Technical inefficiency, abatement cost and substitutability of industrial water pollutants in Jiangsu Province, China

Highlights

• This study analyzed the technical inefficiency of industrial water pollutants.

• The quadratic directional distance function was used to analyze this efficiency.

• The shadow price model was used to calculate the abatement costs of pollutants.

• Morishima elasticity was used to assess the output substitutability of pollutants.

• Differentiated emission reduction policies should be adopted in Jiangsu.

Abstract

Environmental pollution, caused by industrial wastewater discharge, restricts the transformation of China’s industrial economy. In terms of its gross industrial product, Jiangsu Province has always been among the top-ranking areas of China. To ensure the green and sustainable development of Jiangsu’s industry, it is imperative to formulate specific strategies toward the reduction of industrial water pollutant emissions. First, by the parameter quadratic directional output distance function and the shadow price model, this paper estimates the technical inefficiency and marginal abatement costs of industrial chemical oxygen demand (COD) and ammonia nitrogen (NH₃–N) in Jiangsu Province between 2006 and 2017. Second, spatial and temporal distributions as well as the dynamic evolution tendencies of technical inefficiency, and abatement costs of industrial COD and NH₃–N are analyzed. Third, the Morishima elasticity model is utilized to measure the output substitution elasticity of industrial COD and NH₃–N for Jiangsu. The results show that the technical inefficiency of industrial water pollutants in Jiangsu Province followed a decreasing trend from 2006 to 2017; moreover, significant differences were found in technical inefficiency among different regions. Over time, the regional disparity followed a polarization trend. For Jiangsu Province, the marginal abatement cost of industrial COD increased rapidly and continuously, while the marginal abatement cost of industrial NH₃–N showed fluctuating and marginal growth from 2006 to 2017. The marginal abatement cost of industrial COD in southern Jiangsu exceeded that of other regions; however, the marginal abatement cost of industrial NH₃–N in northern Jiangsu was highest. Additionally, the negative sign of the Morishima output elasticity for Jiangsu indicates a “trade-off” relationship between gross industrial output and industrial water pollutants. Finally, according to the empirical results, corresponding policy recommendations are proposed.

Introduction

The shortage of water resources and the deterioration of water environment have become one of the common problems in the world (Zhou et al., 2017). Although 71% of the earth’s surface is covered by water, only 2.5% of the earth’s fresh water resources are available. According to the United Nations World Water Development Report 2018, the water quality of rivers in Africa, Asia and Latin America has been deteriorating since the 1990s, and this deterioration trend will continue for decades in the future, threatening human life, health and the harmonious development of social economy and ecological environment (Cole et al., 2008). Among them, due to the rapid development of population and social economy, developing countries are facing the most serious threat of water environment deterioration.

China is the largest developing country in the world (Hu et al., 2013). Since the reform and opening up, China’s economic development has made remarkable achievements. Industrialization has been a key strategic goal of China’s national economic development, which plays a pivotal role in the transformation of the traditional social economic structure and in tightening urban-rural economic links (Shi et al., 2010). However, industry is also an important sector of pollution. As a byproduct of industrialization, industrial water pollution has become one of the major hindrances to the sustainable development of an economic society (Zhang et al., 2019). It is urgent and necessary to explore low-cost and high-efficiency industrial water pollution reduction methods and balance the contradiction between industrial water pollutants emission and economic growth.

As a first step, the technical efficiency of industrial water pollutants must be estimated. As the basis of future work, this step can help to study the relative efficiency of industrial water pollutant emissions in each region, to promote clean production (Liu and Chen, 2017). Currently, according to the number of indicators involved, efficiency measurements can be divided into single-factor and multi-factor measurements. Single-factor efficiency generally involves only two indicators, usually expressed as the ratio of GDP to pollutant emissions (Yamaji et al., 1993). Although the calculation method of single-factor efficiency is simple, few factors are considered. Many scholars included additional factors for efficiency measurement. Index system method (Garg et al., 2011), data envelopment analysis (DEA) (Yao et al., 2015), stochastic frontier analysis (SFA) (Jin et al., 2018), and directional distance function (DDF) methods (Huang et al., 2016) have been widely used to measure technical efficiency.

