Energy 187 (2019) 115974

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Inflationary and distributional effects of fossil energy price fluctuation

on the Chinese economy
Zhan-Ming Chen a, Pei-Lin Chen a, Zeming Ma a, Shiyun Xu b, c, *, Tasawar Hayat c, d,
Ahmed Alsaedi c
a
School of Applied Economics, Renmin University of China, Beijing, 100872, China
b
China Electric Power Research Institute, Beijing, 100192, China
c
NAAM Group, Faculty of Science, King Abdulaziz University, Jeddah, Saudi Arabia
d
Department of Mathematics, Quaid-i-Azam University, 45320, Islamabad, Pakistan

a r t i c l e i n f o a b s t r a c t

Article history: This study uses a non-competitive input-output model with monthly empirical data from January 2012
Received 4 February 2019 to November 2018 to evaluate the inflationary and distributional effects of fossil energy price fluctuation
Received in revised form on the Chinese economy. Results show that the average monthly responses of consumer price index,
2 July 2019
producer price index, and gross domestic product deflator to fossil energy price change are 0.64%, 1.95%,
Accepted 17 August 2019
and 1.46%, respectively. The inflationary pressures resulted from the price fluctuation of different energy
Available online 19 August 2019
types and sources are identified, which facilitates policymaking targeting different driving factors of
inflation. Opposite progressivity is observed in terms of distributional effect measured by the expendi-
Keywords:
Fossil energy
ture changes of residents at different income levels, i.e., the coal price change leads to regressive impact
Inflationary effect while the crude oil price change leads to progressive one. Policy implications for China's energy market
Distributional effect reform and energy tax design have been put forward based on the results of this study.
Input-output model © 2019 Elsevier Ltd. All rights reserved.

1. Introduction income factors response to the general price level. Trivariate vector
autoregression (VAR) model and Granger causality test were
Playing an essential role in the modern economy, energy is not applied by Cunado et al. [12,13], who concluded that crude oil price
only a fundamental production factor for industries but also an has significant effect on the general price level in the European and
important consumption good for households. Previous studies Asian countries. A computable general equilibrium model has been
confirmed that price fluctuation of energy, especially fossil energy constructed by Lin and Mou [14] to simulate the impact of coal and
which accounts for 90% of the global primary energy consumption oil price change, by testing different price increase scenarios from
in 2017 [1], would bring significant impacts to the economy [2,3]. 20% to 100%. Their results showed that the increase in energy price
The interaction between energy price and economy attracts has a contraction effect on the Chinese economy and the impact of
intensive attention of researchers and policy makers from different coal price dominates the effect. By applying the Phillips curve and
perspectives, two of the most concerned ones are the inflationary the Okun curve, Li and Xing [15] found that 10% increase of crude oil
effect (which indicates efficiency) [4,5] and distributional effect price would lead to 0.36%e0.50% change of the general price level
(which indicates equality) [6e8]. in China. The asymmetric impacts of oil price changes have also
The literature about inflationary effect focuses on crude oil and been observed. Based on a non-competitive input-output model,
coal, because of their importance for the industrialized economy Chen [16] studied the impact of coal price fluctuation on China's
[9,10]. Using the 1990 input-output table of Turkey, Berument and economy under three policy scenarios. In the actual regulatory
Tasci [11] estimated the inflation driven by the increase of crude oil scenario, about 5% of the change in gross domestic product (GDP)
price. Their results suggest that the impact depends on how the deflator and consumer price index (CPI), as well as 25% of producer
price index (PPI) change could be attributed to the rise of coal price
during the research period. Guo et al. [17] examined the relation-
* Corresponding author. China Electric Power Research Institute, Beijing, 100192, ship between coal price and the general price level in China by
China. employing VAR models, Granger causality tests, and impulse
E-mail address: xushiyun@epri.sgcc.com.cn (S. Xu).

