Policy Issue Paper 14-04
Impact of Energy Tax Reform on the Power Market and Sensitivity Analysis
G. S. Park and S. G. Jo
Ⅰ. Background / 1
Ⅱ. Energy Taxes in Korea and Selected Countries / 5
Ⅲ. Methodologies, Models and Results / 16
Ⅵ. Conclusion and Policy Implications / 37
<References> / 42 Contents
Ⅰ. Background
1. Background
□ Since the 9.15 Blackout (September, 2011), the Korean government has been making efforts to introduce a transition in energy policy from supply expansion to demand management.
○ The main idea of this transition is to promote smart and rational energy use, contributing to resolving mismatches between energy supply and demand.
□ As part of the transition, the government adopted excise tax on bituminous coal electricity since July, 2014. This new tax is expected to mitigate excessive electrification.
○ Along with the introduction of a new tax on bituminous coal electricity at 18 won/kg1) on average, a tax cut on LNG was also adopted from 60 won/kg to 42 won/kg, by 18 won/kg.
○ Considered one of the primary causes of global warming, bituminous coal started to be taxed, while LNG and kerosene enjoy tax cuts. The rationale behind this is tax neutrality and a fairer competition between energy sources.
1) Like gasoline, natural gas and bituminous coal are subject to flexible tax rates. The tax rate for natural gas is 42 won/kg, 18 won lower than the basic rate. The tax rate for bituminous coal is 19 won/kg for 5,000kcal or higher of net calorific value and 17 won/kg for less than 5,000kcal net calorific value.
□ There are controversies over the tax rate on bituminous coal and the tax’s effectiveness in controlling electrification. Some argue that nuclear power, the only untaxed energy source, needs to be taxed as well.
○ The tax rate of bituminous coal electricity, 18 won/kg, is lower than that of LNG, 42 won/kg, even though bituminous coal has greater impact on the environment than LNG does.
○ Some argue that the new tax will not be effective in controlling electrification because mid-term and long-term electricity price is highly likely to decrease due to reopening of nuclear power units, additional power capacity installations, and fuel price decrease projections.
○ Alerted by the Fukushima Daiichi nuclear disaster, civic groups and some politicians claim that a new tax on nuclear power is necessary to incorporate external costs for the risk of a nuclear accident.
□ Despite these concerns, few of the previous studies have adopted a quantitative approach to analyze tax reforms on system marginal price (SMP) and electricity tariff.
○ Most of the previous studies on the effects of energy tax reforms have employed accounting methods without accounting for structural issues of the power market and the grid operation in Korea.
2. Objectives and Expected Achievements
□ This study aims to analyze effects of a number of options of tax reform.
○ Particularly it focuses on the electric sector, SMP, the electricity tariff, and tax revenue. Furthermore, we investigate the macro-economic impacts by looking at electricity tariff changes resulting from these energy tax reforms. And also, we examine the income redistribution impacts of various energy tax reforms throughout tax revenue changes.
○ To address the flaws in the previous literature, we used an engineering-based electric market simulator named M-Core model. This electric market simulator is capable of reflecting Korea’s national power grid system more realistically.
○ The results from the M-Core model will be analyzed through computable general equilibrium (CGE) to estimate the effects on the macro economy.
- We acknowledge that we made strong assumptions in attempt to address uncertainties in the real market. However, the model itself reflects better today’s electricity market conditions and structures, offsetting unrealistically simple assumptions.
□ This study set up a series of scenarios of tax reform and analyzed the tax reform’s effects on the power market and the national economy.
○ The new tax reform started this year has limitations when it comes to internalization of costs of external diseconomy.
○ Identifying potential problems of a tax reform is also one of the study’s objectives.
□ This study is expected to make policy suggestions that minimize bad effects and maximize good effects of the tax reform.
Ⅱ. Energy Taxes in Korea and Selected Countries
1. Energy Taxes and Tariffs in Korea
□ Energy Taxes
○ Energy taxes in place in Korea include import tax, individual excise tax, transportation/energy/environmental tax, education tax, local toll tax, and VAT.
- Other charges are also incurred on energy costs as well as taxes.
○ This study closely examine transportation/energy/environmental tax put on energy utilization.
○ Transportation/energy/environmental tax is one of the most sizeable source of tax revenues. For example, in the year of 2012, it totaled 13.8 trillion won. It is put on gasoline, kerosene, and other energy of kind.
- The tax was designed to collect financial resources to invest in transportation infrastructure, such as roads and railways, public transportation, energy and resource-related projects, and environmental projects.
- It was planned to expire in 2012, but it was extended for additional three years in 2013.
- The current tax rate is 475 won/l for gasoline and 340 won/l for
kerosene. The base rate can be adjusted by up to 30 percent through a presidential decree (Act on Transportation/Energy/Environmental Tax Article 2.3).
○ Most energy products except for gasoline and kerosene are subject to individual excise tax. The range of taxable products and rates are defined in Act on Excise tax Article 1.2.
- A tax on bituminous coal was newly added in 2014.
- The tax rate can be adjusted as Act on Transportation/Energy/
Environmental Tax (Act on Excise tax Article 1.2).
- Flexible tax rates were applied to natural gas and bituminous coal, keeping taxes lower at 42 won/kg for natural gas, lower by 18 won/kg than the base rate, and at 19 won/kg for 5,000 kcal or higher bituminous coal and at 17 won/kg for less than 5,000kcal.
□ Energy pricing mechanism
○ Not only taxes, but also sur taxes are on energy.
- Gasoline, diesel, LPG, kerosene, and heavy oil are subject to transportation/energy/environmental tax and 15 percent of individual excise tax. Therefore, the tax revenues from this will go to education business.
- The local toll tax is 36 percent of transportation/energy/environmental tax. All of this goes to the vault of the local governments.
○ [Table II-1] shows energy pricing mechanism composed of taxes and
charges as of December 2013.
- As of December 2013, taxes and charges are relatively high on transportation energy, including gasoline and diesel.
