FIRST OXYGENATED GASOLINE SEASON SHAKES OUT DIFFERENTLY THAN EXPECTED
Charles Dale
Energy Information Administration
Washington, D.C.
John H. Hackworth, Joanne M. Shore
OnLocation Inc.
McLean, Va.
John Ostrich
SAIC Inc.
McLean, Va.
The U.S.'s first oxygenated gasoline season began Nov. 1, 1992. Refiners and marketers achieved compliance with these new specs with little upset to the gasoline production and distribution system. But although the season went smoothly, it did not shake out exactly as projected.
Demand for oxygenated gasoline and, in particular, methyl tertiary butyl ether (MTBE), was lower than expected.
Prior to the season, refiners were concerned that oxygenates might be in short supply. No supply shortages developed, however, and prices of both oxygenates and gasoline decreased during the season.
DEMAND
The Energy information Administration (EIA) - a branch of the U.S. Department of Energy - projected that total seasonal oxygenated gasoline consumption during the full compliance period would be 2.72 million b/d.1 EIA also estimated that the total quantity of oxygenates needed for blending (expressed in MTBE equivalent units) would be 475,000 b/d.
MTBE equivalent units are a measure of oxygen content. Because ethanol contains more oxygen than MTBE, 1 bbl of pure ethanol = 2.04 bbl MTBE. In practice, however, MTBE is about 99% pure and fuel ethanol may contain several percent denatured alcohol. Thus, 1 bbl ethanol = 1.96 bbl MTBE equivalents.
EIA's estimate of MTBE demand was derived by separating oxygenate blending into various components and forecasting them independently. Demand for nonattainment compliance blending was estimated by allocating gasoline consumption to nonattainment areas on a population basis and calculating the amount of oxygenates required to meet this gasoline demand.
The EIA also estimated:
- The quantity of all oxygenates to be consumed in attainment areas, or areas in which oxygenated gasoline is not mandated
- The amount of "over-compliance blending" that would provide a safety margin for meeting oxygen content specifications
- The quantity of ethanol that would be blended, in excess of requirements, to obtain tax subsidies
- Oxygenate consumption outside of, but bordering, nonattairiment areas, sometimes called "spillover."
A nonattainment area is an area designated by the U.S. Environmental Protection Agency (EPA) as being out of compliance with atmospheric carbon monoxide standards. Spillover was assumed to occur as a result of storage and transportation limitations that require distributors and marketers to supply oxygenated gasoline to attainment areas.
Final oxygenated gasoline demand for the full compliance period, based on data in EIA's April 1993 Petroleum Supply Monthly, was estimated to be only 2.2 million b/d. Total seasonal oxygenate demand, in MTBE equivalents, was only about 402, 000 b/d - 15.4% less than forecast. ("Seasonal" is defined as the average consumption during the full compliance period.)
To understand this surprisingly weak demand, the administration of the oxygenation program and other factors, such as spillover, must be examined.
ADMINISTRATION
Under the oxygenated gasoline program, individual states are responsible for implementation. Each state must submit an amended State Implementation Plan to the EPA for approval. As a result, implementation strategies differ.
For example, some states adjusted their gasoline control areas on the basis of sparse population and other factors. ("Control area" denotes everywhere that oxygenated gasoline was mandated by a state. It is frequently more inclusive than the EPA's nonattainment area.) Several states simply opted not to participate in the program this season.
Three nonattainment areas opting out of participation were Boston-Lawrence-Salem, Memphis, and Duluth. Fairbanks, Alas., started out with an oxygenation program but suspended it in December because of motorist complaints (OGJ, Mar. 22, p. 21).
These nonparticipating areas were partly responsible for demand falling short of the forecast. This decrease in consumption, however, was partially offset by greater demand from attainment areas in California and New Jersey.
State control periods also vary. Although all states were required to begin their program no later than Nov. 1, 1992, Texas, Nevada, and Arizona started selling oxygenated gasoline on Oct. 1. In the 1993-94 season, additional areas will start on Oct. 1 and Spokane, Wash., will start on Sept. 1.
This year the control periods ended, at the earliest, Jan. 31. Most ended Feb. 28, however, and a few continued through March or later. Longer control periods in 1993-94 will increase demand for oxygenated gasoline, as compared to this season.
California is, of course, a special case. Some scientists believe that oxygenated gasoline increases NO, emissions, which can lead to higher levels of atmospheric ozone, and thus smog. California applied for a waiver from the EPA regulations to allow them to use gasoline containing 1.8-2.2 wt % oxygen, rather than the 2.7 wt % mandated in other areas.
Because the federal hearing on the waiver was not held until May 1993, California operated according to its proposed plan. This provision for reduced oxygen concentration was included in the original EIA forecast and therefore did not affect the forecast. And because the forecast was for a typical full compliance period, it also was not affected by overlapping control periods and delays.
Of the administrative issues already discussed, only two should have significantly affected oxygenated gasoline consumption, as compared to the forecast. These issues are the noncompliance of the Boston-Lawrence-Salem, Memphis, and Duluth areas and the participation of several attainment areas.
The lack of compliance in some areas, however, is not an adequate explanation for the lower demand. Other factors also contributed.
SPILLOVER
Spillover can result from either distribution problems or the economics of gasoline production. Normal use of oxygenates outside of control areas, such as baseline attainment gasohol sales, are not considered spillover.
EIA assumed baseline nonattainment oxygenated gasoline demand for the full compliance period to be 32.6% of total gasoline demand, based on the proportion of population in participating control areas. Prior to this season, EIA estimated that spillover could be as much as 22% of baseline demand.1
Nevertheless, detailed discussions with state governments and oil companies have indicated that actual spillover was negligible. The distribution system functioned more efficiently than the experts projected.
There was no added margin on the oxygenated gasoline sold in Control areas. This factor provided a strong economic incentive to reduce spillover.
There are more than 1,200 refined petroleum products terminals in the U.S. (Each company with facilities at a tank far is considered a separate terminal.) And less than half of these 1,200 (about 450) are either in a control area or within 100 miles of one.
Terminal operators employed a variety of distribution strategies to minimize spillover, and consequently, storage problems:
- Some terminals stored conventional gasoline in tanks and blended oxygenates, mainly ethanol, by in-line or splash blending. In some cases, the oxygenate even came from a competitor's terminal.
- Discussions with suppliers have indicated that some small refiners simply could not afford to compete in the oxygenated gasoline market. These refineries therefore did not create any spillover.
In spite of these strategies, some spillover was inevitable in regions where a terminal supplied both control and noncontrol areas. Also, sales to an area outside a control area may be so small that they do not justify holding separate stocks of gasoline, splash-blending at the terminal, or making an exchange agreement with a competitor.
Company officials were unanimous in stating that they did everything possible to reduce spillover.
OTHER FACTORS
Prior to this past season, oxygenates were used in attainment areas, primarily as octane boosters and to produce gasohol. Continued demand in attainment areas and overcompliance demand also influenced the market scenario.
The largest source of forecast overcompliance was the use of ethanol in quantities sufficient to receive tax credits. The EIA forecast assumed that producers would add enough ethanol to receive the subsidy, in other words, 10 vol % of gasoline. (Only 7.7 vol % is required to satisfy the oxygen requirement.)
The tax subsidy was changed, however, on Jan. 1, 1993. This ruling prorated the subsidy to include less concentrated ethanol blends, consistent with the oxygen requirements. The result of this change was a reduction in ethanol overcompliance demand.
Although there is a great deal of uncertainty in estimating the effects of these additional factors, they represent a significant portion of oxygenate blending. They also played a role in the difference between forecast and actual demand last winter.
EIA's forecast projected a demand of 55,000 b/d MTBE equivalents for attainment areas and 43,000 b/d for overcompliance demand. This forecast total of 98,000 b/d is 32% of the actual required oxygenate use for nonattainment areas.
The forecast also included 71,000 b/d of spillover. These three factors are, together, equivalent to 55% of nonattainment-area baseline demand (Table 1).
REDUCED DEMAND
The major differences between the forecast (475,000 b/d) and actual demand (402,000 b/d) for blended oxygenate are attributable to spillover, overcompliance blending, continued blending in attainment areas, and the lack of participation by some nonattainment areas.
The data that EIA collected reveal only the total MTBE and ethanol supplied, not the different uses of the oxygenates (such as spillover or overcompliance). As a result, there is no definitive information on how each of these uses contributed to total consumption last winter. Table 1, however, shows a reasonable estimate.
The final estimate of oxygenate requirements for state control regions during the full compliance period was an average 300,000 b/d:
228,000 b/d MTBE and 72,000 b/d ethanol (in MTBE equivalents). This amount is almost identical to the forecast 306,000 b/d (Table 1).
The loss of Boston, Duluth, and Memphis was countered, to some degree, by the blending of more ethanol in state control areas than was estimated, and by the addition of several attainment areas that chose to participate. Thus, the major difference between forecast and actual demand lies in the remaining factors: spillover, overcompliance, and attainment-area baseline use.
First, consider MTBE. One would expect MTBE attainment-area baseline blending to be very small during the season, because additional octane enhancement is not needed and overcompliance was not a major issue with MTBE.
The total use of MTBE averaged 233,000 b/d during the full compliance months. Because control-area usage was 228,000, this leaves only 5,000 b/d for spillover and baseline attainment-area use.
Excess MTBE supplies were available and spot prices low at the end of the season. It is therefore probable that some MTBE was used in attainment areas for octane enhancement. Subtracting this attainment-area usage from total MTBE demand reveals that MTBE spillover likely was less than 2%.
ETHANOL
Actual, estimated usage of 72,000 b/d ethanol in control areas leaves 97,000 b/d for spillover, overcompliance, and attainment-area use (Table 1). Baseline attainment-area ethanol demand was significant, but overcompliance blending was somewhat less than forecast because of the change in tax regulations.
Because ethanol is splash blended, controlling ethanol spillover should be easier than controlling MTBE spillover. As a result, EIA assumes ethanol spillover to be equal to or less than MTBE spillover (< 2%).
Spillover for all oxygenated gasoline last winter was probably less than 2%. Because projected spillover was 71,000 b/d, or almost 15%, this alone would account for most of the difference between forecast and actual oxygenate demand. Differences in overcompliance use of ethanol and estimates of baseline attainment-area use are responsible for the small remaining differences.
OXYGENATE SUPPLY
EIA began collecting monthly data on oxygenate supply in January 1992. Survey responders provide data on oxygenate production, motor gasoline blending, inventories, and imports. Figs. 1 and 2 show monthly ethanol and MTBE production and stocks for January 1992 through June 1993.
oxygenate production in 1992 was greatest for MTBE, followed by ethanol. Production of tertiary amyl methyl ether, tertiary butyl alcohol, and other oxygenates (including other aliphatic alcohols and ethers) accounts for only about 10% of oxygenate volume.
While exports and imports of oxygenates during the first season were not a large factor, they were an important marginal contributor to the excess supply.
One source estimated that MTBE imports ranged from 5,000 b/d in February 1992 to 32,000 b/d in November 1992.2 Net imports supplied approximately 5-10% of MTBE demand during the full compliance months.
Exports in 1992 ranged from only 2,000 b/d to 8,000 b/d, although they increased to 11,000 b/d in January 1993, as opportunities arose for suppliers to reduce excess MTBE inventories.
Table 2 reports operable production capacity, as of Jan. 1, 1993, for several oxygenates. ARCO Chemical Co.'s Corpus Christi, Tex., plant is capable of producing either 9,500 b/d of ethyl tertiary butyl ether (ETBE) or 12,000 b/d of MTBE. As of Jan. 1, the plant was producing ETBE and, therefore, is included as ETBE capacity in the table.
MTBE production is concentrated in the Gulf Coast; ethanol in the Midwest. Petroleum Administration for Defense District (PADD) 3 - the U.S. Gulf Coast - accounted for 94% of U.S. production of methanol, an MTBE feedstock, and 88% of U.S. MTBE production. The MTBE produced in PADD 3 is blended with gasoline and shipped by pipeline.
PADD 2 - the Midwest U.S. - accounted for 95% of U.S. production of fuel ethanol, which was shipped by railroad tank cars and trucks to terminal facilities for splash blending with gasoline.
PADD 3 also dominated the production of all other oxygenates, and PADDs 4 and 5 - the Rocky Mountain and Western U.S. regions - reported the smallest oxygenate production quantities.
MTBE production in 1992 reach a high of 128,000 b/d in November. This represents 75% utilization of the 170,000 b/d operable capacity reported Jan. 1, 1993. Approximately two thirds of the production is supplied by merchant producers, or facilities not associated with refineries. The remainder is produced by petroleum companies for internal use.
Ethanol production has been steady since EIA began collecting monthly data. January 1992 remains the highest-production month on record with 78,000 b/d, followed by March 1993 at 77,000 b/d. Operable capacity equation for ethanol was 84% during January 1993.
As anticipated, the full compliance months saw demand considerably higher than production capacity. During August, prior to the season, 47,000 b/d of MTBE was blended into gasoline. This represents typical baseline needs for MTBE as an octane-enhancing agent.
Blending increased more than 400% during the full compliance months, averaging 233,000 b/d in November through January.
Prior to the season, the proportion of MTBE-blended gasoline to total gasoline production averaged 48% for March through August - the busiest driving months. The remaining production was used to build MTBE inventories. To a lesser extent ethanol inventories also were built up prior to the season, to handle anticipated demand.
Considerable uncertainty regarding the adequacy of MTBE supplies existed during the months preceding the first oxygenation season. EIA estimated that to meet projected MTBE demand during the first winter an oxygenate inventory of 27 million bbl MTBE equivalents would be required.1
MTBE stocks peaked on Aug. 31, 1992, at 23.1 million bbl, while ethanol stocks reached a high of nearly 3.0 million bbl on Sept. 30, 1992. The stock drawdown began in October and continued through February, as shown in Fig. 3.
This pattern of stock buildup and drawdown will be typical of this winter's oxygenate cycle as well. Drawdowns will start earlier, however, as many programs will begin in October rather than November.
While fuel ethanol blended into motor gasoline increased somewhat over the last oxygenated gasoline season, it did not experience the same dramatic blending increases as MTBE. Fuel ethanol blending into motor gasoline in 1992 was about 69,000 b/d. EIA calculates this quantity as "production plus imports minus stock change."
PRICES
In previous years, the price of MTBE was determined by its usefulness as an octane enhancer. MTBE was priced just less than the cost of producing 1 octane number at the refinery .4 This pricing was changed,
However, when ARCO Chemical decided to price MTBE by its oxygenate value. The new pricing formula included rack and spot gasoline prices, along with the spot price of butanes in Mont Belvieu, Tex. The old formula had depended more heavily on pipeline or spot barge gasoline prices.
Because ARCO Chemical is the world's largest MTBE producer and supplies about 60% of the merchant MTBE market, its new pricing mechanism had a significant impact on MTBE pricing. Contract prices rose from less than 90/gal in early 1992 to $1.10/gal in summer 1992. Discussions with industry participants indicate that most MTBE was sold at contract prices.
The spot MTBE market has much less volume than the contract market. Nevertheless, spot prices may be used to approximate marginal MTBE price and, therefore, reflect market pricing pressures. For example, increasing spot prices reflect a tightening market and decreasing spot prices imply a loosening one.
In the first quarter of 1992, it appeared that MTBE demand would be very strong to meet the required fourth quarter implementation of the winter oxygenated-gasoline program. Many major oil companies contracted for large quantities of MTBE at prices well above the traditional octane-value price, to ensure an adequate supply to meet their market needs.
MTBE spot prices on the Gulf Coast reached a high of 98/gal in June 1992 then fell steadily to a low of 71/gal in January 1993. At the end of March 1993, spot MTBE was trading at 70-71/gal.
Price movements for oxygenates during the control season are summarized in Fig. 4. Oxygenate spot prices fell steadily during the control period because of lower-than-expected demand, large stocks, and minimal spillover.
ETHANOL TAX CREDIT
Since the early 1980s, ethanol has grown in importance as a gasoline blending component. Because of its expense, however, ethanol as a blending agent is uneconomical without tax subsidies provided by the U.S. government and some state governments.
Until Jan. 1, 1993, the U.S. government provided a 54/gal tax subsidy for ethanol, if blended at a minimum level of 10 Vol % in gasoline. This amounts to a tax credit of 5.4/gal of gasohol.
According to the U.S. Department of the Treasury's office of tax policy, the subsidy was changed on Jan. 1 to provide a prorated credit for ethanol blended at two other percentages: 7.7 Vol %, which produces 2.7 wt % oxygen in gasoline, and 5.7 Vol %, which produces the 2.0 wt % oxygen level required in California. For example, blending 7.7 Vol % ethanol into gasoline would result in a federal subsidy of 41-58/gal ethanol (77% of 54/gal).
Before January 1993, some refiners used more ethanol than was necessary to produce 2.7 wt % oxygen. Some of those refiners, however found it advantageous to cut back on ethanol usage in January, as shown in Fig. 5.
Until mid-1992, ethanol prices were higher than unleaded gasoline prices by about 54/gal (equivalent to the tax incentive). But after the changes in MTBE contract pricing, ethanol contract prices often were also based on their oxygenate value. 6
Ethanol prices in PADD 5 reached a premium of 78/gal more than unleaded gasoline prices in October 1992, before decreasing to 66/gal more than gasoline in January 1993.
OXYGENATED GASOLINE
The oxygenated gasoline season had greatly different effects on consumers in attainment areas, as compared to nonattainment areas.7
On Nov. 6, 1992, shortly after the start of the season, consumers in the nonattainment areas of Philadelphia, Los Angeles, Long Island, and Denver were paying 2.69-6.13/gal more for self-service regular unleaded gasoline than they had been paying 2 weeks earlier.
An extreme example of the higher cost of oxygenated gasoline was found in Anchorage, Alas. The price of self-service regular unleaded gasoline there increased by 14.5/gal between Oct. 9 and Nov. 6, 1992. In contrast, consumers in the attainment areas of Chicago, Dallas, Rochester, and Salt Lake City, were paying in the range of 1.32/gal more to 0.52/gal less than they had been 2 weeks before the season began.
Table 3 shows national average gasoline prices obtained from an EIA survey. Two points stand out from this table. First, a month before the season started in most areas, prices in nonattainment areas were already 7.5/gal higher than in attainment areas.
There are numerous reasons for regional price variations, including variations in taxes, distribution costs, seasonal demand, and service-station labor, rental, and utilities costs. Thus, differences in gasoline prices between attainment and nonattainment areas can only be attributed to the cost of oxygenates to the extent that it exceeds the underlying baseline difference.
For example, on Jan. 25 1993, prices in nonattainment areas were 116.0/gal, or 13.8/gal higher than in attainment areas. But only 6.3/gal (13.8 - 7.5) could be attributed to the incremental cost of oxygenates if the underlying baseline difference remained at 7.5/gal.
Regional variations in oxygenated gasoline price effects are difficult to determine because the sample size for the EIA survey was chosen to determine national averages. As a result, meaningful conclusions about PADDs can be inferred only for PADDs 1 and 2.
Also, PADD 5 is dominated by California, the whole of which was in control areas, so attainment vs. nonattainment comparisons would not be meaningful there. And finally, even within a PADD, the regional variations described previously can occur.
On Oct. 5, 1992, in PADD 1, the EIA survey showed a 3.5/gal higher retail gasoline price in nonattainment areas, as compared to attainment areas. When the oxygenated gasoline season began, the price difference widened to 8.2/gal on Nov. 16 and 9.7/gal on Dec. 28. During 1991, when there were no oxygenate requirements, the price difference between these areas increased an additional 2/gal between October and December (OGJ-CPC gasoline prices, published weekly).
Taking into account this baseline price difference, it would appear that the incremental difference attributable to oxygenate costs is 3-5/gal. A 3-5/gal differential over baseline also was found to hold true in PADD 2.
The second important point raised by Table 3 is that the prices of all types of gasoline steadily declined during the control period. This decrease helped cushion the impact on consumers of this first oxygenation season.
Declining gasoline prices resulted, in part, because of decreasing crude oil prices. The refiners' composite acquisition price for crude went from $19.49/bbl in October 1992, to $17.10/bbl in January 1993. Crude prices then increased slightly to $18.04 in March.
Oxygenated gasoline prices followed the crude price trend, falling from a high of about 121/gal, on average, in November, to 114/gal at the end of February, to 117/gal in March (Table 3).
Because gasoline prices normally increase during the summer driving season, a seasonal decline in fourth quarter prices is expected, even without a decrease in crude prices. Gasoline prices, however, increased very little in the summer of 1992, so less decline would have been expected in the fourth quarter.
As crude prices fell, the concerns over oxygenate supplies at the beginning of the season may have helped to bolster average retail gasoline prices in both attainment and nonattainment areas. When supply worries disappeared by the end of the year, gasoline prices, both wholesale and retail, drifted downward with crude prices.
On balance, the supply concerns may have given gasoline marketers a few cents of added margin above normal fourth quarter levels, helping to compensate for the high-priced oxygenates purchased on the contract market to build stocks before the season.
REFINERY ECONOMICS
The EIA used refinery models to explore whether refiners were able to pass the incremental costs of oxygenating gasoline along to consumers.
Because oxygenates serve the purpose of enhancing gasoline octane, in addition to adding oxygen, they can replace some high-octane components that would otherwise be blended in the gasoline. The materials being replaced are of higher value than the average finished gasoline. Thus, high-cost oxygenates are essentially substituted for other high-cost materials.
In addition, adding oxygenates increases yields over conventional gasoline production. As a result, the incremental cost to produce oxygenated gasoline is less than the cost of simply replacing gasoline volume with oxygenate volume.
EIA's models showed that the impact of the octane contribution on incremental cost varies with the volume proportion of oxygenated fuel produced by the refiner.
When the percentage of oxygenated gasoline produced is low, the refiner has the greatest latitude to adjust operations to take advantage of the oxygenates' octane contribution while meeting the oxygen content requirement. Thus, for refiners producing low volume percentages of oxygenated gasoline, the incremental cost of producing oxygenated gasoline is lowest.
On the other hand, when the percentage of oxygenated gasoline produced is high (say, more than 70%), the gasoline pool is awash with octane and the refiner cannot compensate much further. At this oxygenate level, the incremental cost of producing oxygenated gasoline approaches the incremental cost of simply substituting oxygenate volume for gasoline volume. That is to say, there is little compensating cost reduction from the oxygenates' octane-enhancement capability.
Fig. 6 shows a typical refiner's incremental cost of producing oxygenated gasoline, using both MTBE and ethanol, as compared to conventional gasoline. The figure illustrates the fact that producing the first barrel of oxygenated gasoline entails the lowest incremental cost.
Fig. 6a shows that, if MTBE were 65/gal, the refiner would incur no additional cost over conventional gasoline cost, to produce the first barrel of oxygenated gasoline. And at an MTBE price of 83/gal, the refiner's additional cost is just more than 2/gal.
The slope of the MTBE cost curve (Fig. 6a) is steeper than that of ethanol because MTBE's lower oxygen content means that about twice as much volume has to be added, as compared to ethanol, thus increasing octane proportionally.
Fig. 6 can be used to estimate the cost of producing oxygenated gasoline this past season. PADD 1 refiners, who supplied a large number of control areas, produced an average of 45 vol % oxygenated gasoline (a range of 20-70%).
The actual oxygenate cost to refiners, however, is unclear.
If a refiner paid, on average, 85-100/gal for MTBE, the incremental oxygenated gasoline cost was 2.5-6.1/gal. This range for PADD 1 implies that, as a group, refiners were probably able to pass through most of their increased costs (3-5/gal).