https://doi.org/10.1016/j.energy.2019.115974
0360-5442/© 2019 Elsevier Ltd. All rights reserved.
2 Z.-M. Chen et al. / Energy 187 (2019) 115974

response functions. Their results confirm that PPI is more sensitive makers concerning the trade-off between economy efficiency and
than CPI to coal price change in China. Morana [18] constructed a equality. Second, by applying empirical monthly price data, this
large-scale time-varying parameter model for the Euro area, which study provides an ex-post examination based on high frequent data
found that net crude oil price increase would always lead to to reflect the real-world impact of major fossil energy price shock
contraction of industrial production but net crude oil price decrease on China. Finally, the inflation pressures resulted from the price
only yielded expansionary effects during early and mid-2000s. fluctuation of different energy types and sources are identified,
In spite of the differentiated methods and research objects which facilitates policymaking targeting different driving factors of
applied in the aforementioned studies, a generally accepted inflation.
conclusion is that energy price change has significant inflationary The rest of this paper is organized as follows. Section 2 in-
effect on the economy [19]. However, the distributional effect troduces the methodological details and describes the employed
brought forward by energy price fluctuation is more complicated, data. Section 3 presents the results regarding to the inflationary and
i.e., the impact progressivity of each case depends on a series of distributional effects. The final section concludes. Policy sugges-
factors such as economic structure, income level, and energy type tions and limitations of this study are also discussed in this section.
[20e22].
Andriamihaja and Vecchi [23] found that a rise in petroleum
price brings forward regressive distributional effect for Madagascar, 2. Methodology and data
which means it would hurt more to the poor, in percentage terms,
than to the rich. They also found that the indirect impact dominates 2.1. Input-output model
the expenditure change instead of direct impact. By specifying and
estimating a consumer demand system with price heterogeneity The input-output model is a systematic structural analytical
between households, Pashardes et al. Pashardes et al. [24] studied method, which reflects the linkage between different economic
the distributional effects of energy price increases in Cyprus. They sectors. It is frequently applied to investigate the price transmission
found that the assumed rise in energy prices would result in a effect, e.g., the inflationary impact of exogenous price shock, at the
nationwide welfare loss of more than EUR 33 million (in 2009 industrial level. Three basic assumptions are commonly adopted in
price) in 2020. More importantly, the impact is regressive, inferring the price transmission analysis [11,16]. First, producer is incapable
that low expenditure households are affected more strongly. When to reduce material input or to adjust input structure via techno-
energy subsidy is usually applied as a policy to reduce energy price, logical improvement in the short-run. Second, value-added of each
the distributional effect of energy subsidy (or removing energy sector keeps constant. Third, the price elasticities of all products are
subsidy) is of special interest to researchers as well as policy zero. The three assumptions simplify the analysis by fully trans-
makers. Rao [25] reported opposite distribution progressivity of the mitting the exogenous price shock to downstream products, which
same subsidy within one country, i.e., kerosene subsidy in India is provides an estimation of the maximum price response to the
regressive for rural residents but progressive for urban residents. shock.
Coady et al. [44] studied the fuel subsidies for five countries and There are two kinds of tables can be used in this paper, one is
confirmed that the distributional effect varies across the cases. single-country input-output table in China and the other is multi-
Using the 2012 input-output table of China, Chen [26] investigated regional input-output table. The former contains more detailed
the distribution of energy subsidy in China and found that the rich sectoral information and higher data quality while the latter can be
urban residents received more welfare transfer, both in absolute used to traced international feedback effect. The single-country
and percentage terms, than the poor. input-output table is applied in this study because the feedback
As the world's largest energy consuming as well as importing effect is deemed to be limited considering economic turnover
country, China relies heavily on the international energy market to through the global economy. There are two kinds of single-country
fill its gap between domestic energy demand and supply [27,28]. input-output models, i.e., the non-competitive one that identifies
Therefore, the price fluctuating nature of energy as a commodity different (domestic or foreign) locations of production and the
will inevitably bring considerable impact on the Chinese economy competitive one that ignores the differences. The identification of
[29]. Moreover, China's increasing energy demand and ongoing imported goods allows the non-competitive single-country input-
energy market reform will amplify the impact. As such, there is output model to separate imported inflation from domestically-
urgent need by the policy makers to understand the impacts of induced effect. Such separation is not trivial for China concerning
energy price volatility on the Chinese economy, both on its size energy-induced inflationary effect because of two reasons. On the
(how large is the general price level change) and its distribution one hand, a large fraction of fossil energy, especially crude oil,
(how different people are affected differently). consumed in China are imported. On the other hand, the Chinese
Recognizing the need, this paper uses a non-competitive input- government has much stronger power on domestic energy market
output model with the monthly empirical data from January 2012 than on international energy market, which means the policy op-
to November 2018 as well as the residential consumption expen- tions to address the inflation driven by domestic and foreign energy
diture data to quantitatively investigate the inflationary and price shocks are different. Therefore, a non-competitive single-
distributional impacts of energy price fluctuation on China. This country input-output model is adopted in this study over the other
paper focuses on the influence of coal and oil prices changes models (see Table 1 for model comparison).
because the two energy types dominate the energy structure of
China. In 2017, coal, crude oil, hydropower, natural gas, nuclear, and
Table 1
other energy types contribute to 61%, 19%, 8%, 7%, 2%, and 3% of the Features of different input-output models.
primary energy input of China.
Global model Single-country model
The marginal contributions of this study stand in the following
three points. First, as an improvement of Chen [16] by taking oil Competitive Non-competitive
price into account and examining the distributional effects, this Feedback effect ✓ Х Х
study does the first time to estimate the inflationary as well as Imported Inflation ✓ Х ✓
distributional effects of energy price shock on China in a consistent High sectoral solution Х ✓ ✓
High data quality Х ✓ ✓
research framework, which provides comparable results for policy
 Z.-M. Chen et al. / Energy 187 (2019) 115974 3

