- Email: [email protected]
The economics of coal power generation in China
Energy Policy 105 (2017) 1–9
Contents lists available at ScienceDirect
Energy Policy journal homepage: www.elsevier.com/locate/enpol
The economics of coal power generation in China a
a
a
b
MARK a
a
Changhong Zhao , Weirong Zhang , Yang Wang , Qilin Liu , Jingsheng Guo , Minpeng Xiong , Jiahai Yuana,c,∗ a b c
School of Economics and Management, North China Electric Power University, Beijing 102206, China Asian Studies Center, University of Pittsburgh, 4400 Posvar Hall, Pittsburgh, PA 15260, USA Research Center for Beijing Energy Development, Beijing 102206, China
A R T I C L E I N F O
A BS T RAC T
Keywords: Coal power LCOE Internal rate of return Economic map China
The Chinese government recently released the 13th FYP (five-year plan) power development plan and proposed a capacity installation target of 1100 GW for coal power. Considering the weak demand growth of coal power since 2014, continuous decline in the annual utilisation hour and the coming market competition, such a planning target is unwelcome and could further the economic deterioration of coal power. In this paper, we employ LCOE (levelised cost of electricity) and project evaluation models to conduct a nationwide survey on the economics of coal power. The economic analysis has clearly indicated that the recent boom of coal power investment in China, which is absurd in many perspectives, is largely the aftermath of uncompleted market reform in the power sector. However, the fundamentals of electricity demand and supply are changing at a speed beyond the imagination of power generators and have foreboded a gloomy prospect for coal power. Our study shows that by 2020, with several exceptions, in most provinces the internal rate of return for coal power will drop below the social average return rate or will even be negative. In this regard, the 13th FYP capacity planning target for coal power is economically untenable and requires radical revision.
1. Introduction
79 GW-capacity projects are currently under construction, which represents significant growth compared to new installation in the previous year. Such a discord in supply and demand is further illustrated by the project scale under the Environment Impact Assessment (EIA) approval announced by either the Ministry of Environment Protection or its provincial counterparts in 2015. The total capacity amounted to 169 GW, of which 159 GW has been granted or pre-granted by the EIA approval (Yuan et al., 2016a, 2016b, 2016c). This represents a significant increase when compared with the total EIA-approved capacity for the same period in 2014—which was 48 GW (Greenpeace, 2015). Although thermal power has enjoyed the best economic return since the 2014 downturn of coal price, discrepancies are apparent in the sector's profitability. In 2015, the thermal power utilisation hour in Yunnan, a province well-known for its rich resources in hydropower, was only 1879 h, while the utilisation hour in Sichuan was 2682 h. In Gansu, a province rich in renewable energy resources, fewer than 3800 h of annual utilisation was recorded, while Jilin documented only 3300 h. The coal power sector fell below the break-even point rapidly in these provinces. The industry institution, China Electricity Council (CEC), expressed its deep worry on the profitability of coal power by publishing a report
With a 2.3% reduction in thermal power generation and only 0.5% growth in total electricity consumption, China's new addition of coal power capacity in 2015 is incompatibly high at 52 GW (CEC, 2016a). Regarding the operation efficiency and profitability of coal power, the paradox is self-evident. The annual utilisation hour of thermal power was only 4329 h in 2015, down by 410 h as of the 2014 level and hit the lowest record since 1969 (CEC, 2016a). But in terms of profitability, the coal power sector appeared to take advantage of the apparent imbalance between coal price and on-grid benchmarking tariff and reaped high profits, reaching a historic record since the 2000s (Polaris Power Net, 2016). In China, the government strictly regulates the ongrid tariff of coal power, although market force largely determines coal price. Though a co-movement mechanism for adjusting on-grid tariff had been formulated by the National Development and Reform Commission (NDRC) since 2004 and updated three times since, it was only loosely and arbitrarily implemented (Polaris Power Net, 2016; NDRC, 2015a). It seems that the interest of power generation companies in investing new coal power projects is strong. A recent study by Greenpeace and CoalSwarm (2016) indicated that approximately 73–
⁎
Corresponding author at: School of Economics and Management, North China Electric Power University, Beijing 102206, China. E-mail address: [email protected] (J. Yuan).
http://dx.doi.org/10.1016/j.enpol.2017.02.020 Received 28 November 2016; Received in revised form 11 January 2017; Accepted 14 February 2017 0301-4215/ © 2017 Elsevier Ltd. All rights reserved.
Energy Policy 105 (2017) 1–9
C. Zhao et al.
investment and financing decision making. IRR refers to the discount rate when the total present value of fund inflow equals the total present value of fund outflow, and the NPV is equal to zero. The advantage of the IRR method is to link the project returns during its lifetime with its total investment and indicate the rate of return of the project to provide a benchmark rate of return to confirm whether the project is worthy of investment. IRR is generally recognized as a profitability indicator for project investment. In project financial appraisal, “full investment” and “proprietary funds” assessment are differentiated. Because the purpose here is the economics of the project, not the profitability of own investment, we use full investment IRR as the indicator, which is consistent with that employed by NEA and NDRC (2016). NEA and NDRC (2016) employed a traffic-light warning system. With return rate of long-term treasury bond as the baseline, a projected IRR below it will get a “red” alert. Those between treasury bond return rate and the average IRR (8% in China) of power generation projects will get an “orange” alert. The projects with IRR higher than 8% will have a “green” reading. Actually, 8% benchmarking return requirement is for proprietary funds, and with a convention of 30% own investment ratio, the sector's benchmarking full investment IRR is 6.6% (NDRC and MOC, 2006). Accordingly, we design the grading scale system as follows (Table 1).
in March 2016 (CEC, 2016b). The National Energy Administration (NEA) and NDRC (2016) subsequently issued a prewarning mechanism, which consists of an economic warning indicator, a capacity adequacy indicator and a resource constraint indicator. The first prewarning is for new projects that will be commissioned by 2019. With a traffic-light reading system, the result shows that the alert status of 28 provincial grid regions are rated as “red”, and only Jiangxi, Anhui and Hainan Province are rated as “green”, while Hubei Province is in the “orange” status. For capacity adequacy indicator, 24 provinces obtain “red alert”, and only Jiangxi, Anhui, Hainan, Southern Hebei, Sichuan and Yunnan obtain “green” pre-warning. For the economic warning indicator, 14 provinces are given “red” alert, while the remaining 17 provinces are read as “green”. Considering the overcapacity in these provinces, the economic warning results are not convincing. In November 2016, the 13th FYP Power Planning was issued (NDRC and NEA, 2016). The 1100 GW planning target for coal power, which requires new installation of 200 GW by 2020, has aroused hot debate in industry observers. The controversies are mainly on two interrelated aspects: the rational capacity target for coal power and the economic base underlying it. For the first point, though the future role of coal in China's energy supply has been extensively discussed without dispute (see, for example, Yuan et al., 2012, 2014; Hao et al., 2015; Tang et al., 2015, 2016; Zhang et al., 2016; among others), academic inquiry on coal's role in China's power system is surprisingly rare and disputable (Na et al., 2015; Hui et al., 2016; Yuan et al., 2016a, 2016b, 2016c). For the second point, to the best of our knowledge, only a recent report by Yuan (2016) studied the economics of coal power in six typical provinces. Because economic return is central to the debate and has direct impact on the perspective of coal power in China, this issue deserves systematic study. This study's purpose is to provide a panoramic overview on coal power's economics in China into 2020 and answer whether the 1,100 GW target proposed in the 13th FYP planning is economically feasible. The paper's structure is organized as follows: Section 2 briefly describes the methodology. Section 3 presents the results and discussions. Section 4 concludes the paper with policy implications.
2.2. Data and estimate process Many variables and parameters are involved in conducting an LCOE estimate and project financial appraisal, which may be divided into four categories: technical and economic variables, operation and maintenance costs variables, taxes and charges and financial variables (Fig. 1). Most are common parameters used in the LCOE model and financial appraisal; however, some parameters are used only in the LCOE model or financial appraisal. Table 2 reports the key common parameters for the estimate, while Table 3 reports the province-specific parameters. Our economic analysis starts from the estimate of LCOEs of a typical 600 MW USC coal power plant installed in case provinces by the end of 2015. By comparing LCOEs with the current actual on-grid tariff levels in each province, we could assess the profitability of coal power in these provinces and term it as 2015 baseline. Then we will project the profitability situation by 2020 by considering the following factors:
2. Methodology 2.1. Economic indicators 2.1.1. LCOE A 600 MW coal power plant is chosen as the objective because currently in China, a 600 MW ultra-supercritical (USC) unit is the mainstream of new installation. Projecting the economy of coal power first necessitates the estimate of generation cost and its dynamics. LCOE refers to the costs of electricity per kWh of power generation during the entire operation period and is a widely recognized and highly transparent calculation method for electricity costs (Branker et al., 2011). This paper will calculate the LCOE by calculating the percentage between the present value of total costs and expenses from initial construction to operation and the economic time value of the energy output during the life of a 600 MW coal-fired plant.
