Tag Archives: demand and supply

Commodity Prices — Who’s Holding the Cards?

Economists have too many hands. On the one hand, they may declare something good. On the other hand, they may say, “Well, not so much.” Some of them may have even a third or fourth hand. My ex-boss, an economist himself, once remarked that he wished he could chop off some of these hands.

In the last couple of months, I plunged right into an ocean of economist hands as I sat down to do a minor research into this troubling phenomenon of skyrocketing food and commodity prices.

The first “hand” pointed out that the demand for food (and energy and commodities in general) has surged due to the increase in the population and changing consumption patterns in the emerging giants of Asia. The well-known demand and supply paradigm explains the price surge, it would seem. Is it as simple as that?

On the other hand, more and more food crops are being diverted into bio-fuel production. Is the bio-fuel demand the root cause? Bio-fuels are attractive because of the astronomical crude oil prices, which drive up the prices of everything. Is the recent OPEC windfall driving the price increases? What about the food subsidies in wealthy nations that skew the market in their favour?

Supply Side Difficulties

When explaining the food prices, one economic opinion puts the blame squarely on the supply side. It points an unwavering finger at the poor weather in food producing countries, and the panic measures imposed on the supply chain, such as export bans and smaller scale hoarding, that drive up the prices.

Looking at the bigger picture, let’s study oil as a proxy commodity and study its dynamics. Because of its effect on the rest of the economy, oil is indeed a good proxy.

In the case of oil, the dearth on the supply side is more structural, it is argued. The production capacity has stagnated over the last thirty years or so [1]. No infrastructural improvements have been made after the seventies. Indeed, new methodological improvements are expensive for all the easy methods have been fully exploited; all the low-hanging fruits have been picked, as it were.

The harder-to-reach “fruits” include deep sea explorations, crude oil from sand and, somewhat more tenuously, bio-fuels. The economic viability of these sources of oil depends on the oil price. Oil from sand, for instance, has an operating cost in the range of $20 to $25, as Shell’s CFO, Peter Voser is quoted as stating [2]. At $100 a barrel, oil from sand clearly becomes an economically viable source. Bio-fuels also are viable at high oil prices.

The huge investments involved in exploiting these new sources and their unpredictable economic viability exert strong upward pressure on oil prices, purely from the supply side, regardless of the demand situation. Once you invest a huge amount banking on a sustained high oil price, and then find that the oil market has softened below your viability level, you have to write off the investment, forcing losses and consequent price hikes.

With the high level of oil prices, investments are moving into infrastructure enhancements that will eventually ease the supply side crunch. But these fixes are slow in coming and are not going to ease the current dearth for about a decade. In other words, the high prices are here to stay. At least, so say the economists subscribing this supply side explanation of things.

Demand Spike

Although I personally find it hard to believe, people assure me that the exponential demand explosion in the emerging economies was completely unforeseen. My friend from a leading investment bank (who used to head their hybrids desk) told me that there was no way they could have anticipated this level of demand. I should probably shelve my scepticism and believe those in the know.

One thing I do know from personal experience is that the dynamics of a demand crash is different in emerging economies for a variety of reasons. First of all, identical movements in fuel prices have different impact in the overall spending pattern depending on the proportion they represent in the purchasing power of an average consumer. A 30% increase in the pump price, for instance, might mean a 5% reduction in the purchasing power to a US consumer, while it might mean 20% reduction for an Indian customer.

Besides, retail fuel prices in India are regulated and supported by government subsidies. Subsidies act as levies delaying the impact crude oil price movements. But when the crude oil prices rise beyond a certain point, the subsidies become untenable and the retail fuel prices surge upward, ushering in instant demand crash.

I came across another view of the skyrocketing oil prices in terms of the Middle-Eastern and American politics. The view was that the Saudi oil capacity is going to increase by about 10% soon and the prices will drop dramatically in the first quarter of 2009. It was argued that the drop will come as boost to the new American president, and the whole show is timed and stage-managed with a clear political motivation.

Speculation

All these different opinions make my head spin. In my untrained view, I always suspected that the speculation in commodities market might be the primary factor driving the prices up. I felt vindicated in my suspicions when I read a recent US senate testimony where a well-known hedge fund manager, Michael Masters [3], shed light on the financial labyrinth of futures transactions and regulatory loopholes through which enormous profits were generated in commodity speculation.

Since speculation is my preferred explanation for the energy and indeed other commodity price movements, I will go over some of the arguments in some detail. I hasten to state that the ideas express in this article are my own personal views (and perhaps those of Michael Masters [3] as well). They do not represent the market views of my employer, their affiliates, the Wilmott Magazine, or anybody else. Besides, some of these views are fairly half-baked and quite likely to be wrong, in which case I reserve the right to disown them and bequeath them to a friend of a friend. (Also, see the box on Biased Opinions).