A schematic single-country non-competitive inputeoutput ta- from Sectors kþ1 to nþm), we have:
ble with n domestic sectors (denoted as Sectors 1 to n) and m
foreign sectors (denoted as Sectors nþ1 to nþm) is illustrated in D1;k ¼ AT1;k D1;k
2 þ3Akþ1;nþm Dkþ1;nþm
T
2 3 (6)
6 ai;1 … ai;k 7
Table 2. The column of each domestic sector indicates that domestic
6 di 7
producer purchases goods from the markets (intermediate input) where Di;j ¼ 6 7 6
4 « 5 and Ai;j ¼ 4 « « « 7
5.
and pays rents for production factors (value-added). The row of dj solved to get:aj;1
Eq. (6) can be … aj;k
each domestic sector indicates that its outputs are sold for domestic
 1
reproduction, domestic consumption, domestic capital formation, D1;k ¼ Ik  AT1;k ATkþ1;nþm Dkþ1;nþm (7)
and foreign use. The monetary balance of each domestic sector is:

X
nþm in which Ik is the identity matrix of size k. Therefore, the relative
Pi Qi ¼ Pj Qj;i þ Vi ði ¼ 1; 2; /; nÞ (1) price response of all endogenous sectors (Sectors 1 to k) to the
j¼1 exogenous price shock Dkþ1;nþm is reported by D1;k.
The relative price change results can be used to calculate the
where Pi and Pj are the prices of output from Sectors i and j, general price level change as well as the expenditure change of
respectively, Qi is the total output (in physical unit) of Sector i, Qj;i is each concerned group as long as the associated commodity basket
the output (in physical unit) from sector i used by Sector j, and Vi is is provided. For example, when Ci and Fi indicate the consumption
the value-added of Sector i. and final demand of output (in monetary unit) from Sector i,
If an external price shock appears, new equilibrium status will respectively, the changes of CPI, PPI, as well as GDP deflator can be
be applied to each sector as: calculated as:
3 2
0 X
nþm
0 ,
P i Qi ¼ P j Qj;i þ Vi ði ¼ 1; 2; /; nÞ (2) C
6 1 7 nþm X
j¼1
T 6
CPI changes ¼ D1;nþm 4 « 5 7 Ci (8)
Cnþm i¼1
0 0
in which P i and P j are the prices of outputs from Sectors i and j in
the new equilibrium. 23
Combine Eq. (1) and Eq. (2) gets ,
6 P1 Q1 7 X n
PPI changes ¼ DT1;n 6
4 « 7
5 Pi Qi (9)
 0  X 0
nþm  Pn Qn i¼1
P i  Pi Qi ¼ P j  Pj Qj;i ði ¼ 1; 2; /; nÞ (3)
j¼1
3 2
0 0
In Eq. (3), ðP i Pi Þ and ðP j Pj Þ refer to output price changes of ,
F X
T 6
17 n
Sectors i and j, respectively. Divide both sides of Eq. (3) by Pi Qi to get 6
GDP deflator changes ¼ D1;n 4 « 57 Fi (10)
the relative price change: Fn i¼1

