Horizontal product differentiation and policy adjustment in the presence of abatement subsidies and emission taxes

Abstract

There are important examples of countries which have implemented policies to promote pollution abatement activities in sectors characterised by some degree of product differentiation. This paper examines the role of product differentiation on optimal policy and industry emissions in a Cournot oligopoly model in the presence of abatement technology, abatement subsidies and emission taxes. The analysis indicates that as products become more differentiated the government can afford a tax increase due to the presence of subsidies and abatement technology. Additionally, highly differentiated industries may experience a rapid increase in emissions and so policies such as research and development (R&D) may be needed to tackle higher emissions. The government adjusts optimal policy as industries become more or less pollution-intensive, and the extent of the adjustment varies across industries characterised by different degrees of product differentiation. The analysis is potentially relevant to industries where firms are taking steps to differentiate their products in order to capture particular market niches, and lower production and abatement costs.

Keywords:

• horizontal product differentiation
• emission taxes
• environmental subsidy
• Cournot

Acknowledgements

I am grateful for constructive and helpful comments by two anonymous reviewers.

Notes

1. See the UK Department of Energy http://www.decc.gov.uk/en/content/cms/what_we_do/uk_supply/energy_mix/ccs/ccs.aspx and a recent EU initiative to promote carbon capture and storage (CCS) technology, the ‘NER 300 call’ at http://ec.europa.eu/clima/funding/ner300/index_en.htm. See also PEW Center of Global Climate Change, http://www.pewclimate.org/technology/factsheet/ccs. Information taken from Massachusetts Institute of Technology (MIT), ‘The Future of Coal: Options for a Carbon-Constrained World’, 2007.

2. Other sectors by NACE classification subject to an energy tax in the EU include NACE Rev. 1.1 Sections A to K and N and Divisions 90, 92 and 93. The complete data-set on environmental taxes in the EU, including pollution taxes, is available at http://epp.eurostat.ec.europa.eu/portal/page/portal/product_details/dataset?p_product_code=ENV_AC_TAXIND. These data show that some of the sectors mentioned, which have applied for NER 300 funding and which include the power generation, iron and steel production and cement sectors, see http://ec.europa.eu/clima/funding/ner300/00001/summary_en.pdf, are also subject to a pollution tax. In terms of product differentiation, other industries which may be relevant to the present analysis and CCS technology include the chemicals and petroleum refineries; see Anderson and Newell (2003) for examples.

3. Real-world examples of industries which can be associated with product differentiation include the chemicals industry whereas the pulp and paper industry exemplifies the case of homogeneous goods (Fujiwara 2009, 240). In addition, Thollander and Ottosson (2008) explain the CO2 tax (see p. 26) firms in the pulp and paper industry in Sweden face as well as some of the potential investment subsidies for energy-efficiency technologies (to reduce energy costs associated to production costs) firms may use to reduce energy and therefore production costs (see pp. 30–31).

4. Requate (2006) and Lambertini (2013) discuss the literature on environmental policy under Cournot oligopoly.

5. One key difference between the present and Poyago-Theotoky's model is that the effective subsidy here is an output subsidy and, consequently, the present work may be understood in this context; whereas in Poyago-Theotoky's model the subsidy is interpreted as an R&D subsidy (with the associated R&D market failure), not an output subsidy.

6. According to a study by the US Economics and Statistics Administration in 2010, the chemicals industry, which is one of the largest pollution intensity coefficients as measured by CO2 per unit of output, has been able to reduce its pollution intensity level between 1998 and 2006 (pp. 12–13). See US Economics and Statistics Administration (2010) ‘US Carbon dioxide emissions and intensities over time: A detailed accounting of industries, government and households’. US Department of Commerce, April 2010’. Link: http://www.esa.doc.gov/sites/default/files/reports/documents/co2reportfinal.pdf

7. Lambertini (2013) presents an overview of Fujiwara's model (see pp. 37–40).

8. This structure follows Fujiwara (2009). The specific utility function comes from Cellini, Lambertinib, and Ottavianoc (2004). The assumption of rules out the possibility of complements.

9. The comparative statics effect hold under a general functional form for the cost function.

10. In order to find the change in emissions via the adjustment in policy in the case of an end-of pipe cost function and quadratic damage function I use the closed-form solution for emissions, .

11. The data presented in US Economics and Statistics Administration (2010) show some evidence of reductions in intensity coefficients in some sectors (e.g., transportation, government) being associated to increases in emissions. However, the analysis does not show data indicating the channel whereby this correlation works.

12. The pollution intensity coefficient, , is given by ; see Lahiri and Symeonidis (2007).

13. Formally for any , at , which indicates a positive optimal tax under the concavity assumption of the function.

14. Indeed, in the case of no abatement the effect of the tax via output, , vanishes and, as a result, (i) the optimal subsidy is zero (see the expression in (24(24) )), and the closed-form solution for the tax, assuming , is given by , which is analogous to Fujiwara (2009) and therefore analogous results are obtained.