1) The decrease of annual utilisation hour Given the irrationally high scale of coal power projects under construction and planning and the weak demand growth prospective during the 13th FYP period, a pessimistic prospective on the annual utilisation of coal power is predicted (Yuan et al., 2016a, 2016b, 2016c; Greenpeace, 2016). National average utilisation hours are predicted to drop to 3600 h by 2020 (Yuan et al., 2016a, 2016b, 2016c). We then calibrate the estimate for coal power in different provinces by considering the differences of national, regional and provincial demand growth rates, as well as the differences in new capacity under construction. 2) Higher generation cost incurred by more stringent environment regulation and the national carbon market
2.1.2. Full investment internal rate of return (IRR) and its grading scale Appraising the economics of new coal power involves project financial appraisal. An economic appraisal method analyses the investment, costs, revenues, taxes and profits of the engineering projects under an existing accounting system and tax regulations and price system of the state (Fu and Quan, 1996). It involves a study of the profitability, solvency and financial viability of the project after being put into operation, and judges the financial economics of the project based upon such an appraisal. In addition to specifying the value of the engineering project to the financial entity and the contribution to investors, the project financial appraisal also provides a basis for
Table 1 the grading scale for the economics of coal power.
2
Reading
Grading scale
Illustration
Dark Red Red Orange Yellow Green Deep green
IRR < 0 0≤IRR < 4.2% 4.2%≤IRR < 6.6% 6.6%≤IRR < 8% 8%≤IRR < 12% IRR≥12%
totally unacceptable very serious, below social risk-free return serious, merely above social risk-free return marginal, the sector's average return satisfactory, above the sector's average return more than satisfactory, super return
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 1. Model variables and parameters.
“encouraging the establishment of long-term stable transaction mechanism and construction of the long-term stable bilateral market mode”. The steady implementation of the No.9 Document means the market-oriented purchase and sale of electricity will be promoted faster (The State Council of China, 2015), implying that the wholesale tariff of coal power is expected to fall sharply in the environment of overcapacity. A survey on the pilot in several provinces indicates a range of 0.3-0.12 yuan/KWh reduction in the bilateral trading prices compared to the benchmarking on-grid tariff levels regulated by NDRC (China Energy Newspaper, 2015). Though the introduction of the electricity market will take time, under the current co-movement mechanism between on-grid tariff and coal price, with a big drop in coal price, NDRC (2015c) declared to lower the national average of the benchmarking on-grid tariff for coal power by 0.03 yuan/KWh from January 2016. In the near term, because of the Chinese government's strong determination to lower electricity tariff to activate the economy's vitality, a co-movement mechanism would not be kicked off even if coal price rebounded to the threshold (The State Council of China, 2016), which implies that generators will have to bear the cost increase by themselves. In the long term, with the deepening of power sector reform, the co-movement mechanism will be abandoned and give way to market force.
In terms of environment regulation, we consider the declared target in the Implementation Plan of Ultra-low Emission and Energy Saving Retrofitting of Coal-fired Plants promulgated by NEA (2015), which requires that ultra-low emission retrofitting of coal-fired units should be completed by 2017, 2018 and 2020 in eastern, central and western regions, respectively. In June 2015, China submitted its Intended Nationally Determined Contributions to the United Nations (NDRC, 2015b). This plan commits to peak greenhouse emissions by 2030 and to reach it as quickly as possible. The 13th FYP is the crucial stage for the implementation of China's greenhouse gas policies, and China will launch a national carbon emission trading system in 2017 (NDRC, 2016). The rigidity of carbon price will inevitably increase the costs of coal power. In such an oversupply market environment, power-generation enterprises will and must internally bear considerable percentage of the anticipated carbon costs. 3) Price reduction, bilateral trading and electricity market In March 2015, the issuance of the Several Opinions on Further Deepening Electric Power System Reform (“No.9 Document”) kicked off the new round of power system reform (The State Council of China, 2015). The No.9 Document sets forth the recent key tasks of reform, including “the realization of the market-oriented power generation and retail price (except for public welfare undertakings)”, “guiding the market entities to carry out multi-party direct transaction” and
Table 2 key common parameters for the economic analysis. Common parameters
Value
Common parameters
Value
Unit Investment Costs (YUAN/kW) Proprietary Funds Ratio (%) Term of Loan (Year) Annual Interest Rate (%) Operation Life (Year) Ratio of Residual Value of Asset (%) Discount Rate (%) Depreciation Rate (%) Capital IRR (%) Coal Consumption Rate of Power Generation (grams standard coal/kWh) Reduction Rate of Coal Consumption in Generation (%) Emission Charge (yuan/tonne) Water Consumption Rate in Generation (kg/kWh) Auxiliary Power Consumption Rate (%)
3590 30 15 6 30 5 8 5 8 286 0.10 1260 1.6 5
Vat(%) Income Tax (%) Housing Property Tax (%) Urban Maintenance and Construction Tax (%) Education Surcharge (%) VAT for Water and Fuel (%) VAT for Materials (%) Overhaul Fee Rate (%) Premium Rate (%) Labor Cost (yuan/Year) Materials Costs and Other Expenses (yuan/kWh) Escalation Rate of Materials Costs and Other Expenses (%) Escalation Rate of Employees’ Salary(%) Pollution Control Costs (yuan/kWh)
17 25 1.2 5 0.5 13 17 2 0.25 80000 0.02 2 6 0.006
3
Energy Policy 105 (2017) 1–9
C. Zhao et al.
3. Results
Table 3 the province-specific parameters for the economic analysis.
Beijing Tianjin Hebei Shandong Shanxi Inner Mongolia Shanghai Zhejiang Jiangsu Anhui Fujian Hubei Henan Hunan Jiangxi Sichuan Chongqing Liaoning Jilin Heilongjiang Shaanxi Gansu Qinghai Ningxia Xinjiang Guangdong Guangxi Yunnan Guizhou Hainan
Utilisation Hour (h)
Coal price
2015
2020
（Yuan/ tonne,5000 kcal steam coal）
4158 4519 4846 4924 4100 4979 3716 3950 5125 4541 3872 4024 4025 3452 4927 2682 3708 4343 3326 4081 4690 3778 4958 5422 4730 4028 3193 1879 4304 5586
NA 4409 4449 4172 3350 4403 3700 3700 4755 4000 3872 3641 3950 3355 4733 2603 3504 4054 3226 3835 4422 3700 4550 4900 3750 3467 2937 1879 4000 5500
337 312 286 375 194 215 374 405 363 387 366 371 336 401 431 360 391 371 344 352 259 257 399 214 167 399 524 395 322 417
3.1. 2015 case
On-grid benchmarking tariff （Yuan/kWh）
Coal price and utilisation hour are the two factors that have largest impact on the LCOEs of coal power. In recent years, because of persistent oversupply, coal price has been on a continuous decline. According to the steam coal (5000 kcal) price index issued by Price Monitor Center, NDRC (2016), the price dropped by 36% in May 2016 relative to the level in January 2014 and returned to the 2004 level. Only in Zhejiang, Guangxi, Hunan and Jiangxi, coal price above 400 Yuan/tonne is observed, while in Shanxi and Xinjiang it is well below 200 Yuan/tonne. Because fuel accounts for roughly 50–60% of generation cost, low coal price will certainly lead to lower generation cost. Utilisation hour is another key factor. According to the sector's practice, if coal units’ annual utilisation hour is higher than 5500 h, then new capacity installation is necessary; if the utilisation hour is fewer than 4500 h, then overcapacity is detected and no new installation is needed (Polaris Power Net, 2015). The national average in 2015 was 4329 h, while in most provinces, the number was lower. With 5500 h as a reference, the utilisation rate of coal power was roughly 78% in 2015, indicating serious overcapacity. Fig. 2 reports the estimate LCOEs of coal power for each province. The regional pattern is high at eastern regions while low at western regions, high at southern regions while low at northern regions, which is consistent with the regional endowment of coal resources. The LCOE is the highest at 0.4 yuan/KWh in provinces as Yunnan and Guangxi. In 16 provinces, such as Sichuan, the LCOE ranges between 0.3 and 0.4 yuan/KWh. In the other 12 provinces, the LCOE is below 0.3 yuan/ KWh. In particular, in four coal-based provinces (Shanxi, Inner Mongolia, Xinjiang and Ningxia), the LCOE is well below 0.25 yuan/ KWh. Note: 1) The study does not cover Tibet and Taiwan. 2) Beijing is not included because of its policy to close all coal power plants. NDRC declared to reduce the on-grid tariff of coal power by 0.03 yuan/KWh in 2016 (NDRC, 2015c). Fig. 3 reports the price gap
0.35 0.35 0.36 0.37 0.3 0.29 0.4 0.42 0.38 0.37 0.37 0.4 0.36 0.45 0.4 0.4 0.38 0.37 0.37 0.37 0.33 0.3 0.32 0.32 0.26 0.45 0.41 0.34 0.34 0.42
Fig. 2. The estimate of LCOEs for coal power at provincial level, 2015.
4
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 3. Price gaps between on-grid tariff and LCOE for coal power at provincial level, 2015.
Fig. 4. The estimate of IRRs for coal power at provincial level, 2015.
average of the price gap is 0.06 Yuan/KWh. At the provincial level, the highest gap (0.13 Yuan/KWh) is in Guangdong and Hainan; the second highest (0.10–0.11 Yuan/KWh) is in Jiangsu, Jiangxi, Ningxia and Hunan provinces. In all the other provinces, the price gap is less than 0.10 Yuan/KWh. Marginal price gap (0.01–0.03 Yuan/KWh) is in
between the effective on-grid tariff and LCOE at each province. According to the concept of LCOE and the pricing rules of on-grid tariff in China, zero gap between them means that the generator can have the benchmarking return while positive gap means that the generator can enjoy extra profit. According to our study, the national 5
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 5. The estimate of IRRs for coal power at outlook scenario 1, 2020.
