Modelling wood quality, productivity, demands and supplies in the sawmilling industry : British Columbia coast and Pacific Northwest westside

The main purpose of this thesis is to empirically evaluate the impacts of sawlog scarcity, revealed in the increasing prices and decreasing quality, on the sawmilling industries of the British Columbia (BC) Coast and the United States Pacific Northwest (PNW) Westside. The objectives of the research are: (1) to test whether wood quality has declined overtime; (2) to evaluate the extent to which lagging productivity performance of the industry can be explained through declining wood quality; and, (3) to model the industry econometrically in order to measure short-run input and output responses to changes in wood price and quality. Wood quality is defined as the ratio of a quantity index of the sawlog input to the total volume of wood consumed. The quantity index is an aggregate of grades and species harvested and traded. Quality change is decomposed into two effects: a grade effect and a species effect. It is shown that from 1925 through 1980/1982 the quality of wood harvested and traded in BC declined on average, with both the species and grade effects contributing to the quality decline. From 1957 through 1982 wood quality remained stable in BC while it declined considerably in the PNW. The quality level of the wood traded is considerably higher, on average, in the PNW, but this region is losing its quality advantage relative to BC. Regional differences in average wood prices are completely explained away by quality differences, and so there is no evidence that the BC industry benefits from a wood cost advantage relative to the PNW one. A measure of total factor productivity is developed from a production function. It is shown that in the PNW the rate of technical progress is clearly higher when the wood quality decline is accounted for. When rates of technical progress are compared, it is found that BC lags behind the PNW, and that the relative productivity losses in BC are even greater when wood quality changes are considered. When productivity levels are compared, it is found that the PNW industry is the most efficient of the two regions. However, when wood quality differences are accounted for, BC becomes the most efficient of the two regions on average. The regional industries are modelled econometrically using a static translog restricted profit function. Lumber and pulpchips are variable outputs and sawlogs and labour are variable inputs, while capital is a fixed factor. Wood quality and technical progress are exogenous shifters. Annual observations from 1957 through 1981 are utilized for estimation. The PNW model outperforms the BC model in terms of its theoretical consistency with the profit maximizing hypothesis. The hypotheses that wood quality does not belong in the model, or that it is sufficient to use quality adjusted prices, are both statistically rejected. Constant returns to scale can not be rejected in the PNW, while increasing returns are found in BC. Constant output demand and constant input supply elasticities are computed in addition to total short-run elasticities. It is found that short-run input substitution effects are negligible when compared with the very large output effects. The proportion of pulpchips in output is price responsive. In both regions, a decline in wood quality moves the industry to a more pulpchip intensive region of the output space and to a more labour intensive region of the input space. In the PNW, the estimated rate of technical progress is three times lower when wood quality is ommitted from the model, but in BC, where wood quality changed less, the rate of technical progress is not noticeably affected.Forestry, Faculty ofGraduat