· Unlike other energy, transportation energy has a high ratio of taxes and charges: Gasoline 49.6 percent, diesel 41.1 percent.
- The excise tax rate of heavy oil is remarkably low at 11.2 percent.
- Secondary energy, including electric power and thermal energy, is not taxable, so the ratio of tax to price is low.
· However, considering LNG and oil inputs in electric and thermal power, the ratio of tax increases significantly.
Gasoli ne(won
/ℓ)
Domest ic Kerosen e(won/ℓ
)
Diesel (won/ℓ)
Heavy Oil (B-C) (won/ℓ)
LPG
(won/kg) LNG (won/kg
)
Briqu ette (won/
unit)
Electricity (won/kWh)
Therma l (won/
10thous andkcal
) Profan
e Butan Resid
ential Night
Import Duties
Basic 3% 3% 3% - - - -
Allocati on
Crude oil(for naphta) 0%, Products are subject to allocation, but to lower basic
rates.
Product 0%, Crude Oil 0%
2%
(Winter season)
- - - -
Individu al Excise
Tax
Basic - 90 - 17 20 252 60 - - - -
Flexibl
e - 90 - 17 14 275 60 - - - -
Transp ortation
Energy Tax
Basic 475 - 340 - - - - - - - -
Flexibl
e 529 - 375 - - - - - - - -
Education Tax 79.35 13.5 56.25 2.55 - 41.25 - - - - -
Local Toll Tax 137.54 - 95.50 - - - - - - - -
VAT 171.01 122.93 154.46 84.51 189.9 208.2 102.9 - 10% of Fixed and variable rates
Import Charge 16 16 16 16 - - 24.2 - - - -
Quality
Inspection Fee 0.47 0.47 0.47 0.47 0.027 0.027 - - - - -
Safety Management
Charge - - - - 5.0 5.0 5.5 - - - -
Sales Charge
High quality (36)
- - - - 62.28 - - - - -
Electricity
Industry Fund - - - - - - - -
3.7% of power bills
and VATs -
Total
Tax Subtotal (Excludin g Import Duties)
933.4 242.9 699.7 104.5 215.0 591.7 192.6 - 19.4 8.4 8.4
Ratio of Tax to
Price
49.6 18.0 41.2 11.2 10.3 25.1 17.0 0 12.1 12.1 9.1
Price (December
2013)
1,881.1 1,352.2 1,699.1 929.6 2,094.3 2,357.3 1,132.1 391.3 125.9 55.4 91.9 Note: Taxes as of January 22, 2014, Prices as of December 2013. Price indices for gasoline, domestic
kerosene, and diesel are from gas station prices, heavy oil from sales offices in November 2013, LPG from sales offices, LNG from average KOGAS wholesale prices in January 2014 converted in MJ, thermal from residential space heating rate (basic rate not incorporated).
Source: Opinet.co.kr, Ministry of Strategy and Finance, Ministry of Industry and commerce, National Legal Information Center, KEPCO website, Korea District Heating Corporation, and Energy
[Table Ⅱ-1] Energy Taxes and Prices as of December 2013
□ Issues in current energy taxes
○ Current energy taxes do not fully internalize external costs.
- Current transportation/energy/environmental tax and excise taxdo not internalize external costs such as pollutants.
- There is no available study that estimate social costs of pollutants at this point.
- Current energy taxes have not been built upon the careful and thorough analysis on social costs of energy use, but upon tax revenue creation.
This approach needs to be revised.
○ The principle of tax equity is not being properly respected.
- To properly internalize external costs, taxing high on energy that pollutes more is a logical course of measure. However, in Korea, tax rates are decided regardless of polluting potential.
- The current practice is that all the energy sources use the same calorific unit to determine tax rates and the Korean government adjusts them in a way not to undermine industrial competitiveness.
· For example, heavy oil has a higher pollutant coefficient than kerosene and LNG, but is subject to a lower tax rate.
- The fact that the current taxes do not address external costs properly also threatens the principal of tax equity.
○ Types of taxes are more than needed and complicated.
- As already shown previously, taxes not related to energy, including
education tax, local toll tax, and VAT, are attached to energy prices.
· This implies that energy taxes are not built upon internalization of external costs, but upon maximizing tax revenues and diversifying revenue uses.
- To address the issues mentioned above, an amendment took place in 2014, coordinating tax rates of certain energy sources. This amendment is expected to better reflect social costs of energy and improve equity.
· However, many experts think that the effects would still remain minimal.
2. Energy Taxes and Tariffs in Selected Countries
□ We categorized countries into three: taxing energy for power generation only, taxing electricity only, and taxing both.
○ Israel and Mexico are one of those who tax energy use for power generation only, whereas Germany and Hungary tax electricity only.
○ Japan and Italy tax both energy inputs for power generation and electricity itself.
○ This research covers the countries that tax electricity regardless of taxing energy inputs for power generation.
□ The majority of countries that tax electricity adopt excise taxand VAT together.
○ Tax rates vary a lot, but in general, European countries have higher tax rates. Non-European countries tend not to put tax on electricity or even if so, tax rates are low.
- In Denmark and Germany, tax rates for residential power use are 145.85
$/MWh and 131.79$/MWh respectively, significantly higher than the average 23.2 $/MWh.
- Denmark shows the largest difference in tax rates between residential use and industrial use, which are 145. 85$/MWh and 3.84 $/MWh.
- Similarly, Germany has a tax rate for industrial use half of that for residential use.
- In Sweden, residential tax rate is 42.71 $/MWh, showing a big difference compared to industrial tax rate, 0.77 $/MWh.
- To the contrary, Italy’s tax rate for industrial power use is 123.89
$/MWh, higher than residential tax rate 77.64$/MWh.
○ Except the cases aforementioned, industrial power tax rates are slightly lower than residential.
- Japan and Poland places the same tax rates on residential and industrial.
○ All the governments that place excise taxon electricity also levy VAT at the same time. The U.S., Canada, and Australia neither levy excise taxnor VAT on electricity.
