Economics favor GTL projects with condensate coproduction

GTL OUTLOOK-Conclusion

Michael J. Corke
Purvin & Gertz Inc.
London

A moderate scale (20,000 b/d) gas-to-liquids (GTL) project can have viable economics if condensate revenues are credited to the project.

Many reserves contain valuable condensates and other liquids as well as natural gas. Without such coproduction, near-to-medium term economics are challenging.

A few projects of this size will pave the way for larger projects that benefit from economies of scale. The larger projects will have viable economics without the benefits of liquid coproduction. Liquid coproduction gives the larger projects even stronger economics.

These economics and further technology developments create strong prospects for a GTL industry. The most difficult challenge is taking the first steps. Flexibility will be needed on the part of host governments with regard to feed-gas pricing, the allocation of revenues from coproduced liquids, and tax and royalty questions. Project sponsors may have to accept greater risks and lower returns than they ideally wish. The prize for success, however, is substantial.

The first part of this two-part series (OGJ. Sept. 21, 1998, pp. 71) reviewed the development of GTL technology and the status of existing and proposed GTL projects. This second part considers the opportunities for new GTL projects based on economics. The availability of suitable gas resources and the drivers for GTL developments are discussed first.

Opportunities for GTL projects

World proven reserves of natural gas grew from less than 50 trillion cu m (tcm) in 1971 to about 145 tcm at the end of 1997-an average annual growth of 4%/year. This increase occurred despite an increasing demand for gas.

Of current reserves, nearly three-quarters are found in the Former Soviet Union (FSU) and the Middle East, often at great distances from where the world's major demand is found. The remaining reserves are distributed among Organization for Economic Cooperation & Development (OECD) countries and the rest of the world. A summary of the distribution of the world's gas reserves at the end of 1997 is shown in Fig. 1 [104,748 bytes].

As noted above, remaining reserves at the end of 1997 were estimated at about 145 tcm. World production in 1997 was about 2.3 tcm. Thus, the overall reserve/production ratio is about 64 years.

Although consumption in all regions is expected to increase, reserves are also expected to rise. Thus, the huge overhang of resources in many regions is likely to persist (Fig. 2 [90,712 bytes]).

Conventional gas dispositions

In many countries, major incentives exist to produce more gas, but this has frequently been frustrated by the lack of a market outlet at an economic price.

To date, with a few exceptions, gas transportation has taken place via pipeline or via the production and bulk shipping of liquefied natural gas (LNG).

Gas pipelines run from producing areas to consuming areas. The end users typically fall into one of three categories: industrial, domestic/commercial, or power generation. In each of these market sectors, gas is mainly used as a heating fuel. The same end users also receive regasified LNG when this transportation route is more economic or appropriate.

One of the main difficulties faced by LNG project developers is the large investment required for liquefaction, shipping, off-loading, and regasification. In the past, large national buyers from countries such as Japan, South Korea, and Taiwan have taken the entire production from a train of LNG. Although this is still possible, many LNG buyers are now looking for smaller supply tranches.

Buyers are also increasingly reluctant to enter into long-term volume and price agreements, fearing that increasing competition may leave them exposed. This is making the development of greenfield LNG projects more difficult.

Large-scale, long-distance pipeline projects involve large investments and can be difficult to mount. To achieve economies of scale, large volumes must be transported, which creates a need for sales to a number of buyers. Transit issues may also create political or commercial barriers. Similar to LNG projects, however, the ultimate barrier to the success of a major pipeline project may be simply that the market does not need additional supplies when the producer wishes to develop them.

Alternative gas dispositions

In areas where there are large gas reserves that do not have markets, alternative methods of utilization have to be considered. Gas may be used to produce ammonia, urea, methanol, or liquid hydrocarbons.

Gas used as a petrochemical feedstock, particularly for the production of ammonia or urea, must generally be sold at a low price so that the final product can compete in world markets. For example, fertilizer products are available at low prices from a number of sources, particularly from the FSU and the Middle East. Perhaps more importantly, the sizes of these markets are limited, which limits the opportunity for using natural gas in this way.

Gas can also be used as a feedstock to produce methanol to be sold on world markets. Again, for this to be successful, low prices for feed gas are necessary. The methanol market is also relatively small and thus few opportunities for new plants exist.

The limited natural gas fuel and petrochemical feedstock markets limit gas developments. In this scenario, GTL technology may be a viable alternative for monetising stranded gas reserves.

Just as for LNG and petrochemical projects, GTL projects require competitively priced gas for a commercially viable plant. GTL products, however, can be sold in the huge international crude/refined oil markets.

World crude oil production is about 70 million b/d, and commercial GTL plants will likely produce up to 50,000 b/d each. Thus, many GTL plants would have to be built to have a significant impact on world petroleum markets.