Secondly, China’s regions must recognize their own marginal abatement cost (MAC) when promoting emission reduction of industrial pollutants. The shadow price is an important means of marginal abatement cost accounting and can be used as a reference value for setting environmental tax rate and emission trading pricing. Two methods are currently used to estimate the shadow price. The first is a non-parametric method, which is characterized by no constructed form. It is mainly used to build a production possibility set based on DEA (Zhou et al., 2008). In the calculation of the shadow price based on DEA (Lee et al., 2014), the calculation results are sensitive to outliers, which may ultimately affect the calculation accuracy. The second method is the parametric method. Early research mainly used production functions (Aigner and Chu, 1968) and cost function (Pittman, 1981) to estimate the shadow price. Then parametric DDF (Shephard, 1970) is widely used in shadow price calculation. In contrast to the non-parametric method, parametric DDF can choose the different directional vector to ensure the expansion of the desirable output and the reduction of the undesirable output, and can overcome the negative externalities the negative externality of the undesirable output (Lee et al., 2015). This is more in line with the requirements of a green economy (Feng et al., 2018).

In addition, with the increasingly strict control of pollutant emissions, the determination of the elasticity of substitutions between desirable and undesirable outputs of industrial production will help a region to identify and control the emissions of high-cost pollutants to achieve specified pollutant control objectives. Several scholars have studied the elasticity of output substitution. For example, Blackorby and Russell (1989) constructed the Morishima elasticity of substitution. Based on their work, Färe et al. (2006) estimated the substitutability between generating capacity and SO₂. Kumar and Managi (2011) calculated the substitution elasticity between the output of goods of India’s industry and various water pollutants. Bonilla et al. (2018) measured the Morishima output elasticity of the substitution between Sweden’s useful energy, CO₂, and NO_X.

Previous studies have provided a rich theoretical basis for discussions on pollutant reduction. However, most of the literature focuses on air pollutants, while little research is available on water pollutant emission reduction. Moreover, the available research on water pollutant discharge focused on the national level (Tang et al., 2016), without sound in-depth comparative research on emission inefficiency, shadow prices, and substitutability of industrial water pollutants in prefecture-level cities throughout a province. China has a vast territory, and both the economic development and water resource endowment of different regions vary significantly. Compared with the western region, the eastern coastal region is rich in water resources and has a more developed industry. It is of great significant to study industrial water pollution in regions with different characteristics.

This paper uses Jiangsu Province as the study area and analyzes the technical inefficiency, abatement cost, and substitutability of its main industrial water pollutants (COD and NH₃–N) during the 11th Five-Year Plan (FYP), the 12th FYP, and the ongoing 13th FYP (2016–2020). The main contribution of this research can be summarized in three aspects: (1) This study contributes to the understanding of the industrial water pollution emission trend and abatement situation in a developed region with high industrialization level and rich water resources by using provincial representative data. As a representative of China’s rapid urbanization and industrialization, Jiangsu’s industrial wastewater discharge reached 1652.13 million tons in 2017, ranking first in China (Fig. 1). These findings in Jiangsu can provide an effective reference for the industrial water pollution reduction policy design in other regions that face a similar development trajectory. (2) A parameter quadratic directional output distance function (PQDODF) is applied to estimate the technical inefficiency of industrial water pollutants (IWPTIE) and marginal abatement cost of industrial COD (MACIC) and NH₃–N (MACIN), which conforms to the concept of clean production. This can help policy makers to clearly understand the emission reduction potential of the three regions of Jiangsu and promote the allocation of water pollutant emission rights in different regions. (3) This paper analyzes the relationship between industrial water pollutants abatement and economic growth in Jiangsu Province by the Morishima elasticity method. On the premise of substitutability, the dual goals of industrial economic growth and industrial water pollutants abatement can be achieved by changing the combination of industrial desirable outputs and undesirable outputs.