- In certain cases, including Korea’s, there is no excise taxon electricity, but VAT.
Residential Electricity Industrial Electricity VAT(%)2) Excise tax3) (USD/MWh) Excise tax
(USD/MWh)
Australia ‥ ‥ ‥
Austria 20.0 43.79 29.95
Belgium 20.0 24.97 22.92
Canada ‥ ‥ ‥
Chile 19.0 - -
Czech 21.0 1.49 1.41
Denmark 25.0 149.85 3.84
Estonia 20.0 18.01 17.74
Finland 24.0 23.23 9.59
France 17.5 38.89 26.78
Germany 19.0 131.79 65.35
Greece 13.0 29.20 26.47
Hungary 27.0 2.17 9.67
Ireland 13.5 - -
Israel 17.3 - -
Italy 10.0 77.64 123.89
Japan 5.0 3.63 3.63
Korea4) 10 - -
Luxemburg 6.0 18.01 5.87
Mexico 16.0 - -
NetherlandsThe 21.0 15.54 12.14
New Zealand 15.0 - -
Norway 25.0 18.90 -
Poland 23.0 6.53 6.53
Portugal 23.0 - -
Slovakia 20.0 - -
Slovenia 22.0 23.08 15.08
Spain 21.0 11.94 7.09
Sweden 25.0 42.71 0.77
Switzerland 8.0 5.01 5.01
Turkey 18.0 10.24 4.10
UK 5.0 - 3.98
USA ‥ ‥ ‥
Average 17.6 23.2 13.4
[Table Ⅱ-2] Energy Taxes by Country as of the 1st Quarter of 2014(unit:
USD/MWh)
2) Goods and service tax in Australia, Canada, and New Zealand; Sales tax in the U.S.; Consumption tax in Japan.
3) Excise tax is all the taxes on energy except VAT.
4) Korea has an additional 3.7% of Power Industry Fund contribution, a type of quasi-tax.
- Note: The most recent figures. (.. no data; x not applicable; - non-existent; c – not being public) - Source: IEA, Energy Prices and Taxes. 2014. 1stQ; Kyung-sun Huh(Feb. 2014), Finance Forum. Feb.
2014, Issue 212, p.34.
3. Nuclear Power Taxes in Selected Countries
□ Countries that already put in place or plan to put in place taxes on nuclear power include Germany, Sweden, Belgium, Slovenia, Japan, France, and Brazil.
○ In June, 2011, even before the moratorium on nuclear power plans, Germany started to tax nuclear fuels. In exchange of permission on operation of nuclear power units that expire the life time, the German government taxed fuels for nuclear power units.
- The tax is on nuclear fuel rod at the rate of €145/g, equivalent to
$195/g.
- However, Fukushima Nuclear accident nudged the Merkel administration into declaring the moratorium on nuclear power5), which practically led to a repeal of nuclear fuel rod tax.
○ The Swedish government applies differentiated tax rates on nuclear power. As of the end of 2013, Sweden has 10 units of nuclear power on operation.
- The tax rate is 0.67 eurocents/kWh, approximately one third of the nuclear operating cost.
5) The decision to shut down all the nuclear power plants in Germany by the year 2022.
○ In Belgium, on condition that an extended life time is allowed to obsolete nuclear power units, nuclear power producers pay special tax of 0.5 euro cents/kWh. The special tax totals up to 215 to €245 million per year.
- The tax revenues are used for renewables and energy demand management.
○ In Slovenia, €0.30 has been imposed on every kWh of nuclear power sold domestically since 2008.
○ The French and Brazilian governments are discussing taxing nuclear power units. The expected tax revenues are planned to promote renewables in France and to go into local governments’ budgets in Brazil.
- Brazil plans to impose 10 percent of nuclear power sales as tax and give the tax revenues to local governments that operate nuclear power units as token of compensation.
○ In Japan, nuclear fuel tax and after-use nuclear fuel tax are in place.
The former was introduced first in 1976 in Fukui Prefecture and now in place in 13 prefectures, while the latter was first adopted in 2003 in a few districts and towns.
- The purposes of these taxes are environment and welfare improvements and in neighborhoods of nuclear power units. The initial tax rate was roughly 5 percent of the price of nuclear fuel, but now is 8.5-13 percent, reflecting inflation (refer to [Table II-2]).
Tax Prefecture/
District
Introduction Year and Initial
Tax Rate Tax Rate at Present Remarks
Nuclear Fuel Tax
Fukui 1976(5%)
① 8.5% of nuclear fuel price
② 45,750 yen/thousand kWh(for 3 months)
New taxes introduced on nuclear power operation in 2011
Fukushima 1977(5%) 10% of price 8,000 yen/kg
Ehime 1979(5%) 13% of nuclear fuel price Saga 1979(5%) 13% of nuclear fuel price Shimane 1980(5%) 13% of nuclear fuel price Shizuoka 1980(5%) 13% of nuclear fuel price Gagoshima 1973(7%) 12% of nuclear fuel price Miyagi 1983(7%) 12% of nuclear fuel price Niigata 1984(7%) 14.5% of nuclear fuel
price
Hotkaido 1988(7%) 12% of nuclear fuel price Ishigawa 1992(7%) 12% of nuclear fuel price
Nuclear Material Treatmen
t Tax
Ibaragi 1978(5%)
① 13% of nuclear fuel price
② 46,000 yen/kg
③ 1,219,000 yen/m3
④ 1,219,000yen/unit
⑤ 81,100yen/㎥
⑥ 3,900yen/㎥
Nuclear fuel tax changed to nuclear material treatment tax in 1999 and extended to four items;
extended to six items in 2011
Aomori 2004(10%)
① 19,000yen/kg
② 9,000yen/thousand kW(3months)
③ 13% of nuclear fuel price
④ 19,400yen/kg
⑤ 1,300yen kg(8,300yen/kg)
⑥ 27,500yen/㎥
⑦ 845,400yen/unit
Extended from four items(2004) to seven items recently
[Table Ⅱ-3] Nuclear Fuel Taxes in Japan as of 2014
- Source: Byoung-ho Choi et al. (Nov. 2013), p.64; Kyung-sun Huh (Feb. 2014), p.33
Ⅲ. Methodologies, Models and Results
1. Methodologies and Models
□ We set up eleven energy tax reform scenarios based on the previous literature.
