A- A+
Alt. Display

# Evaluation of a CO2 Tax in Chile: Emissions Reduction or Design Problems?

#### Cristian Mardones,

##### University of Concepción, CL

Cristian Mardones is associate professor in the Department of Industrial Engineering, University of Concepción. He received a PhD in economics from the University of Chile in 2009. He has been active in the areas of environmental economics and economic modeling for more than five years, with publications in Energy Policy, Journal of Environmental Management, Economic Systems Research, Energy & Environment, Environment Development & Sustainability, Journal Economía Chilena, CEPAL Review, and Latin American Journal of Economics, among others. His professional experience includes consultancies for the Ministry of Environment in Chile. His current research involves studying the economic effects of carbon taxes.

#### Belén Flores

##### University of Concepción, CL
Belén Flores graduated with a degree in industrial engineering from University of Concepción (Chile) in 2015.

## Abstract

In 2014, Chile introduced a tax reform on carbon dioxide (CO2) emissions that began to be collected in 2017; the reform is restricted to large industrial and power generation sources with thermal power greater than 50 megawatts (MW). Therefore, this study evaluates each industrial source’s option to reduce its taxes by switching to cleaner fuels (investing in new combustion equipment, such as boilers and dual burners). The results show that a tax of US$5 per ton of CO2 for industrial sources of more than 50 MW of thermal power is wholly ineffective in reducing emissions. If a carbon tax is applied to all independent sources of power, only a few industrial sources are predicted to change their current fuel, mainly changing coal to biomass. The conclusion is that the carbon tax serves to raise tax revenues rather than reduce emissions. Resumen En la reciente reforma tributaria de 2014 Chile introdujo un impuesto a las emisiones de CO2, el cual comenzó a ser cobrado en 2017 y está restringido a las grandes fuentes industriales y de generación eléctrica con potencia térmica mayor a 50 MW. Este estudio evalúa la opción de cada una de las fuentes industriales para reducir su carga tributaria al cambiarse a combustibles más limpios invirtiendo en nuevos equipos de combustión, tales como calderas y quemadores duales. Los resultados reflejan que un impuesto de US$5 por tonelada de CO2 para fuentes industriales con potencia térmica mayor a 50 MW es completamente infectivo para reducir las emisiones. Si un impuesto es aplicado a todas las fuentes, independiente de su potencia, solo unas pocas fuentes industriales cambiarían su combustible actual, principalmente de carbón a biomasa. Se concluye que el impuesto sirve para recaudar recursos más que para reducir las emisiones.

How to Cite: Mardones, C., & Flores, B. (2017). Evaluation of a CO2 Tax in Chile: Emissions Reduction or Design Problems?. Latin American Research Review, 52(3), 334–343. DOI: http://doi.org/10.25222/larr.33
Published on 22 Sep 2017
Accepted on 20 Jan 2016            Submitted on 03 Mar 2015

The Kyoto Protocol is an international agreement submitted by the Convention of the United Nations to reduce the emissions of greenhouse gases. The protocol entered into force on February 16, 2005, after being ratified by 55 of the 181 UN member countries, including Chile; its objective was to reduce 5 percent of the countries’ emissions between 2008 and 2012. At the UN Climate Change Conference in Paris (COP21) in December 2015, 195 countries signed a new climate change agreement. Specifically, Chile committed to reducing its emissions intensity by 30 percent by 2030 from its 2007 level.

The countries that emit higher quantities of CO2 include China, the United States, India, Russia, and Japan, which produce 55 percent of CO2 emissions (OECD 2012). The human action source that most contributes to CO2 emissions is the use of fossil fuels. Globally, the three main economic sectors that generate CO2 emissions are the power generation and heat production sector (energy), the transport sector, and the industrial sector. CO2 is mainly emitted to the environment by the energy and industrial sectors by means of boilers and furnaces.

