Every week, the U.S. Energy Information Administration publishes an estimate of the amount of natural gas in storage. These estimates provide the basis for judging the adequacy of gas supplies in North America. As of June 1, 2003, gas storage levels were 500 billion cubic feet below the five-year average for the date. By the end of 2004, storage levels were about 350 Bcf above the five-year average for the date, an apparent increase of some 850 Bcf. However, despite ample supplies, wellhead prices increased 15 percent between June 2003 and December 2004, according to EIA. Why did the additional gas not reduce prices? Some observers believe that the puzzling behavior of gas prices in the last two years is evidence of market manipulation. This possibility should not be ignored by those responsible for market oversight. There is, however, an alternative explanation. It?s one analysts have known since the beginning of 2004?that the apparent gas surplus was likely temporary and would not last. Here?s how we knew, if you?ll bear with me. During summer, gas supplies are more than enough to satisfy consumption, and the excess goes into storage. In winter, when heating loads are high, supplies are inadequate and gas is taken out of storage to meet the demand. In a general way, the storage levels tell us something about supply adequacy. But the changes in storage reported every week by EIA are almost meaningless without additional information. Gas is consumed for a variety of uses. Some gas consumption is relatively constant over time?heat and feedstock for industries such as cement production, restaurants, plastics, and fertilizer, for example. Some consumption is quite variable?providing heat in winter and power plant fuel to run air conditioners in summer, for example. From weather data it is possible to estimate quite accurately the variable component of consumption?the amount used for heating and cooling. When that component is added to the weekly storage data, we know rather precisely how much of the gas supply was not used for industry and so forth. That is, we know on a weekly basis how much gas is available for heating and cooling and for storage. As mentioned above, storage levels have increased since mid-2003, indicating that more gas was available than was needed for heating and cooling. However, a closer look shows that the surplus of available gas actually increased only for the first six months of that period. In the twelve months of 2004, the surplus was shrinking even though there was a net surplus that added to storage. Since the beginning of 2004, the amount of gas available for heating and cooling or storage has been abundant but decreasing. The surplus has been shrinking and recently became a deficit once again. Moreover, the deficit is projected to grow rapidly. Heating and cooling consumption averages about 158 billion cubic feet per week throughout the year. As of today, the gas supply available for these uses is estimated at only 142 Bcf per week. If current levels were to continue throughout the year, storage would decrease by about 830 Bcf. To make matters worse, available supplies are projected to decrease still further, dropping to 132 Bcf per week by the end of March. Unless this trend turns around, storage levels will be seriously depleted by next winter. Fear of shortages can be expected to cause prices to rise again soon. Traders who believed that the gas surplus was temporary would have been willing to buy the surplus and put it into storage, awaiting the deficit that appears to be developing this year. Their buying may have been sufficient to support prices despite the apparent surplus. Although they paid rather high prices during the surplus period, they will profit if prices rise again this year. So now you have two theories that explain the paradox of high prices in a time of surplus gas?market manipulation and the temporary nature of the surplus. I?m positive that there are others being discussed by folks to whom gas prices really matter. If you have a good explanation, please send me an e-mail at the address below. ?Dr. Rich Ferguson, Research Director, CEERT, rich@ceert.org