Archives

Search Archives
In the Spotlight

Transforming Secondary Education: New $100 million initiative to improve education quality across the nation.
Learn More »

Recent Spotlights »

View all Archives - Environment and Development »

Competition in the U.S. Energy Industry







each size category of firms for the years 1958, 1963, and 1967. Using the survivor technique it is apparent that only the two largest size categories are clearly economic. Other size categories are losing share of output. Even among the two largest categories, only the largest eight firms gained as a share of the industry output from each census to the next census.

Table B-5 also shows the amount of value added per manhour and value added per dollar of investment. As can be seen, value added per manhour is inversely correlated with the size of the company. The largest companies have significantly higher value added per manhour than do smaller. This, of course, may be due to a higher level of investment per manhour for larger companies and may not permit them to have significantly lower costs than smaller operators. But as can be seen on the basis of one year's figures, value added per dollar of capital expenditures while lower for the biggest eight companies than for many of the smaller categories, is just equal to the industry average. It seems unlikely that the efficiency in terms of manhours is entirely offset by less value added per unit of capital.

PETROLEUM REFINING

A considerable number of estimates have been made of economies of scale in petroleum refining. Surprisingly enough, the estimates are reasonably consistent, which gives confidence in the figures. However, most of the data and estimates are for the refinery and not for companies.

One of the earliest estimates was made by Bain on the basis of engineering data gathered from published sources and from industry interviews. For the petroleum refining industry, he found that a plant of the minimum efficient size—120,000 barrels per day—would provide 1.75 percent of industry capacity and would have cost in 1951 between $225 and $250 million including transportation facilities [Bain, pp. 72, 158, and 233]. After adjusting for inflation, it would take about $400 million today to construct such a refinery. Bain also claimed that there were no economies from owning more than one refinery.

In a speech before the Economics Club of Detroit in March of 1973, Rawleigh Warner, Jr., Chairman of Mobil Oil Corporation, asserted that the minimum economic size for a new refinery was about 160,000 barrels a day and would cost in the neighborhood of a quarter of a billion dollars. This size refinery would be about 1.2 percent of industry capacity. These figures are reasonably consistent with Bain's estimates for two decades earlier.

The inventor of the survivor technique, George Stigler, applied it to petroleum refining. He found that the plant size between 0.5 and 2.5 of industry capacity had grown relatively rapidly between 1947 and 1954, and were therefore efficient [Stigler, p. 69]. These figures bracket Bain's estimates. For companies, Stigler found that the efficiency sizes ranged from 0.5 to 10 percent