lows, with reported warehouse stocks literally disappearing. There is now less than 100,000 tonnes of copper available in LME warehouses. At the average rate of drawdown so far this year, this could be gone by mid-November. There have been a few sizable but sporadic inward deliveries — most notably in Asia at Singapore and South Korea, in response to a widening backwardation. This has left the market nervous over potential large quantities of unreported material being placed on warrant. However, with today’s high premiums and freight rates, an even wider backwardation for a sustained period is required for any excess material to emerge. At present, the backwardation is especially large for nearby dates (October). It will be crucial to see if this attracts material into warehouses. If not, a further sharp price spike looks likely.

The Western World market is seriously short of copper. This is clear from high physical premiums. And in line with sharp drawdowns of LME stocks, there are only 40,000 tonnes left in Comex warehouses.

And in China, the Shanghai Futures Exchange has pretty much already run out of copper. In a market that consumes 3.5 million tonnes a year, it holds only 20,000 tonnes of cathode! Copper stocks in bonded warehouses that are non-customs-cleared have also been rapidly brought down. We believe they are close to zero at present. Also, selling by the China State Reserve Bureau has dried up, and the bureau is reportedly keen to refill its strategic stockpile of copper.

The result of all this is that the stock-to-consumption ratio has fallen to only about four weeks on a total reported basis, including estimated inventories at producers and consumers. This is critically low, and we think neither demand nor supply is likely to adjust enough to refill empty warehouses in the near future. These low inventories will maintain upward pressure on copper prices.

In the current environment of low inventories, the copper price is extremely sensitive to changes in both supply and demand; only small differences in supply and demand assumptions can have large implications on one’s view on the copper price. That is why there are so many differing views on the outlook for copper prices in the market.

Looking at consensus price forecasts, there is now a huge split between bulls and bears. According to Reuters’ latest analyst poll, for example, price forecasts for 2005 show the widest divergence ever. And perhaps such a divergence is not so surprising given the historically high level of current prices. Set against low inventories, the outlook for prices is extremely vulnerable to different economic outlooks.

What about prices when growth is slowing?

So, low stockpiles are currently a key feature to the copper market. The second hot topic for discussions during this week will be if the copper price can rise even more during a period of slowing growth.

There is no doubt economic growth has slowed recently. The traditional use of leading economic indicators would suggest it is time to take profits in industrial metals. Moreover, market participants are watching with concern as car sales slow and motor companies announce output cuts, semi-conductor sales slow, and the housing markets ease. We conclude that demand for copper is heading down. But even though demand growth is slowing, growth remains positive (which is significant for a deficit market with already critically low inventories).

In fact, from a demand perspective, we would argue that this cycle is different. Why is that? It is because fixed capital spending has been the major driver of U.S. demand fixed capital investment growth in this rally, rather than consumer spending. And this is partly why this is not a traditional price rally.

The flood of investment capital into the information technology sector from 1996 until 2001 caused a drought in investment in the unglamorous “old economy” sector. The subsequent collapse in capital expenditure following the bubble further worsened this drought. The net result is a widespread condition of under-investment in the “old economy.
“

Rising metal prices are partly attributable to this long phase of under-investment in the industrial (as opposed to information) economy. And this shortfall needs to be addressed if the global economy is to continue growing. This requirement is rendered all the more compelling by the secular demand shock from the Chinese and Indian industrialization. And as evident from the rise in metals prices, the global economy is completely unprepared for an increase in demand on the scale of that experienced in China over the past couple of years.

In light of this, the non-residential fixed investment component of U.S. gross domestic product rose an impressive 12.5% in the second quarter. This is almost China-style growth! We appreciate this is extraordinarily strong, however, and these growth rates are unlikely to be sustained.

So what about Chinese demand? De-stocking was a key feature of the market in the second quarter, but it appears to have been relatively short-lived. The underlying trend in copper consumption remains strong. Even though it has slowed in line with fixed-asset investment and industrial production, copper semis production continues to register double-digit growth. It’s up about 26% so far this year. However, continued strong economic activity is probably the largest threat to Chinese metals consumption; it means credits could be further tightened and threatens to slow down the economy abruptly.

