In the face of the current energy situation, we are aware of the need for accurate and quickly obtained information on refinery data and stocks and other energy information. New methods are currently under review in the Executive branch and we expect to forward our recommendations to the Congress shortly. Pending this action, Administrator Simon has announced that the Federal Energy Administration has begun its own audits of refiners stocks as well as compliance with federal price ceilings.

In the case of natural gas, the Bureau's annual survey on natural gas consumption accounts for approximately 95 percent of the annual production reported by State agencies.

Coal

Eleven industry coal canvasses are conducted, 3 monthly, 1 quarterly, and 7 on an annual basis. Approximately 6,000 mining establishments are contacted for the Bituminous Coal and Lignite Survey. Response rate is poor, less than 30 percent, but those companies reporting produce over 85 percent of U.S. coal production. Data are verified by State reports.

Results of the surveys are published in 3 monthly, 1 quarterly, and 4 annual Mineral Industry Surveys. Two weekly Mineral Industry Surveys report bituminous and anthracite production as estimated from railcar loading information. Data also appear in the annual "Mineral Yearbook" and "Mineral Facts and Problems."

The Geological Survey and the Bureau of Mines estimate coal resources and reserves from a variety of data sources. A part of the data is from industry records, but the majority comes from original Survey and Bureau investigations, some of which date back to the turn of the century.

VI. OTHER SOURCES OF ENERGY DATA

A number of data sources other than Bureau of Mines conducted surveys are used in compiling energy statistics.

Atomic Energy Commission

For statistics on domestic uranium production, resources and reserves the Bureau relies on the AEC data. Until 1962 the AEC was the only market for uranium which it purchased from industry under negotiated contracts. These contracts required that the contractor reveal cost and other information to the Commission. After 1962 Government purchased only that uranium which had been developed into reserves by 1958. The owners submitted their detailed justification of their claimed reserves to the Commission and Government and industry geologists worked together to develop the final allocation. Through this procedure, an industry wide standard basis for reserve calculations was developed.

Although the last purchase contracts expired in 1970, the system developed in the late 1950's and early 1960's has continued on a voluntary basis to the present. Reserve and cost data are submitted on a regular schedule to the AEC and calculations are jointly reviewed by Government and industry. Strict confidentiality is maintained.

Selling prices are determined by a voluntary canvass of the utility industry which reports average annual contract prices for Uз08 on a 10-year forward basis. The AEC thus predicts price trends and estimates reserves at varying prices.

The AEC maintains a staff of about 65 at Grand Junction, Colorado, to operate its program. Resource estimates from nonoperating lands are prepared for the AEC by its contractor, Lucius Pitkin, Inc. (LPI). LPI, staffed by over 100 professional and support personnel, has a continuing contract for updating resource information.

American Petroleum Institute and American Gas Association

For domestic oil and gas reserves the Bureau relies on the annual studies of the American Petroleum Institute (API) and the American Gas Association (AGA). These two organizations each have created a reserve committee of senior executives from major companies to oversee the gathering of data and publishing of reserve estimates. The API has a 15-member committee and each member appoints one or more subcommittees for the purpose of preparing reserve estimates. The subcommittees are composed of approximately 175 geolo

gists and engineers who (1) represent segments of the industry having prominent ownership holdings in the assigned area; (2) have broad experience in estimating reserves; and (3) have intimate knowledge of the fields assigned to them.

The AGA operates in a similar manner, but with about 100 subcommittee members. Thus, some 300 industry executive and professionals with a close association with the problem are involved in the oil and gas reserve estimates. Confidentiality of proprietary information is carefully guarded. The estimates are believed to be of high quality. A senior employee of the Bureau of Mines is currently a member of the 15-man AGA reserve committee.

State agencies

Coal, crude oil, and gas production data are compiled from State reports by the Bureau of Mines specialists. Although the coal mining industry is canvassed, the data obtained are used for determining mining methods, equipment used, and transportation modes rather than production statistics. These State reports are considered to be adequate.

Other Federal agencies

Largely for consumption data but also for imports and exports, the Bureau supplements its statistics with data from the following Federal Departments and Agencies: 1. Federal Power Commission; 2. Department of Commerce; 3. Department of Transportation; 4. Department of Defense; and 5. The Department of the Treasury.