○ The major studies that we referred to include ‘Research on Energy Tax Reform Basic Plan’ of Korea Institute of Public Finance, Tae-hun Kim et al.(Jan., 2014), the 2nd Energy Basic Plan (Jan. 2014, "2nd Basic Plan”
hereinafter)’, Kyung-sun Huh (Feb., 2014), and the ‘Amendment to Excise Tax Act (Feb., 2014). Based on the reports and the studies, eleven scenarios were laid out (refer to [Table III-1]).
- One of the objectives of this study is to analyze the impacts of energy tax reforms on SMPs, electricity tariffs, and tax revenues. Therefore, we accounted for the upper and lower limits of flexible tax rates of 18 won/kg and 30 won/kg for bituminous coal electricity and 60 won/kg of flat tax rate for the same coal, which was set at the same level with pre-reform LNG excise tax rate.
- Pre-reform LNG tax rate is 60 won/kg and post-reform 42 won/kg.
- We also accounted for excise tax on nuclear power generation at 5.75 won/kWh, into which hidden costs of generating electricity with nuclear power is incorporated. The hidden external costs were determined based on the ‘2th National Energy Basic Plan’ established in Jan. 2014.
○ We designated the ‘6th Power Supply and Demand Basic Plan (‘6th Basic Plan’ hereinafter)’ as a reference scenario in this study. Our forecast period is from 2014 to 2027, coincided with the 6th Basic Plan.
Scenario
Bituminous Coal Excise Tax
(won/kg)
LNG Excise Tax
(won/kg) Nuclear Excise Tax (won/kg) Reference
Scenario
No tax on
nuclear 0 60* 0
Reference Scenario-1
Tax on
nuclear 0 60* 5.75
Scenario 1 No tax on
nuclear 18 42 0
Scenario 1-1 No tax on
nuclear 18 60* 0
Scenario 1-2 Tax on
nuclear 18 42 5.75
Scenario 2 No tax on
nuclear 30 42 0
Scenario 2-1 No tax on
nuclear 30 60* 0
Scenario 2-2 Tax on
nuclear 30 42 5.75
Scenario 3 No tax on
nuclear 60 42 0
Scenario 3-1 No tax on
nuclear 60 60* 0
Scenario 3-2 Tax on
nuclear 60 42 5.75
[Table Ⅲ-1] Excise Tax Rate Assumptions by Scenario
* : Assumed to be the same as the current LNG excise tax rate (before July, 2014)
□ To analyze the impact of the tax reform on SMPs, we compared SMPs in the scenarios with one in the reference scenario.
○ We employed the M-Core model to estimate long-term SMPs. We used the basic input data reported by the ‘6th Basic Plan’
□ In analyzing changes in electricity tariffs, we calculate the adjusted settlement price (adjusted wholesale price of electricity, which are considered as a proxy for electric tariff in this study).
○ In Korea’s electricity pricing mechanism, it is nearly impossible to identify electricity purchase price because the final electricity bill is adjusted by coefficients after the power has been used. Furthermore, it is even harder to deduce future trends of power purchase prices because there are a lot of factors that intervene in, including carbon abatement costs, renewable purchase prices, investment in transmission and distribution, resident compensation, and investment in demand-side management, such as smart meter promotion.
○ We maintained a strong assumption that the level of uncertainties remain constant across the scenarios so that we can make the changes in electricity tariffs proportional to those in adjusted settlement prices.
○ We believe that the results will be still significant despite this unrealistically strong assumption, because the objective of this study is to estimate the changes in electricity tariffs, not the absolute levels of them.
□ The primary tool that this study employed is the engineering-based electric market simulator named M-Core model.
○ The M-Core model can estimate SMPs by hour, which helps identify SMP and adjusted settlement price of each generator by hour based on
the information of power production volume of each generator by hour.
○ Further information about the optimizing engine algorithm and user guidelines of the M-Core model can be found in ‘Space of Artisans’
(2011) and Hyung-tae Kim et al. (Oct. 19, 2012).
□ The basic input data for the simulator model is mostly from the '6th Power Supply and Demand Basic Plan'.
○ The mid- and long-term power demand is deduced by applying scaling factors that reflect past power consumption patterns on the forecast target demand in the ‘6th Basic Plan'.
○ The estimates of power capacity took into account expansion or shutdown plans of the ‘6th Basic Plan’, including 2,000MW of coal-fired power plants and 1,820MW of LNG power plants, which are originally planned in preparation for future uncertainties.
- Also we assumed that the Shin Wolsong unit 2, the Shin Gori unit 3 and 4 will start operation in July, 2014, September, 2015 and September, 2015 respectively, which come later than planned, because there have been corruption scandals around nuclear power in Korea, which would clearly hamper progress of constructions underway of nuclear power units.
- The Gori unit 1 and the Wolsong unit 1, of which the former is now operating and the latter is being reviewed to operate beyond their lifetimes, are assumed to shut down late June, 2027 and late December, 2022 respectively.
- For a sensitivity analysis, we set up two scenarios, in one of which roughly 9,000MW coal power is delayed for 10 months and in the other of which approximately 5,000MW of LNG power is delayed for 8 months.
○ Data inputs for specifications of power generation units, fuel costs, grid constraints, fuel constraints, MOR, EFOR, auxiliary power ratio, O&M costs coincide all with the 6th Basic Plan.
○ TLFs were determined based on reports of Korea Power Exchange (KPX). For new power units, TLFs were set similar to those of the most likely power units.
○ Inputs for hydro/pumped-storage hydro power production, generation patterns, and stocks on bid were selected from the database of Korea Electric Power Corporation (KEPCO), Korea Electric Power Statistics, and past records of Korea Hydro Nuclear Power and K-Water.