Copper would obviously not be immune to such a slowdown. However, we attribute a low probability to a sharp slowdown in Chinese industrial activity over the foreseeable future. And second, a large amount of copper consumed in China goes into power transmission — about 40%, we estimate. Output of power-generating equipment is up sharply so far this year, by more than 100%. And this is a sector in which investment is still being aggressively encouraged, with a recent announcement of plans to spend a massive US$560-billion on power generation and transmission by 2010. On past consumption patterns, this will require around half a million tonnes of copper. But this is only one driver of rising Chinese copper demand. In addition to this, living standards are rising fast, along with urbanization, and there are ever-growing infrastructure requirements and demand for consumer durables. All of which we believe should sustain a strong underlying trend rate in copper demand growth.

There is compelling evidence that the Chinese copper buyer has now returned after a period of de-stocking this summer. Physical premiums and Shanghai prices are on the rise again. In line with low stocks, as discussed earlier, metal availability has definitely tightened. Codelco’s recent announcement that contract premiums for next year’s deliveries are at record highs of US$115 per tonne provides further support for a strong market outlook. Large requirements from China and insufficient domestic production will continue to encourage large quantities of imports. We expect Chinese refined imports to increase again through the fourth quarter.

So, against this background of strong demand, let’s look at the prospect of supply rising sufficiently fast to bring the market back to balance as soon as next year. This will be the third key issue for copper over the coming year.

No question, there has been a strong reaction in production to these high prices. Mine output has risen sharply in recent months, and concentrates availability has improved and pushed treatment and refining charges rapidly higher.

We estimate that about 200,000-300,000 tonnes of high-cost mine production have been reactivated this year. However, the recent sharp rise in mine output must be compared with the very depressed levels of last year, when the world’s largest mines, Escondida and Grasberg, were operating well below capacity. These operations are now ramping up satisfactorily. Scrap availability has also improved, as can be expected in a high price environment. And we think an even sharper price spike during the first quarter of this year was partly prevented from improved scrap supply.

In any event, we believe at least part of the recent strong growth in production is a temporary phenomenon. There is still a distinct lack of large new projects coming on-stream in coming years. In this strong price environment, it is remarkable that only two new mines have started production this year. This is entirely in line with the low capex into new capacity in recent years. Having said that, there are naturally a number of expansions at existing operations scheduled to come on-stream next year. These are often underestimated.

Production will continue to grow, as long as there are no major unforeseen production problems. There have been many labour disputes this year, which have also added to the upside pressure on prices. A large number of labour disputes should in fact not be too surprising. There is a clear relationship between copper prices and labour costs. Labour unions are watching higher copper prices carefully in order to argue for sizable increases in wages. This also provides an example of how the trend in real production costs is pressured higher.

But a major talking point at the moment is smelter capacity. With rising concentrates output, some are talking about a smelter bottleneck. However, we do not agree. The market has been worried about insufficient smelter capacity many times in the past, and they have never happened. Today, global smelter utilization rates have moved up above the average of the past 15 years but are still not even 80%. In other words, there will be enough smelter capacity to treat rising concentrates production, not the least in India and China, and especially now, when treatment charges are more attractive.

Smelter stocks are now being replenished. It will take some time for the new concentrates to give relief to a tight refined market. And we expect the global refined market will remain in deficit next year. Assuming global demand growth of 4.8% next year, total supply would have to grow about 9% to remove this year’s expected deficit of about 600,000 tonnes. We think refined supply is likely to rise about 7%, keeping a deficit of about 340,000 tonnes.

Copper prices have already exceeded not just the March 2004 peak but also the peak in January 1995. Given the current momentum, low inventory levels, underexposed consumers, and under-invested speculators, we believe the scene is set for a challenge of the peaks seen in 1988-1989. In December 1988, cash copper prices averaged nearly US$3,500 tonnes; a spike to such levels is possible.

However, we then expect copper prices to correct back from such peak levels, as cyclically positive factors ease back in strength. But, we are strong believers that copper prices will remain above historical averages, buoyed by the structural phenomenon of below-average growth in copper mine output following a decade of under-investment, set against above-average trend growth fuelled by ongoing industrialization and urbanization in China and other emerging markets.

This phenomenon requires substantial investment in new copper production capacity, and the location for such growth appears likely to entail higher geopolitical risks than in the past. This will require, in our view, a move up in the long-term average copper price — probably at least US$1 per lb. (US$2,200 per tonne), if not higher.

— The preceding is an edited transcript of a presentation given by the author the London Metal Exchange Metals Seminar in London, U.K., earlier this month. The opinions presented are Ms. Sternby’s and do not necessarily represent those of the Barclays group. For access to all of Barclays’ economic, foreign-exchange and fixed-income research, go to the web site at barclayscapital.com. Queries may be submitted to the author at ingrid.sternby@barcap.com