Masters points out that there is no real supply crunch. Unlike the Arab Oil Embargo time in 1973, there are no long lines at the gas pump. Food supplies are also healthy. So some new mechanism must be at work that drives up the commodity demand despite the price level.

Masters blames the institutional investors (pension funds, sovereign wealth funds, university endowments etc.) for the unreasonable demand on commodity futures. Since futures prices are the benchmark for actual physical commodities, this hoarding of the futures contracts immediately reflects in the physical spot prices and in the real economy. And as the prices climb, the investors smell blood and invest more heavily, stoking a deadly vicious cycle. Masters points out that the speculative position in petroleum is roughly the same as the increase in demand from China, debunking the popular notion that it is the demand spike from the emerging giants of Asia that is driving the oil price. Similarly, bio-fuel is not the driver in food prices — the speculators have stockpiled enough corn futures to power the entire US ethanol industry for a year.

Although quants are not terribly interested in the transient economic drivers of market dynamics or trading psychology, here is an interesting thought from Mike Master’s testimony. A typical commodity trader initiating a new trade is pretty much insensitive to the price of the underlying. He has, say, a billion dollars to “put to work,” and doesn’t care if the position he ends up holding has five million or ten million barrels of oil. He never intends to take delivery. This price-insensitivity amplifies his impact on the market, and the investor appetite for commodities increases as the prices go up.

Most trading positions are directional views, not merely on the spot price, but on volatility. In a world of long and short Vega positions, we cannot expect to get a full picture of trading pressures exerted on oil prices by studying the single dimension of spot. Is there a correlation between the oil prices and its price volatility?

Figure 1
Figure 1. Scatter-plot of WTI Spot prices in Dollar and its volatility. Although the plot shows random clusters at low spot levels, at price > $75 (highlighted in the purple box), there appears to be a structure with significant correlation.

Figure 1 shows a scatter plot of the WTI spot price vs. the annualized volatility from publicly available WTI spot prices data [4]. Note than my definition of volatility may be different from yours [5]. At first glance, there appears to be little correlation between the spot price and volatility. Indeed the computed correlation over all the data is about -0.3.

However, the highlighted part of the figure tells a different story. As the spot price climbed over $75 per barrel, the volatility started showing a remarkable correlation (of 0.7) with it. Was the trading activity responsible for the concerted move on both prices and volatility? That is my theory, and Michael Masters may agree.

Hidden Currency Theory

Here is a dangerous thought — could it be that traders are pricing oil in a currency other than the once mighty dollar? This thought is dangerous because international armed conflicts may have arisen out of precisely such ideas. But an intrepid columnist is expected to have a high level of controversy affinity, so here goes…

We keep hearing that the oil price is down on the back of a strong dollar. There is little doubt that the oil prices are highly correlated to the strength of the dollar in 2007 and 2008, as shown in Table 1. Let’s look at the oil prices in Euro, the challenging heavy-weight currency.

Figure 1
Figure 2. Time evolution of the WTI spot price in Dollar and Euro. The Euro price looks more stable.

At first sight, Figure 2 does appear to show that the price is more stable when viewed in Euro, as expected. But does it mean that the traders are secretly pricing their positions in Euro, while quoting in Dollar? Or is it just the natural tandem movement of the Euro and WTI spots?

If the hidden currency theory is to hold water, I would expect stability in the price levels when priced in that currency. But, more directly, I would naively expect less volatility when the price is expressed in the hidden currency.

Figure 1
Figure 3. WTI Volatilities measured in Dollar and Euro. They are nearly identical.
Figure 1
Figure 4. Scatter-plot of WTI volatilities in Dollar and Euro. The excess population above the dividing line of equal volatilities implies that the WTI spot is more volatile when measured in Euro.

Figure 3 shows the WTI volatilities in Dollar and Euro. They look pretty much identical, which is why I replotted them as a scatter-plot of one against the other in Figure 4. If the Dollar volatility is higher, we will find more points below the red line, which we don’t. So it should mean that the hidden currency theory is probably wrong [6].

A good thing too, for nobody would be tempted to bomb me back to the stone ages now.

Human Costs

The real reasons behind the food and commodity price crisis are likely to be a combination of all these economic factors. But the crisis itself is a silent tsunami sweeping the world, as the UN World Food Program puts it.

Increase in the food prices, though unpleasant, is not such a big deal for a large number of us. With our first world income, most of us spend about 20% of our salary on food. If it becomes 30% as a result of a 50% increase in the prices, we certainly won’t like it, but we won’t suffer that much. We may have to cut down on the taxi rides, or fine-dining, but it is not the end of our world.

If we are in the top 10% income bracket (as the readers of this magazine tend to be), we may not even notice the increase. The impact of the high food prices on our lifestyle will be minimal — say, a business-class holiday instead of a first-class one.

