本研究探討台灣廠商之能源運用效率，以2002 年至2018 年我國電子零組件
製造業為研究對象，建構廠商之資本生產力、勞動生產力、中間財生產力與能源
生產力。接著分析生產力與廠商規模之關係，探討非中性生產力對要素密集度的
影響，並將能源生產力與能源價格進行連結。
本研究以Zhang (2019) 之估計方法為基礎，在生產函數中加入能源要素投入，
同時參考Berry et al. (1995)、Raval (2019) 和Aw and Lee (2023) 使用之外生工具變
數，以一般動差法(Generalized Method of Moments, GMM) 進行估計，並建構廠商
之非中性生產力。
實證結果顯示，能源投入之替代性低於資本、勞動與中間財。電子零組件製
造業之能源生產力於廠商間最具異質性，且於2002 年至2018 年間持續下降，資
本生產力與勞動生產力僅些微下降，中間財生產力些微上升，使資本能源相對生
產力、勞動能源相對生產力與中間財能源相對生產力上升，導致廠商之能源消費
逐年相對上升。因我國之能源供應屬國營事業，能源價格於研究期間並無明顯提
升，使廠商缺乏提升能源生產力之動力。而大型廠商之資本與勞動生產力較高，
且能源生產力低於中型廠商，為造成能源消費相對上升之主因。

This study investigates the energy efficiency of Taiwanese firms, focusing on Taiwanese
electronic firms from 2002 to 2018. We construct measurements of capital-augmenting
productivity, labor-augmenting productivity, material-augmenting productivity and energyaugmenting
productivity for each firm. We analyze the relationship between factor augmenting
productivity and firm size, as well as elucidating the impact of non-neutral productivity
on factor intensity. Moreover, we discuss the connection between energy-augmenting
productivity and energy prices.
Based on Zhang’s (2019) estimation method, this study includes energy input into the
production function. Additionally, referring to Berry et al. (1995), Raval (2019) and Aw
and Lee (2023), we adopt exogenous instrumental variables within Generalized Method
of Moments estimation. Afterwards, we construct non-neutral productivity for each firm.
Our empirical results indicate that the substitutability of energy input is lower compared
to capital, labor, and materials. The energy-augmenting productivity of Taiwanese
electronics firms exhibits the highest heterogeneity among firms, and had been continuously
declining from 2002 to 2018. In contrast, capital-augmenting productivity and
labor-augmenting productivity had only slightly decreased, while material-augmenting
productivity had slightly increased. This has led to an increase in the relative productivity
of capital to energy, labor to energy, and material to energy, resulting in an sustaining
relative rise in firms’ energy consumption.
Given that energy supply in Taiwan is predominantly managed by state-owned enterprises,
energy prices hadn’t significantly increase during the study period, leading to a
lack of incentive for firms to improve their energy-augmenting productivity. Large firms
exhibit higher capital and labor augmenting productivity but lower energy-augmenting productivity compared to medium-sized firms, which is the primary reason for the relative
increase in energy consumption.

[1] Ackerberg, D. A., Caves, K., and Frazer, G. Identification properties of recent production
function estimators. Econometrica 83, 6 (2015), 2411–2451.
[2] Aw, B. Y., and Lee, Y. R&d investments, outsourcing and non-neutral productivity
growth. Working Paper (2023).
[3] Aw, B. Y., Roberts, M. J., and Xu, D. Y. R&d investment, exporting, and productivity
dynamics. American Economic Review 101, 4 (2011), 1312–1344.
[4] Berry, S., Levinsohn, J., and Pakes, A. Automobile prices in market equilibrium.
Econometrica 63, 4 (1995), 841–890.
[5] Demirer, M. Production function estimation with factor-augmenting technology: An
application to markups. Job Market Paper (2020).
[6] Dewitte, R., Dumont, M., Merlevede, B., Rayp, G., and Verschelde, M. Firmheterogeneous
biased technological change: A nonparametric approach under endogeneity.
European Journal of Operational Research 283, 3 (2020), 1172–1182.
[7] Doraszelski, U., and Jaumandreu, J. R&d and productivity: Estimating endogenous
productivity. Review of economic studies 80, 4 (2013), 1338–1383.
[8] Doraszelski, U., and Jaumandreu, J. Measuring the bias of technological change.
Journal of Political Economy 126, 3 (2018), 1027–1084.
[9] Gao, Q., Zhang, J., and Chen, G. Firm heterogeneity, biased technological change,
and total factor productivity: Evidence from china. Journal of Productivity Analysis
60, 2 (2023), 147–177.
[10] Griliches, Z., and Mairesse, J. Production functions: the search for identification.
Working Paper (1995).
[11] Griliches, Z., and Ringstad, V. Economies of scale and the form of the production
function: An econometric study of Norwegian manufacturing establishment data.
North-Holland Amsterdam, (1971).
[12] Levinsohn, J., and Petrin, A. Estimating production functions using inputs to control
for unobservables. The review of economic studies 70, 2 (2003), 317–341.
[13] Liu, J. Multinational production and innovation in tandem. Tech. rep., Working
Paper, (2024).
[14] Marschak, J., and Andrews, W. H. Random simultaneous equations and the theory
of production. Econometrica, Journal of the Econometric Society (1944), 143–205.
[15] Mundlak, Y., and Hoch, I. Consequences of alternative specifications in estimation
of cobb-douglas production functions. Econometrica: Journal of the Econometric
Society (1965), 814–828.
[16] Nerlove, M., et al. Returns to scale in electricity supply. Institute for mathematical
studies in the social sciences, (1961).
[17] Olley, G. S., and Pakes, A. The dynamics of productivity in the telecommunications
equipment industry. Econometrica 64, 6 (1996), 1263–1297.
[18] Raval, D. R. The micro elasticity of substitution and non-neutral technology. The
RAND Journal of Economics 50, 1 (2019), 147–167.
[19] Zhang, H. Non-neutral technology, firm heterogeneity, and labor demand. Journal
of Development Economics 140 (2019), 145–168.