Fig. 6. The estimate of IRRs for coal power at outlook scenario 2, 2020.
provincial level. With the only exception of Yunnan province, in all other provinces the IRR is higher than the sector's benchmarking return rate (6.6%). In 21 provinces, the IRR of coal power is higher than 12%, among which in provinces such as Hebei, Jiangsu,
Guangxi, Qinghai and Gansu. Here we confirm again that the generous profit space by the price gap is an important economic factor driving the investment bubble and overcapacity Figs. 4–6. We further estimate the full investment IRRs of coal power at the
6
Energy Policy 105 (2017) 1–9
C. Zhao et al.
the sector's average return rate (8%), including Guangdong, Hainan, Hebei, Hunan, Jiangsu and Shaanxi. Remember that scenario 1 is rather optimistic and unrealistic because utilisation hour is assumed to be the same as 2015 levels. Nonetheless, radical deterioration in the economic return can clearly demonstrate the myopia of the investment behaviours of Chinese generators. In scenario 2, when the impact of new installation on utilisation hour is considered, a deep fall in IRRs would be expected. Another five provinces (Henan, Inner Mongolia, Shandong, Shanxi and Anhui) will be added to the list of “red” alert, while Ningxia and Guangxi will get negative IRRs. Only three provinces (Guangdong, Hainan and Hunan) can enjoy more than 8% IRRs because these provinces have fewer new projects. Our projection explicitly reveals a highly negative correlation between the scale of new projects and return expectation.
Guangdong, Jiangxi, Shaanxi and Hainan, the IRR is larger than 15%. Generally speaking, under the present environment, the return on investment in coal power is very impressive, which can largely explain record high new projects despite weak demand growth and deteriorated annual utilisation. 3.2. 2020 case We consider two scenarios when making the projection into 2020. The first scenario takes the ultra-low emission requirement, national carbon market and electricity market reform into the estimate. To comply with the ultra-low emission requirement, an extra 110 million Yuan will be invested. The denitration rate and desulfurization rate will rise from 80% to 95% and 90%, respectively, while the operation of denitration and desulfurization facilities will lead to efficiency penalty in heat rate. For the carbon market, we assume a low carbon price (30 Yuan/tonne) by 2020, while the generator will bear 30% of it. For electricity market reform, we anticipate that the share of direct power purchase in total electricity consumption will rise from 10% at 2015 to 80% at 2020, and the settlement price will decrease radically. For the first scenario, it is assumed that the utilisation hour would hold constant at the 2015 level. Scenario 2 considers the impact of new installation on annual utilisation hour. In this way, the difference between scenario 1 and scenario 2 is solely the impact of new projects. The LCOEs of coal power are expected to increase in the two 2020 scenarios. On the national average, an increase of at least 0.02 Yuan/ KWh is expected in scenario 1. In a province such as Yunnan, the increase is substantial at 0.04 Yuan/KWh. It is evident that with the deepening of power sector reform and the implementation of a series of clean and low-carbon policies, the cost of coal power generation will step into an uptrend channel. By 2020, 22 provinces will have LCOEs ranging between 0.3 and 0.4 Yuan/KWh under scenario 1, while the number is 16 in 2015. In southwest provinces (Yunnan and Guangxi) with rich hydropower resources, the LCOEs of coal power will reach above 0.4 Yuan/KWh. Provinces with lower LCOEs are still those with abundant coal resources, including Shanxi, Inner Mongolia, Shaanxi, Xinjiang, Gansu, Hebei and Henan. In scenario 2, with the commission of more new coal power plants, the LCOEs will continue to rise. Relative to scenario 1, an increase of 0.01–0.02 Yuan/KWh will be expected. The expected increase is not uniform across the provincial level. In provinces with high new capacity under construction (Xinjiang, Shanxi, Shandong, Inner Mongolia and Guangdong), the increase ranges between 0.02 and 0.03 Yuan/KWh. In scenario 2, by 2020, 23 provinces will have LCOEs ranging between 0.3 and 0.4 Yuan/KWh. Sichuan will have LCOE above 0.4 Yuan/KWh largely because of serious deterioration in utilisation hour. Though the upward trend of LCOE is moderate on average, the drop in IRR will be significant, by 6.95% in scenario 1 as to 2015 level on national average. In 10 provinces (Chongqing, Gansu, Guangxi, Guizhou, Jilin, Ningxia, Qinghai, Sichuan, Yunnan and Xinjiang), the IRRs will drop below 4.2%, the long-term treasury bond rate of the present period (Ministry of Finance, 2016). Note that most of these provinces are southwest provinces with plenty of hydropower resources or northwest provinces with weak demand growth and rich renewable resources. It is also worth noting that Xinjiang is on the list of “red” alert, largely because of the huge scale of new installation despite small and weak local demand. In particular, in three provinces (Guangxi, Qinghai and Yunnan), the IRRs are expected to be negative, implying the most serious impact of unchecked coal power. Eight provinces (Liaoning, Heilongjiang, Inner Mongolia, Shanxi, Tianjin, Anhui, Henan and Fujian) will have “orange” alert in scenario 1. Note that two most important coal-power-based provinces, Shanxi and Inner Mongolia, are on the list of “orange” alert under scenario 1. In five provinces (Shandong, Hubei, Shanghai, Jiangxi and Zhejiang) the IRRs will fall to the marginal zone. Note that in 2015 the IRRs are well above 12% in these provinces. Only six provinces can enjoy IRRs higher than
3.3. Discussions and sensitivity analysis The impact of three factors on the prospective of coal power economics deserves further discussions, the first being pollutant emissions tax, the second carbon price and the last coal price. Currently, legislation on pollutant emissions tax is underway in China. Though in the initial stage it is only a switch from an emissions fee to an emissions tax, an increase in the emissions tax seems unavoidable. We predict that coal-fired plants will face stricter pressure in pollutant control, even accounting for any subsidies that may be available to the plants upon ultra-low emission retrofitting, which will rise substantially in the future. The current emission charge level of 1260 Yuan/tonne equivalent for main pollutants as dust, NOx and SO2 is much lower than the social cost of emissions (Yuan and Cheng, 2011). The local standard is 10,000 Yuan/tonne equivalent in Beijing, and Shanghai is planning to raise the level to 6000–8000 Yuan/tonne equivalent in the coming years. A higher emissions tax will lead to an increase in the cost of coal-power generation. The assumption of carbon price level in our scenario study is rather conservative. According to Ouyang (2016), it is reported that the Chinese government envisions the carbon price starting from a low level at 30–50 Yuan/tonne by 2020 to an ideal level at about 200–300 Yuan/tonne when the market is mature. However, because of the vast uncertainty on when carbon price will reach such a level, in our study we set a more likely price level at 30 Yuan/tonne for 2020. Our projection may underestimate the impact of carbon price on the economic return of coal power. The unexpected rise of coal price in 2016 has already resulted in great loss of coal power in many provinces. In provinces such as Zhejiang, Jiangsu, Guangdong, Gansu and Shanxi, the price of steam coal price increased by 80–100 Yuan/tonne from January to September 2016. In provinces such as Anhui and Shandong, the increase is by more than 120 Yuan/tonne (Price Monitor Center of NDRC, 2016). Coal price alone has pushed up the generation cost by 0.04–0.06 Yuan/KWh in 2016; however, under oversupply direct power purchasing is still cutting down the price. Therefore, though the long-term trend of coal price is highly uncertain, the instant impact is that what we expect for 2020 (serious economic deterioration of coal power) has already appeared in advance. To ensure the robustness of our results, here we further conduct sensitivity analysis based on scenario 2 of 2020. Because under any changing conditions, the dark red provinces are impossible to become economically attractive, here the interest is on whether variations in some critical factors will change the rating of those provinces— for example, between “red” and “orange”, or between “orange” and “yellow”. Here we conduct sensitivity analysis for four case provinces, including two “red” provinces Shanxi and Heilongjiang, one “yellow” province Hubei and one “green” province Jiangsu. As illustrated in Figs. 7–10, the most sensitive factor is utilisation hour, and coal price and reduction in direct purchase tariff are next. For Shanxi province, 7
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 10. Sensitivity of IRR in Jiangsu province.
Fig. 7. Sensitivity of IRR in Shanxi province.
the rating scale would change from “red to ”orange” with at least 15% increase in annual utilisation hour (500 h) (Fig. 7). However, it is almost impossible when considering the weak demand growth and large new projects under construction and planning (30 GW). For Heilongjiang province, the rating scale will change from “red” to “orange” with 5% reduction in coal price or 5% increase in annual utilisation hour (190 h) (Fig. 8). Though possible, the improvement is only marginal. For Hubei province, a 15% increase in annual utilisation hour (546 h) would change the rating scale from “yellow” to “green” (Fig. 9). But it requires a utilisation hour number higher than the 2015 level and seems unrealistic. A 15% reduction in coal price has a similar impact, but it seems impossible. For Jiangsu province, a 7.5% reduction in utilisation hour (356 h) will change the rating from “green” to “yellow” (Fig. 10). Considering the change trend of annual utilisation hour in this province, we think this is highly possible. In short, sensitivity analysis on these case provinces indicates that the results of our projection are credible, and at least we do not overestimate the severity of the issue. 4. Conclusion and policy implications
Fig. 8. Sensitivity of IRR in Heilongjiang province.
