Review of Industrial Organization
12: 519–536, 1997.
1997 Kluwer Academic Publishers. Printed in the Netherlands.
Interstate Long Distance Rates: Search Costs,
Switching Costs, and Market Power
CHRISTOPHER R. KNITTEL
Department of Economics, University of California, Berkeley, CA, U.S.A.
Address for correspondence: 2539 Channing Way, Berkeley, CA 94720, U.S.A.
Abstract. A number of authors have argued that the divestiture of AT&T did not reduce the rates
of long distance telephone companies as often believed. However, the literature has offered few
explanations as to why competition has not lowered rates. This study argues that rates have failed to
fall due to the importance of search and switching costs in the industry. Using a panel data set of rates
for the three largest long distance carriers, stretching from 1984 to 1993, a reduced form equation is
speciﬁed to empirically test for the inﬂuence of search and switching costs on the price cost margin
of the three carriers. The results illustrate that both search and switching costs have provided long
distance carriers with market power.
Key words: Market power, search costs, switching costs.
The 1984 divestiture of AT&T brought about many structural changes for long
distance telephone service. AT&T was broken up vertically into local servers and
a long distance provider, the regulatory methods were changed, and entry was
encouraged. The market has since seen an inﬂux of ﬁrms and a decrease in the
market share of AT&T. This new competition and largely deregulated environment
was expected to bring large reductions in the market power of ﬁrms in the industry.
This, however, has not occurred. Although a brief glance at interstate rates would
lead one to believe competition has succeededin reducing market power, this would
be incorrect. After controlling for reductions in access charges, this study ﬁnds that
the Lerner Index has actually increased during the years studied.
I would like to thank Jason Albright, Severin Borenstein, Victor Stango, and two anonymous
referees for their invaluable suggestions and comments. This paper also beneﬁted from conversations
with John Constantine, Andrew Lee, and Stefan Palmqvist. Of course, all remaining errors are
attributed to the author.
Access charges are the per minute charge paid by the long distance carriers to the local Bell’s.
The fee covers line termination, intercept, local switching, local transport, and common carrier line
charges. These charges are set by the FCC to offset the expense incurred by the local telephone
companies when long distance calls are routed to the consumer.
Recall that the Lerner Index for period
is deﬁned as follows:
controlling for reductions in access charges, I am implicitly deﬁning marginal cost to equal only
access charges. This is not a large assumption. In fact, MacAvoy (1995) estimates the remaining of