VII. PROBLEM AREAS

The design of a basic, practical, and effective information system to assess the consequences of policies, existing and potential, is a major problem. The system must be responsive to change and yet maintain continuity. The standardization of definitions, the format of statistical reports that limits items that can be collected, and the organization of primary data to improve its retrievability and usefulness are difficult problems because of the complexity of the energy industries.

Problems arise with the sorting of fact from conjecture in the statistical area. This is especially true in reserve reporting. Collection of data has been necessarily confined to the collection of information where there is a recognition of standardized definitions and of procedures such as are reported in the measured and indicated categories.

The capability to conduct analytical studies that could provide energy information essential to an adequate data system is seriously inadequate.

The matching of statistical information is a difficulty that confronts and often precludes good analysis. Current information systems have essentially been designed to serve a purpose rather than being multipurpose. This limits the utilization of collected data. Improvements are being made but greater improvements are essential.

A recognition of a requirement for the need of proprietary information results in some limit to data collection. This is especially true in the cost/price and reserve information areas.

The cost of obtaining specific information as well as the time required are major problems. Although timeliness is essential, faster compilation of reliable indepth data on a national basis even for emergency purposes, is impractical under current reporting concepts when the magnitude of the energy industries is considered. Improvements also are needed in the relation of energy to smaller geographical areas.

Data relating to international operations of energy industries are seriously inadequate. Improvements in data collection, processing, analysis, and publication must be made. Standardization of definitions and procedures or improvements in matching methods is a must in this area.

VIII. SUMMARY

In conclusion Mr. Chairman, the problems are complex and we in Interior are determined to rearrange our priorities so that every possible effort is made to obtain accuurate and timely information, not only for the energy crisis in which we now find ourselves, but also, to avert, if possible, any parallel devel

oping situation with respect to the non-fuel minerals on which our economy depends so entirely.

Chairman PROXMIRE. Mr. Shiskin, please proceed.

STATEMENT OF HON. JULIUS SHISKIN, COMMISSIONER, BUREAU OF LABOR STATISTICS, DEPARTMENT OF LABOR, ACCOMPANIED BY JANET NORWOOD, DEPUTY COMMISSIONER FOR DATA ANALYSIS; AND MARGARET STOTZ, CHIEF, WHOLESALE PRICE DIVISION

Mr. SHISKIN. Thank you, Senator Proxmire.

Like Mr. Rigg, I have a prepared statement, which I will read only part of, only an abstract of.

Also, I have with me Mrs. Janet Norwood, who is the Deputy Commissioner for Data Analysis of the Bureau of Labor Statistics, and Mrs. Margaret Stotz, who is Chief of our Wholesale Price Division.

As you may know, the BLS has great tradition for having distinguished women professionals in high positions and I am very pleased that the hearing today offered me an opportunity to bring

two of them with me.

There are also other members of the staff present to support us in some of the more detailed questions that may come up.

I will now read my summary statement which will take about 10

minutes.

Mr. Chairman and members of the subcommittee, I welcome this opportunity to appear before your committee to discuss the efforts of the Bureau of Labor Statistics to improve wholesale and retail price data for petroleum products.

I found when I came to BLS last August that the staff was very much aware of the need for better information on petroleum prices and had been struggling hard to obtain it. However, neither they, nor I, anticipated early enough the speed with which these problems would become urgent.

This morning I would like to describe current BLS procedures for compiling the wholesale price indexes for refined and crude petroleum products, components of the wholesale price index-ŴPI— and efforts to improve them. In addition, I will describe briefly recent improvements in the accuracy and timeliness of retail gasoline price data.

The BLS wholesale price indexes for the major components of the refined petroleum index-gasoline, light distillate, middle distillate, and residual fuel oil-have for years been based on prices quoted in trade publications. These secondary source data have been used with great reluctance, since BLS policy is to collect data directly from companies wherever possible. Although we have long recognized the risks of using secondary data, we were unable to introduce direct collection for two reasons. First, the BLS relies completely on voluntary reporting in its price programs. We have no legal authority to compel companies to report price data to the Bureau.

BLS efforts to achieve the voluntary cooperation of the petroleum companies in past years have been unsuccessful.

Their reluctance was explained on the grounds that, until recently, the spot market data used in the WPI constituted a reasonably good indication of the price trend for refined petroleum products.

Second, secondary data is far cheaper than direct price collection, and the BLS has not been able to secure, as a part of its regular budget, adequate funds for the WPI.

