The issue on the economics of methane hydrate can be viewed from several perspectives. One approach pertains to economic calculations in terms of technical costs for exploration, wells, facilities, transport etc, and revenues for produced and marketed gas. This approach will only be dealt with briefly, because to date there are no real-world examples of marketed hydrate gas and hence all calculations must be based on speculative data. The emphasis here is rather on taking a broad-brush, long-term look at the economic justification of spending money on research now, in the hope of realizing future profits.

Several factors currently appear to preclude commercial exploitation of natural methane hydrate. In the first place, it is not known what the best locations are in technical terms like saturation and areal extent. This is partly due to the immaturity of surveying techniques for methane hydrate (including their lack of quantitativeness), and partly to the fact that few dedicated surveys using equipment for hydrate imaging (e.g. Max and Miles, 1999; Miles and Max, 1999) have been carried out to date.

Technically, gas could be produced from "solid" natural methane hydrate occurrences by thermal stimulation (possibly in combination with depressurization), either by using geothermal brine, or by injecting artificially heated hot water or steam. However, with reasonable assumptions on field size and quality, the total gas production costs would range between US$12 and US$30 per barrel of oil equivalent. Hence, this would essentially be a loss-making enterprise. On the other hand, with the current state of technology, the production of free methane from below a hydrate occurrence could probably be done profitably; however, it would not be as profitable as conventional gas production in otherwise similar conditions. This was concluded from a proprietary study by Shell, in which cost engineering and dynamic reservoir models were run on hypothetical, though not unrealistic, natural methane hydrate reservoirs, offshore a large country with little indigenous hydrocarbon resources. More elaborate examples of such calculations can be found in e.g. Kuuskraa (1985) and Godbole and Ehlig-Economides (1985).

In view of the fact that world-wide, large sums of governmental money are spent on research for methane hydrate production, oil and gas companies have addressed the question as to how much budget (if any) they should currently devote to this area. The answer would depend on the view of such a company on uncertainties such as the timing of methane hydrate becoming commercially viable, the size of the potential benefit, etc.

To quantify in economic terms the effects of these uncertainties on the appropriate spending level for the next few (10) years, an analysis such as the following could be useful. This analysis would be most appropriate for a major oil / gas company that normally holds the resources that it produces. It is based on a number of "base case" assumptions that have been listed in Table 1 below. Naturally, these assumptions are very much debatable. The cost and statistically most probable revenues have been calculated (using a discount rate of 10%/year) for three possible spending levels in the period prior to the commercial viability of methane hydrate production, i.e.:

Probably, by the time that methane hydrate becomes commercially viable, one would want to increase the research effort. Therefore, each of the options includes a "mandatory" step-up effort ofUS$ 1,000,000 and US$2,000,000 in the year before, and the year of commercial viability, respectively.

Number of years (from 2000) before methane hydrate will begin to be commercially viable (years) |
10, 20 or 30 |

Chance of hydrate being commercially feasible in 2010 |
7.0% |

Chance of hydrate being commercially feasible in 2020 (cumulative) |
21.0% |

Chance of hydrate being commercially feasible in 2030 (cumulative) |
27.0% |

Chance of the company to be the first if spending US$ 0 pa, including the probability of an adequate step-up timing |
1.0% |

Chance of the company to be the first if spending US$ 100,000 pa, including the probability of an adequate step-up timing |
5.0% |

Chance of the company to be the first if spending US$ 500,000 pa, including the probability of an adequate step-up timing |
25.0% |

Company share of hydrate gas in the first 10 years if they are NOT the first |
3.0% |

Company share of hydrate gas in the first 10 years if they ARE the first |
6.0% |

Period that competitive advantage of being the first to be ready fully lasts before falling back to zero (years) |
10 |

Annual increase in market share of hydrate gas in the first 10 years of commercial viability (fraction of total gas production) |
0.50% |

%age growth of worldwide gas consumption (equivalent to trebling in 30 years) |
3.73% |

Net revenue for produced gas per barrel of oil equivalent (US$ MOTD) |
1.00 |

Average annual discount rate |
10% |

Table 1. Base-case assumptions for the economic analysis

Table 1. Base-case assumptions for the economic analysis

In addition to the assumptions in Table 1, for simplicity it has been assumed that:

• Decisions on spending levels pertain to blocks of 10 years.

• Hydrate become commercially viable in a year ending in 0, i.e. 2010, 2020 or 2030. (Obviously, it is possible that hydrate hasn't yet become commercial by 2030, but the analysis has shown that in that case, from a commercial point of view one should certainly not invest money now.) The assumed probabilities in the table have been derived from the offshore curve in Fig. 3.

• The chance of hydrate becoming commercial at any given moment does not depend on the activity level of the company concerned. It will mostly be determined by the market price of gas (driven by supply-demand balance), and by the influence on methane hydrate production cost of technological advances originating from governmentally-paid research programs.

