|
ONLINE ENERGY
|
 |
 
 
 
|
|||||||||||||||||||||||||||||||||||||||
 October 29, 2004 Summary Since our last report, the prompt NYMEX natural gas contract has risen by almost $2.00 per mcf. The gains have come about despite near-record storage levels going into the upcoming heating season due largely to natural gas “catching up” with ever-escalating crude oil prices. The price rise paralleled a move by large commodity funds who report to the CFTC from net short to net long positions. Most recently, relatively constructive storage data had tried to pull the prompt natural contract up further, but an almost $5.00 per barrel decline in WTI from its recent peak precluded natural gas from setting new highs for 2004. The
culprit in the recent crude oil price decline was initially
attributed to bearish weekly crude oil data, but in our view
the market had fully discounted the upcoming winter and began
focusing on the interest rate hike in China, deflating at least
for time being one of the “themes” that had helped
pull hydrocarbon prices above what we have termed an “adjusted” fundamental
value. The adjusted value would take into account the fundamental “pull” by
China as well as the relative tightness in sweet crude oil
availability this year. The fact that actual fundamental analysis
on this heterogeneous country is so difficult to perform, however,
has allowed China to impact energy in as much or more a psychological
fashion as fundamental. Turning back to the natural gas picture itself, however, our updated analysis continues to point to a relatively constructive balance for 2005. Despite the fact we have made little change to our volume numbers, we have concluded that some upward revaluation in our forecast average natural gas price is warranted, assuming the recent “re-linking” between oil and gas prices remains in place. As such, we now anticipate that Henry Hub will average about $6.20 per mcf in 2005, an upward revision of about $1.00 from our previous report, and a price that is more consistent with our forecast WTI average for next year of $40.00 per barrel.
The gains have come about despite a move toward near-record working storage levels going into the upcoming heating season. Natural gas finally decided to “catch up” with ever-escalating crude oil prices, as the prompt NYMEX crude oil contract traded in excess of $55.00 per barrel until its recent retracement. The price rise paralleled a move by large commodity funds who report to the CFTC to move from net short to net long positions. On September 1, with the prompt NYMEX contract settling at $5.608 per mcf, managed funds held net short positions of 25,574 contracts, the largest net short position since February 24. On October 19 with the prompt NYMEX contract settling at $7.123 per mcf, managed funds held net long positions totaling 15,554 contracts. Therefore a swing in fund positions of 41,128 contracts corresponded with a price move in excess of $1.50 per mcf. Recently, constructive storage data relative to consensus expectations had tried to pull up the prompt NYMEX natural gas contract up further, but an almost $5.00 per barrel drop in WTI from its recent peak in excess of $55.00 per barrel precluded natural gas from setting new highs for 2004. The blame for the recent crude oil price decline was first assigned to bearish weekly crude oil data. However, in our view the market had fully discounted the upcoming winter and there was no one left to buy. A day later the market began focusing on the interest rate hike in China, which may have been sensed the day before when the weekly data were released. In any event, the China news has deflated, at least for the time being, one of the “themes” that had helped pull hydrocarbon prices above what we have termed the new “adjusted” fundamental value. Our adjusted value would take into account the fundamental “pull” by China of Atlantic Basin sweet crudes as well as the relative overall tightness in sweet crude oil availability this year. The adjusted value would lie above what historical relationships to inventory levels would imply by perhaps $5.00+ per barrel or so. However, the fact that actual fundamental analysis on this heterogeneous country is so difficult to perform has allowed China to impact energy prices in as much or more a psychological manner as fundamental. Turning back to the natural gas picture itself, our updated analysis continues to point to a relatively constructive balance for 2005. Despite the fact we have trimmed modestly our natural gas demand outlook for next year to reflect higher prices, we have concluded that some upward revaluation in our forecast average natural gas price is warranted, assuming the recent “re-linking” between oil and gas prices remains in place. As such, we now anticipate that Henry Hub will average around $6.20 per mcf in 2005, an upward revision of about $1.00 from our previous report, a price that is more consistent with our forecast WTI average for next year of $40.00 per barrel.
Demand: Incorporating the most recent data from the DOE, we estimate that U.S. natural gas consumption will gain by 0.2%, or about 160 bcf this year, a slight upward revision from our last report. The revision reflects modestly more constructive demand in both the electric utility and industrial sectors in the second quarter than previously estimated. The latest data on industrial production released by the Federal Reserve Board revealed that manufacturing activity in September rose by 5.3% over the previous year, about in line with the assumption imbedded in our natural gas demand model. The most gas-intensive industries continued to improve, on average, more than the aggregate manufacturing index with the exception of food and the petroleum industry. In terms of the latter, the data reflected downtime as a result of Hurricane Ivan during the month. In general, however, over the summer there appears to have been a continuation of the trend in reduced gas consumption per unit of output in the industrial sector. For 2005, assuming normal weather, we are looking for a rise in total U.S. natural gas demand of 1.6%, or about 355 bcf. Underlying this growth is a forecast rise in manufacturing activity of 2.5%, in contrast to an estimated gain in 2004 of 5.1%. While this deceleration for next year may strike some as too conservative, the relative maturity of the current economic cycle combined with the adverse impact on growth of higher energy prices lead us to this conclusion. Winter weather is obviously on everyone’s mind these days, and as such we wish to provide two sensitivities to demand for the first quarter which would primarily, but not exclusively, impact the residential and commercial sectors. Our Base Case under normal heating degree days anticipates first quarter U.S. natural gas consumption to rise by 2.9%, or about 205 bcf. Under a colder than normal winter, we ran our model assuming 10% more heating degree days weighted by residential gas furnaces for the entire first quarter. This scenario yields a rise in total U.S. first quarter gas consumption of 8.2%, or about 590 bcf, some 385 bcf above our Base Case. Alternatively, under a 10% warmer than normal first quarter, total U.S. gas demand declines by 2.5%, or about 180 bcf below the first quarter of 2003, also some 385 bcf below our Base Case. We have already assumed some conservation by residential consumers in our Base Case, and the symmetry in the colder and warmer than normal cases reflects no further adjustments to consumption per heating degree day. In reality, however, we might expect some further conservation
under the colder than normal case as consumers face even higher
heating bills, while they may relax a bit under warmer than normal
conditions which would provide some relief from current prices.