The economic analysis reported in this paper has clearly indicated that the recent boom of coal power investment in China, which is absurd in many perspectives, is largely the aftermath of uncompleted market reform in the power sector. On the one hand, low coal price has pulled down the cost of power generation. On the other hand, the regulated on-grid tariffs, which are remarkably higher than the levelised cost of energy, provide strong economic motivation for investing in new projects. The decentralisation of project approval rights from central to provincial government catalyses the investment boom. However, though improper price regulation has provided a distorted economic signal, the fundamentals of electricity demand and supply are changing at a speed beyond the imagination of power generators and have foreboded coal power's gloomy prospect. For Chinese generators who are accustomed to price protection in the planning system, fierce competition in the electricity market is the new normal that they are not prepared for. In this regard, the 13th FYP capacity planning target for coal power is economically untenable and requires radical revision. The first policy implication of our study is that NEA will need a more accurate economic alerting system to inform the power sector. The main pitfall of the existing system is that it is static rather than dynamic. It largely overlooks the impact of new projects on the annual utilisation hour. Another pitfall is that demand growth during the 13th FYP period could be overstated. A parallel suggestion is that NDRC should improve the work on generation cost estimating and timely
Fig. 9. Sensitivity of IRR in Hubei province.
8
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fu, J., Quan, Y., 1996. Industrial Technology Economics. Tsinghua University Press, Beijing. Greenpeace, 2015. Conséq. Coal Invest. China. Greenpeace, 2016. New coal power project database (2012 to June 2016). Greenpeace and CoalSwarm, 2016. Boom and Bust-Tracking Global Coal Power Plants. 〈http://www.greenpeace.org.cn/boom-and-bust-2016-report/〉. Hao, Y., Zhang, Z.-Y., Liao, H., Wei, Y.-M., 2015. China's farewell to coal: a forecast of coal consumption through 2020. Energy Policy 86, 444–455. Hui, J., Cai, W., Wang, C., Ye, M., 2016. Analyzing the penetration barriers of clean generation technologies in China's power sector using a multi-region optimization model. Appl. Energy. http://dx.doi.org/10.1016/j.apenergy.2016.02.034. Ministry of Finance, 2016. The announcement of Ministry of Finance (No.53 of2016). 〈http://gks.mof.gov.cn/redianzhuanti/guozaiguanli/gzfxpzs/201605/t20160503_ 1974782.html〉. Na, C., Yuan, J., Xu, Y., Hu, Z., 2015. Penetration of clean coal technology and its impact on China's power industry. Energy Strategy Rev. 7, 1–8. NEA, 2015. Full implementation of ultra-low emission and energy saving transformation of coal-fired power plants. NEA, NDRC, 2016. Notice on the establishment of pre-warning system for the planning and construction of coal power and the release of 2019 pre-warning information. NDRC, Ministry of Construction (MOC), 2006. Notice on printing and distributing the economic evaluation methods and parameters of construction projects (investment2006[1325]). NDRC, 2015a. Notice on improving the relevant matters of coal price linkage mechanism. 〈http://jgs.ndrc.gov.cn/zcfg/201512/t20151231_770446.html〉. NDRC, 2015b..China Submitted Intended Nationally Determined Contributions file. 〈http://www.sdpc.gov.cn/xwzx/xwfb/201506/t20150630_710204.html〉. NDRC, 2015c. Notice on reducing the on-grid tariff for coal power and the retail electricity prices for industrial and commercial users (price2015[3105]). 〈http://jgs. ndrc.gov.cn/zcfg/201512/t20151230_769552.html〉. NDRC, 2016. Notice on implementing key work to start the national carbon emissions trading market. NDRC, NEA, 2016. The 13th FYP plan for power development. Ouyang, K., 2016. 200-300 yuan/tonne is the ideal carbon price in China. (Zhaoli Jiang) Natl. Bus. Dly. 21 (June). Price Monitoring Center, 2016. NDRC, Steam coal price index. Polaris Power Net, 2015. Thermal power utilization hours has dropped to the lowest in 36 years there was a serious surplus. 〈http://news.bjx.com.cn/html/20150123/ 584268-2.shtml〉. Polaris Power Net, 2016. We are concerned about investment opportunities from the reduction of coal power price. 〈http://news.bjx.com.cn/html/20160111/699767-5. shtml〉. The State Council of China, 2015. Some opinions on further deepening of the reform of electric power system. The State Council of China, 2016. Notice on the work plan to reducing cost in real economy enterprises. 〈http://www.gov.cn/zhengce/content/2016-08/22/content_ 5101282.htm〉. Tang, X., Snowden, S., McLellan, B.C., Höök, M., 2015. Clean coal use in China: challenges and policy implications. Energy Policy 87, 517–523. Tang, X., Jin, Y., McLellan, B.C., Wang, J., Li, S., 2016. China's coal consumption declining—Impermanent or permanent? Resour., Conserv. Recycl.. http:// dx.doi.org/10.1016/j.resconrec.2016.07.018. Yuan, J., Lei, Q., Xiong, M., Guo, J., et al., 2016a. The prospective of coal power in China: will it reach a plateau in the coming decade? Energy Policy 98, 495–504. Yuan, J., Li, P., Wang, Y., Liu, Q., et al., 2016b. Coal power overcapacity and investment bubble in China during 2015–2020. Energy Policy 97, 136–144. Yuan J., Na C., Lei, Q., Xiong M., , et al.2016c. Coal use for power generation in China. Resources, Conservation and Recycling, 〈http://dx.doi.org/10.1016/j.resconrec. 2016.03.021〉. Yuan J., 2016. The Economics of Coal-fired Power Generation in China (Greenpeace report). Yuan, J., Xu, Y., Hu, Z., 2012. Delivering power system transition in China. Energy Policy 50, 751–772. Yuan, J., Xu, Y., Hu, Z., et al., 2014. Peak energy consumption and CO2 emissions in China. Energy Policy 68, 508–523. Yuan, P., Cheng, S., 2011. Shadow price estimation of industrial pollutants in China. Stat. Res. 28 (9), 66–73. Zhang M., Bai C., Zhou M., 2016. Decomposition Analysis for Assessing the Progress in Decoupling Relationship between Coal Consumption and Economic Growth in China. Resources, Conservation and Recycling, 〈http://dx.doi.org/10.1016/j. resconrec.2016.06.021〉.
adjust the on-grid tariff. It is important that before the functioning of the electricity market, NDRC should function properly on price regulation to properly signal the investment decision of power generation projects. The second policy implication is that a harder brake policy should be implemented to strictly regulate new capacity installation. For 16 provinces with “red” alert, it is strongly recommended that no new projects should be approved; all the projects under construction should be postponed, while all the pending projects should be cancelled. For seven provinces with “orange” alert, all the projects under construction should be postponed by 2018, and all the pending projects should be cancelled. For four provinces with “yellow” alert, strict control on the new projects is recommended to curb potential capacity excess. The third implication is on power planning. Relevant national authorities are advised to work out power development planning adapted to the new economic normal as soon as possible. Such planning should facilitate low-carbon power sector transition, leave sufficient lead time for the completion of 20% non-fossil energy target by 2030 and rein the irrational growth of coal-fired power projects. In this regard, the recently issued 13th FYP planning target for coal power, 1100 GW by 2020 or 200 GW new installation, is incongruent with our microeconomic analysis. We thus strongly recommend formulating rolling planning and updating the capacity target for coal power. Last but not least, speeding up the power market reform is the ultimate solution. Only real market competition may break the stubborn expectation of power generation enterprises on utilisation hour and on-grid tariff to gradually establish a truly market-oriented investment mechanism. It is advised that no annual generation quota for any new coal-fired power projects commissioned after 2016 will be approved, and all such projects should directly participate in the market. It is also advised that market reform should be steadily promoted in accordance with the established timetable. Besides, in the market reform process, the government should gradually rectify the negative externality of coal-fired power generation by operating an effective national carbon market, increasing pollution tax standards and others to provide a level playground for renewable energy development. Acknowledgments The authors acknowledge the funding of National Natural Science Foundation of China (71673085). The usual caveats apply. References Branker, K., Pathak, M.J.M., Pearce, J.M., 2011. A review of solar photovoltaic levelized cost of electricity. Renew. Sustain. Energy Rev. 15, 4470–4482. China Electricity Council (CEC), 2016a. Analysis and prediction report on national electricity demand and supply in 2016. 〈http://www.cec.org.cn/yaowenkuaidi/201602-03/148763.html〉. China Electricity Council, 2016b. Analysis and prediction report on national electricity demand and supply in the first quarter of 2016. 〈http://www.cec.org.cn/ yaowenkuaidi/2016-04-27/152156.html〉. China Energy Newspaper, 2015. Direct large customer purchase is leveraging on the new round power sector reform. 〈http://paper.people.com.cn/zgnyb/html/2015-04/20/ content_1556583.htm〉.