• The probability of success of the company concerned is only determined by what they did in the most recent 10 years, not by what they did before that period. (The probability of success is defined as the probability that hydrate becomes commercial at a certain time, multiplied by the probability that the company can determine this moment far enough ahead to step-up its hydrate research activities in a timely manner and reap the rewards for its efforts).

The possible branches from decision points in the decision tree are indicated with uppercase letters, as in Fig. 4. In this figure, the percentages quoted correspond to the base case assumptions as in Table 1. Note that the probabilities of commercial feasibility are stated in a non-cumulative way in this figure, as opposed to the table.

It is assumed that the company will make some money (i.e. 3% market share, see Table 1) from hydrate methane even if they make no research effort now, because at some time in the future there will be an industry-wide capability available to produce methane from hydrate. The additional value for the company is in the possibility of making more money than their competitors by "being the first" (although this strategy also carries greater risk than being the "first to follow"). The additional revenues generated in this way can be accredited to a prior research investment and are hereafter referred to as "revenues".

The potential revenues of each particular path through the decision tree are calculated and then multiplied with its probability of success, and the appropriate costs are subtracted. That way, it is possible to arrive at the Net Present Value for each of the possible decision paths.

From the detailed calculations for above base case, which are not presented here, it is very clear that from a commercial point of view, the best option at this moment is to invest no money at all in hydrate production research (decision path CFI in Fig. 4). This is mainly caused by the relatively low probability of success estimates.

Figure 4. Decision tree showing the possible branches for which cost-revenue calculations were made.

Remarkably, all Net Present Values appear to be negative; i.e. the company would lose money even if the spending level were US$0 per annum. This is, in a way, an artifact of the algorithm that requires a step-up in spending level towards the end of the considered term, as indicated earlier. If the "mandatory" step up in spending level would be omitted, cost and revenues for the path CFI would both be zero, as would therefore the Net Present Value. If we would add more 10-year blocks to the calculation (which was not done to prevent the calculation from becoming overly complicated), the preference for zero-spending mode would probably extend more into the future. So while there is indeed some artificiality in the way the calculations have been set up, the conclusion is robust.

The calculations were carried out in a way that is convenient for perturbation analysis. It appeared possible to tune the input assumptions such that spending US$ 100,000 pa in the period 2000-2010 comes out as the best option within the decision tree. This could be achieved by various sets of assumptions; e.g. when making all of the following changes:

• Chance of hydrate being commercially feasible in 2010, 2020 and 2030 would be 10%, 20% and 30%, respectively;

• Chance of the company to be the first if spending US$ 100,000 and US$ 500,000 would be 20% and 35%, respectively;

• Period that competitive advantage lasts would be 15 years;

• Net revenue for produced gas would be US$ 2 per barrel of oil equivalent.

This set of assumptions can still be viewed as realistic, and leads to a (positive) Net Present Value of US$ 538,000 for option B (higher than any other). Hence, one might be tempted to conclude that this option, i.e. spending US$ 100,000 pa is reasonably justifiable. It must be noted, though, that simply "making money" is not enough. A company would have to weigh this profitgenerating option against other investment opportunities, some of which may be in non-energy related areas, that may yield a better return per dollar invested.

It appeared to be very difficult to make investing US$ 500,000 pa from 2000 the best option by tuning the base case assumptions. The only way to achieve this by tuning a single assumption is to change the "net revenue for produced gas per barrel of oil equivalent" to more than US$69. Obviously, this is utterly unrealistic because oil at that price would have a serious impact upon the world economy and change the setting of the decision process. Also, simultaneous tuning of several assumptions did not lead to a reasonably realistic set of assumptions that would justify spending US$ 500,000 pa. From this it can also be concluded that spending more than US$ 500,000 pa from 2000 would be even less feasible.

All of the foregoing analysis would be most appropriate for relatively large oil or gas companies. A smaller, maybe more entrepreneurial or risk-taking type of company would have a different perspective:

• They probably could move quicker, possibly by taking up a consultant or advisory role.

• They would typically have no ownership of resources, and would run fewer risks because their costs would be recovered as a "fee for service".

• In some countries (including the US) tax laws would allow to fiscally write off research and development costs.

• In some countries tax laws allow distributions to directors of some companies to be free of tax if the research and development was carried out in that country (e.g. Ireland).

• Government provided funding for methane hydrate research in the US and elsewhere may also flow to private companies.

It can be concluded that excellent commercial opportunities may exist for a company that can develop technology that fills a niche or is otherwise a (potential) key component of hydrate recovery schemes. Even though the large oil and gas companies of this world may be reluctant to invest now, they definitely have an interest in the fuel of the future.

Was this article helpful?

## Post a comment