Prior studies of petroleum production costs have been limited to cost data averaged over several producing basins and, therefore, have not been able producing basins and, therefore, have not been able to associate the relevant geological and engineering determinants with discovery and production costs. With cost data developed at the field level, however, the production characteristics of these fields can be lined with the costs of lifting crude oil or producing natural gas. The cost data used in this producing natural gas. The cost data used in this analysis were gathered by special interviews with the individual respondent companies. As a caveat, it should be understood that these cost figures do not include the nonproductive costs, e.g., dry holes and central office management, because they cannot be associated directly with these fields even though such costs are necessary to the discovery and production of oil and gas. production of oil and gas.


The data utilized in this analysis are the finding and production cost data developed from company accounting records for representative leases in 56 fields within the defined area (see Table 1). The quality of the cost data was enhanced by three specific measures. First, a uniform set of cost element definitions and procedures were applied to standardize the measure costs across companies. Second, where appropriate, the historical costs were brought to a 1974 current basis so that depreciation and depletion charges would be comparable across companies. Third, data were gathered from a randomly selected set of leases. Fields within the Anadarko, Hugoton-Panhandle, and Frio basins were stratified by depth, size, and product, and randomly selected for the study; then leases within these representative fields were selected randomly. Consistent engineering and geological data also are available for each of these fields. The geological and engineering data drawn from the Petroleum Data Systems files at The U. of Oklahoma were compared with company records where possible.


In this study, we developed two separate cost equations. First, we specified a lease operating cost function that included geological and engineering factors that affect production costs as independent variables, and second, we specified a total cost function, including 1974 depreciation and depletion costs, with the same independent variables.

The variables hypothesized to affect oil and gas costs by field were the following:

  1. secondary recovery (SECR), a dummy variable that assumed the value of one for fields using secondary recovery methods, zero otherwise;

  2. number of wells (WELLS), a scalar to reduce the fields to a comparable size basis;

  3. median depth (LNDEPTH) of wells within the field, stated in the natural logarithm to accommodate a hypothesized nonlinear relationship between costs of production and depth;

  4. date of field discovery (DATE), to account for changing production and development techniques;

  5. (5–6) dummy variables to account for differences across the three geological basins in the study, the variable for the Hugoton-Panhandle basin (HUGO) and the Anadarko (ANADK) basin were assigned the value of one for fields in those basins making the comparison of cost relative to the Frio Trend area, the third area in the analysis;

  6. the hydrocarbon trap type (STST), where fields with traps classified as structural and stratigraphic were assigned the value of one;

  7. the output variable of natural gas and crude oil converted to equivalent barrels (EQBBL) on the basis of value;

  8. the square of equivalent barrels (EQBBL2), to test the hypothesis that there might be a nonlinear relationship between cost and output.


The results of lease operating cost (LOC) and total production cost (TPC) estimations are given below (t-statistics are in parentheses). These results were derived from a step-wise linear regression where all variables with a t-statistic greater than unity were allowed to enter. This criterion maximizes the explanatory power of the equation.

This content is only available via PDF.
You can access this article if you purchase or spend a download.