PRICING BEHAVIOR
The pricing hypotheses examined in this chapter include:
(1) has the U.S. established a price umbrella under which
imports have been free to erode domestic markets; (2) have
domestic steel firms "administered" prices and (3) have
Japanese or European steel firms dumped steel? These hypotheses
. ~~ .
are examined in an effort to determine if the pricing policies
of the U.S. steel firms and their foreign rivals have contributed to the observed pattern of imported steel over the past
20 years.
The chapter is divided into three sections: (I) the longrun pricing policy of the U.S. steel industry, (II) cyclical
pricing practices and (III) international differences in
pr icing practices.
I. LONGRUN PRICING BEHAVIOR
Confl ict ing Theor ies
There are many theories of the pricing behavior of the U.S.
steel industry. Adams and Dirlam (1, pp. 638, 639) have argued
that:
Cost movements and target returns--more so than
demand--are prime determinants of price policy. . .
.
Pricing policy of the steel industry, and of its price
leaders, includes therefore, not only the achievement
of goals set for individual prices, but also maintenance of the proportions, ratios and margins between
prices at successivê levels of production that seem to be in the best interest of the industry. .
When a strategic price is threatened, however, the
threat must be dealt with at almost any cost, not
-152-
simply because of the necessity for re-evaluating
the specific price, but because of the danger to
the market structure dependent upon it.
They conclude that import erosion of domestic markets,
especially wire rods, was directly attributable to the industry
pricing policies. The pricing policy is in turn attributaole
to the oligopolistic and vertically integrated nature of the
industry.
The Council on Wage and Pr ice Stabi1.,ty'.s "A Study of
Steel Pr ices" (5) has presented a model of steel pr icing similar
to Adams and Dirlam. The report claims (5, pp. 7, 8):
Pr ices are set by industry leaders in an effort
to cover full costs and generate a desired rate
of return at a level of output less than capacity
. . . . The customary situation is price leadership by larger mills, and fOllowing by the other
major inills.
The report oy the Council on Wage and Pr ice Stability, however,
qualifies the above theory by stating that limitations such as
imports and alternate suppliers, including minimills, might
cause the industry to fail to achieve its goals (5, pp. 8, 9).
On the other hand, in analyzing the historical position
of the U.S. steel industry (the first quarter of the 20th
century and possibly later), Stigler (32) has concluded that
the U.S. Steel Corporation was a dominant firm, whereas Parsons
and Ray (22, p. 208) have characterized the industry as a
"dominant cartel."
However, the Gaskins model of limit pricing (8) implies
that even if U.S. Steel were a dominant firm (or the industry
-153-
a dominant cartel or group of joint profit maximizing
oligopolists) it would still price in a manner to slow the
import erosion of its markets.
Moreover, there are authors such as Rowley (25) and Mancke
(17) who regard the pricing practices of the U.S. steel industry
as quasi-competitive; i.e., the observed pr icing pattern is
similar to that which would occur in a competitive industry.
Structura Change in pricing Around 196~"
There is considerable evidence to suggest that the longrun
pricing policy of the U.S. steel industry changed arouna 1960.
In this section some of this evidence is presented, but its
explanation is left to following sections.
Employing quarterly data for the perioa 1952(2) to 1968(2),
Rippe (24) estimated the rate of change in steel mill product
pr ices as a function of (i) the rate of change in employment
costs per man-hour, ECMH; (ii) the ratio of unfilled oròers to
shipments, UO; and (iii) capacity utilization, CU; yielding: ŠP = -2.84 + .59 ECMH + .34 UO + .024 CU
P (-4.6) (9.8)ECMH (2.7)5 (2.8)
R2 = .74, D.W. = 1.59, t values in parentheses.
He found that a Chow test rejpcted, at the one percent significance level, the hypothesis that there was no difference
between the pre- and post-1959 steel strike periods. Regarding
the pre- and post-1959 periOdS, Rippe concluded that .we have
strong evidence that there has been a change in the steel
price-setting mechanism."
-154-
On the basis of a ser ies of regressions which he ran,
Mancke (17) has also found a structural shift in pricing in
the pre- anã post-1959 per iods. He found that Capacity
utilization rates affected pr ices after 1959 but not before
1959. Mancke (17, pp. 154, 155) states: "we must conclude
that from 1947-58 steel prices tended to rise independently
of demand relative to supply but not from 1959-65. Instead
. ~¡¡ .
its price behavior began to parrot that of a more competitive industry since pr ices could now be raised only when
ãernand pr essed upon capac i ty."
Finally, the well known studies by Eckstein and Fromm
(4J and Bailey (3J calculated the impact of steel price
increases and steel value added increases on other products
in the wholesale Price Index. Recently, Ornstein and Eckard
(21) have recalculated the impact of steel value added on other
sectors. Llke Bailey and ECkstein-Fromm, OrsteinEckard useã
an input-output oaseã method of estimation; they calculated
what the ~holesale Pc ice Index (and Impl icit Pr ice Deflator)
woulã have been if the change in steel value added were the
same as the change in the value added in all other manufacturing. As the' ear 1 ier author s found, if steel value added had
increased at the same rate as non-steel value added, the WPI
would have increaseà by a smaller amount dur ing the 1947-58
period. However, the opposite result is true for the 1958-74
per iod.
-155-
While overstating their case,
Ornstein and Eckard (21,
p. 15) conclude:
Steel has had very litle effect on aggregate
pr ice increases dur ing the strong inflat ionary
period since 1965. This is in marked contrast
to the results of Eckstein and Fromm for the
late 1940 i sand 1950' s. Whatever the basis and
reliability of their conclusion on the monopolistic pricing power of the steel industry,
there is no basis for a similar conclusion for
the 1958-1974 period.
The Rippe and Mancke studies menti~~~d above are efforts
to determine the -nature of the steel pr ice-setting mechanism.
-ê~ \;;;
On the other hand, the Eckstein-Fromm, Bailey, and OrnsteinEcklard studies are efforts at ascertaining the relationship
between steel prices and the prices of other sectors. Since
this s~udy is concerned with pricing policy, it is the former
studies which are most relevant for our purposes. Nonetheless,
it is interesting to note, via the latter studies, that the
steel sector's impact on inflation in other sectors subsided
in the 1959-74 period relative to 1947-58.
Change in the Pr ic ing Behav ior of the U. S. Steel Cor por at ion
Data on the market shares of each of the top eight u.s.
steel companies are reported in chapter 2, for the years
1938-76. These data as well as some other pivotal concentration
ratios are summarized here (table 4.1) to provide background
for the pr ic ing behav ior discuss ion wh ich fol lows.