○ Transmission, fuel, and thermal constraints were rebuilt through a computable model based on estimates of KEPCO and KPX and past records.
○ Forecasts on fuel prices were taken from fuel price by source and calorific value given in the 6th Basic Plan.
○ In practice, hourly CP coefficients and regional CP coefficients vary year by year, but this study coincides regional CP coefficients and hourly CP coefficients with 2013 standard CPs.
○ As for settlement coefficients, we applied nuclear and bituminous coal adjustment coefficients of the first quarter of the year 2014, while keeping the adjustment coefficient of general power generators, such as LNG combined cycle and petroleum), at one. This set of inputs remain constant until the year of 2027.
2. Results (1): Impact on SMPs
□ Imposing excise tax on bituminous coal for power and reducing excise tax on LNG for tax neutrality, in the short term, SMP will go down further than the reference scenario, which is a case of no tax reform (refer to [table III-3]).
○ The reason behind this is that SMPs are determined by LNG combined cycle most of the times, as a rising power demand already the reduced power reserve ratio. Changes in variable costs of bituminous coal power end up not affecting SMPs.
- The fact that decreases in SMPs will move adjusted settlement prices in the same direction signifies that the government’s attempt to control electrification by taxing bituminous coal is mistaken.
- Decreases in SMPs will place downward pressure on electricity tariffs, thus leading to increases in power demand.
- However, new coal power plants are anticipated to enter the market over the mid and the long haul, the impact of coal-fired power on
SMPs will increase. As a result, taxing bituminous coal will put an upward pressure on SMPs.
○ In order to meet the goal of curbing power demand increases, the right course of action is to tax bituminous coal in the short term and adjust tax rates on LNG in the long term, when new bituminous coal-fired power units enter the market.
□ It is anticipated that nuclear power tax will have a miniscule impact on SMPs.
○ SMPs barely change in comparing the reference scenario with the nuclear-fuel taxscenario(Reference Scenario-1) and comparing within energy tax reform scenarios – e.g. Scenario 1 vs. Scenario 1-1 vs.
Scenario 1—2).
- This is because taxing 5.75 won/kWh does not really help increase nuclear power production costs to a comparable level to other types of power production and nuclear rarely has a decisive effect on SMPs during any time of the day.
□ When the excise tax on coal is levied by 18Won/kg, SMP will rise by 13% compared to the reference scenario. When the tax rate rise by 30Won/kg, then SMP will increase by 22 percent.
○ Likewise, when the tax rate on coal rise by 60Won/kg, SMP will increase by 43 percent (76.36 won/kWh by 2024) compared to the reference case.
○ A tax increase on bituminous coal will of course lead to an increase in SMPs, however, the magnitude of the increase is far less than 132.89 won/kWh in the 2014 reference scenario.
○ The implication of this is that the tax reform consisting of 24 won/kg of base tax rate and 18 won/kg of flexible tax rate on bituminous coal will have a limited impact on abating electrification in the long run.
○ Therefore, it is essential to impose a higher tax rate on bituminous coal in order to ensure the tax’s role of relieving environmental degradation and limiting electrification.
- Considering that LNG tax rate is 60 won/kg, a tax on bituminous coal, which is more polluting than LNG, is supposed to be higher than the current level.
○ Accounting for the factors that are not reflected in the analysis, such as the expected wider use of renewables and carbon credit prices, there exists a chance of electricity rates going up. However, assuming that the forecasts of the 6th Basic Plan would hold, the tax reform does not much effect on limiting electrification.
24 Policy Issue Paper 14-04 Year Reference
Scenario-1 Scenario
1 Scenario
1-1 Scenario
1-2 Scenario
2 Scenario
2-1 Scenario
2-2 Scenario 3
2014 0.00 -1.66 0.10 -1.66 -1.58 0.18 -1.58 -1.39
2015 0.00 -1.78 0.17 -1.78 -1.64 0.32 -1.64 -1.36
2016 0.00 -0.91 0.93 -0.91 -0.24 1.60 -0.23 1.41
2017 0.02 -0.09 1.66 -0.08 1.09 2.84 1.11 4.04
2018 0.03 0.37 2.07 0.40 1.84 3.54 1.87 5.34
2019 0.08 2.40 3.85 2.48 5.05 6.51 5.13 11.47
2020 0.34 7.27 8.12 7.60 12.87 13.71 13.20 26.61
2021 0.64 9.94 10.45 10.56 17.17 17.67 17.79 34.60
2022 0.84 12.65 12.83 13.50 21.27 21.45 22.12 42.10
2023 0.83 12.66 12.83 13.49 21.42 21.57 22.24 42.42
2024 1.00 13.06 13.18 14.06 22.01 22.10 23.01 43.42
2025 1.22 12.90 13.03 14.12 21.97 22.05 23.19 42.82
2026 1.25 12.81 12.90 14.04 21.72 21.80 22.95 42.14
2027 1.19 12.49 12.63 13.69 21.34 21.41 22.54 40.73
[Table Ⅲ-3] Percent Changes in SMPs Compared to the Reference Scenario
3. Results (2): Impact on Electricity Tariffs
□ In analyzing changes in electricity tariffs, we calculate the adjusted settlement price (adjusted electricity purchased costs by electricity wholesaler, i.e. KEPCO, which are considered as the proxy for the electric tariff in this study).
○ We faced a lot of uncertainty factors, including Vesting Contract, renewable credit prices, carbon emission credit prices, and power transmission and distribution costs, which seem quite unpredictable due to the ongoing process of the Act on Compensation and Supports for Residents near Transmission and Distribution Lines. Therefore, it is nearly impossible to estimate variables that affect power tariffs in future.
□ The effects of imposing tax on coal on electricity tariff are similar with ones of SMP. If the excise tax rate on coal is 18Won/kg on average, while the tax on LNG is reduced from 60Won/kg to 42Won/kg, electricity tariff will increase by 7.6 percent compared to the reference case.