GTL project economics

As noted, stranded gas reserves are widely available, and the technology exists to convert them into marketable liquid hydrocarbons. Historically, the barrier to GTL development has been economic. It now appears that GTL conversion may be economic in favorable circumstances.

The economics of a representative commercial scale GTL project have been evaluated using discounted cash flow analyses. A range of cases have been evaluated. The principle variables are investment cost, feed-gas price, and the production (or lack of production) of associated LPG or condensate. Sensitivity cases have also been run to ascertain the effect of variations in financing assumptions (debt/equity split, interest rate, discount rate) and operating costs.

Investment and operating costs

Project investment costs have been estimated based on data published in industry journals and discussions with industry representatives. The information used includes information published by Bechtel Corp. based on a study sponsored by the U.S. Department of Energy (DOE).

The Bechtel study gives an overall investment cost of about $38,000 per b/d for a 45,000 b/d plant in a U.S. location in 1993. The facilities included syngas production, F-T conversion, and full product work-up. The technical basis of this work is conservative in the light of recent developments.

Both Sasol and Shell have recently indicated that an investment cost of around $30,000 per b/d for a 20,000 b/d facility should be achievable, which appears reasonable given the technology developments of recent years described in the first part of this series. Exxon and others have indicated a lower unit cost for larger facilities. For this analysis, the above cost of $30,000 per b/d has been used as the base case investment cost. Sensitivities to variations in investment costs have also been checked. Comparative investment costs are illustrated in Fig. 3 [73,754 bytes].

Economies of scale are important in this evaluation. A unit investment cost of $30,000 per b/d at 20,000 b/d capacity would be reduced to about $22,000 per b/d at a capacity of 50,000 b/d and about $17,000 per b/d at a capacity of 100,000 b/d. Even allowing for additional costs associated with construction in remote locations, high levels of limited recourse debt finance, or the risk of reduced economies of scale, investment costs at or below the $30,000 per b/d should be achievable for future larger-scale projects.

Operating costs are expected to be similar to those for comparable oil refinery or petrochemical plant operations. Costs accounted for include costs for labor, maintenance, rates/insurance/taxes, general, and administration.

Operating life

The project has been assessed based on an operating life of 25 years. The investment costs have been spread over a 4-year construction period, based on a schedule similar to that found for the construction of a grassroots oil refinery or petrochemical complex. No allowance has been given for the residual value of this project after 25 years of operation. The value of future cash flows beyond this period, however, is generally minor unless a very low discount rate is applied in the calculation.

Feed-gas price

The main materials delivered to the plant are natural gas, raw water, catalysts, chemicals, and normal butane. Normal butane is used as a feed for the C4 isomerization and subsequent alkylation processes which form part of the product work-up section of the plant.

Specifically, a plant designed to produce 20,000 b/d of liquid products would use about 180 MMscfd of natural gas, 10 million gal/day of raw water, and 150 b/d of normal butane. Excess steam, a by-product, can be used in a steam turbine to produce approximately 10 megawatts of electricity.

For gas projects which are entirely owned by a producing state, the state may be content to receive the cash flow which represents an acceptable rate of return on the project, particularly in the early years. Where foreign equity participation is involved, however, the state will typically wish to see some positive value for the gas extracted from the ground. In the cases studied here, gas prices to the GTL plant of from $0.50 to $2.00/million BTU have been used.

A feed gas price of $0.50/million BTU is broadly in line with recent Middle East LNG and other project experience.

This price may seem inadequate to cover the investment and operating costs of providing the gas; that is, costs associated with exploration, field development and production, gas gathering and processing, and gas transportation to the plant, even without taking into account a separate resource gas value or royalty.

In many cases, however, substantial volumes of LPG and/or condensate are produced in association with the gas, and the implicit philosophy is that the costs of delivering the gas to the plant are covered by the revenues from associated liquids sales. An alternative approach would be to charge a higher feed-gas price, but to include revenues associated with the sale of LPG and/or condensate within the GTL project revenues.

Products markets

The product mix for the representative GTL plant analyzed, which reflects the results of the above-referenced Bechtel study, is primarily made of gasoline and diesel-blending stocks. In addition, some LPG is produced. A 20,000 b/d unit would produce about 7,500 b/d of gasoline, 11,600 b/d of diesel, and 750 b/d of propane.

The yields for any project will reflect its design basis. A plant directed towards distillate production and with limited product work-up complexity could yield about 80% diesel fuel with most of the remaining production as naphtha.

The main products of the GTL plant, gasoline and diesel, are environmentally superior to existing products because they have no sulfur associated with them. The diesel would also have a very high cetane index, making it a desirable blendstock.

The market for these products will be large and diverse. Demand for products from the GTL plant would be particularly strong in areas where environmental emissions are a major consideration. Currently, this would apply to most of North America, Europe, Japan, and Australasia. It is likely, however, to also be a major concern for other developing areas.