In Chile, energy consumption is mainly based on petroleum, which represents 54.8 percent of the total secondary consumption.1 The second most used energy source is electricity, which represents 19.9 percent of final consumption; next is biomass, with 18.1 percent of consumption, and natural gas with a 5.4 percent of consumption (Ministry of Energy, 2015). The industrial sector uses 23.2 percent of the total energy consumption of the country, which corresponds to 1406.8 trillion BTUs.

## Methodology

### Information Sources

For Chile, it is possible to obtain industrial data on employment, raw material and energy consumption, and physical production from the Annual National Industrial Survey (ENIA). For this study, variables including identification information, the International Standard Industrial Classification (ISIC) economic activity code, region, fuel consumption, fuel measurement units, fuel expenditure, and working days (Table A2 in Appendix) were used. In particular, we used the 2011 survey because it was the latest available at the time of the study’s termination (at the beginning of 2015).

To estimate the energy used by each industrial source, we used the data from ENIA on the quantity of each consumed fuel, as well as the fuel density and heating values obtained from the National Energy Balance from the Ministry of Energy and the US Environmental Protection Agency (EPA 2009). However, the ENIA survey does not directly identify which industry sources have thermal power greater than 50 MW; this can be inferred only indirectly by annual fuel consumption. Therefore, the following descriptive analysis presents the behavior of all industrial sources (independent of MW of thermal power) that are included in the ENIA survey (3,223 industrial sources).

The following industry sectors consume high quantities of coal: the metallurgical industry, with 66 percent of total fuel consumption (129,712 million tons per year); the nonmetallic minerals industry, with 20 percent of total consumption (39,307 million tons year); and the food industry, with 13 percent of total consumption (25,549 million tons per year).

The largest industry sector that mainly consumes fuel oil and diesel is the food industry, with 53 percent of the total (650 million tons per year); this is followed by the paper industry, with 22 percent (270 million tons per year), and the nonmetallic minerals industry, with 5 percent (61 million tons per year).

The industry sector that consumes the most gasoline is the food industry, with 40 percent of the total coal consumption (7,429 tons per year); this is followed by the metal-mechanical industry, with 12 percent (2,229 tons year), and the metal products industry, with 10 percent (1,857 tons per year).

The highest-consuming liquefied petroleum gas (LPG) industry sectors are the food industry, with 39 percent (41,509 tons per year), followed by the chemical industry, with 16 percent (17,029 tons per year), and the paper industry, with 14 percent (14,901 tons per year).

Natural gas is primarily consumed in the chemical industry, with a consumption of 34 percent (175,174 tons per year); this is followed by the food industry, with 19 percent (98,892 tons per year), and the paper industry, with 19 percent (98,892 tons per year).

The largest biomass consumer in the industry sector is the paper industry, with 64 percent (726,980 million tons per year); this is followed by the wood industry, with 22 percent (249,900 million tons per year), and the food industry, with 14 percent (159,027 million tons per year).

### Estimation of CO2 Emissions

The following fossil-fuel sources were considered in order to evaluate the effect of applying a CO2 tax on industrial sources; coal, diesel oil, fuel oil, natural gas, and biomass. Other fuels, such as gasoline and kerosene, are not primarily used in the production processes of industries, whereas LPG is mainly used for room and office heating. For those industrial sources that consume more than one fuel, they were divided into a source for each fuel (coal, biomass, petroleum No. 6, petroleum No. 2, and natural gas); thus, the research has information from 3,223 sources. In the baseline scenario, all these sources emit a total of 5,198,826 tons of CO2.

To evaluate a potential change of fuel in case of a CO2 tax, it is necessary to know the quantity of fuel required for each industrial source to generate the same quantity of energy (BTU). If the payable tax amount under the original fuel for an industrial source is greater than the sum of the annualized cost of technology investments and the increase of spending associated with the new fuel, the industrial source should then reduce its emissions.

To estimate the quantities of CO2 emitted by each industrial source, the consumed quantities of energy must be multiplied by the factors of CO2 emissions according to the type of fuel. These data are obtained from “IPCC Guidelines 2006 for the National Greenhouse Gases Inventories” (Table 1).

Table 1

CO2 emission factors per fuels.