○ Likewise, if the tax rate on coal sets to 30Won/kg combined with 42Won/kg tax rate on LNG, then the electricity tariff will rise by 12.8 percent in 2026 year. In case of 60Won/kg tax rate on coal with 42Won/kg on LNG, the electric tariff will rise further by 25.0 percent in 2026 year compared to the reference case.
○ The tax reform of imposing 18 won/kg, 30 won/kg, and 60 won/kg on
bituminous coal, while maintaining 60 won/kg on LNG, turned out to bring about a higher change in electricity tariffs compared to the reform of the same tax rates on bituminous coal except a lower tax rate on LNG. However, in the long run, the changes do not differ greatly.
- This is because LNG combined cycle is anticipated to decline in use in the mid- and long-term, which makes the impact of LNG tax miniscule on power tariffs.
□ Scenarios that involve nuclear-fuel taxes do not show strong changes in SMPs, but in electricity prices. Adjusted settlement prices seem to play a great role in this result.
○ Nuclear-fuel tax up to 5.75 won/kg, proposed by the 2nd Basic Plan after incorporating the hidden costs such as the cost of accident-risk response and political costs, is analyzed to increase electricity tariff about 1.23 to 1.75 times higher compared to not taxing on nuclear power generation.
□ The tax reform of imposing 18 won/kg on bituminous coal electricity, which is to take effect in July this year, is expected to raise power tariffs by 1.1(2014)-7.6(2026) percent. Its impact on the power market is anticipated to be limited until the year 2015.
○ A tax rate of 30 won/kg on bituminous coal is expected to raise power tariffs by 2.6-5.2 percent compared to the reference scenario in the short term. Although the short-term effect seems insignificant, the long-term change in power tariffs is as great as 12.8 percent (in case of no
nuclear-fuel tax) and 18.5 percent (in case of having nuclear-fuel tax).
○ Increasing bituminous coal tax to 60 won/kg is expected to push up the power tariffs by 24.9 percent (2026 in scenario 3) to 30.6 percent (2026 in scenario3-2).
- Only if electricity prices will eventually decline over the long run and offset the increases made by the tax reform, 60 won/kg of tax on bituminous coal is also worth considering.
28 Policy Issue Paper 14-04 Year Reference
Scenario-1 Scenario
1 Scenario
1-1
Scenario 1-2
Scenario 2
Scenario 2-1
Scenario 2-2
Scenario 3
2014 1.57 1.07 2.27 2.64 2.64 3.84 4.21 6.34
2015 1.98 1.45 2.63 3.42 3.25 4.44 5.23 7.52
2016 2.44 2.44 3.41 4.88 4.80 5.77 7.24 10.39
2017 2.75 3.04 3.89 5.80 5.74 6.59 8.49 12.16
2018 2.98 3.27 4.08 6.24 6.09 6.91 9.07 12.79
2019 3.21 4.28 4.93 7.49 7.66 8.31 10.87 15.72
2020 3.68 5.85 6.21 9.53 10.13 10.48 13.80 20.43
2021 4.24 6.66 6.88 10.89 11.40 11.61 15.63 22.73
2022 4.77 7.43 7.52 12.20 12.51 12.59 17.28 24.67
2023 4.87 7.48 7.56 12.35 12.66 12.73 17.53 24.96
2024 5.33 7.54 7.60 12.87 12.71 12.76 18.04 24.99
2025 5.66 7.49 7.55 13.14 12.72 12.76 18.37 24.77
2026 5.72 7.58 7.63 13.29 12.83 12.88 18.54 24.92
2027 5.70 7.52 7.59 13.22 12.80 12.85 18.50 24.61
[Table Ⅲ-4] Percent Changes in Adjusted Settlement Prices Compared to the Reference Scenario
□ The sensitivity analysis of delays of power plant constructions in the 6th Basic Plan and endogenous changes in power demand shows that SMPs will rise by 5.42 percent compared to the reference scenario.
○ Delays in power plant construction are very likely because the national plan for transmission and distribution is under uncertainties and the power market and public sentiments will change a lot over a period of time. If there are delays, SMPs and electricity prices could go even higher than estimated in this study.
□ Our sensitivity test results incorporating endogenous factors of power demand and tariffs indicate that the mid- and long-term power demand, re-estimated SMPs and electricity tariffs will be higher than the corresponding forecasts of the 6th Basic Plan(refer to [Table III-5]).
○ SMPs and adjusted settlement prices decline over time even assuming re-estimated power demand.
Year
SMP(won/kWh) Adjusted Settlement Prices(won/kWh)
ChangeSMP Rate(%)
Adjusted Settlement
Prices Change Rate
(%) Reference
Scenario (6th Basic Plan)
Power Demand
Change (Scenario
)
Reference Scenario (6th Basic
Plan)
Power Demand
Change (Scenario)
Change Rate Compared
to the Reference
Scenario
Change Rate Compared to the Reference
Scenario
2014 132.89 132.89 91.84 91.84 0.00 0.00
2015 109.03 115.20 80.24 83.33 5.66 3.85
2016 95.90 102.36 73.14 76.10 6.74 4.05
2017 87.05 95.80 69.54 73.00 10.05 4.98
2018 83.70 93.42 68.54 72.44 11.61 5.69
2019 75.06 87.83 65.93 70.53 17.01 6.98
2020 61.93 79.19 63.03 68.54 27.87 8.74
2021 56.39 75.56 60.84 67.04 34.00 10.19
2022 52.91 69.80 59.46 64.62 31.92 8.68
2023 53.53 68.73 59.93 64.40 28.40 7.46
2024 53.24 66.19 59.60 63.12 24.32 5.91
2025 53.38 66.39 59.42 63.06 24.37 6.13
2026 54.11 66.74 59.80 63.36 23.34 5.95
2027 54.93 66.95 60.40 63.90 21.88 5.79
[Table Ⅲ-5] Sensitivity Results of SMPs and Adjusted Settlement Prices
4. Results (3): Impact on Tax Revenues
□ According to the estimates of tax revenues by energy tax reform scenario, the total tax revenues of the reference case might be 1,101.1 billion won in 2014, and 471.5 billion won in 2027.