Transportation of GTL products would be straightforward, as no special shipping, storage, or handling facilities would be required. For remote GTL projects, it may be more attractive to export syncrude and avoid the investments and other costs required for product work-up.

Products pricing

The products produced from the modeled GTL plant were valued as mainstream refined products. In practice, some price premia would be expected, reflecting their superior blending quantities. Opportunities may also exist to obtain higher prices by selling into niche markets or by conversion to specialty products.

The product prices were related to market crude oil prices using Purvin & Gertz' price analysis and forecasting methodology to express the product prices in terms of the equivalent marker crude oil prices. In this way, the crude oil price levels needed to generate certain rates of return were established.

Project financing

For the purposes of this analysis, it has been assumed that a commercial-scale GTL plant would be built using limited-recourse project finance. In view of the limited current development of the GTL industry, and in particular, the absence of experience of commercial scale slurry bed F-T reactors, lenders would probably view the project as involving a significant technology risk.

Market risk, however, is less significant if the plant's products can be sold into the worldwide petroleum market if no better opportunity exists. This balance between technology and market risk is the reverse of that generally perceived for LNG projects. It seems likely, however, that around 70% loan finance would be available for a project involving sound sponsors and correspondingly weighty guarantees.

In the economic analysis, a loan interest rate of 8% has been assumed, and a 10-year principal repayment period commences immediately following start-up. In reality, it is likely that various tranches of debt with various repayment periods would be used, including export-credit agency-backed loans, other commercial loans, and bond funding.

Economic analysis

A summary of the main parameters used in the base case, cash-flow model is shown in Table 1 [35,400 bytes].

Taking all of the above factors into account, the base case economics give a crude oil equivalent (COE) figure of about $16/bbl expressed in terms of a Brent fob crude price. Thus, a 15% real after tax return for the defined GTL project should be achieved at a Brent crude oil fob price of $16/bbl.

To translate these prices into the prices of other marker crudes, for the 2000-2010 period, Dubai fob would be expected to average about $1.70/bbl below Brent fob, and WTI at Cushing would be expected to average about $1.10/bbl above Brent fob.

Variations to base case

A number of variations to the base case have been reviewed. The most important factors for determining the rate of return that can be achieved are the cost of feed gas and the capital investment. Variations in the resultant COE required to obtain a 15% real after tax return are illustrated by Fig. 4 [101,380 bytes].

During 1997, the Brent crude fob price averaged around $19/bbl, which means that a number of scenarios shown in the table above would have been feasible based on the assumptions in the base case economic model. Brent prices during the first half of 1998 have averaged below $14/bbl, a level which would clearly present more serious challenges to the viability of a GTL project.

While Purvin & Gertz expects market to remain fairly soft through several years as a result of the Asian financial and economic crises, burgeoning supply capacity, and other factors, price levels are expected to average over $16/bbl within 5 years or so.

Coproduction of liquids

Many of the undeveloped stranded gas reserves around the world are likely to have associated condensates and natural gas liquids. These liquids can only be produced if the gas is produced. They can make an important contribution to overall project revenues. Condensate and LPG are valuable commodities, and in general, condensate is sold at a premium to crude oil.

Considerable volumes of condensates are being produced or will be produced in connection with Qatar's LNG and other North field-based projects. Oman's LNG and other gas-consuming projects will also release substantial condensate flows.

Condensate revenues may contribute directly to a GTL project's revenues if the project "ring fence" is drawn widely. Even where the liquid revenues go to a state-owned production company rather than to the project itself, an indirect benefit to the project is available. As a quid pro quo for such revenues, the state might feel able to supply feed gas at a low price, to provide an extended tax holiday, or to otherwise assist the project.

The effect of condensate production on the representative GTL project economics has been evaluated based on the following assumptions:

Production rate of 5,000 b/d of condensate per 100 MMscfd of gas
Condensate priced at a premium vs. Brent fob of approximately $0.50/bbl
Condensate royalty (not available to enhance GTL project revenues) at 10% of production.

This additional income stream has been added to the base case model, and the same review of required COE price for a range of feed-gas price and capital cost cases has been carried out. The results of this analysis are illustrated in Fig. 5 [90,843 bytes].

Comparing the results of this analysis with the results of the base (dry gas) case shows that a reduction in the COE price of around $4/bbl can be achieved. This makes a GTL project in this situation very attractive. The base case investment ($30,000 per b/d) and feed gas price ($0.50/million BTU) results in a 15% rate of return even with the price of Brent crude fob at $12/bbl. The impact is even greater for cases in which the investment cost or the feed gas price is higher.

Of course, the results of the economic analysis are dependent on many factors in addition to the key factors shown. Variations (within reasonable ranges) in the amount of debt financing, in interest rates, and in operating costs would typically change the required COE by $1-2/bbl.

The discount rate applied in the analysis is also an important variable, with the $16/bbl Brent based COE for the base case ranging from $13-18/bbl for a discount range from 10 to 17.5%.