 0  X 0
nþm  
More methodological details including the theory and calcula-
P i  Pi Pi ¼ P j  Pj Qj;i ðPi Qi Þ ði ¼ 1; 2; /; nÞ (4)
j¼1
tions can be referred to the literature [30e33].
0
Introduce di ¼ ðP i  Pi Þ=Pi to denote the relative price change 2.2. Data
and aj;i ¼ Pj Qj;i =ðPi Qi Þ to denote the direct consumption coefficient,
we have: 2.2.1. Input-output data
The 139-sector input-output table of China for 2012 (The [34] is
X
nþm
used in this study. It is the latest available input-output table of
di ¼ aj;i dj ði ¼ 1; 2; /; nÞ (5)
China with the highest data quality and sectoral solution. However,
j¼1
the original table is competitive, which classifies imported goods
Consider a model to estimate the price response of some do- according to domestic sector definition and aggregates domestic
mestic sectors (for simplicity, assume they are Sectors 1 to k, where and imported flows together. In order to construct a non-
1  k  n) to exogenous price shock (price change of the outputs competitive input-output table, we disaggregate intermediate

Table 2
Schematic non-competitive inputeoutput table for a typical open national economy.

Intermediate use Final demand

Sector 1 Sector 2 … Sector n Domestic use Foreign use

Domestic inputs Sector 1

Sector 2
…
Sector n
Foreign inputs Sector nþ1
Sector nþ2
…
Sector nþm
Value-added Wages
Taxes
Depreciation
Surplus
4 Z.-M. Chen et al. / Energy 187 (2019) 115974