9
Contents lists available at ScienceDirect
Energy Policy journal homepage: www.elsevier.com/locate/enpol
The economics of coal power generation in China a
a
a
b
MARK a
a
Changhong Zhao , Weirong Zhang , Yang Wang , Qilin Liu , Jingsheng Guo , Minpeng Xiong , Jiahai Yuana,c,∗ a b c
School of Economics and Management, North China Electric Power University, Beijing 102206, China Asian Studies Center, University of Pittsburgh, 4400 Posvar Hall, Pittsburgh, PA 15260, USA Research Center for Beijing Energy Development, Beijing 102206, China
A R T I C L E I N F O
A BS T RAC T
Keywords: Coal power LCOE Internal rate of return Economic map China
The Chinese government recently released the 13th FYP (five-year plan) power development plan and proposed a capacity installation target of 1100 GW for coal power. Considering the weak demand growth of coal power since 2014, continuous decline in the annual utilisation hour and the coming market competition, such a planning target is unwelcome and could further the economic deterioration of coal power. In this paper, we employ LCOE (levelised cost of electricity) and project evaluation models to conduct a nationwide survey on the economics of coal power. The economic analysis has clearly indicated that the recent boom of coal power investment in China, which is absurd in many perspectives, is largely the aftermath of uncompleted market reform in the power sector. However, the fundamentals of electricity demand and supply are changing at a speed beyond the imagination of power generators and have foreboded a gloomy prospect for coal power. Our study shows that by 2020, with several exceptions, in most provinces the internal rate of return for coal power will drop below the social average return rate or will even be negative. In this regard, the 13th FYP capacity planning target for coal power is economically untenable and requires radical revision.
1. Introduction
79 GW-capacity projects are currently under construction, which represents significant growth compared to new installation in the previous year. Such a discord in supply and demand is further illustrated by the project scale under the Environment Impact Assessment (EIA) approval announced by either the Ministry of Environment Protection or its provincial counterparts in 2015. The total capacity amounted to 169 GW, of which 159 GW has been granted or pre-granted by the EIA approval (Yuan et al., 2016a, 2016b, 2016c). This represents a significant increase when compared with the total EIA-approved capacity for the same period in 2014—which was 48 GW (Greenpeace, 2015). Although thermal power has enjoyed the best economic return since the 2014 downturn of coal price, discrepancies are apparent in the sector's profitability. In 2015, the thermal power utilisation hour in Yunnan, a province well-known for its rich resources in hydropower, was only 1879 h, while the utilisation hour in Sichuan was 2682 h. In Gansu, a province rich in renewable energy resources, fewer than 3800 h of annual utilisation was recorded, while Jilin documented only 3300 h. The coal power sector fell below the break-even point rapidly in these provinces. The industry institution, China Electricity Council (CEC), expressed its deep worry on the profitability of coal power by publishing a report
With a 2.3% reduction in thermal power generation and only 0.5% growth in total electricity consumption, China's new addition of coal power capacity in 2015 is incompatibly high at 52 GW (CEC, 2016a). Regarding the operation efficiency and profitability of coal power, the paradox is self-evident. The annual utilisation hour of thermal power was only 4329 h in 2015, down by 410 h as of the 2014 level and hit the lowest record since 1969 (CEC, 2016a). But in terms of profitability, the coal power sector appeared to take advantage of the apparent imbalance between coal price and on-grid benchmarking tariff and reaped high profits, reaching a historic record since the 2000s (Polaris Power Net, 2016). In China, the government strictly regulates the ongrid tariff of coal power, although market force largely determines coal price. Though a co-movement mechanism for adjusting on-grid tariff had been formulated by the National Development and Reform Commission (NDRC) since 2004 and updated three times since, it was only loosely and arbitrarily implemented (Polaris Power Net, 2016; NDRC, 2015a). It seems that the interest of power generation companies in investing new coal power projects is strong. A recent study by Greenpeace and CoalSwarm (2016) indicated that approximately 73–
⁎
Corresponding author at: School of Economics and Management, North China Electric Power University, Beijing 102206, China. E-mail address: [email protected] (J. Yuan).
http://dx.doi.org/10.1016/j.enpol.2017.02.020 Received 28 November 2016; Received in revised form 11 January 2017; Accepted 14 February 2017 0301-4215/ © 2017 Elsevier Ltd. All rights reserved.
Energy Policy 105 (2017) 1–9
C. Zhao et al.
investment and financing decision making. IRR refers to the discount rate when the total present value of fund inflow equals the total present value of fund outflow, and the NPV is equal to zero. The advantage of the IRR method is to link the project returns during its lifetime with its total investment and indicate the rate of return of the project to provide a benchmark rate of return to confirm whether the project is worthy of investment. IRR is generally recognized as a profitability indicator for project investment. In project financial appraisal, “full investment” and “proprietary funds” assessment are differentiated. Because the purpose here is the economics of the project, not the profitability of own investment, we use full investment IRR as the indicator, which is consistent with that employed by NEA and NDRC (2016). NEA and NDRC (2016) employed a traffic-light warning system. With return rate of long-term treasury bond as the baseline, a projected IRR below it will get a “red” alert. Those between treasury bond return rate and the average IRR (8% in China) of power generation projects will get an “orange” alert. The projects with IRR higher than 8% will have a “green” reading. Actually, 8% benchmarking return requirement is for proprietary funds, and with a convention of 30% own investment ratio, the sector's benchmarking full investment IRR is 6.6% (NDRC and MOC, 2006). Accordingly, we design the grading scale system as follows (Table 1).
in March 2016 (CEC, 2016b). The National Energy Administration (NEA) and NDRC (2016) subsequently issued a prewarning mechanism, which consists of an economic warning indicator, a capacity adequacy indicator and a resource constraint indicator. The first prewarning is for new projects that will be commissioned by 2019. With a traffic-light reading system, the result shows that the alert status of 28 provincial grid regions are rated as “red”, and only Jiangxi, Anhui and Hainan Province are rated as “green”, while Hubei Province is in the “orange” status. For capacity adequacy indicator, 24 provinces obtain “red alert”, and only Jiangxi, Anhui, Hainan, Southern Hebei, Sichuan and Yunnan obtain “green” pre-warning. For the economic warning indicator, 14 provinces are given “red” alert, while the remaining 17 provinces are read as “green”. Considering the overcapacity in these provinces, the economic warning results are not convincing. In November 2016, the 13th FYP Power Planning was issued (NDRC and NEA, 2016). The 1100 GW planning target for coal power, which requires new installation of 200 GW by 2020, has aroused hot debate in industry observers. The controversies are mainly on two interrelated aspects: the rational capacity target for coal power and the economic base underlying it. For the first point, though the future role of coal in China's energy supply has been extensively discussed without dispute (see, for example, Yuan et al., 2012, 2014; Hao et al., 2015; Tang et al., 2015, 2016; Zhang et al., 2016; among others), academic inquiry on coal's role in China's power system is surprisingly rare and disputable (Na et al., 2015; Hui et al., 2016; Yuan et al., 2016a, 2016b, 2016c). For the second point, to the best of our knowledge, only a recent report by Yuan (2016) studied the economics of coal power in six typical provinces. Because economic return is central to the debate and has direct impact on the perspective of coal power in China, this issue deserves systematic study. This study's purpose is to provide a panoramic overview on coal power's economics in China into 2020 and answer whether the 1,100 GW target proposed in the 13th FYP planning is economically feasible. The paper's structure is organized as follows: Section 2 briefly describes the methodology. Section 3 presents the results and discussions. Section 4 concludes the paper with policy implications.
2.2. Data and estimate process Many variables and parameters are involved in conducting an LCOE estimate and project financial appraisal, which may be divided into four categories: technical and economic variables, operation and maintenance costs variables, taxes and charges and financial variables (Fig. 1). Most are common parameters used in the LCOE model and financial appraisal; however, some parameters are used only in the LCOE model or financial appraisal. Table 2 reports the key common parameters for the estimate, while Table 3 reports the province-specific parameters. Our economic analysis starts from the estimate of LCOEs of a typical 600 MW USC coal power plant installed in case provinces by the end of 2015. By comparing LCOEs with the current actual on-grid tariff levels in each province, we could assess the profitability of coal power in these provinces and term it as 2015 baseline. Then we will project the profitability situation by 2020 by considering the following factors:
2. Methodology 2.1. Economic indicators 2.1.1. LCOE A 600 MW coal power plant is chosen as the objective because currently in China, a 600 MW ultra-supercritical (USC) unit is the mainstream of new installation. Projecting the economy of coal power first necessitates the estimate of generation cost and its dynamics. LCOE refers to the costs of electricity per kWh of power generation during the entire operation period and is a widely recognized and highly transparent calculation method for electricity costs (Branker et al., 2011). This paper will calculate the LCOE by calculating the percentage between the present value of total costs and expenses from initial construction to operation and the economic time value of the energy output during the life of a 600 MW coal-fired plant.