Fuel Factor Unit (MMBTU = 106 BTU)

Petroleum 77.2 Kg CO2/ MMBTU
Coal 97.6 Kg CO2/ MMBTU
Gasoline 72.2 Kg CO2/ MMBTU
LPG 65.8 Kg CO2/ MMBTU
Natural gas 58.8 Kg CO2/ MMBTU
Biomass 0.0 Kg CO2/ MMBTU

Source: IPCC 2006.

It is important to note that the CO2 emission factor for biomass is zero because trees capture CO2 from the atmosphere during growth; then, in the process of burning the same quantity of wood, the CO2 captured during the growth phase is released. Although this course has been questioned because the recapture of vegetation may vary from years to centuries, it is valid to assume zero emissions in this study because the tax in Chile excludes emission sources that use biomass as fuel.

### Technology Options

Of the industrial sources to be analyzed, 303 consume solid fuels (225 consume biomass and 78 consume coal), 2,114 consume liquid fuels (1,403 consume petroleum No. 2 and 711 consume petroleum No. 6), and 806 consume natural gas.

The technological requirements that are necessary to change the fuel currently used in the production processes from industrial sources are presented in Table 2. Sources that consume liquid fuels will need to change to dual burners, which operate on gas and petroleum. Conversely, the solid fuel sources that want to use liquid or gaseous fuels will require a complete change of equipment, because the fuel supply to the boiler and in the area in which the fuel is burned differs.

Table 2

Technological requirements.

Current fuel Fuel to be evaluated Technology

Liquid Solid Boiler change
Liquid Gas Burner change
Solid Liquid Boiler change
Solid Gas Boiler change
Gas Liquid Burner change
Gas Solid Boiler change

Source: Own elaboration from Mechanical Engineering Department, University of Concepción.

However, the ENIA survey does not specify the technical characteristics of the combustion equipment of the sources; in particular, it does not include thermal powers in MW (PotMW) that are relevant to determining the equipment cost. Then, and based on the Industrial Emissions Inventory from Metropolitan Concepción (UDT-PROTERM 2011), it was possible to obtain a sample from industrial sources that included information on fuel consumption and boiler power; this information was used to estimate functions to simulate the boiler power for each industrial source that was included in the ENIA survey (see Table A1 in Appendix). The estimated functions, the standard errors of the coefficients, and R2 are as follows:

(1)
(2)
(3)
$\begin{array}{c}{\mathit{\text{PotMW}}}_{\mathrm{petroleum}/\mathrm{gas}}=0.0036547*\mathit{\text{Petroleum\hspace{0.17em}or\hspace{0.17em}gas\hspace{0.17em}consumption\hspace{0.17em}}}\left({m}^{3}\right){\text{R}}^{\text{2}}=0.70\\ \left(0.0004095\right)\end{array}$

### The Economic Costs of Technological Options

To perform a comparative analysis of different alternatives for the industrial sources once a CO2 tax on emissions is applied, it is necessary to know the data regarding the energy consumed by the different sources and the power in MW of each piece of combustion equipment, the quantity of technological equipment used in each source,5 the value of the new combustion equipment, the fuel prices per power unit, and the number of tons of CO2 emitted from the fuel being evaluated.

The fuel prices were obtained from the ENIA survey as the average value per MMBTU (one million BTU) paid by sources in each region of Chile. In the regions without fuel price information, the average price of the nearest regions is reported. Table 3 contains price information for each type of fuel.

Table 3

## Results and Analysis

A total of 3,223 industrial sources that consume different fuels and emit approximately 5.2 million tons of CO2 in the baseline scenario was evaluated. When the study solely considers sources of thermal generation greater than 50 MW, the quantity is reduced to 436 sources that emit a total of 3.3 million tons of CO2, which corresponds to 63 percent of the total.

In the case of a US$5 tax per ton of CO2 emitted, no industrial source finds it economically attractive to switch from its current fuel; therefore, it is concluded that no reduction in CO2 emissions would be observed, and tax collection would correspond to$16.3 million.11 This result is robust to different errors in estimating the costs of adjustment in fuel switching (to a maximum of 25 percent of the adjustment cost).