input and non-export final demand into domestic and imported be seen that the fluctuation of coal prices are smaller than that of
sources. It is assumed that for the same goods, there is a constant oil. For the same type of energy, the price fluctuation of imported
ratio of imported to domestic sources for every use purposes except and domestic ones are similar.
export. For example, if the imported farming products contribute to
10% of household consumption, it will be assumed that 10% of the 2.2.3. Household consumption expenditure and other data
farming products used for intermediate input, government con- In addition to the impact of fossil energy price fluctuation on the
sumption, and capital formation are imported. Accordingly, we general price level, this paper also studies how residents belonging
construct a non-competitive input-output table for China with 278 to different groups, i.e., seven urban groups with different income
sectors (139 domestic and 139 foreign sectors, see Table A1). In the levels and one rural group (see Table 3), are affected. Detailed
current study, the input-output data are also used to construct the procedures to estimate the consumption baskets and to calculate
consumption, industrial output, and final demand baskets for the the expenditure changes of different groups can be referred to Chen
estimation of general price change. [16,26].
Fig. 2 shows the share of household expenditure on energy
2.2.2. Energy price (mainly coal, electricity, natural gas and refined oil) by the eight
This study evaluates the impact of price shocks from domesti- income groups. It is interesting to find that the urban groups spend
cally produced coal (domestic coal hereafter), imported coal, similar portions of their incomes on energy commodities (3.6%e
domestically produced crude oil (domestic crude oil hereafter), and 4.2% for the seven groups). However, there is a significant energy
imported crude oil from January 2012 to November 2018. The structure change when income increases. On the one hand, richer
Chinese coal price index compiled by the Chinese Coal Industry households tend to expend more, in both absolute and percentage
Association is used to calculate the price change of domestic coal. terms, on petroleum products, which might be explained by the
When the original index is released weekly, the monthly domestic fact that they own more cars and drive more. On the other hand, the
coal price change is calculated by assuming: (1) the monthly price is residents with higher income spend less on the other energy (coal,
the average of daily price; and (2) the daily price within one week gas, electricity, and heat), which means the income elasticity of
keeps constant. The monthly domestic crude oil price is calculated demand is less than one. An intuitive explanation is that the
based on the monthly spot price of crude oil produced by the cooking, heating, and lighting demands do not change much along
Daqing oilfield and the Shengli oilfield, who contribute to around with income change.
40% of China's domestic crude oil production. The two prices are
weighted according to the outputs of the two oilfields from 2012 to 3. Results
2017. The price data are obtained from the Wind Database [35]
while the output data are obtained from China National Petroleum 3.1. Inflationary effect
Corporation Yearbook [43], the Annual Report of Shengli Oilfield
Branch of China Petrochemical Co., Ltd. [36e38], and news reports Fig. 3 depicts, from January 2012 to November 2018, the infla-
[39,40]. The monthly imported coal price and imported crude oil tion pressure caused by the fluctuation of coal price and/or crude oil
price, which are measured in US dollar per ton, are published by the price in terms of general price level change. Positive value indicates
General Administration of Customs of the PR China [41]. The prices that energy price change drives up inflation, while negative one
are transformed into Chinese yuan per ton based on the exchange means it contributes to deflation. The lasting decline in coal price
rate extracted from The People's Bank of China. between June 2012 and July 2016 leads to a steady deflationary
Fig. 1 shows the price indexes of domestic coal, domestic oil, effect (average pressures to CPI, PPI, and GDP deflator are 0.22%,
imported coal and imported oil from January 2012 to November 0.99%, and 0.71%, respectively). In the next 15 months, the recovery
2018, which decreased by 8.32%e45.60% during the period. It can of coal price boosts the general price level. The inflation pressure

120

100
PRICEINDEXES (JAN-2012 = 100)

80

60

40

20

0
Oct-18
Jul-12
Oct-12

Oct-13

Jul-14
Oct-14

Jul-15
Oct-15

Jul-16
Oct-16

Jul-17
Oct-17

Jul-18
Jul-13

Apr-15

Apr-16

Apr-17

Apr-18
Apr-12

Apr-13

Apr-14

Jan-17

Jan-18
Jan-12

Jan-13

Jan-14

Jan-15

Jan-16

domes c coal domes c oil imported coal imported oil

Fig. 1. Fossil energy price indexes from January 2012 to November 2018.
 Z.-M. Chen et al. / Energy 187 (2019) 115974 5

Table 3
Monthly per capita consumption expenditures of households from different income groups in 2012.

Household group Division standard Average expenditure (￥)

Rural The entire rural households 586.80

Urban lowest First decile in the urban households 608.45
Urban low Second decile in the urban households 800.89
Urban lower middle Second quintile in the urban households 1023.40
Urban middle Third quintile in the urban households 1310.00
Urban upper middle Fourth quintile in the urban households 1652.51
Urban high Ninth decile in the urban households 2149.74
Urban highest Tenth decile in the urban households 3138.47

Source: Infobank [42].

4.5%
4.0%
3.5%
3.0%
2.5%
2.0%
1.5%
1.0%
0.5%
0.0%
Rural Urban Urban low Urban lower Urban Urban Urban high Urban
lowest middle middle upper highest
middle

Refined petroleum and nuclear fuel processed products

Coal mining products
Gas produc on and supply
Electricity, heat produc on and supply

Fig. 2. Budget share of expenditure on household energy demand.

