The Persistence of "High Rates of Return" in High-Stable Concentration Industries

PROFESSORS MacAvoy, McKie, and Preston believe that they have supported the "empirical observation . . . that high and stable levels of concentration are associated with persistently high levels of profitability."' They select a few industries from each of the three samples I analyzed,2 present rate of return data for these, and conclude that, "Without exception, these results are consistent with the proposition that high and stable concentration levels and persistently above-normal profits tend to be associated."3 Their own selected data refute Professors MacAvoy, McKie, and Preston. The nine industries they chose from the Bain sample converge on the mean. In 1936-1940, the nine were 320 basis points above the mean. By 1953-1957, the return declined to only 140 points above the mean of the forty-two industry sample, an insignificant difference (Table 1). The twelve industries they selected from the Stigler sample to show "persistently above-normal profits" also converge on the mean. The twelve industries' average return was 240 points above the all manufacturing mean for 1953-1957. It declined to only 100 points above the all manufacturing mean in 1962-1966, again an insignificant difference (Table 1). The seven industries they selected from the Mann sample converge on the mean. The average return was 280 points above the all manufacturing mean for 1950-1960. It declined to eighty points above the all manufacturing mean for 1961-1966, once again an insignificant difference (Table 1). These results are based on small samples. Enlarging the samples of highstable concentration industries, using at least eighteen, we do not find "high" returns in the high-stable concentration industries much less "persistently