Natural Gas: Bulls, Bears, and Bankers
http://www.energypulse.net/centers/article/article_print.cfm?a_id=1291
Let’s start with the punch line: As of mid-June, natural gas in storage is roughly 450 Bcf (or 22%) above the year-ago level and roughly 650 Bcf (or 35%) above the five-year average. Without an extremely disruptive Gulf of Mexico hurricane season, Henry Hub natural gas prices will decline significantly by mid-summer and regional basis discounts will widen much further, especially in the Rocky Mountain regions. Hot weather, increased industrial demand, NGL liquids stripping, and fuel switching are nowhere near enough to work off the storage surplus without a major hurricane disruption or a major cut-back in production. Investors need to be very cautious about short-term natural gas prices.
By now the bulls’ case and the bears’ case are well known. The bankers’ case is much less well known but no less important. This article summarizes these three views of the natural gas markets and offers some ideas about where the market is headed.
The Bulls’ Case
First, at $70 per barrel, crude oil is trading at more than 10:1 versus the price of natural gas. The long-run average is closer to 6:1. More importantly, natural gas at $6.50/mmBtu (Henry Hub) is trading about 10% below 3% sulfur residual fuel on a Btu basis. This is unusual since residual fuel typically sets the floor price for natural gas. Fuel switching should add a couple of Bcf/d to load.
Second, with basis discounts already exceeding $1.00/mmBtu in the Rockies and parts of the mid-continent and approaching $1.50/mmBtu for late summer futures in several major producing regions, price-sensitive industrial load unrelated to fuel switching will return, adding another couple of Bcf/d to load. Liquids stripping in the NGL industry will absorb another couple of Bcf/d versus long-run averages at current and future relative oil / gas prices.
Third, as summer heat kicks in and peaking plants crank up for the ever larger housing stock, utilities will turn to gas in larger quantities. Additionally, the low price of natural gas will encourage utilities to use more gas and less coal to conserve below-average coal inventories. The combined effect of fuel switching, industrial load pick-up, NGL liquids stripping, and summer peaking demand will absorb at least 5 Bcf/d more than average through the summer, allowing storage to trend towards normal levels by the end of injection season even without a hurricane.
Fourth, the futures markets are saying that Henry Hub gas this winter (and at least the next three winters) will be closer to $10/mmBtu versus today’s $6-7 range. That means the futures markets do not believe we will enter the winter with a sufficiently large storage surplus to keep prices depressed.
Fifth, the natural gas strip is set by winter prices, which are set by heating loads, which are set in the residential sector. The U.S. is rapidly increasing its natural gas-heated housing stock, making five-year average storage levels unrepresentative of the levels required to meet winter heating loads. The storage surplus is nowhere near as large, relative to normal winter loads, as it appears. The excess storage today is a byproduct of last winter’s unseasonably warm weather. Don’t count on it happening again.
Sixth, North American natural gas producers have dramatically increased exploration and production spending in the past few years with little or no net impact on total production. Even a slight pull-back in capital expenditures, without any production shut-ins, will put the already severe depletion curves back into play. Any surplus will disappear in short-order.
Finally, Hurricanes Katrina and Rita combined to shut-in almost 800 Bcf offshore and probably another 100 Bcf onshore and are continuing to shut-in more than 1 Bcf/d, nearly a year later. Hurricane Ivan shut-in hundreds of Bcf the year before. The hurricane season currently forecast, even if only at the Ivan level and far below the Katrina-Rita level, is more than enough to eliminate the current storage surplus.
The bottom line for the bulls: natural gas prices are near a bottom relative to competing fuels, loads are going to pick-up from multiple industrial and power generation uses at current prices, and the storage surplus – which isn’t as big as it appears assuming only average weather – is going to be largely absorbed by the end of injection season without any further price declines. Even a moderately active hurricane season will send spot prices back into double-digits and create a serious problem this winter.
The Bears’ Case
First, total switchable load (industrial and power generation) is no more than about 2-3 Bcf/d and most of the load that can switch has switched because natural gas has been priced below residual fuel oil for months now. Similarly, NGL liquids stripping, which can vary by several Bcf/d depending on relative prices, should already be near a maximum for the same pricing reasons. While these two factors could maximally account for 20-30 Bcf per week of gas diverted from storage, the pricing relationships have been in place long enough that very little price-sensitive switching or stripping is left.