Fig. 3. The inflationary effects of coal price changes (a), oil price changes (b) and price changes of both coal and oil (c) on economy.
6 Z.-M. Chen et al. / Energy 187 (2019) 115974

Table 4 driven by crude oil price fluctuation dives rapidly in the second half
Inflation pressures caused by different energy price fluctuation. of 2014 because of the international oil price plunge. Afterward, an
CPI PPI GDP deflator impulse shape increase of the inflation pressure occurs along with
Domestic coal 0.22% 0.99% 0.71%
the crude oil price recovery. Indicated by GDP deflator, the price
Domestic oil 0.42% 1.15% 0.91% volatility of coal and crude oil prices contributes to alleviating
Imported coal 0.04% 0.14% 0.10% inflation between June 2012 and August 2016 but stimulating
Imported oil 0.42% 0.86% 0.71% inflation during the rest of the period. To sum up, the monthly

Fig. 4. Distributional effects of coal price fluctuation: (a) in absolute value, (b) in relative value.
 Z.-M. Chen et al. / Energy 187 (2019) 115974 7

average responses of CPI, PPI, and GDP deflator to fossil energy might be of special interest to the separation of imported energy
price fluctuation are 0.64%, 1.95%, and 1.46%, respectively. price shocks from other factors. Table 4 separates the inflation
Regarding the energy-driven inflation in China, policy makers pressures owing to the price fluctuations of domestic coal,

150

100

50

-50

-100
Jul-12
Oct-12

Jul-13
Oct-13

Jul-14
Oct-14
Jan-12
Apr-12

Jul-15
Oct-15

Jul-16
Jan-13
Apr-13

Oct-16
Jan-14
Apr-14

Jul-17
Oct-17
Jan-15
Apr-15

Jul-18
Oct-18
Jan-16
Apr-16

Jan-17
Apr-17

Jan-18
Apr-18
(a)

5%
4.0%
4%

3% 3.0%

2% 2.0%

1%

0%

-1%

-2%

-3%
Jul-12
Oct-12

Jul-13
Oct-13

Jul-14
Oct-14
Jan-12
Apr-12

Jul-15
Oct-15
Jan-13
Apr-13

Jul-16
Oct-16
Jan-14
Apr-14

Jul-17
Oct-17
Jan-15
Apr-15

Jul-18
Oct-18
Jan-16
Apr-16

Jan-17
Apr-17

Jan-18
Apr-18

(b)

Fig. 5. Distributional effects of crude oil price fluctuation: (a) in absolute value, (b) in relative value.
8 Z.-M. Chen et al. / Energy 187 (2019) 115974

imported coal, domestic crude oil, and imported crude oil. Ac- affected by coal price fluctuation. It can be found that, the absolute
cording to the table, the inflation pressure caused by domestic expenditure of rural residents is less affected by coal price change
crude oil price change is larger than those caused by the other three comparing with that of urban residents. However, the relative
sources. impact, measured by the ratio of expenditure change to total
expenditure, shows that rural residents are affected as much as the
residents of urban middle income group. More importantly, the
3.2. Distributional effect richer urban residents tend to be affected heavier in terms of ab-
solute expenditure change, while the impacts on the relative
While the efficiency concern measured by the inflationary effect expenditure are opposite, i.e., richer ones are less affected in terms
reflects the overall impact of energy price fluctuation, the equality of percentage expenditure change. This means the impact of coal
concern pays more attention to the heterogeneous impact to indi- price fluctuation is regressive to urban residents. This point should
vidual. Fig. 4 shows how the expenditures of different groups are

Fig. 6. Distributional effects of coal and crude oil price fluctuation: (a) in absolute value, (b) in relative value.
 Z.-M. Chen et al. / Energy 187 (2019) 115974 9