We address the overall price implications of these alternative
scenarios in the last section of this report.
Supply: The impact of Ivan continues to reflect in the gas production data. As of October 28, 1.227 BCF/D remained shut in, with the cumulative impact of Ivan reaching 105.757 BCF. Nonetheless, the market will enter the 2004-2005 heating season with more than adequate gas volumes under normal weather conditions. We estimate that, depending upon next week’s data, end-October working storage is likely to reach the highest levels since 1991. Nonetheless, managed funds and some gas traders still expect storage to decline dramatically under the assumption that colder than normal weather will characterize the upcoming winter. Under our Base Case, end-first quarter working storage will stand at about 1.0 tcf, slightly higher than the level forecast in last month’s report. If reasonable, it would lie some 55-60 bcf below the end of the first quarter of this year. Granted, under our 10% colder than normal case for the first quarter, things get pretty tight. Working storage at the end of March would decline to only 615 bcf, the lowest level in history, and over 100 bcf below the tight conditions of end-March 2003. Alternatively, our 10% warmer than normal case leads to end-March 2005 working storage standing at almost 1.4 tcf. If this occurs, it would not be an all-time record high for the end of the first quarter. In fact, it would be lower than the 1.518 tcf level reached at the end of March just three years previously, in 2002. Turning back to our Base Case for 2005 as a whole, our balances imply a net draw in working storage next year of 225 bcf, a smaller draw than that implied by our last report by 55 bcf.
Implications
for Price As we issue this report, prompt NYMEX crude oil is trying to hold at around $51.00 per barrel, while prompt NYMEX heating oil is hovering around $1.45 per gallon. In contrast, the December NYMEX natural contract is trading at about $8.60 per mcf. Are these relationships consistent with our pricing model?
Let us assume for the moment that current oil price levels prevail into December, which may not beunreasonable depending upon the weather following what may be further short-term weakness. If so, our model suggests a target for the NYMEX contract in December (January basis) incorporating the oil premium of $7.30 per mcf. This price target suggests that the January NYMEX contract is about $1.90 per mcf overvalued relative to current oil prices and anticipated storage levels. We had previously argued that natural gas was undervalued relative to oil, and that was clearly the case. After a flurry by managed funds to cover shorts and go long, however, gas has now gone too far relative to oil prices. This brings us to our quick look at gas/oil price correlations for this year. We can conclude that compared to previous years, the relationship between oil and gas prices, as well as the individual prices themselves, have witnessed greater volatility than we have seen in some time. The table below illustrates for discreet time periods in 2004
simple correlations between the prompt NYMEX crude oil contract
and the prompt NYMEX natural gas contract.
Clearly, there has been no extended time period this year when the relationship was extremely close. What stands out most, however, is the third quarter, when the relationship was virtually nonexistent during most of the period. This reflects the significant move up in crude oil prices after the impact of Ivan. Natural gas continued to lag as managed funds retained net short positions, probably in part to hedge their rising crude length. Ultimately, however, lagging gas prices recovered as funds covered shorts and went net long, reestablishing a somewhat closer relationship since the beginning of the fourth quarter. As we suggested earlier, however, gas prices now strike us as overvalued relative to oil after lagging behind for so long. We believe there will be a somewhat closer correlation between gas and oil prices going forward, and as such under our $40.00 per barrel WTI price assumption for next year and since our gas balance is relatively constructive, we feel compelled to raise our forecast Henry Hub average for 2005 to about $6.20 per mcf, about $1.00 per mcf higher than our previous forecast. Even so, however, first quarter NYMEX gas prices appear to be discounting a colder than normal winter already. To examine this possibility, we ran our pricing model under out alternative weather scenarios. Our current Base Case anticipates that first quarter Henry Hub will average about $6.95 per mcf, some $1.50 per mcf below the current arithmetic average of the February, March, and April NYMEX contracts. Under a 10% colder than normal weather scenario for the entire first quarter, our model yields a first quarter average of $11.30 per mcf, a full $4.35 above our Base Case. Alternatively, our 10% warmer than normal scenario yields a first quarter average of $5.00 per mcf, $1.95 per mcf below our Base Case. The severe asymmetry in price sensitivity comes about through the combined influence of our constructive gas balance in the context of the impact of oil prices on natural gas values. It is highly unlikely that the entire first quarter will end up either 10% colder or 10% warmer than normal. The price spike that would result under any extended period of cold weather in the first quarter, however, is clearly what the consensus is focusing on at the present time, right or wrong.
| Summary | Viewpoint | Demand | Supply | Implications | |
|||||||||||||||||||||||||||||||||||||||
