Sugar Stock Levels: Myth and Reality?

The global sugar market is facing significant stress. A combination of long run events, such as the reform of Europe’s sugar regime, the destabilising impact of the credit crunch on Brazil’s heavily-leveraged sugar industry and shorter term weather events are testing the market’s limits on a daily basis. Though the fundamental reasons for higher sugar prices remain firmly intact, the rise in prices is driving an ongoing re-evaluation of assumptions in particular stock levels. Our view is that prices are not only reflecting the bullish statistical balance, but also much lower stock levels than many analysts would recognise. This also indicates that sugar prices have historically discounted supply chain risks in a way that now appears set to change.

The issue of sugar stocks is normally seen as an extremely dry topic and one best left to analysts to debate in private! However, it has always caused contention. One of the peculiar factors about the sugar balance sheet has been the high levels of aggregate stocks that are recorded by some well-regarded analysts, which are in contrast to commercial trends. During the 1990s a focus on cost reduction and just-in-time supply systems drove businesses to scale back on stock levels in order to increase efficiency. This efficiency drive also resulted in a greater correlation between trends in local markets and the world market as replacements costs flow through the supply chain. Though time lags exist, we can observe this price relationship in action today.

Based upon this price behaviour our view is that some estimates of aggregate stocks have become disconnected from market reality as a result of the inherent difficulties that official agencies face when verifying commodity data. This is especially true for consumption where there is a natural tendency to underestimate demand. While the detail is not that impactive in the short-term it becomes a problem in the longer-term as aggregate stock levels become inflated. As a result published stock estimates today range from between 30m mtrv to 70m mtrv, while our analysis indicates that global stock levels are around 20m mtrv.

The Problem with the Data
At the individual level it is very difficult to quantify the exact volume of sugar consumed as sugar is available in an enormous variety of forms and products. In addition not all sugar movements are correctly reported as a result of variations in national price levels encouraging illicit trade. For analysts seeking to measure consumption data, these problems can be addressed within individual crop years and for this reason we, and other analysts, make an allowance for ‘unrecorded disappearance’ when looking at annual consumption data. However, as estimates of total global stocks are based upon the aggregation of this annual data, the errors that are inherent within annual consumption data end up being compounded.

In sugar, stock levels are calculated by taking the difference between annual production and consumption estimates. At Czarnikow, we have always taken the view that the most prudent way to interpret this data is to look at annual changes in stock in isolation and we have consciously taken a decision not to aggregate this data. However, given the interest in using stock to use ratios to assess commodity trends, there has been a tendency for analysts to aggregate these annual differences. The aggregation of this data has consequently led to a focus on ‘top-down’ estimates of global stocks. And invariably these models indicate higher levels of sugar stocks than can be supported by a ‘bottom-up’ based analysis of individual markets.

The Real Cost of Stocks
All physical trading businesses are acutely aware of the cost and market risk implications of holding stocks. In virtually all markets, stock management is carried out by producers and traders, with government intervention and support rarely available. As carrying stock is a business cost, efforts have been focused on ensuring minimum levels of carry-over stocks. Minimal stock levels have become a necessity given forgone revenue implications and the difficulty of marketing ‘old crop’ sugars to buyers.

For trade houses holding physical stocks there was also a market incentive to minimize levels. Prior to the growth in passive commodity index investment products, which have affected the structure of the futures board, the futures market was typically in backwardation. This backwardated structure was a function of the concentration of demand in the spot position with the deferred contracts dominated by producer hedging. And the implication of this structure for physical traders was that stock levels should be kept to a minimum as the cost of rolling hedges forward would result in physical sugar being sold at a loss!

In summary, business practice and market signals have discouraged both traders and producers from carrying forward large physical positions, enforcing the long term trend to low commercial stocks.

Estimating Stocks
An additional problem in looking at stocks is the difference in methodology used by different analysts. Czarnikow data, for example, is based upon the aggregation of individual national crop years for producing countries while other analysis create theoretical crop years (say October – September) and then fit the production data to this period. As a consequence stock levels are not reported at the low point of each individual country’s balance sheet, which we view as the critical data point for measuring commercial supply risk. With stock levels today becoming much more topical we have created a bottom-up analysis of aggregate closing stocks on this basis, which in our view is a better reflection of reality.