○ The reason behind the continuous decrease in tax revenues is that the electricity mix would change in the long-run causing the capacity factor of the combined cycle generator to decline sharply.
□ In case of the 2014 amendment, the tax revenues will be 2,018.0 billion won in 2014, 2,224.3 billion won in 2019, and 1,756.7 billion won in 2027.
○ The government might collect more energy tax revenues 1 trillion won to 1.7 trillion won after the 2014 amendment.
□ When the government levies a tax up to 5.75 won/kWh on nuclear power generation, the revenues could increase by 1.5 trillion won compared to when the government does not levy a tax on it in 2027.
○ These estimated results might be understated and are likely to increase in the long-run because there exits various uncertainty such as the policy changes of generation mix, power demand fluctuations, environmental policy changes, grid construction and power plant construction delays and so forth. Depending on a variety of uncertainty variables, the potential tax revenues will increase further.
□ [Table III-6] and [Table III-7] give a glimpse on tax revenue estimates and comparisons with the reference scenario.
Korea Energy Economics Institute 33 Year Reference
Scenario
Reference Scenario-1
Scenario
1
Scenario
1-1
Scenario
1-2
Scenario
2
Scenario
2-1
Scenario
2-2
Scenario
3 2014 11,011 20,213 20,180 23,479 29,382 28,492 31,791 37,694 49,270 2015 10,003 19,734 20,125 23,126 29,857 28,874 31,875 38,605 50,746
2016 7,223 17,818 20,519 22,686 31,114 30,828 32,994 41,422 56,597
2017 6,114 17,112 21,227 23,061 32,226 32,524 34,358 43,523 60,767
2018 6,342 17,931 21,585 23,488 33,174 33,016 34,918 44,605 61,590
2019 5,766 17,584 22,243 23,973 34,061 34,381 36,111 46,199 64,722
2020 4,823 17,248 20,805 22,252 33,231 32,424 33,871 44,849 61,469
2021 4,751 17,938 19,639 21,064 32,825 30,514 31,939 43,700 57,701
2022 4,681 18,559 17,991 19,395 31,868 27,800 29,204 41,677 52,319
2023 4,662 18,498 17,859 19,257 31,695 27,589 28,987 41,425 51,909
2024 4,670 19,162 16,856 18,257 31,350 25,915 27,315 40,408 48,555
2025 4,672 19,673 16,741 18,142 31,743 25,721 27,122 40,723 48,164
2026 4,673 19,669 17,082 18,484 32,078 26,290 27,691 41,286 49,300
2027 4,715 19,602 17,276 18,690 32,163 26,591 28,006 41,479 49,859
[Table Ⅲ-6] Tax Revenue Estimates by Scenario (unit: 100 million won)
- Note: Fuel consumption excerpted from estimates of power plants’ operational plans, while power generation from estimates of transmission and distribution lines.
34 Policy Issue Paper 14-04
Year Reference
Scenario-1 Scenario1
Scenario
1-1
Scenario
1-2
Scenario
2
Scenario
2-1
Scenario
2-2
Scenario
3
2014 9,202 9,168 12,468 18,370 17,480 20,779 26,682 38,259
2015 9,731 10,122 13,123 19,853 18,871 21,872 28,602 40,743
2016 10,595 13,296 15,463 23,891 23,605 25,772 34,199 49,374
2017 10,998 15,113 16,947 26,112 26,410 28,244 37,409 54,653
2018 11,589 15,243 17,146 26,832 26,673 28,576 38,262 55,248
2019 11,818 16,477 18,207 28,296 28,615 30,345 40,433 58,957
2020 12,425 15,983 17,429 28,408 27,601 29,048 40,027 56,646
2021 13,187 14,888 16,313 28,074 25,763 27,188 38,949 52,950
2022 13,878 13,310 14,714 27,187 23,119 24,523 36,996 47,638
2023 13,836 13,197 14,595 27,033 22,927 24,325 36,763 47,247
2024 14,492 12,187 13,588 26,680 21,245 22,646 35,738 43,885
2025 15,002 12,069 13,471 27,071 21,049 22,451 36,051 43,492
2026 14,996 12,409 13,811 27,405 21,617 23,018 36,613 44,627
2027 14,887 12,560 13,975 27,447 21,876 23,290 36,763 45,143
[Table Ⅲ-7] Differences in Tax Revenue Estimates Compared with the Reference Scenario (unit: 100 million won)
5. Results (4): Impact on Macroeconomy and Income Redistribution
□ The impacts of the energy tax changes on macroeconomy might be somewhat negative. They kick in the economy through increases in power prices, increases in consumer prices, decreases in consumption and investment, and decreases in GDP.
○ If the tax rate enacted this year in the Amendment to Excise Tax Act will be maintained, GDP is forecasted to decrease by 0.03 percent compared to the reference scenario in 2014, and the decline will be further expanded by 0.26 percent in 2027.
○ In the short-run, the nominal income level would be also reduced and the effect will be expanded further in the longer run.
○ By income quintiles, the income of first income quintile would decrease the most. One who has higher income is more likely to see a smaller negative impact on income. As a result, the energy tax reform brings about a regressive effect to lower income households.
□ According to the estimates of Gini coefficient, the 2014 amendment could aggravate the income redistribution from 0.3815 to 0.3818 (refer to [Table III-8]).
○ These negative consequence for the macroeconomic indicators and income redistribution is deemed to have occurred because we do not assume the tax neutrality in this study.