1) The decrease of annual utilisation hour Given the irrationally high scale of coal power projects under construction and planning and the weak demand growth prospective during the 13th FYP period, a pessimistic prospective on the annual utilisation of coal power is predicted (Yuan et al., 2016a, 2016b, 2016c; Greenpeace, 2016). National average utilisation hours are predicted to drop to 3600 h by 2020 (Yuan et al., 2016a, 2016b, 2016c). We then calibrate the estimate for coal power in different provinces by considering the differences of national, regional and provincial demand growth rates, as well as the differences in new capacity under construction. 2) Higher generation cost incurred by more stringent environment regulation and the national carbon market
2.1.2. Full investment internal rate of return (IRR) and its grading scale Appraising the economics of new coal power involves project financial appraisal. An economic appraisal method analyses the investment, costs, revenues, taxes and profits of the engineering projects under an existing accounting system and tax regulations and price system of the state (Fu and Quan, 1996). It involves a study of the profitability, solvency and financial viability of the project after being put into operation, and judges the financial economics of the project based upon such an appraisal. In addition to specifying the value of the engineering project to the financial entity and the contribution to investors, the project financial appraisal also provides a basis for
Table 1 the grading scale for the economics of coal power.
2
Reading
Grading scale
Illustration
Dark Red Red Orange Yellow Green Deep green
IRR < 0 0≤IRR < 4.2% 4.2%≤IRR < 6.6% 6.6%≤IRR < 8% 8%≤IRR < 12% IRR≥12%
totally unacceptable very serious, below social risk-free return serious, merely above social risk-free return marginal, the sector's average return satisfactory, above the sector's average return more than satisfactory, super return
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 1. Model variables and parameters.
“encouraging the establishment of long-term stable transaction mechanism and construction of the long-term stable bilateral market mode”. The steady implementation of the No.9 Document means the market-oriented purchase and sale of electricity will be promoted faster (The State Council of China, 2015), implying that the wholesale tariff of coal power is expected to fall sharply in the environment of overcapacity. A survey on the pilot in several provinces indicates a range of 0.3-0.12 yuan/KWh reduction in the bilateral trading prices compared to the benchmarking on-grid tariff levels regulated by NDRC (China Energy Newspaper, 2015). Though the introduction of the electricity market will take time, under the current co-movement mechanism between on-grid tariff and coal price, with a big drop in coal price, NDRC (2015c) declared to lower the national average of the benchmarking on-grid tariff for coal power by 0.03 yuan/KWh from January 2016. In the near term, because of the Chinese government's strong determination to lower electricity tariff to activate the economy's vitality, a co-movement mechanism would not be kicked off even if coal price rebounded to the threshold (The State Council of China, 2016), which implies that generators will have to bear the cost increase by themselves. In the long term, with the deepening of power sector reform, the co-movement mechanism will be abandoned and give way to market force.
In terms of environment regulation, we consider the declared target in the Implementation Plan of Ultra-low Emission and Energy Saving Retrofitting of Coal-fired Plants promulgated by NEA (2015), which requires that ultra-low emission retrofitting of coal-fired units should be completed by 2017, 2018 and 2020 in eastern, central and western regions, respectively. In June 2015, China submitted its Intended Nationally Determined Contributions to the United Nations (NDRC, 2015b). This plan commits to peak greenhouse emissions by 2030 and to reach it as quickly as possible. The 13th FYP is the crucial stage for the implementation of China's greenhouse gas policies, and China will launch a national carbon emission trading system in 2017 (NDRC, 2016). The rigidity of carbon price will inevitably increase the costs of coal power. In such an oversupply market environment, power-generation enterprises will and must internally bear considerable percentage of the anticipated carbon costs. 3) Price reduction, bilateral trading and electricity market In March 2015, the issuance of the Several Opinions on Further Deepening Electric Power System Reform (“No.9 Document”) kicked off the new round of power system reform (The State Council of China, 2015). The No.9 Document sets forth the recent key tasks of reform, including “the realization of the market-oriented power generation and retail price (except for public welfare undertakings)”, “guiding the market entities to carry out multi-party direct transaction” and
Table 2 key common parameters for the economic analysis. Common parameters
Value
Common parameters
Value
Unit Investment Costs (YUAN/kW) Proprietary Funds Ratio (%) Term of Loan (Year) Annual Interest Rate (%) Operation Life (Year) Ratio of Residual Value of Asset (%) Discount Rate (%) Depreciation Rate (%) Capital IRR (%) Coal Consumption Rate of Power Generation (grams standard coal/kWh) Reduction Rate of Coal Consumption in Generation (%) Emission Charge (yuan/tonne) Water Consumption Rate in Generation (kg/kWh) Auxiliary Power Consumption Rate (%)
3590 30 15 6 30 5 8 5 8 286 0.10 1260 1.6 5
Vat(%) Income Tax (%) Housing Property Tax (%) Urban Maintenance and Construction Tax (%) Education Surcharge (%) VAT for Water and Fuel (%) VAT for Materials (%) Overhaul Fee Rate (%) Premium Rate (%) Labor Cost (yuan/Year) Materials Costs and Other Expenses (yuan/kWh) Escalation Rate of Materials Costs and Other Expenses (%) Escalation Rate of Employees’ Salary(%) Pollution Control Costs (yuan/kWh)
17 25 1.2 5 0.5 13 17 2 0.25 80000 0.02 2 6 0.006
3
Energy Policy 105 (2017) 1–9
C. Zhao et al.
3. Results
Table 3 the province-specific parameters for the economic analysis.
Beijing Tianjin Hebei Shandong Shanxi Inner Mongolia Shanghai Zhejiang Jiangsu Anhui Fujian Hubei Henan Hunan Jiangxi Sichuan Chongqing Liaoning Jilin Heilongjiang Shaanxi Gansu Qinghai Ningxia Xinjiang Guangdong Guangxi Yunnan Guizhou Hainan
Utilisation Hour (h)
Coal price
2015
2020
（Yuan/ tonne,5000 kcal steam coal）
4158 4519 4846 4924 4100 4979 3716 3950 5125 4541 3872 4024 4025 3452 4927 2682 3708 4343 3326 4081 4690 3778 4958 5422 4730 4028 3193 1879 4304 5586
NA 4409 4449 4172 3350 4403 3700 3700 4755 4000 3872 3641 3950 3355 4733 2603 3504 4054 3226 3835 4422 3700 4550 4900 3750 3467 2937 1879 4000 5500
337 312 286 375 194 215 374 405 363 387 366 371 336 401 431 360 391 371 344 352 259 257 399 214 167 399 524 395 322 417
3.1. 2015 case
On-grid benchmarking tariff （Yuan/kWh）
Coal price and utilisation hour are the two factors that have largest impact on the LCOEs of coal power. In recent years, because of persistent oversupply, coal price has been on a continuous decline. According to the steam coal (5000 kcal) price index issued by Price Monitor Center, NDRC (2016), the price dropped by 36% in May 2016 relative to the level in January 2014 and returned to the 2004 level. Only in Zhejiang, Guangxi, Hunan and Jiangxi, coal price above 400 Yuan/tonne is observed, while in Shanxi and Xinjiang it is well below 200 Yuan/tonne. Because fuel accounts for roughly 50–60% of generation cost, low coal price will certainly lead to lower generation cost. Utilisation hour is another key factor. According to the sector's practice, if coal units’ annual utilisation hour is higher than 5500 h, then new capacity installation is necessary; if the utilisation hour is fewer than 4500 h, then overcapacity is detected and no new installation is needed (Polaris Power Net, 2015). The national average in 2015 was 4329 h, while in most provinces, the number was lower. With 5500 h as a reference, the utilisation rate of coal power was roughly 78% in 2015, indicating serious overcapacity. Fig. 2 reports the estimate LCOEs of coal power for each province. The regional pattern is high at eastern regions while low at western regions, high at southern regions while low at northern regions, which is consistent with the regional endowment of coal resources. The LCOE is the highest at 0.4 yuan/KWh in provinces as Yunnan and Guangxi. In 16 provinces, such as Sichuan, the LCOE ranges between 0.3 and 0.4 yuan/KWh. In the other 12 provinces, the LCOE is below 0.3 yuan/ KWh. In particular, in four coal-based provinces (Shanxi, Inner Mongolia, Xinjiang and Ningxia), the LCOE is well below 0.25 yuan/ KWh. Note: 1) The study does not cover Tibet and Taiwan. 2) Beijing is not included because of its policy to close all coal power plants. NDRC declared to reduce the on-grid tariff of coal power by 0.03 yuan/KWh in 2016 (NDRC, 2015c). Fig. 3 reports the price gap
0.35 0.35 0.36 0.37 0.3 0.29 0.4 0.42 0.38 0.37 0.37 0.4 0.36 0.45 0.4 0.4 0.38 0.37 0.37 0.37 0.33 0.3 0.32 0.32 0.26 0.45 0.41 0.34 0.34 0.42
Fig. 2. The estimate of LCOEs for coal power at provincial level, 2015.
4
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 3. Price gaps between on-grid tariff and LCOE for coal power at provincial level, 2015.
Fig. 4. The estimate of IRRs for coal power at provincial level, 2015.
average of the price gap is 0.06 Yuan/KWh. At the provincial level, the highest gap (0.13 Yuan/KWh) is in Guangdong and Hainan; the second highest (0.10–0.11 Yuan/KWh) is in Jiangsu, Jiangxi, Ningxia and Hunan provinces. In all the other provinces, the price gap is less than 0.10 Yuan/KWh. Marginal price gap (0.01–0.03 Yuan/KWh) is in
between the effective on-grid tariff and LCOE at each province. According to the concept of LCOE and the pricing rules of on-grid tariff in China, zero gap between them means that the generator can have the benchmarking return while positive gap means that the generator can enjoy extra profit. According to our study, the national 5
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 5. The estimate of IRRs for coal power at outlook scenario 1, 2020.