Data used in BLS price indexes should be directly collected at the primary source. As is well illustrated by the present situation in the prices of refined petroleum products, shortcuts to save money on key statistical indicators do not pay in the long run.

Historically, the trend of the WPI for refined petroleum products, based on spot market quotations, was similar to that of retail prices. Events during 1973, however, appear to have changed the relationship of the spot market data to the total market. As shortages developed, domestic spot market transactions began to represent a declining portion of the total refined petroleum market and, thus, became a less reliable indicator of the total market.

I have a chart in my prepared statement which shows how wholesale prices and retail products moved quite similarly from 1967 to 1972 and then departed markedly in 1973.

We have called attention to our concerns over this index in the November and December WPI press releases, and will continue to do so until a new refined petroleum index is available.

In order to get an idea of the possible magnitudes involved, we have calculated a special WPI which removes the direct impact of price changes for refined petroleum products on the WPI. This special index increased 14.6 percent from December 1972 to December 1973, compared with an 18.2-percent rise in the published WPI.

The BLS began to develop a better index for refined petroleum products in 1972, with funds-and specifications-provided by the Office of Emergency Preparedness-OEP. This program, funded by OEP in fiscal years 1972 and 1973, was designed primarily to meet the contractor's OEP requirements. The OEP specifications were designed to yield both monthly average unit prices and price indexes.

It required the selected companies to adjust their recordkeeping.

Many companies were unwilling to make such adjustments and would not participate in the program. In addition, some companies were reluctant to report because they had reservations about the use of these data to compute average unit prices.

Many of the difficulties in this program have stemmed from our attempt to adapt and implement, on a voluntary basis, a program to meet OEP needs which would also be suitable for the WPI.

BLS options in this situation were limited because of the needs of the sponsoring agency, the attitude of the companies, and inadequate funds for the WPI.

It may be useful to note the BLS has received in its fiscal year 1974 appropriation an increase of $450,000 for the WPI. This increase of almost 50 percent of the total appropriation for the WPI will enable us to make many improvements.

In early 1973, the BLS tried to reduce the reporting burden and to provide ime for companies to adjust their accounting sys

tems to the program's requirements. With OEP concurrence, the BLS concentrated on the collection of national data and temporarily discontinued regional data. Unhappily, the response rate did not improve markedly. In mid-1973, the OEP contract authorization expired.

In view of the public interest in these data, the BLS diverted fiscal 1974 funds from other WPI activities in order to continue this program, and made another effort to increase cooperation.

To pave the way for better reporting, a decision was made to concentrate on improving the measures of price change and to postpone development of average unit prices.

In addition, requests for data for some products were discontinued and some product descriptions were modified.

The BLS gave a top priority to this program and hoped to have had it operational by this time.

However, serious problems still remain. The most important is the timelag in the reporting of the data to the BLS. The WPI is released in the first week of the month following the month covered by the data, for example, on January 8 figures for December 1973 were released.

Volume and revenue data cannot be available until the month is over. Because of this time schedule, cooperating companies would, therefore, be unable to provide refined petroleum data without a timelag of at least 1 month. In fact, however, in most cases data are being reported with a 2-month lag.

Coverage of the industry has been another problem. At the present time only about half of the companies selected for the probability sample of producers have agreed to furnish the national data required.

Why do we request revenue and volume data covering the whole month, which results in lag, rather than average prices for a particular day as is the practice in most other parts of the WPI?

We are working intensively with the companies to resolve the coverage problem and to reduce the lag in reporting. We hope to introduce a better refined petroleum component in the spring, but I cannot specify any completion date today.

As soon as we can secure sufficient coverage to assure reasonable reliability and prompt reporting, we shall introduce the new data into the index. The timeliness requirement would necessitate that the cooperating companies currently reporting data speed up their reports from about 40 to 50 days from the end of the reference month to 15 days. And this would still leave us with a 1-month lag; that is, the report issued at the beginning of March covering February would use January refined petroleum products data.

We shall also make a more comprehensive study of direct price reporting and if that approach proves feasible and more timely we shall switch to it. About 2.7 percent of the weight in the WPI lags by 1 month and 0.5 percent by 2 months.

In addition to its work on refined petroleum, the BLS also initiated a program last year to expand and improve the pricing of crude petroleum for the WPI. For many years, BLS has directly priced domestic crude petroleum in six major producing areas of the