It is a different story near the bottom. If we earn less than $1000 a month, and we are forced to spend $750 instead of $500 on food, it may mean a choice between a bus ride and legging it. At that level, the increase in food prices does hurt us, and our choices become grim.

But there are people in this world who face a much harsher reality as the prices shoot up with no end in sight. Their choices are often as terrible as Sophie’s Choice. Which child goes to sleep hungry tonight? Medicine for the sick one or food for the rest?

We are all powerless against the juggernaut of market forces creating the food crisis. Although we cannot realistically change the course of this silent tsunami, let’s at least try not to exacerbate the situation through waste. Buy only what you will use, and use only what you need to. Even if we cannot help those who will invariably go hungry, let’s not insult them by throwing away what they will die yearning for. Hunger is a terrible thing. If you don’t believe me, try fasting for a day. Well, try it even if you do — for it may help someone somewhere.

Conclusions

Commodity speculation by institutional investors is one of the driving factors of this silent tsunami of rising food prices. Their trading strategies have been compared to virtual hoarding in the futures market, driving up real prices of physical commodities and profiting from it.

I don’t mean to portray institutional investors and commodity traders as criminal masterminds hiding behind their multiple monitors and hatching plots to swindle the world. The ones I have discussed with do agree on the need to curtail the potential abuse of the system by closing the regulatory loopholes and setting new accountability frameworks. However, we are still on the rising edge of this influx of institutional funds into this lucrative asset class. Perhaps the time is not ripe enough for robust regulations yet. Let us make a bit more money first!

Reference and Endnotes

[1] Jeffrey Currie et al. “The Revenge of the Old ‘Political’ Economy” Commodities (Goldman Sachs Commodities Research), March 14, 2008.
[2] Business Times, “Shell counts rising cost of squeezing oil from sand in Canada,” March 18, 2008. http://business.timesonline.co.uk/tol/business/article3572646.ece
[3] Testimony of Michael W. Masters (Managing Member / Portfolio Manager, Masters Capital Management, LLC) before the Committee on Homeland Security and Governmental Affairs. May 20, 2008. http://hsgac.senate.gov/public/_files/052008Masters.pdf
[4] Cushing, OK WTI Spot Price FOB (Dollars per Barrel) Data source: Energy Information Administration. http://tonto.eia.doe.gov/dnav/pet/hist/rwtcd.htm
[5] I define the WTI volatility on a particular day as the standard deviation of the spot price returns over 31 days around that day, annualized by the appropriate factor. Using standard notations, the volatility on a day t is defined as:
sigma (t) = sqrt {frac{1}{{31}}sumlimits_{t - 15}^{t + 15} {left( {ln left[ {frac{{S(t)}}{{S(t - 1)}}} right] - mu } right)^2 frac{{252}}{{31}}} }
[6] Given that the correlation between EUR/USD and WTI Spot is positive (in 2007 and 2008), the volatility, when measured in Euro, is expected to be smaller than the volatility in Dollar. The expected difference is tiny (about 0.3% absolute) because the EUR/USD volatility (defined as in [5]) is about 2%. The reason for the counter-intuitive finding in Figure 4 is probably in my definition of the volatility as in [5].
[7] Monwhea Jeng, “A selected history of expectation bias in physics,” American Journal of Physics, July 2006, Volume 74, Issue 7, pp. 578-583. http://arxiv.org/pdf/physics/0508199

Box: Biased Opinions

As an ex-experimental physicist, I am well aware of the effect of bias. Once you have a favoured view, you can never be free of bias. It is not that you actively misrepresent the data to push your view. But you tend to critically analyze the data points that do not match your view, and tend to be lenient on the ones that do.

For instance, suppose I do an experiment to measure a quantity that Richard Feynman predicted to be, say, 1.37. I repeat the experiment three times and get values 1.34, 1.30 and 1.21. The right thing to do is to report a measurement of 1.29 with an error of 0.06. But, knowing the Feynman prediction (and, more importantly, knowing who Feynman is), I would take a hard look at the 1.21 trial. If I find anything wrong with it (which I will, because no experiment is perfect), I might repeat it and possibly get a number closer to 1.37. It is biases of this kind that physicists try very hard to avoid. See [7] for an interesting study on biases in physics.

In this column, I do have a favoured view — that the main driver of the commodity price inflation is speculation. In order to avoid pushing my view and shaping my readers’ opinion, I state clearly that there is a potential of bias in this column. The view that I have chosen to favour has no special reason for being right. It is just one of the many “hands” popular among economists.

About the Author
The author is a scientist from the European Organization for Nuclear Research (CERN), who currently works as a senior quantitative developer at Standard Chartered in Singapore. The views expressed in this column are only his personal views, which have not been influenced by considerations of the firm’s business or client relationships. More information about the author can be found at his web site: http://www.Thulasidas.com.