Fig. 6. The estimate of IRRs for coal power at outlook scenario 2, 2020.
provincial level. With the only exception of Yunnan province, in all other provinces the IRR is higher than the sector's benchmarking return rate (6.6%). In 21 provinces, the IRR of coal power is higher than 12%, among which in provinces such as Hebei, Jiangsu,
Guangxi, Qinghai and Gansu. Here we confirm again that the generous profit space by the price gap is an important economic factor driving the investment bubble and overcapacity Figs. 4–6. We further estimate the full investment IRRs of coal power at the
6
Energy Policy 105 (2017) 1–9
C. Zhao et al.
the sector's average return rate (8%), including Guangdong, Hainan, Hebei, Hunan, Jiangsu and Shaanxi. Remember that scenario 1 is rather optimistic and unrealistic because utilisation hour is assumed to be the same as 2015 levels. Nonetheless, radical deterioration in the economic return can clearly demonstrate the myopia of the investment behaviours of Chinese generators. In scenario 2, when the impact of new installation on utilisation hour is considered, a deep fall in IRRs would be expected. Another five provinces (Henan, Inner Mongolia, Shandong, Shanxi and Anhui) will be added to the list of “red” alert, while Ningxia and Guangxi will get negative IRRs. Only three provinces (Guangdong, Hainan and Hunan) can enjoy more than 8% IRRs because these provinces have fewer new projects. Our projection explicitly reveals a highly negative correlation between the scale of new projects and return expectation.
Guangdong, Jiangxi, Shaanxi and Hainan, the IRR is larger than 15%. Generally speaking, under the present environment, the return on investment in coal power is very impressive, which can largely explain record high new projects despite weak demand growth and deteriorated annual utilisation. 3.2. 2020 case We consider two scenarios when making the projection into 2020. The first scenario takes the ultra-low emission requirement, national carbon market and electricity market reform into the estimate. To comply with the ultra-low emission requirement, an extra 110 million Yuan will be invested. The denitration rate and desulfurization rate will rise from 80% to 95% and 90%, respectively, while the operation of denitration and desulfurization facilities will lead to efficiency penalty in heat rate. For the carbon market, we assume a low carbon price (30 Yuan/tonne) by 2020, while the generator will bear 30% of it. For electricity market reform, we anticipate that the share of direct power purchase in total electricity consumption will rise from 10% at 2015 to 80% at 2020, and the settlement price will decrease radically. For the first scenario, it is assumed that the utilisation hour would hold constant at the 2015 level. Scenario 2 considers the impact of new installation on annual utilisation hour. In this way, the difference between scenario 1 and scenario 2 is solely the impact of new projects. The LCOEs of coal power are expected to increase in the two 2020 scenarios. On the national average, an increase of at least 0.02 Yuan/ KWh is expected in scenario 1. In a province such as Yunnan, the increase is substantial at 0.04 Yuan/KWh. It is evident that with the deepening of power sector reform and the implementation of a series of clean and low-carbon policies, the cost of coal power generation will step into an uptrend channel. By 2020, 22 provinces will have LCOEs ranging between 0.3 and 0.4 Yuan/KWh under scenario 1, while the number is 16 in 2015. In southwest provinces (Yunnan and Guangxi) with rich hydropower resources, the LCOEs of coal power will reach above 0.4 Yuan/KWh. Provinces with lower LCOEs are still those with abundant coal resources, including Shanxi, Inner Mongolia, Shaanxi, Xinjiang, Gansu, Hebei and Henan. In scenario 2, with the commission of more new coal power plants, the LCOEs will continue to rise. Relative to scenario 1, an increase of 0.01–0.02 Yuan/KWh will be expected. The expected increase is not uniform across the provincial level. In provinces with high new capacity under construction (Xinjiang, Shanxi, Shandong, Inner Mongolia and Guangdong), the increase ranges between 0.02 and 0.03 Yuan/KWh. In scenario 2, by 2020, 23 provinces will have LCOEs ranging between 0.3 and 0.4 Yuan/KWh. Sichuan will have LCOE above 0.4 Yuan/KWh largely because of serious deterioration in utilisation hour. Though the upward trend of LCOE is moderate on average, the drop in IRR will be significant, by 6.95% in scenario 1 as to 2015 level on national average. In 10 provinces (Chongqing, Gansu, Guangxi, Guizhou, Jilin, Ningxia, Qinghai, Sichuan, Yunnan and Xinjiang), the IRRs will drop below 4.2%, the long-term treasury bond rate of the present period (Ministry of Finance, 2016). Note that most of these provinces are southwest provinces with plenty of hydropower resources or northwest provinces with weak demand growth and rich renewable resources. It is also worth noting that Xinjiang is on the list of “red” alert, largely because of the huge scale of new installation despite small and weak local demand. In particular, in three provinces (Guangxi, Qinghai and Yunnan), the IRRs are expected to be negative, implying the most serious impact of unchecked coal power. Eight provinces (Liaoning, Heilongjiang, Inner Mongolia, Shanxi, Tianjin, Anhui, Henan and Fujian) will have “orange” alert in scenario 1. Note that two most important coal-power-based provinces, Shanxi and Inner Mongolia, are on the list of “orange” alert under scenario 1. In five provinces (Shandong, Hubei, Shanghai, Jiangxi and Zhejiang) the IRRs will fall to the marginal zone. Note that in 2015 the IRRs are well above 12% in these provinces. Only six provinces can enjoy IRRs higher than
3.3. Discussions and sensitivity analysis The impact of three factors on the prospective of coal power economics deserves further discussions, the first being pollutant emissions tax, the second carbon price and the last coal price. Currently, legislation on pollutant emissions tax is underway in China. Though in the initial stage it is only a switch from an emissions fee to an emissions tax, an increase in the emissions tax seems unavoidable. We predict that coal-fired plants will face stricter pressure in pollutant control, even accounting for any subsidies that may be available to the plants upon ultra-low emission retrofitting, which will rise substantially in the future. The current emission charge level of 1260 Yuan/tonne equivalent for main pollutants as dust, NOx and SO2 is much lower than the social cost of emissions (Yuan and Cheng, 2011). The local standard is 10,000 Yuan/tonne equivalent in Beijing, and Shanghai is planning to raise the level to 6000–8000 Yuan/tonne equivalent in the coming years. A higher emissions tax will lead to an increase in the cost of coal-power generation. The assumption of carbon price level in our scenario study is rather conservative. According to Ouyang (2016), it is reported that the Chinese government envisions the carbon price starting from a low level at 30–50 Yuan/tonne by 2020 to an ideal level at about 200–300 Yuan/tonne when the market is mature. However, because of the vast uncertainty on when carbon price will reach such a level, in our study we set a more likely price level at 30 Yuan/tonne for 2020. Our projection may underestimate the impact of carbon price on the economic return of coal power. The unexpected rise of coal price in 2016 has already resulted in great loss of coal power in many provinces. In provinces such as Zhejiang, Jiangsu, Guangdong, Gansu and Shanxi, the price of steam coal price increased by 80–100 Yuan/tonne from January to September 2016. In provinces such as Anhui and Shandong, the increase is by more than 120 Yuan/tonne (Price Monitor Center of NDRC, 2016). Coal price alone has pushed up the generation cost by 0.04–0.06 Yuan/KWh in 2016; however, under oversupply direct power purchasing is still cutting down the price. Therefore, though the long-term trend of coal price is highly uncertain, the instant impact is that what we expect for 2020 (serious economic deterioration of coal power) has already appeared in advance. To ensure the robustness of our results, here we further conduct sensitivity analysis based on scenario 2 of 2020. Because under any changing conditions, the dark red provinces are impossible to become economically attractive, here the interest is on whether variations in some critical factors will change the rating of those provinces— for example, between “red” and “orange”, or between “orange” and “yellow”. Here we conduct sensitivity analysis for four case provinces, including two “red” provinces Shanxi and Heilongjiang, one “yellow” province Hubei and one “green” province Jiangsu. As illustrated in Figs. 7–10, the most sensitive factor is utilisation hour, and coal price and reduction in direct purchase tariff are next. For Shanxi province, 7
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fig. 10. Sensitivity of IRR in Jiangsu province.
Fig. 7. Sensitivity of IRR in Shanxi province.
the rating scale would change from “red to ”orange” with at least 15% increase in annual utilisation hour (500 h) (Fig. 7). However, it is almost impossible when considering the weak demand growth and large new projects under construction and planning (30 GW). For Heilongjiang province, the rating scale will change from “red” to “orange” with 5% reduction in coal price or 5% increase in annual utilisation hour (190 h) (Fig. 8). Though possible, the improvement is only marginal. For Hubei province, a 15% increase in annual utilisation hour (546 h) would change the rating scale from “yellow” to “green” (Fig. 9). But it requires a utilisation hour number higher than the 2015 level and seems unrealistic. A 15% reduction in coal price has a similar impact, but it seems impossible. For Jiangsu province, a 7.5% reduction in utilisation hour (356 h) will change the rating from “green” to “yellow” (Fig. 10). Considering the change trend of annual utilisation hour in this province, we think this is highly possible. In short, sensitivity analysis on these case provinces indicates that the results of our projection are credible, and at least we do not overestimate the severity of the issue. 4. Conclusion and policy implications
Fig. 8. Sensitivity of IRR in Heilongjiang province.
