Using a panel data set on the local telecommunications companies reporting to the FCC
for 1988-95, this paper investigates the subadditivity of the cost function, as well as
technical and allocative inefficiency of the U.S. local telephone industry. The subadditivity
test on the estimated translog cost function indicates that certain subdivisions of the
monopolized regional markets between two hypothetical firms might lower total cost.
However, the evidence is not as clear cut as in an earlier study by Shin and Ying (1992,
RAND), with savings from a two-firm industry being negative on average over all possible
two-firm output vector combinations. The results of the subadditivity test suggest that
companies with a relatively high share of residential customers experience higher degrees
of scale inefficiency. Specification of technical inefficiency as fixed company-specific
effects results in a different efficiency ranking than the specification with random effects.
The estimation results for the generalized (non-minimum) cost model suggest that capital
is being under-employed relative to residual inputs. This finding does not support the
theoretical prediction that an industry under rate of return regulation tends to over-employ capital relative to other inputs. The subadditivity test for the generalized cost function that accounts for technical and allocative inefficiency generated a much more favorable estimates of cost reductions from the subdivision of the monopolized markets than the test on the conventional specification of the cost function. The estimated losses from technical, allocative and scale inefficiency reflect potential gains from competition. / Graduation date: 1999
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/33870 |
| Date | 12 February 1999 |
| Creators | Gainutdinova, Olesya |
| Contributors | O'Sullivan, Arthur |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.001 seconds