If the application of a tax of US$10 per ton of CO2 emitted is evaluated, one industry source that uses petroleum No. 6 would change to biomass. The quantity of tons of CO2 emitted would be 3,191,685, which implies a reduction of 2 percent of the total tonnage currently emitted by thermal power sources over 50 MW. The reduction in tax revenues associated with the fuel change is reflected by the same percentage. The result is not robust at different levels of errors in adjustment costs, particularly for errors greater than 5 percent in the adjustment costs in which no source would change to biomass. These results reflect the ineffectiveness of this CO2 tax, as currently proposed in the tax reform of 2014, to reduce industrial emissions. Thus, it is necessary to evaluate other, more restrictive settings to determine whether a tax actually can help reduce industrial emissions in Chile. Therefore, an evaluation of the application of the tax to all industrial sources regardless of their installed thermal power is proposed. That is, a tax is proposed on the industrial use of fuels that emit CO2; the amount of the tax is US$5 per ton of CO2. Table 4 shows the results after raising awareness of different error scenarios in estimating the adjustment costs.

Table 4

Simulation of applying a CO2 tax of US$5 per ton emitted. Variable Percentage of error in adjustment costs 0% 5% 10% 15% 20% 25% Tax collection (in US$ millions) 25.13 25.13 25.13 25.13 25.13 25.13
Reduction of tax collection 3.31% 3.31% 3.31% 3.31% 3.31% 3.31%
Tons of CO2 baseline situation 5,198,826 5,198,826 5,198,826 5,198,826 5,198,826 5,198,826
Tons of CO2 carbon tax situation 5,026,701 5,026,701 5,026,701 5,026,701 5,026,701 5,026,701
Reduction of tons of CO2 3.31% 3.31% 3.31% 3.31% 3.31% 3.31%
Number of changes 22 22 22 22 22 22
% change biomass sources 9.78% 9.78% 9.78% 9.78% 9.78% 9.78%
% change coal sources –28.21% –28.21% –28.81% –28.81% –28.81% –28.81%
% change petroleum No. 6 sources 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
% change petroleum No. 2 sources 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
% change natural gas sources 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
Power range (in MW) 1–26 MW 1–26 MW 1–26 MW 1–26 MW 1–26 MW 1–26 MW

Industrial sources that are willing to make a fuel change under a $5 tax scenario are heterogeneous, with thermal power from 1 MW to 26 MW, whereas large thermal power sources (over 50 MW) would not change their fuel. Of the sources currently using coal, 28.8 percent would change to biomass; however, sources that use fuel that pollutes less than coal would decide not to make changes in their fuel because of their relatively low emissions and payable tax amount. The following industries would adopt a new fuel: food (50 percent), plastics (20 percent), leather (20 percent), metal products (5 percent), and other (5 percent). In all simulated scenarios, tax collections would be$25.1 million and the reduction of emissions, 3.3 percent.

Although it may seem counterintuitive that including a larger group of small sources increases the effectiveness of the tax, this result is logical because smaller firms have lower investment costs for a change of fuel. In contrast, investment costs are significantly higher for the largest sources and do not generate sufficient incentives to switch to cleaner fuels if taxes are low. Furthermore, because emission factors by fuel type are not affected, this result could be generalized to other Latin American countries with similar technological fuel change options to reduce emissions. However, the results could vary if other relative fuel prices exist or if the relative prices change of technologies related to changing boilers.

Applying a tax of US$10 per ton of CO2 emitted does not generate major changes in fuel switching decisions on industrial sources (Table 5). Adjustment costs remain for twenty-two sources that changed from coal to biomass for all other scenarios; CO2 emissions regarding the base situation are reduced by 3.3 percent, and tax collections could reach$50.3 million.

