Thursday, March 26, 2009

Complexity and uncertainty in the global agricultural commodity matrix

In a time of great uncertainty, one thing is certain: 2008 will go down as a year of extremes.

On the surface, the year started off so strong in so many sectors before experiencing the virtual free-fall that left no industry untouched. As we near the end of the first quarter in 2009, the downward trend has continued, and uncertainty is escalating. Agricultural commodities were not spared from this trend reversal, with many contracts going from record highs to low prices not seen in years, in a span that covered a few months. With fiscal pressures mounting, coupled with decreased confidence in the robustness of the global financial system, it follows that growth in sectors supported by non-essential items will suffer the most. However, unlike many other sectors of the global economies of goods and services, food and energy are two commodities that society can not do without. Globally, the population living a middle class lifestyle continues to increase, and along with this rise and increased purchasing power comes the demand for more and better sources of food and protein. So if demand for food staples such as grains, sugar and oilseeds remains strong, why then are prices sinking with the rest of the market?

This addresses some of the factors that I spend quite a bit of my time thinking about, and firmly believe that these issues will significantly influence the global commodity complex in 2009 and 2010 in ags/energy as well as the ancillary markets, with particular emphasis on the supply side of the S&D equation. Taking agriculture, supply side disruption factors, including weather, will affect agricultural production in the major origins, and the goal for analysts at this stage in the year should be to establish a comprehensive view of global crop potential at the starting point of the food supply chain. This, however, does not mean that the next 18 months need to be viewed with a fixed view of the global supply forecast. Forecasts for crop sizes, yields, volumes, acreage, etc., are always, to some degree, going to be wrong. Even with perfect information at the start of a season, there will surely be something to move the market tomorrow, which was not anticipated by the majority of market participants today. That notion can be taken one step further. It is acknowledged that there are always going to be unforeseen events that will cause our idea of what we think happen in a market at the outset to be different from reality (see the WTI white paper on Climate Black Swans). While many trades are executed as the result of the output of an algorithm, at the end of the day, it is still a human driving the bus. There is no model which can accurately account for the spectrum of human emotions that move markets for many different reasons. Putting this in a macro perspective, Taleb has argued that the entire framework that measures financial risk, employed by nearly every bank and desk around the world, is flawed (a viewpoint that this author agrees with) and should therefore not be trusted. While those in futures analysis and commodity trading need to always be mindful of the technical signals, they should also be wary of a strategy that is not nimble enough to quickly change direction, even if the models, or the supporting team of PhD quants, are telling them otherwise. So, if as stated above, a forecast is always wrong, why attempt to construct a balance sheet? The intent of this piece is not to dismiss long range outlooks or forecasts in the traditional sense; instead, it should frame future discussions surrounding the larger issues that can affect global agricultural commodities over the next two crop years. Crop outlooks are part of this complex, but the broad range of possibilities should be included, rather than a fixed point forecast. Some, most, or maybe even all of the long range issues identified may take place. The point then is to be as prepared as possible, because just as a cold snap in the midst of a warming trend can move markets in unexpected ways, climate black swans will occur, and when they do, one certainty is market volatility.

So the 2009/10 crop year coincides with a critical juncture for agricultural commodities. Pressures ranging from food demand to foreign exchange to biofuel potential will all help shape the next 18 months in corn, wheat, soybeans and also sugar, coffee and cocoa. So it is important to understand the spectrum of potential issues now. Some of the events might not make a dent in futures behavior for 9-12 months. But if a proactive desk is at least thinking about the potential issues and taking a forward view on the suite of potential scenarios (both directions), if and when the event triggers, the result should be positive financial performance. We live in a time where the economic stability of any given country is more closely related to the extent of foreign trade relationships than it has been during any other time in recent history; as such, the role of science and technology in developing and fostering these relationships, for better or worse, has become increasingly important. One scientific tool used more and more by the agricultural community is the employment of longer range weather and crop yield models to assess agricultural potential ahead of the majority of market participants. Using weather outlooks that assess the potential weather a couple seasons ahead, for example, allows for the identification of where and when there may be the highest potential for crop problems. As stated above, this is not saying that using long range guidance will or will not allow us to definitively identify or avoid a pending crop issue, but it does put more detailed metrics around where the likelihood is surrounding potential crop problems. This approach is much better than either of the commonly accepted measures for estimating crop potential today: (1) assuming ‘normal’ weather and associated crop response, or (2) extending the trend line for another year, thus believing the coming year will see conditions that are similar to those seen in the prior year. Both of these methods can be right, but more often than not, they will miss the pattern and significantly miscalculate the risk to the underlying commodity.

Early warning systems and crop projections that focus on the assessing the long range potential for food production/security are necessary from an economic, environmental and most of all a humanitarian perspective. The uncertainty that accompanies the projections should not be hidden, and serious discussion of the full range of possibilities needs to be part of the dialogue. Future posts will address these ideas in more detail, separately and in relation to each other.

No comments: