POSITIVE NOTES APPEAR IN U.S. GAS PROCESSING OUTLOOK FOR THIS DECADE

R. R. Haun, E. E. Ellington, J. R. Hawes
Purvin & Gertz Inc.
Dallas

U.S. gas processors will experience more favorable times during the next 10 years than during 1988 and most of 1989. The recovery will be gradual, however, and seasonal changes in profitability will continue.

On the supply side, there is uncertainty as to the future volumes, compositions, and locations of natural gas available for processing. The effects of future volumes of LPG imports are also uncertain.

On the demand side, the future for n-butane is the subject of much debate. The extent to which ethane demand will be sufficient to justify its extraction is also debatable.

U.S. GAS SUPPLY

The total U.S. gas supply available for processing ("marketed production") will likely decline from about 17.6 tcf in 1989 to 16.2 tcf in 2000 (Fig. 1). The volume of gas processed will decline at a slightly greater rate, from about 13.5 tcf in 1989 to 11.9 tcf in 2000.

The greater decline in gas processed will result from a higher percentage of lean, unprocessable gas being produced by the end of the century. The gas processed will also become leaner, as discussed later.

In addition to declining supply, the location of gas production will likely shift somewhat (Fig. 2). Production in the major gas-producing states, Texas, Louisiana, and Oklahoma, will decline, while production in New Mexico, the north central region (primarily Kansas, North Dakota, and Michigan), and the Rocky Mountain region will likely increase.

The shifting production may cause some new plant construction because new production will not be available to existing plants with surplus capacity. The shifting gas production also may cause minor disruptions in NGL logistics.

NGL SUPPLY, DEMAND

During the 1986-1988 period, production of NGL from natural-gas processing plants increased about 5% as the total volume of gas processed went up. The volume of gas processed in 1989 was about the same as 1988, but NGL recovery dropped to 1986 levels.

The decline in NGL volume was due mainly to increased ethane rejection as ethane liquid prices were below residue-gas prices in many parts of the country.

A gradual increase in the total NGL volume recovered from gas plants should occur through 1992 (Table 1). A small drop in production is likely in 1995, but by 2000 a significant reduction will occur. By then the volume and quality of gas processed will have declined significantly from present levels.

Although the long-term forecast indicates that less natural gas will be processed in the U.S., there are expected to be areas (such as the Overthrust) that will see a significant increase in the volume of gas processed.

Purvin & Gertz' projection of NGL demand by component is shown in Table 2.

