Abstract:
In order to estimate the effects of potential reductions of
timber availability upon employment in Oregon, a model based upon
marginal sector analysis is appropriate. This study utilizes marginal
analysis based upon a homothetic Constant Elasticity of Substitution
(CES) production function to estimate expansion path condition models
for predicting the effects of factor cost changes and/or reduced
factor availability upon employment in Oregon's forest products
industries, as defined by the Standard Industrial Classification
system (SIC) designations 241 (logging), 242 (lumber mill operations),
243 (veneer and panel products) and 26 (pulp, paper and allied
products). The SIC 241 and 243 models are two factor models (capital
and labor), while the SIC 242 and 26 models are three factor models
(capital, labor and raw materials).
The data used to estimate the models were collected at two
levels of aggregation. The SIC 241 and 242 models were estimated
using county level cross-sectional data from the years 1967, 1972,
1977 and 1982. The SIC 243 and 26 models were estimated from
state-wide time series data, with the SIC 243 series stretching from
1956-1962 and the SIC 26 series ranging from 1965-1982. A second SIC
242 model was estimated, at the county level, and pertained only to
those Oregon counties west of the Cascade mountains.
None of the elasticities of substitution calculated differed
significantly from unity, save the capital-labor elasticity ([sigma]kl) for SIC 241, which equalled 1.34. The elasticities found in this study
are higher than most of the elasticities found in prior studies at the
national level. This difference could be attributed to the
cross-sectional data utilization for the SIC 241 and 242 and/or to
aging capital allowing greater opportunities for substitution in all
of the SICs.
An insignificant technical change bias favoring raw material
over capital usage was found in all of the SIC 242 and 26 models. A
significant technical change bias was found favoring capital usage
over capital stock usage.
The expansion path models created by this study proved effective
for tracking historical labor use and show potential for predicting
the effects of changing factor conditions in the future.