Individual refiners, however, may have experienced different results. For example, a refiner that contracted most of its MTBE supply at 100/gal or more would have seen lower margins on oxygenated gasoline than on conventional gasoline. A refiner with more favorable MTBE cost terms and a need to produce lower percentages of oxygenated gasoline may have increased its margins.
OUTLOOK
Because the 1993-94 season will differ from the first in terms of participating areas, control periods, and possibly demand, refiners cannot relax.
An investigation of what is needed for the new season requires an explanation of the overall gasoline market, as well as specific oxygenated gasoline market factors.
This past winter, the U.S. experienced higher distillate demand than in recent years.
Because of this increase, more gasoline was produced. In addition, oxygenates were being added to the gasoline, contributing another 250,000 b/d to gasoline supply.
The consequence was significantly higher gasoline stocks going into the 1993 summer gasoline season than have been seen for at least 5 years. The high stocks should lessen the likelihood of tight supply in the summer and dampen prospects for strong summer prices.
All gasoline prices fell throughout the first oxygenation season, in part because of declining crude prices. It is unlikely that crude prices will increase substantially in the upcoming winter season.
Crude price declines, however, may not occur to cushion the oxygenated gasoline price increases as they did this year.
The main change in the new season will be higher overall oxygenated gasoline demand.
The two primary reasons for this increase are the earlier participation of some areas (beginning Oct. 1) and the probable inclusion of Boston, Memphis, and Duluth.
Uncertainty still exists over the oxygen content that will be required in California, and over which attainment areas will choose to participate.
Inventory levels necessary at the end of August to meet upcoming demand can be estimated by assuming some production level and approximating the new higher MTBE demands.
If overall gasoline consumption remains about the same, the additional participants will increase seasonal MTBE demand by about 20,000 b/d. Further, if new capacity comes on stream as planned, production averages 150,000 b/d, and imports average about 16,000 b/d, end-of-August inventories should have been about 23 million bbl. This is the same inventory level achieved last year at the end of September.
This inventory could be achieved only if preseason production plus imports averaged 140,000 b/d and baseline MTBE demand is about 50,000 b/d. This leaves about 90,000 b/d, or 2.8 million bbl/month, to be added to inventory. Because MTBE inventories at the end of July were about 16 million bbl, the excess supply over consumption appears adequate to meet the inventory requirements.
PRICES
Prices also may be different for MTBE this winter. MTBE prices weakened this past year as the MTBE oversupply situation unfolded.
Prudent refiners that committed early to large contract coverage paid more for their MTBE than those that waited and relied more on spot markets. This year, there is less uncertainty about supply capability - particularly storage availability - and refiners will probably be less conservative in their supply arrangements.
This behavior could result in a tighter supply situation and higher prices. Exactly how much higher, however, is still unclear.
Presumably, if no shortages develop, spot prices will return to values closer to the contract values seen earlier this past season, or about 100/gal.
REFERENCES
- Energy Information Administration, Monthly Energy Review, DOE/EIA-0035 (92/08), August 1992, Washington, D.C., pp. 2-5.
- Oil Market Listener, "Oxygenate Overhang Thwarts Ability of U.S. Refiners to Recoup Production Costs," Mar. 19, 1993, pp. 1-3.
- Energy Information Administration, Petroleum Supply Annual 1992, and Petroleum Supply Monthly, various months, 1993.
- Octane Week, New MTBE Contracts to Factor in Oxygen Value," Vol. 35, Jan. 21, 1991, p. 1ff.
- Platt's Oilgram Price Report, Energy Information Administration, Octane Week, various issues, 1992-1993.
- Octane Week, "Oxygenate Pricing Heading Toward Marginal Costs of Production," Vol. 36, Jan. 25, 1993, p. 1ff.
- Luhdberg Letter, "How Did Oxy Mandates Affect Prices?," XIX:23, Dec. 11, 1992.
Copyright 1993 Oil & Gas Journal. All Rights Reserved.