Ethane. The demand for ethane in the U.S. is primarily as feedstock for the production of ethylene. Due to the tremendous flexibility that exists at U.S. ethylene plants to consume a wide range of feedstocks, the projection of future ethane consumption is difficult.
Plants with flexibility base their feedstock decisions on price and by-product demand. If ethane is not priced competitively, then its demand will fall. This will affect gas processors' decisions concerning ethane rejection or recovery.
Demand for ethane through 1995 will likely firm. Beyond then, consumption is likely to decline as a result of a substantial decrease in availability of supply from U.S. gas plants.
This forecast assumes that about 20,000 b/d of ethane will be available from Canada in 2000. If Canada is able to supply more than 20,000 b/d in 2000, then ethane demand will increase accordingly.
The possibility of waterborne imports of ethane into the U.S. during the forecast period still creates some uncertainty. The gas processor with ethane-rejection capabilities will be best equipped to maximize income as demand (and therefore price) because ethane moves up and down with the fortunes of the ethylene producers.
Propane. As the largest component of NGL demand, propane is used in a number of applications including space heating, cooking, water heating, industrial furnaces, engine fuel, and chemical feed. Only the residential and commercial market and the chemical feed market show any signs of growth.
The residential and commercial market is growing at a slow rate for the U.S. in total. Demand in some portions of the country is growing at a healthy rate, while demand in other areas is either declining or stagnant. Total growth rate for the U.S. will average 0.5-1.5%/year throughout the forecast period.
The chemical feed market includes propane consumed in ethylene crackers. And it includes propane used in ethane/propane crackers that have little or no flexibility to crack other feedstocks as well as propane used in flexible crackers that purchase propane only if the price is attractive.
Significant growth in the chemical feed market is likely as new NGL-based ethylene plants are built to satisfy the increasing demand for ethylene. A continuing requirement to develop price-sensitive markets to clear available supplies may also have a significant impact on propane consumed as chemical feedstock.
The only other market for propane that seems to have growth potential is engine fuel. Although the current forecast assumes a stagnant market, there is a high potential for growth if the right conditions exist. As a general rule only high-mileage fleet vehicles can justify conversion to propane today, and many of these have been reluctant to convert.
In light of current developments regarding clean-air initiatives, it is possible that propane auto fuel could some day receive additional government incentives (or even mandates) for use. If this happens, growth in the market will occur.
As shown in Table 2, total demand for propane is expected to grow about 8% from 1990 to 1992 but only about 2% from 1992 to 2000.
Butane. The U.S. demand for butane is dominated by the refinery and chemical end-use sectors.
The refiner's net demand for n-butane is likely to decline in 1992 with the expected EPA-mandated Phase 11 Reid vapor pressure (Rvp) reduction for gasoline. This will offset the firm requirements for isobutane and should result in a decline in outside purchases of butane by refiners in 1992.
Continued growth in gasoline-blending demand and increasing alkylation requirements will likely result in increased demand by refiners for normal and isobutane, respectively, by 1995 and throughout the end of the century.
Growth in maleic anhydride, propylene oxide, and methyl tertiary butyl ether (MTBE) will result in an increase in the underlying base demand for butane as a petrochemical feedstock. The increased seasonal availability of domestically produced n-butane (as a result of the Rvp reduction) and emerging international butane supplies are also likely to result in the significant development of price-sensitive butane demand as olefin plant feedstock through at least 1995.
The emerging role of MTBE as a key component in U.S. reformulated gasoline will likely have a significant effect on world butane supply and demand balances by the mid-to-late 1990s.
Primary MTBE-supply potential from currently identifiable sources of butane will be sufficient to satisfy the minimum U.S. MTBE-demand outlook (3% MTBE in total U.S. gasoline pool by 2005; Fig. 3). New secondary sources of butane, however, will be required if MTBE usage is more than 5% of the pool.
At the maximum level being considered (15% MTBE in gasoline), new synthetic sources of butane would be required.
Natural gasoline. The demand for natural gasoline is almost totally dominated by refinery and chemical end-use markets. Refiners consume natural gasoline for the manufacture of motor gasoline, while chemical companies consume the material for the manufacture of ethylene.

Little change in ethylene plant demand for natural gasoline seems likely. Refinery demand should be sufficient to consume available remaining supplies from gas plants.

Future refinery demand is somewhat uncertain, however, in light of current developments regarding reformulated gasoline. If reformulated gasoline eventually makes a major impact on the motor-fuel market, then demand for natural gasoline as isomerization-unit feedstock could accelerate. If this happens, some ethylene plant demand could be shifted to refinery demand.

From a gas-processing standpoint, however, all natural gasoline produced is expected to have a market and the price (relative to motor gasoline) should improve over the forecast period.

INTERNATIONAL LPG

Historically, the U.S. natural-gas processing industry has been influenced by developments in the international LPG industry. Review of the outlook for the international LPG industry and its probable impact on the U.S. market is therefore appropriate before an examination of the economics of U.S. gas processing.

World LPG supply is expected to grow from 124 million tons (4.01 million b/d) in 1989 to 160 million tons (5.17 million b/d) by 2000. The most significant events affecting short-term LPG supply are increased Middle East crude-oil production and the emergence of Iraq as an exporter.

Middle East LPG export availability is likely to increase by about 35% by 1993 (Fig. 4). Further increases in crude production and the expected re-emergence of Iran in the world LPG trade will result in additional LPG exports from the Middle East during the 1993-2000 period. Increased production from Africa and Canada will also contribute to increased LPG availabilities from the major world exporters (Fig. 5).

While premium LPG demand growth will increase from 115 million tons (3.72 million b/d) in 1989 to 144 million tons (4.66 million b/d) by 2000, a continuing excess supply will require further development of price-sensitive petrochemical feedstock markets (Fig. 6). This situation is particularly true for propane.

Price-sensitive butane market development (as olefin plant feedstock) will also be required through 1995. Depending on the requirements for reformulated gasoline in the U.S. and possibly throughout the world, the current price-sensitive market developed for butane could be displaced by a significant increase in base demand as MTBE feedstock in the mid-to late-1990s.

Throughout the near-term, all major regions of the world (North America, Europe, and the Far East) will participate in the development of the price-sensitive markets.