Second, price-sensitive industrial load that hasn’t returned to natural gas (previous paragraph) isn’t coming back until prices are much lower, stay that way for an extended period of time, and occur in an otherwise favorable long-term business environment. Industrial loads are not ramped up or down weekly or monthly as if linearly tied to gas prices. It will take a great deal of time and business confidence to restore the multiple Bcf/d of industrial load lost to last year’s very high prices and it won’t take place in the next month or two. Meanwhile, literal demand destruction from Katrina doesn’t reappear simply because gas is “only” $6/mmBtu. It’s gone. Finally, production of certain other gas-intensive commodities, like ammonia, has permanently moved to locations like the Middle East and the Caribbean to take advantage of much lower feedstock costs (e.g., $2/mmBtu).
Third, coal inventories at power plants were low in the winter because of rail problems. Those problems have mostly been resolved and coal inventories are acceptable. Nuclear and hydro are operating at higher capacity factors than last year and total electric power generation is down versus last year. There is some deterioration in the average heat rates of the natural gas plants but it’s worth no more than 1-2 Bcf/d. If early-August heat (the hottest part of the summer) were in place from early-July through early-September, the incremental gas absorption for power generation and air conditioning would be around 100 Bcf. This is only 20% of the current surplus. The next inflection point for meaningful incremental gas demand is substitution for coal. For this to take place, natural gas has to be in the range of $4/mmBtu, not $6/mmBtu.
Fourth, the futures markets are not going to sustain a $3-4 winter premium to spot if the current injection and storage patterns hold much longer. From the perspective of the hedge funds, the futures markets are in extreme contango, which costs the funds substantial amounts of money each month on negative roll yield. From the perspective of the producers, the incentive to sell the strip forward or at least hedge the winter months at a $3-4 premium is going to become irresistible as operational flow orders loom. As of mid-June, storage is almost 2.5Tcf. In the past eight years, the earliest date that storage approached 2.5Tcf was the third week of July in 2002. Not coincidentally, that was also the last time natural gas was below $3/mmBtu and the last time the oil/gas price ratio hit 10:1. If the current storage trends continue for even another few weeks, operational flow orders from the pipelines will force production shut-ins. Without shut-ins and hurricanes, average summer injections would take storage to the traditional maximum fill of about 3.3Tcf by the end of August. This would leave no place to put September and October injections – typically totaling almost 600 Bcf.
Fifth, winter prices may spike in February or March if it’s cold but this has little meaning over the next few months if storage is at 3.3Tcf in August or early September. While a very cold winter (e.g., mid-January temperatures from mid-December through mid-February) would absorb about 350-400 Bcf more than a normal winter, there is no evidence that the upcoming winter will be unusually cold. Moreover, the combination of a hot summer and a cold winter only represents about 450-500 Bcf more than average loads. Normally, this would be an enormous increment and raise important questions about storage adequacy and winter deliverability, but this year it’s no more than the year-over-year storage surplus.
Sixth, a significant pull-back in capital expenditures on exploration and production will absolutely put the North American depletion curves back in play. Gas production will decline. However, the issue is the supply / demand balance in the summer of 2006, not next year or the year after, assuming a near-term cut-back in exploration expenditures. Meanwhile, the big increase in expenditures in the past few years has at least temporarily reversed the production decline. Production in 2006 is up versus 2005. It’s not much in the context of the other supply / demand imbalances but it is adding to the storage and pricing pressure.
Finally, a hurricane having an impact between that of Ivan in 2004 and Katrina in 2005 would definitely absorb the storage surplus for at least a short period of time. However, Katrina and Rita eliminated close to 900 Bcf (offshore and onshore) and ten months later we’re 500 Bcf above average. Mild weather since the hurricanes cannot explain more than about 1/3 of that swing. Price-sensitive demand destruction and literal demand destruction were and are the larger reasons for the surplus.
The bottom line for the bears: without production shut-ins, natural gas prices will drop towards $5 at the Henry Hub and $4 or lower in regions with the biggest supply / demand imbalances, notably the Rockies. Major price-sensitive demand from utilities firing coal won’t kick-in until that point. The idea that oil-to-gas fuel switching, industrial load, NGL liquids stripping, and utility peaking load at $6/mmBtu can reliably absorb 5 Bcf/d above current levels is wishful thinking. It may be possible to add 1-2 Bcf/d after a period of time where prices remain stable and industrial demand remains strong but the cumulative effect between now and the end of injection season is potentially no more than 100 Bcf. Only a hurricane that can shut-in hundreds of Bcf of production - without further destroying demand via high prices or literal destruction - can prevent comparably large shut-ins this summer.