The economic analysis reported in this paper has clearly indicated that the recent boom of coal power investment in China, which is absurd in many perspectives, is largely the aftermath of uncompleted market reform in the power sector. On the one hand, low coal price has pulled down the cost of power generation. On the other hand, the regulated on-grid tariffs, which are remarkably higher than the levelised cost of energy, provide strong economic motivation for investing in new projects. The decentralisation of project approval rights from central to provincial government catalyses the investment boom. However, though improper price regulation has provided a distorted economic signal, the fundamentals of electricity demand and supply are changing at a speed beyond the imagination of power generators and have foreboded coal power's gloomy prospect. For Chinese generators who are accustomed to price protection in the planning system, fierce competition in the electricity market is the new normal that they are not prepared for. In this regard, the 13th FYP capacity planning target for coal power is economically untenable and requires radical revision. The first policy implication of our study is that NEA will need a more accurate economic alerting system to inform the power sector. The main pitfall of the existing system is that it is static rather than dynamic. It largely overlooks the impact of new projects on the annual utilisation hour. Another pitfall is that demand growth during the 13th FYP period could be overstated. A parallel suggestion is that NDRC should improve the work on generation cost estimating and timely
Fig. 9. Sensitivity of IRR in Hubei province.
8
Energy Policy 105 (2017) 1–9
C. Zhao et al.
Fu, J., Quan, Y., 1996. Industrial Technology Economics. Tsinghua University Press, Beijing. Greenpeace, 2015. Conséq. Coal Invest. China. Greenpeace, 2016. New coal power project database (2012 to June 2016). Greenpeace and CoalSwarm, 2016. Boom and Bust-Tracking Global Coal Power Plants. 〈http://www.greenpeace.org.cn/boom-and-bust-2016-report/〉. Hao, Y., Zhang, Z.-Y., Liao, H., Wei, Y.-M., 2015. China's farewell to coal: a forecast of coal consumption through 2020. Energy Policy 86, 444–455. Hui, J., Cai, W., Wang, C., Ye, M., 2016. Analyzing the penetration barriers of clean generation technologies in China's power sector using a multi-region optimization model. Appl. Energy. http://dx.doi.org/10.1016/j.apenergy.2016.02.034. Ministry of Finance, 2016. The announcement of Ministry of Finance (No.53 of2016). 〈http://gks.mof.gov.cn/redianzhuanti/guozaiguanli/gzfxpzs/201605/t20160503_ 1974782.html〉. Na, C., Yuan, J., Xu, Y., Hu, Z., 2015. Penetration of clean coal technology and its impact on China's power industry. Energy Strategy Rev. 7, 1–8. NEA, 2015. Full implementation of ultra-low emission and energy saving transformation of coal-fired power plants. NEA, NDRC, 2016. Notice on the establishment of pre-warning system for the planning and construction of coal power and the release of 2019 pre-warning information. NDRC, Ministry of Construction (MOC), 2006. Notice on printing and distributing the economic evaluation methods and parameters of construction projects (investment2006[1325]). NDRC, 2015a. Notice on improving the relevant matters of coal price linkage mechanism. 〈http://jgs.ndrc.gov.cn/zcfg/201512/t20151231_770446.html〉. NDRC, 2015b..China Submitted Intended Nationally Determined Contributions file. 〈http://www.sdpc.gov.cn/xwzx/xwfb/201506/t20150630_710204.html〉. NDRC, 2015c. Notice on reducing the on-grid tariff for coal power and the retail electricity prices for industrial and commercial users (price2015[3105]). 〈http://jgs. ndrc.gov.cn/zcfg/201512/t20151230_769552.html〉. NDRC, 2016. Notice on implementing key work to start the national carbon emissions trading market. NDRC, NEA, 2016. The 13th FYP plan for power development. Ouyang, K., 2016. 200-300 yuan/tonne is the ideal carbon price in China. (Zhaoli Jiang) Natl. Bus. Dly. 21 (June). Price Monitoring Center, 2016. NDRC, Steam coal price index. Polaris Power Net, 2015. Thermal power utilization hours has dropped to the lowest in 36 years there was a serious surplus. 〈http://news.bjx.com.cn/html/20150123/ 584268-2.shtml〉. Polaris Power Net, 2016. We are concerned about investment opportunities from the reduction of coal power price. 〈http://news.bjx.com.cn/html/20160111/699767-5. shtml〉. The State Council of China, 2015. Some opinions on further deepening of the reform of electric power system. The State Council of China, 2016. Notice on the work plan to reducing cost in real economy enterprises. 〈http://www.gov.cn/zhengce/content/2016-08/22/content_ 5101282.htm〉. Tang, X., Snowden, S., McLellan, B.C., Höök, M., 2015. Clean coal use in China: challenges and policy implications. Energy Policy 87, 517–523. Tang, X., Jin, Y., McLellan, B.C., Wang, J., Li, S., 2016. China's coal consumption declining—Impermanent or permanent? Resour., Conserv. Recycl.. http:// dx.doi.org/10.1016/j.resconrec.2016.07.018. Yuan, J., Lei, Q., Xiong, M., Guo, J., et al., 2016a. The prospective of coal power in China: will it reach a plateau in the coming decade? Energy Policy 98, 495–504. Yuan, J., Li, P., Wang, Y., Liu, Q., et al., 2016b. Coal power overcapacity and investment bubble in China during 2015–2020. Energy Policy 97, 136–144. Yuan J., Na C., Lei, Q., Xiong M., , et al.2016c. Coal use for power generation in China. Resources, Conservation and Recycling, 〈http://dx.doi.org/10.1016/j.resconrec. 2016.03.021〉. Yuan J., 2016. The Economics of Coal-fired Power Generation in China (Greenpeace report). Yuan, J., Xu, Y., Hu, Z., 2012. Delivering power system transition in China. Energy Policy 50, 751–772. Yuan, J., Xu, Y., Hu, Z., et al., 2014. Peak energy consumption and CO2 emissions in China. Energy Policy 68, 508–523. Yuan, P., Cheng, S., 2011. Shadow price estimation of industrial pollutants in China. Stat. Res. 28 (9), 66–73. Zhang M., Bai C., Zhou M., 2016. Decomposition Analysis for Assessing the Progress in Decoupling Relationship between Coal Consumption and Economic Growth in China. Resources, Conservation and Recycling, 〈http://dx.doi.org/10.1016/j. resconrec.2016.06.021〉.
adjust the on-grid tariff. It is important that before the functioning of the electricity market, NDRC should function properly on price regulation to properly signal the investment decision of power generation projects. The second policy implication is that a harder brake policy should be implemented to strictly regulate new capacity installation. For 16 provinces with “red” alert, it is strongly recommended that no new projects should be approved; all the projects under construction should be postponed, while all the pending projects should be cancelled. For seven provinces with “orange” alert, all the projects under construction should be postponed by 2018, and all the pending projects should be cancelled. For four provinces with “yellow” alert, strict control on the new projects is recommended to curb potential capacity excess. The third implication is on power planning. Relevant national authorities are advised to work out power development planning adapted to the new economic normal as soon as possible. Such planning should facilitate low-carbon power sector transition, leave sufficient lead time for the completion of 20% non-fossil energy target by 2030 and rein the irrational growth of coal-fired power projects. In this regard, the recently issued 13th FYP planning target for coal power, 1100 GW by 2020 or 200 GW new installation, is incongruent with our microeconomic analysis. We thus strongly recommend formulating rolling planning and updating the capacity target for coal power. Last but not least, speeding up the power market reform is the ultimate solution. Only real market competition may break the stubborn expectation of power generation enterprises on utilisation hour and on-grid tariff to gradually establish a truly market-oriented investment mechanism. It is advised that no annual generation quota for any new coal-fired power projects commissioned after 2016 will be approved, and all such projects should directly participate in the market. It is also advised that market reform should be steadily promoted in accordance with the established timetable. Besides, in the market reform process, the government should gradually rectify the negative externality of coal-fired power generation by operating an effective national carbon market, increasing pollution tax standards and others to provide a level playground for renewable energy development. Acknowledgments The authors acknowledge the funding of National Natural Science Foundation of China (71673085). The usual caveats apply. References Branker, K., Pathak, M.J.M., Pearce, J.M., 2011. A review of solar photovoltaic levelized cost of electricity. Renew. Sustain. Energy Rev. 15, 4470–4482. China Electricity Council (CEC), 2016a. Analysis and prediction report on national electricity demand and supply in 2016. 〈http://www.cec.org.cn/yaowenkuaidi/201602-03/148763.html〉. China Electricity Council, 2016b. Analysis and prediction report on national electricity demand and supply in the first quarter of 2016. 〈http://www.cec.org.cn/ yaowenkuaidi/2016-04-27/152156.html〉. China Energy Newspaper, 2015. Direct large customer purchase is leveraging on the new round power sector reform. 〈http://paper.people.com.cn/zgnyb/html/2015-04/20/ content_1556583.htm〉.
9