Thus, interpreting concentration ratios in different industries is not a simple task. The same figure, say 70 percent for the top eight firms, may have very different meanings in different industries. In the energy industry, competition may be determined more by institutions than by concentration levels. An interesting question is raised by the discussion above concerning concentration and certain institutions. It is possible that an industry's market structure is determined to an important degree by certain institutional features. If so, policy makers, when considering institutional changes for whatever reasons, must be aware that such changes affect market structure and the degree of competition. The impact on structure could either help or hinder competition, but it should be kept in mind that economic theory is relatively clear on one point: if an industry's structure is noncompetitive, in the long run performance will suffer.

Three, the relationship between market structure and conduct is not well established in economics. Relatively little research has been conducted on the relationship of a firm's behavior and internal organization to market structure. Certainly, additional knowledge of this relationship is essential to a more complete understanding of market processes.

Despite this shortcoming, an analysis of market structure remains a necessary step in evaluating competition in an industry. Such an analysis, in combination with an awareness of how institutional features can affect conduct and performance, provides a sound basis for a rational policy of competition in the energy industry.

Past studies have generally analyzed structure from the perspective of a single fuel. This emphasis is not incorrect, but it tends to obscure the fact that competition in energy production takes two forms: (1) interfuel and (2) intrafuel. Complete analysis requires that both forms be included.