The Bankers' Case
The bankers are aware of the tug-of-war between the bulls and bears over the storage numbers, the factors that could generate incremental load growth, the recent and ongoing expansion of North American E&P capital spending, incremental LNG imports, the producers’ collective desire to “drill through” the current price weakness and, of course, the risk of hurricanes. The bankers are also aware of the producers’ budgeted 2006 oil & gas price decks (about $56/bbl. and $7/mmBtu) and the levels at which spending would be significantly cut back (about $42/bbl. and $5/mmBtu) assuming the lower prices were in place for at least three to six months. (Ref. 1)
The bankers are also aware that statistically (based on crude and products prices, gas storage levels, and seasonal factors) spot natural gas prices actually should be closer to the double-digit futures prices for the winter months than the current $6-7 level. These statistical inferences aren’t reliable in the current situation, however. Storage levels are far outside the historical range used to establish the pricing relationships and short-term gas demand is demonstrably less price-elastic than the models assume. These factors wreak havoc with models that require continuous substitutability between residual oil and gas; a relationship that has obviously broken down in the face of the current storage surplus and the current oil / gas price relationship. In other words, it’s more prudent to believe the physical surplus than the regression models for now.
According to a survey of 41 bankers reported in the June issue of Oil and Gas Investor magazine, the forecast mean Henry Hub natural gas price for the second through fourth quarters of 2006 is $6.66/mmBtu. For 2007, the figure is $6.32. For 2008 and later years, it’s below $6/mmBtu. The sensitivity case downside values are $5.37, $4.84, and mid $4s for those same periods, respectively. (Ref. 2)
The bankers’ price forecasts do not take aim at the bull-bear argument between $6 summer gas and $10 winter gas. They take aim at the argument between $6 gas and $70 oil. The same banker survey puts the crude oil forecast at roughly $48 (WTI) for the second through fourth quarters of 2006, $44 for 2007, and $40 or below thereafter. The downside mean values are roughly 20% lower.
With the important caveat that the market views of bankers and oil & gas producers reflect very different risk-reward dynamics, the implication of the bankers’ forecasts versus current strip prices is greater price risk in the crude strip than the gas strip. This has significant implications for the bull-bear gas argument. If relative oil prices slip like the bankers’ forecast then the case for gas-for-oil substitution and NGL liquids stripping becomes weaker and natural gas inventories expand, at least until the forecast trend towards backwardation in the futures market pressures the gas strip.
Additionally, the bankers’ oil price deck versus current world prices is consistent with either or both 1) a U.S. and probably worldwide macroeconomic slowdown and reduction in demand or at least demand growth, and 2) a large increase in oil supply arising from the large recent increase in capital expenditures on exploration and production. Through some combination of a shift in the demand curve downward or a shift in the supply curve upward, the equilibrium price for oil would decline.
With respect to natural gas (though not oil), the long-term validity of these relatively conservative forecasts is subject to dispute for one gigantic reason – the unstoppable acceleration of North American depletion rates over time. (This will be the subject of a future article.) For the short-term, however, the bankers’ current view of the world challenges one and possibly both of the major price-related demand underpinnings from the bulls’ case: 1) gas-for-oil substitution at the industrial level, NGL liquids stripping, and gas-for-oil and ultimately gas-for-coal at the utility level, and 2) absolute price-sensitive load increases at the industrial level. The position of the bankers with respect to supporting natural gas prices at current levels this summer without large production shut-ins can thus be summed up in one word: hurricanes.
Conclusion
Natural gas storage levels are so far ahead of historical averages that there is no longer any non-hurricane alternative to price declines and production cut-backs. For example, everything else equal and assuming average injection rates, natural gas storage would reach traditional maximum levels around 3.3 Tcf about two months ahead of schedule and absolute physical maximum storage around 3.5 Tcf more than one month ahead of schedule. There will simply be nowhere to put the gas normally injected in September and October. Second, a 450 Bcf storage surplus is approximately equal to incremental gas usage from a back-to-back hot summer and cold winter. This is approaching the definition of a short-term gas bubble. Apart from lower prices and production cut-backs within a month or two, the only force that can change this dynamic is a highly disruptive Gulf of Mexico hurricane season. In the long-run, the gas bubble collapses as North American depletion curves dominate all other variables. But not this summer. For this summer, it’s either much lower prices and voluntary production shut-ins in the regions with the most extreme supply / demand imbalances or hurricanes.
References:
1. Lehman Brothers, Oil Services & Drilling Original E&P Spending Survey, June 21, 2006.
2. Oil and Gas Investor magazine, Tristone Capital’s Energy Lender Price Survey, June, 2006, p. 11.
6.30.06 Harry Chernoff, Principal, Pathfinder Capital Advisors, LLC
News 
Market Data 
Discover 
