MANAGEMENT PERSPECTIVE NEW TECHNOLOGY FOR GAS FINDING: HOW IMPORTANT HAS IT BEEN?

Phillip A. Ellis
Vice-President
Booz Allen & Hamilton
Dallas

This is the second of occasional articles on petroleum management issues. The first, on gas price forecasts, appeared in OGJ, Sept. 2, p. 44.

Finding costs for natural gas in the U.S. were more than halved in real terms from 94 cents/Mcf in 1983 to 44 cents in 1989.

A number of industry analysts and oil company executives recently have suggested that technology improvements contributed significantly to this improvement.

This claim is an especially important one because its proponents often use it to support the view that the gas glut in the U.S. will persist. In this view, new technology will push finding costs lower and help sustain deliverability, further prolonging the bubble.

Managers who are convinced of this position will want to invest in the people, software, and equipment that is supposed to be behind this improved performance. But they also are more likely to defer drilling and acquisitions of companies or producing leases to avoid adding to the supply overhang.

So how important was new technology in this trend? And how likely is it that technology will keep finding costs down?

TECHNOLOGY'S ROLE

Our view is that application of new or improved technology has been very important in specific circumstances, such as horizontal drilling in the Austin chalk and elsewhere, 3-D seismic and its role in reading structural geology in the Gulf of Mexico, improvements in fracturing tight sands, and elsewhere.

Technology has not been the leading factor, however, in driving down finding costs in the U.S. as a whole during the 1980s.

Instead, the main reason for declining U.S. finding costs has been the deflationary impact of failing gas prices on well costs. As fewer wells were drilled, the service industry cut its prices, people and pipe became less expensive, and costs per well declined.

The second most important factor has been highgrading. That is, as drilling outlays dried up in the mid and late 1980s, oil and gas companies fell back on their best prospects. As they did so, reserves per well increased, and finding costs declined.

Technology and other factors, such as better teamwork among disciplines and other organizational improvements, had a smaller but probably a still important effect on declining finding costs.

BEHIND THE CONCLUSIONS

How did we arrive at those conclusions?

First, it is clear that finding costs are very closely tied to the number of wells drilled in a given year (Fig. 1). There is a close correlation between well numbers and finding costs during the past two decades, a relationship that holds in times of rising and falling prices.

While the data cover gas wells, the analysis also holds for oil wells and therefore for all wells in general.

In 1983, 14,505 gas wells were drilled, and finding costs were 94 cents/Mcf. In 1989, 9,708 gas wells were drilled, and finding costs were 44 cents/Mcf. Both costs are in 1989 cents.

The number of wells drilled drives the two elements of the finding cost calculation: cost per well and reserves per well.

As the number of wells falls, the cost per well declines due, again, to deflation of the well's cost components (Fig. 2).

This analysis suggests that the number of wells drilled--and therefore the strength or weakness of the drilling service industry--explains 78% of the variation in cost per well.

Feet drilled per well explains another 6%, leaving 16% or so to other factors.

We tested the effect of improved drilling technologies and methods on finding costs by using feet drilled per rig-day but found no correlation.

At the same time, the number of wells also explains most of the variation over time in reserves discovered per well: 53% of the variation in one analysis (Fig. 3).

This can be described as the effect of highgrading. More reserves were discovered per well because fewer wells--the better prospects--were drilled, and they were drilled by more successful companies.

The rest of the variation came from other factors, including difficulties with the data, such as the timing of reserve bookings, technology improvements, and the like.

How much of the decline in finding costs can we attribute to reduced well costs, highgrading, and technology and other factors?

The reduction in cost per well during 1983-89 accounted for 62% of the decline (table), while the increase in find size per well caused 38% of the improvement.

The data also show that 52% of the fall was due to deflation in the oil industry, 20% to highgrading, and 28% to other factors, including technology.

For example, because the decrease in drilling, measured by the number of wells, accounts for 53% of the increase in find size, 20% (.38 x .53) of the decline in finding cost was driven by high-grading prospects.

The remaining 18% of the fall in finding costs--about 9 cents/Mcf--is not accounted for by the analysis we have done to date and can be attributed to technology improvements and other factors.

MAKING DECISIONS

What are the implications for company decision-making in the 1990s?

First, a company can do nothing to affect overall gas prices, and gas prices drive the number of wells drilled, which drives cost per well.

Second, nothing beats a good prospect--and the more you have to choose from the better off you will be. Maximizing your internal and external opportunity flow is critical so you have more opportunities from which to set priorities.

Third, technology may be an important factor, but it has accounted for no more than a 14 cents improvement in overall finding costs. That is not to say its importance will not increase or that it is not the critical factor on specific prospects or plays-and 14 cents/Mcf is important.

In addition, technology is one of the few factors management can influence. Highgrading and organizational improvement are the others.

Finally, these conclusions suggest that replacing reserves in the U.S. is an uphill battle. We evidently are able to generate a limited number of good prospects at a given time. When we push beyond that number, basic geology bites us.

(Thank you to Jon Rasich, who was the key analyst for this article.)