Tip of the Week #98 Tip Index

Go to the Prior Tip "Portfolio Management: What is the Contribution to Shareholder Value?"
Go to the Next Tip Neural Network Control System for Steam Injection
Return to MaxValue Home Page

Risk, Uncertainty and Investment Decision-Making in the Upstream Oil and Gas Industry

by Fiona Macmillan, Ph.D. Thesis, 2000, University of Aberdeen, Scotland, U.K.

I have been pleased to read a copy of Fiona Macmillan's doctoral thesis. She has done a splendid job of examining the evaluation practices in the upstream oil & gas industry and determining the association between evaluation practices and company performance.

She carefully designed her study around three Research Questions:

Which techniques are the most appropriate for companies to utilize in their investment decision-making? (literature search)
Which techniques do companies use to make investment decisions and how are they used in the investment decision-making process? (semi-structured interviews)
Is there a relationship between using decision analysis (DA) techniques in investment appraisal decision-making and good organizational performance? (statistical analysis).

Previously, I had attempted a limited analysis along in the area of the third question. I presented "Best Practices in Project Evaluation and Influence on Company Performance," at the SPE Hydrocarbon Economics and Evaluation Symposium, March 16-18, 1997 (Society of Petroleum Engineers, Dallas, SPE paper no. 37945; a revised version in appeared in JPT (formerly the Journal of Petroleum Technology), August 1997, pp. 818-823). As a decision analysis consultant and trainer, I wanted a way to demonstrate the value of decision analysis. Dr. Macmillan picked up on the idea and executed a scholarly investigation program . Where I had simply sent out a few questionnaires, she reviewed the literature, conducted semi-structured interview, and did more-rigorous statistical analysis.

Dr. Macmillan received tremendous cooperation and interest in response to her requests for interviews. She interviewed 27 of the 31 companies who were operators in the U.K.'s upstream oil and gas industry in March 1998. As most of these companies operate internationally, this is a very reasonable sample of the worldwide E&P industry.

Her statistical results are summarized in the following table. The R value is the Spearman correlation coefficient (0 means no correlation and +1 would be a perfect, positive correlation). The n value is the number of data pairs available.

ROE (Return on Equity) seems to me the best indication of company performance, and there is a significant 0.25 correlation.
PR (Proved Reserves) and NOE (Number of Employees) might better represent old successes since company size is largely the result of growth.
TBV (Total Base Value, a Wood Mackenzie assessment of company value, including exploration positions), PE (Price/Earnings Ratio), MC (Market Capitalization), and PSR (Prudential Securities' Ranking) include judgments about the companies' growth outlook. These indicators would reflect both historical and recent performance in the context of adding value.

PR	R=0.701, n=14
MC	R=0.538, n=13
TBV	R=0.655, n=16
NOE	R=0.3823, n=17
ROE	R=0.252, n=17
PE	R=0.296, n=13
PSR	R=0.6, n=9

The results are significant. For me, the key findings are:

A. Firms using decision analysis (DA) tend to be more profitable (evidenced by ROE and, perhaps, other criteria). We cannot conclude that using DA causes companies to be more profitable, on average. It may be that more-successful companies have more analysis resources and tend to use DA.

B. Larger firms are those most likely to be using the more-sophisticated analysis techniques. At the time of her study, however, "option, portfolio and preference theories are hardly used at all by any firm."

Her thesis reports that much of the acceptance and adoption of DA techniques had occurred in the five years prior to her interviews. Today, in a follow-on discussion, she says that change continues to be slow and progressive.

Popularity of Decision Criteria

Dr. Macmillan's thesis contains a summary of seven surveys of decision criteria (adapted from A. Buckley, 2000, Multinational Finance, 4th edition, Prentice Hall, England). These surveys date variously from 1980 to 1996. They included two (in my humble opinion) "junk" criteria: ARR (accounting rate of return) and PO (payout or payback period). Of course, the principle criteria are NPV (net present value) and IRR (internal rate of return).

What shocks and amazes me is that IRR was universally more popular than NPV. IRR has enough trouble with the multiple-roots problem and works poorly with probabilities. NPV, however, works well with probability-weighting in the expected value (EV) calculation. [EV NPV = EMV, expected monetary value. This is the leading decision criterion when using decision analysis.] I'm doing what I can to discredit the usefulness of IRR. If we must rank for a capital constraint, a better criterion is something like PI (profitability index) = EMV / EV Investment.

Fiona has been graciously making PDF file copies of her thesis available upon request. Contact her at:

She now works in economic modeling for Palantir Economic Solutions, Aberdeen, Scotland. http://www.palantirsolutions.com.

—John Schuyler, Nov. 2003.