Table 5

Simulation of CO2 tax applied to US$10 per ton emitted. Variable Percentage of error in adjustment costs 0% 5% 10% 15% 20% 25% Tax collection (in US$ millions) 49.58 50.27 50.27 50.27 50.27 50.27
Reduction of tax collection 4.62% 3.31% 3.31% 3.31% 3.31% 3.31%
Tons of CO2 baseline situation 5,198,826 5,198,826 5,198,826 5,198,826 5,198,826 5,198,826
Tons of CO2 carbon tax situation 4,958,443 5,026,701 5,026,701 5,026,701 5,026,701 5,026,701
Reduction of tons of CO2 4.62% 3.31% 3.31% 3.31% 3.31% 3.31%
Number of changes 24 22 22 22 22 22
% change biomass sources 10.67% 9.78% 9.78% 9.78% 9.78% 9.78%
% change coal sources –28.21% –28.21% –28.21% –28.21% –28.21% –28.21%
% change petroleum N°6 sources –0.28% 0.00% 0.00% 0.00% 0.00% 0.00%
% change petroleum N°2 sources 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
% change natural gas sources 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
Power range (in MW) 1–50 MW 1–26 MW 1–26 MW 1–26 MW 1–26 MW 1–26 MW

According to unreported additional estimations, a tax of $13.73 per ton of CO2 applied to all industrial sources could meet the goal of reducing industrial CO2 emissions by 20 percent. However, this result is very sensitive to errors in estimating the adjustment costs. If adjustment costs achieve 25 percent, the tax required to meet the target would rise to$51.47 per ton of CO2. Thus, the required tax is consistent with those applied in the United Kingdom ($16), Australia ($22), British Columbia and Ireland ($28), Denmark ($31), and Finland ($48). ## Conclusion In analyzing the effects on Chile of the tax reform, which includes a tax of US$5 per ton of CO2 emitted for thermal generating sources over 50 MW, we conclude that no industrial source will decide to opt for cleaner fuels; the quantity of CO2 tons actually emitted remains the same.

By analyzing more restrictive settings, it can be concluded that a tax of $5 to$10 per ton of the CO2 emitted by all industrial, independent sources of thermal power boilers does not generate major changes in the use of fuels. The reduction in emissions is 3.3 percent (172,125 tons of CO2) in nearly all the simulated scenarios, which is why industrial sources would find it more expensive to change their production systems to consume a lower contaminant fuel than to pay the tax.

## Author Information

Cristian Mardones is associate professor in the Department of Industrial Engineering, University of Concepción. He received a PhD in economics from the University of Chile in 2009. He has been active in the areas of environmental economics and economic modeling for more than five years, with publications in Energy Policy, Journal of Environmental Management, Economic Systems Research, Energy & Environment, Environment Development & Sustainability, Journal Economía Chilena, CEPAL Review, and Latin American Journal of Economics, among others. His professional experience includes consultancies for the Ministry of Environment in Chile. His current research involves studying the economic effects of carbon taxes.

Belén Flores graduated with a degree in industrial engineering from University of Concepción (Chile) in 2015.

## References

1. Agnolucci, Paolo. 2004. “Ex Post Evaluations of CO2-Based Taxes: A Survey.” Working Paper No. 52, Tyndall Centre for Climate Change Research, Norwich, United Kingdom.

2. Bjorner, Thomas, and Henrik Jensen. 2002. “Energy Taxes, Voluntary Agreements and Investment Subsidies: A Micro-Panel Analysis of the Effect on Danish Industrial Companies’ Energy Demand.” Resource and Energy Economics 24 (3): 229–249. DOI: https://doi.org/10.1016/S0928-7655(01)00049-5

3. EPA. 2009. “AP 42, Fifth Edition Compilation of Air Pollutant Emission Factors, Volume 1: Stationary Point and Area Sources.” https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-compilation-air-emission-factors.

4. Escobar, Carlos. 2013. “Evaluación técnico económica calderas auxiliares y de biomasa CMPC Pacífico.” http://cybertesis.ubiobio.cl/tesis/2013/escobar_c/doc/escobar_c.pdf.