Where are the Stocks?
On a bottom-up basis we estimate that global stock levels are around 20m mtrv taking individual country closing balances for the 08/09 season. The five countries holding the largest volume of stocks are China, India, Brazil, Europe and the US, which in total account for 65% of all stocks. However, aside from Europe and China, which are carrying surplus stock, most markets are running short of sugar. This is even the case in the top five stock holding countries, as both India and the US need to increase imports in 2010 to prevent demand overrunning supply.

Russian prices have rallied due to low stocks and higher replacement costs
 

Looking around the world today it is clear that stock levels are vulnerable in most markets. During the past fortnight Russian domestic prices have rallied by around $65 / tonne as the local market adjusts to a much tighter supply balance and ending commercial stocks potentially falling to 350,000 tonnes*, which is less than one month’s supply. Prices have also been rising sharply in importing countries in the Middle East with prices in Iran now above $700 / tonne as stocks have been drawn down as buyers have struggled to come to market.

Of the larger stock holders, Indian stock levels are today particularly vulnerable, which is being reflected in the continued rise in local prices. Despite a strong raw sugar import programme this year, lower production and disappointing white sugar import demand seem set to result in stock levels falling below the minimum level of 3 months supply, which is regarded as a benchmark to prevent disruption. At the current rate of demand India requires around 6m tonnes to meet this benchmark while we estimate that stocks levels could be as low as 2.2m tonnes during the Diwali festival in mid-October, prior to new season sugars becoming available. To the north, stock levels in Pakistan are also low after the PSMA persuaded the government to halt its delayed import programme earlier this year on the grounds that high prices had already curtailed demand. With local prices continuing to strengthen indications are that stock levels are falling to minimal levels and could potential force Pakistan to resume imports prior to new crop sugar becoming available.

India’s import needs are reflected in rising domestic prices

While there might be a temptation to view the reduction in sugar stocks as an issue facing developing markets, where funding costs are high, it is also clear that levels are very vulnerable in the US and Mexico, which under NAFTA have combined to create the world’s largest sweetener market. Prices in both markets have been rising as a result of tight supply and with US stock-to-use ratios falling to 10% this year (and in theory becoming negative in 2009/10) it is apparent that there is no possibility of meeting current levels of domestic demand without a rise in imports. However, with the production balance in deficit once again in 2009/10 it is impossible to see how this additional import demand can be met without some forced adjustment of consumption.

Though global sugar prices have risen sharply this year there is still little sign of higher prices discouraging demand. While import demand has been affected in some instances it is not clear whether this has been a simple reduction in the supply pipeline, which in part reflects the tighter credit environment, or a response from consumers. However, most indications are that actual consumer demand has been robust. In Russia, for example, which is the market where demand is most seen at risk, the recent rally in prices is an indication that consumption has actually been stronger than traders had feared. Tight commercial stocks have meant that prices have had to move higher in many markets to beyond levels that consumers have been used to. However, sugar still remains an affordable necessity for most consumers while in developed markets the price is largely irrelevant.

As a consequence, though prices in some countries are now lagging the strength seen in the global futures market it is just a matter of time before values once again converge as tight stock levels force price levels higher.

Conclusion
The rise in local prices that we have seen this year enforces our belief that global stock levels are lower than the market has generally accepted. We consider that this is a result of the inherent difficulties in measuring consumption and commercial trends towards minimum stocks levels. Due to the potential for error that exists in aggregating annual stock changes we have intentionally avoided running a global stock balance structured in this way. An additional complexity is the difference in methodology used by different analysts which has meant that stock levels have not been reported at the low point of each individual county’s balance sheet, which we would view as the critical data point for measuring supply risk.

In order to provide a clearer view of the global stock balance we have therefore considered a bottom-up analysis of individual country closing stocks, which is where supply is most stressed. On this basis we estimate that global stocks are around 20m mtrv with the real stock-to-use ratio around 12%. With the production balance forecast to be in deficit during 2009/10 the sugar market is now facing the risk of supply being insufficient to meet current consumption levels.

© All information is copyright Czarnikow Group Limited 2009