Continued short-term shipping and commercial constraints are likely to result in the withholding of significant international LPG production (from 1.0 to 2.5 million tons/year or 32,000 to 80,000 b/d) from the world markets. Unacceptable netback pricing to producers may result in LPG products being flared, fueled, or reinjected in the short term. Shipping constraints, however, may start easing by 1993 as newly constructed shipping capacity enters the international market.

The new Middle East f.o.b.-term market LPG pricing policy, implemented in November 1989, will result in stable prices relative to crude oil in the Far East markets. This outlook is a dramatic change from historical price relationships and should sustain premium market-growth prospects in the Far East.

Significant discounts from firm f.o.b.-term market pricing levels are required to develop seasonal price-sensitive spot markets in the world including the U.S., which will likely play an increasing role in clearing marginal LPG supplies.

The U.S. LPG supply and demand balance likely through 2000 is presented in Table 3. Total U.S. imports by 2000 are likely to be more than double 1989 levels in order to meet future end-use demand as shown by component in Table 4. With Canadian LPG imports leveling out at about 145,000 b/d, an increasing requirement for waterborne imports is probable.

Future development of the international LPG industry will result in U.S. propane continuing to be priced relative to alternate feedstocks (primarily naphtha) through 2000. Through at least 1993, n-butane pricing in the U.S. is also expected to be influenced by alternate olefin-plant feedstock economics on a seasonal basis.

By the latter half of the 1990s, a firming of n-butane pricing is probable depending on the ultimate development of MTBE as a component in U.S. reformulated gasoline. Future isobutane pricing is expected to reflect isomerization economics to support increasing demand.

GAS-PROCESSING ECONOMICS

Purvin & Gertz follows the profitability (NGL revenue less fuel and shrinkage costs and operating expenses) of a hypothetical 50 MMcfd expander plant in West Texas.

This plant operates at capacity and processes a medium-quality gas stream (i.e., inlet composition of about 3 gal of ethane and heavier NGL per Mcf of inlet gas).

All NGL recovered from the gas is moved as a pipeline mix to Mont Belvieu, Tex., for fractionation. The plant receives the Mont Belvieu spot price for each component less a 4.5/gal transportation and fractionation fee.

The gas-plant operator receives all revenues from the sale of NGL and pays the producer for fuel and shrinkage at residue gas value (equal to West Texas spot natural-gas price). The operator must also cover plant-operating expenses. And the plant has ethane-rejection capabilities.

Purvin & Gertz began tracking the economics of this plant in 1987. As shown in Fig. 7, the profitability was very good in 1987, but lower NGL prices and higher fuel and shrinkage costs in 1988 and 1989 caused profitability to decline significantly.

The plant's profitability in 1988 and 1989 was roughly one third of the level in 1987. The large increase in NGL prices that occurred in December 1989 and January 1990 caused profitability to improve dramatically in those months to levels well above that in 1987. NGL prices have since declined, however, along with profitability.

Purvin & Gertz projects the overall profitability in 1990 to be better than in 1988 and 1989 but still only about 65% of the level in 1987. This improvement over 1988 and 1989 is based on a belief that NGL prices will increase more than natural-gas prices in West Texas in 1990.

It was economical for the reference plant to recover ethane during 5 months of 1989. Based on Purvin & Gertz' ethane and residue-gas price forecasts, ethane recovery will not be profitable for this plant in 1990.

Over the short to medium-term (1991-1993), profitability of this plant is likely to be slightly below the 1990 level but still well above the levels in 1988 and 1989.

Beyond 1993, profitability is projected gradually to improve, and by 1996 the plant should reach near-1987 profitability levels and then remain stable through 2000.

If this plant were located in the Midcontinent area and processed gas under the same terms as just described for its West Texas location (but sold NGL at Conway and paid for fuel and shrinkage at Midcontinent residue gas prices), the profitability trends would be similar to those described.

A little slower recovery in gas-processing profitability in the Overthrust area seems likely as a result of more rapid increase in natural-gas prices there compared with West Texas during the early years of the forecast. Between 1995 and 2000, however, the profitability should be favorable and at least as good as in 1987.

Ethane recovery in the Overthrust will likely not be profitable until the end of the forecast period; that assessment assumes a transportation and fractionation fee for delivery of NGL to Mont Belvieu of 9/gal (1990 dollars).