台灣紙漿、造紙工業之生產者行為與生產技術︰無母數分析

A Nonparametric Analysis of Producer Behavior and Technology in the Pulp and Paper Industry of Taiwan

Nonparametric tests of fundamental economic assumptions were conducted for Taiwan’s pulp and paper industry. The method permitted the use of disaggregate data and avoided the imposition of a particular functional form. The validity of the following hypotheses was investigated: cost minimization, profit maximization, technical change (Hicks-neutral or biased), and separability of inputs and outputs. Both deterministic and stochastic tests were applied. The results suggested that cost minimization, or profit maximization in the presence of technical change, could be assumed for the industry, in the long run and the short run. Manifold separability tests suggested that high levels of aggregation for inputs and outputs were possible in modeling the industry. The hypothesis of Hicks-neutral technical change seemed questionable for the Taiwan’s pulp an paper industry. 本文利用無母數分析法檢測台灣紙漿、造紙工業的基本經濟假說。無母數法容許使用低總合度（disaggregate）的數據資料，並可避免設立特定函數的限制。包括成本最小化、利潤最大化、技術變動（技術中立或偏差）、以及投入與產出分離性（separability）等假說的正當性皆以無母數法探究。限定（deterministic）與隨機（stochastic）檢測皆被應用。結果顯示︰在長期或短期下，該產業呈現成本最小化及利潤最大化的假設可被接受。多種分離性檢測結果表示︰高總合度的投入與產出模型適用於台灣紙漿、造紙工業。席克斯技術中立變動（Hicks-neutral technical change）的假設則為不適當。