36 Policy Issue Paper 14-04
Nominal Income Direct Impact on Electricity
Direct and Indirect Impact on
Electricity
Space Heating
Energy Transportation
Energy
Reference Scenario 0.3815 0.3847 0.3847 0.3904 0.3827
Year 2015 2025 2015 2025 2015 2025 2015 2025 2015
Reference Scenario
1 0.3818 0.3822 0.3847 0.3848 0.3850 0.3855 0.3907 0.3913 0.3829
Scenario 1 0.3817 0.3824 0.3847 0.3849 0.3849 0.3858 0.3907 0.3916 0.3829
Scenario 1-1 0.3818 0.3825 0.3847 0.3849 0.3850 0.3858 0.3908 0.3916 0.3830
Scenario 1-2 0.3818 0.3830 0.3848 0.3850 0.3851 0.3865 0.3909 0.3923 0.3830
Scenario 2 0.3818 0.3830 0.3848 0.3850 0.3851 0.3865 0.3909 0.3923 0.3830
Scenario 2-1 0.3819 0.3830 0.3848 0.3850 0.3852 0.3865 0.3910 0.3923 0.3831
Scenario 2-2 0.3820 0.3835 0.3848 0.3852 0.3853 0.3872 0.3911 0.3930 0.3831
Scenario 3 0.3821 0.3841 0.3849 0.3854 0.3855 0.3880 0.3913 0.3938 0.3833
Scenario 3-1 0.3822 0.3841 0.3849 0.3854 0.3856 0.3880 0.3914 0.3939 0.3834
Scenario 3-2 0.3822 0.3846 0.3850 0.3856 0.3857 0.3887 0.3915 0.3945 0.3834
[Tabke Ⅲ-8] Changes in GINI Index by Scenario
Ⅳ. Conclusion and Policy Implications
□ Amendment to Excise Tax Act of 2014 could help slowing down the electricity dependent phenomenon that appears recently in Korea by increasing the electricity tariff.
○ However, the optimal strategy in this tax reform would be to wait until 2017 to adjust LNG tax rates, because before the year 2017, there is a possibility that lowering LNG tax rates could place a downward pressure on electricity tariffs when SMPs and adjusted settlement prices are synchronized.
□ The excise tax on coal of flexible rate 18Won/kg and basic rate 24 won/kg from the 2014 amendment might be not enough to blunt the electricity dependent phenomenon, especially when a declining trend of SMPs is expected.
○ Since the purpose of imposing tax on coal is for internalizing the external effects such as pollution price, the tax rate on coal should be higher than one on LNG (60 Won/kg).
□ According to the sensitivity test based on the slightly upward adjusted electricity demand forecasts of the 6th Basic Plan, SMPs and adjusted settlement prices are forecast to stay higher than the reference scenario.
However, the overall trend of SMPs and adjusted settlement prices is expected to be downward.
○ This points to the fact that the current version of the tax reform will still have a limited impact on controlling electrification even if there will be higher demand for electricity.
○ Therefore, in the long term, it is necessary to adjust upward the tax rate on coal, although it might be hard to correct the tax rate in the short term.
○ And yet, attempts to estimate accurate external should be made first before starting tax reform discussions.
□ The analysis based on the assumption that electricity demand will be higher than the 6th Basic Plan implies that the expected power reserve ratio of the 6th Basic Plan is unlikely to be met.
○ Our estimation indicates that electricity demand in 2027 would be higher than that of the 6th Basic Plan by 16.12 percent. Accordingly re-estimated SMPs and adjusted settlement prices still cannot turn the downward trend in electricity price.
○ This reveals the fundamental problem of the methodology employed in the national power plan, which is to treat power prices as an exogenous variable and lay out power supply plans based on them.
○ Power price should be treated as an endogenous factor because it clearly comes under influence of power demand, supply capacity, and electricity generation mix.
○ The current methodology is prone to estimation errors in power
demand, could cause situations in which power capacity runs short or power reserve gets too large.
□ This study extended to cover nuclear-fuel tax in an attempt to incorporate all possible electricity sources in Korea and to take into consideration controversies over the necessity for nuclear-fuel tax.
Including nuclear-fuel tax in the analysis does not necessarily mean that we recommend nuclear-fuel tax.
○ It seems beyond the scope of this study to discuss whether to introduce nuclear-fuel tax or not.
○ Nevertheless, if there is social consensus that nuclear-fuel tax is needed, then we will need a full extensive discussion among interest groups.
○ We should carefully choose between tax and levy as a form of taxation.
In our opinion, nuclear-fuel tax would be better to take form of tax instead of levy.
○ This is because external costs of nuclear power such as accident-risk response costs incur irregularly, rather periodically.
○ When external costs of nuclear power are incorporated as a form of excise tax, the tax revenues would be subject to general budget every year. In case of a serious nuclear accident, the government will have to issue national bonds to cover the costs. This is understood as double taxation.
□ The tax reform was assessed to have impact on macroscopic variables
and income redistribution, by increasing prices and reducing GDP.
○ Lower income brackets are more likely to suffer decreases in nominal income, which contributes to regressive redistribution.
○ Considering that excise tax is an indirect tax, a negative effect on income distribution is nothing surprising. Although, the principle of tax neutrality, which this study didn’t uphold, would have mitigated the negative impact to some extent.
○ The most efficient way of taxing while upholding tax neutrality would be a future research subject. Our suggestion is that since the energy tax reform could bring about regressive redistribution, we need to resolve these negative effects on the macroeconomy and income redistribution by providing income supports or offsetting other taxes by energy taxes.
□ What distinguishes this study from previous ones is the realistic analytic tool that accounts for Korea’s electricity market operational architecture.
○ In addition, one of the greatest strengths of this study is the year-by-year approach, which enables a mid- and long-term policy planning.
○ One of the limitations of this research is that the power market in Korea is under much influence of political decisions and that the research cannot expect political uncertainties.
- For example, it is extremely hard to forecast a series of factors that
determine electricity wholesale prices, including adjusted settlement prices, power policies, and power tariff regulations.
○ Also, we acknowledge that the research did not incorporate electricity policy programs being discussed, such as Vesting Contract.
□ We are not able to control several uncertain variables such as the power plant and grid construction delays, renewable energy and greenhouse gas reduction policy, nuclear power plant construction and life extension policy direction, and power supply and demand policy, and energy and environmental policy conditions etc. In a further research, these limitations may have to be dealt with sufficiently.