5. Gallardo, Alvaro, and Cristian Mardones. 2013. “Environmentally Extended Social Accounting Matrix for Chile.” Environment, Development and Sustainability 15 (4): 1099–1127. DOI: https://doi.org/10.1007/s10668-012-9430-0

6. Gerbelová, Hana, Filipa Amorima, André Pinaa, Mário Melob, Christos Ioakimidisc, and Paulo Ferrãoa. 2014. “Potential of CO2 (Carbon Dioxide) Taxes as a Policy Measure towards Low-Carbon Portuguese Electricity Sector by 2050.” Energy Policy 69 (1): 113–119. DOI: https://doi.org/10.1016/j.energy.2014.01.011

7. INE. 2014. “Encuesta Nacional Industrial Anual 2011, ENIA.” Santiago de Chile: Instituto Nacional de Estadísticas.

8. IPCC. 2006. “Guidelines for National Greenhouse Gas Inventories.” Prepared by the National Greenhouse Gas Inventories Programme, and edited by Simon Eggleston, Leandro Buendia, Kyoko Miwa, Todd Ngara, and Kiyoto Tanabe. Institute for Global Environmental Strategies, Japan.

9. Larsen, Bodil, and Runa Nesbakken. 1997. “Norwegian Emissions of CO2 1987–1994.” Environmental and Resource Economics 9 (3): 275–290. DOI: https://doi.org/10.1007/BF02441400

10. Lin, Boqiang, and Xuehui Li. 2011. “The Effect of a Carbon Tax on Per Capita CO2 Emissions.” Energy Policy 39 (9): 5137–5146. DOI: https://doi.org/10.1016/j.enpol.2011.05.050

11. Mardones, Cristian, and Tamara Muñoz. 2017. “Impuesto al CO2 en el sector eléctrico chileno: Efectividad y efectos macroeconómicos.” Economía Chilena 20 (1): 4–25.

12. Ministry of Energy. 2015. “Energy Balances.” http://energiaabierta.cl/visualizaciones/balance-de-energia/.

13. Morley, Bruce. 2012. “Empirical Evidence on the Effectiveness of Environmental Taxes.” Applied Economics Letters 19 (18): 1817–1820. DOI: https://doi.org/10.1080/13504851.2011.650324

14. Muñoz, Tamara, and Cristian Mardones. 2016. “Simulation of a CO2e Tax to Mitigate Impacts from Chilean Agricultural and Livestock Sectors on Climate Change.” Agrociencia 50 (3): 271–285.

15. OECD. 2012. Energy, OECD Green Growth Studies. OECD Publishing. DOI: https://doi.org/10.1787/9789264115118-en

16. Pigou, Arthur. 1920. The Economics of Welfare. London: Macmillan.

17. Symons, Elizabeth, John Proops, and Philip Gay. 1994. “Carbon Taxes, Consumer Demand and Carbon Dioxide Emissions: A Simulation Analysis for the UK.” Fiscal Studies 15 (2): 19–43. DOI: https://doi.org/10.1111/j.1475-5890.1994.tb00195.x

18. UDT-PROTERM. 2011. “Inventario de emisiones para el Gran Concepción años base 2008 (fuentes fijas).” Prepared for the Environmental Ministry of Bío Bío Region, Chile.

19. Vera, Sonia, and Enzo Sauma. 2015. “Does a Carbon Tax Make Sense in Countries with Still a High Potential for Energy Efficiency? Comparison between the Reducing-Emissions Effects of Carbon Tax and Energy Efficiency Measures in the Chilean Case.” Energy 88: 478–488. DOI: https://doi.org/10.1016/j.energy.2015.05.067

20. Wissema, Wiepke, and Rob Dellink. 2007. “AGE Analysis of the Impact of a Carbon Energy Tax on the Irish Economy.” Ecological Economics 61 (4): 671–683. DOI: https://doi.org/10.1016/j.ecolecon.2006.07.034

21. World Bank. 2014. State and Trends of Carbon Pricing 2014. Washington, DC: World Bank.