be paid special attention regarding policy design for resource tax imported crude oil (0.71%), domestic coal (0.71%), and imported
and carbon tax, which are in risk of enlarging social inequality. coal (0.10%).
Illustrated in Fig. 5 are the expenditure changes caused by crude In terms of the distributional effect, the rising of energy price
oil price fluctuation of the eight resident groups. Richer urban tends to cost more money for the urban residents than the rural
residents expend (save) more when crude oil price goes up (down) ones. The results also suggest that the coal price rise leads to
than the poorer ones. At the same time, rural residents are affected regressive impact for urban residents of different income levels, i.e.,
less than their urban neighbors, both in absolute and percentage poorer one suffers more in percentage term. On the contrary, the
terms. Therefore, the impact of crude oil price fluctuation is pro- crude oil price rise causes progressive effect, i.e., richer one suffers
gressive, which suggests that the increase of crude oil price con- more in percentage term. China has implemented a series of fiscal
tributes to enhancing social equality via welfare redistribution. policies (e.g., resource tax, energy consumption tax, and renewable
The accumulated distributional impacts driven by both coal and energy subsidy) to achieve the goals of controlling carbon emission
crude oil prices fluctuation are shown in Fig. 6. The absolute and developing clean energy. According to the findings of this pa-
expenditure of richer urban residents still responses more sensi- per, taxes on coal would be regressive while taxes on oil product
tively to energy price change than that of poorer urban residents would be progressive. So when imposing carbon taxations, the
and rural residents. However, the progressivity in terms of relative government could set tax rates on coal and oil products differently
expenditure change for urban residents becomes more neutral as a to balance the progressivity of the taxation system.
result of the compromise of the opposite impacts of coal and crude The present research could benefit from further studies. For
oil prices changes. example, the non-competitive input-output model applied in this
study does not take into account the imperfect price transmission,
4. Concluding remarks which thus explains the theoretical maximum response to energy
price fluctuation. Therefore, further investigation on the price
The inflationary and distributional impacts originated from transmission capacity should provide a simulation closer to the
energy price fluctuation to the Chinese economy are studied based real-world situation. Another concern comes from the timeliness of
on a non-competitive single-country input-output model. The data. This study builds upon the input-output and household
model is applied using the monthly empirical coal and crude oil expenditure data of 2012, which does not reflect the up-to-date
price data from January 2012 to November 2018 and the Chinese production and consumption structures. Besides, in this paper, it
input-output table of 2012. The inflationary impacts in terms of is assumed that the input-output structure in China does not vary
prominent general price level indexes, i.e., CPI, PPI, and GDP significantly during the research period, thus can be represented by
deflator, are calculated while the effects driven by price changes of that of 2012. In fact, considering the rapid development of the
different energy types and sources are identified. Besides, the Chinese economy, this assumption will cause bias in the result.
distributional impacts are investigated by analyzing the heteroge- However, it is believed that this study provides the most convincing
neous responses of eight resident groups, i.e., one rural group and quantitative and qualitative results unless updated input-output
seven urban groups. statistics are available.
As an improvement of Chen [16] which investigates the infla-
tionary effect of coal price fluctuation on China, this study considers
Acknowledgements
both coal and oil prices changes and further examines the distri-
butional effect. In terms of the inflationary effect, this study finds
This study has been supported by the Natural Science Founda-
that fossil energy price fluctuations have significantly pressure on
tion of Beijing (No. 9192012), the Research Funds for Graduate
the general price level of China during the research period. The
Student of School of Applied Economics, Renmin University of
inflation rate measured by PPI is the most sensitive (monthly
China, and Fund for Building World-class Universities (Disciplines)
average response of 1.95%) to fossil energy price change, followed
of Renmin University of China (No. KYGJ2019002).
by that measured by GDP deflator (1.46%) and then by CPI (0.64%).
The relative response is consistent with the findings of previous
literature [16, 17]. Further analyses reveal that the price fluctuation Appendix
of domestic crude oil brings larger inflation pressure (monthly
average adjustment pressure on GDP deflator is 0.91%) than

Table A 1
The split and adjusted 2012 non-competitive Chinese input-output table

Sector name Sector code Consumer expenditure items

Domestic Imported

Farming 1 140 Foodstuff

Forestry 2 141 Water, electricity, fuel and others for
households
Animal Production 3 142 Eggs
Fishery 4 143 Aquatic products
Support Services to Farming, Forestry, Animal Production and Fishery 5 144 e
Mining of Ferrous Metal Ores 6 145 e
Mining of Non-Ferrous Metal Ores 7 146 e
Mining and Quarrying of Nonmetallic Mineral 8 147 e
Mining Support Activities and Other Mining and Quarrying n. e. c. 9 148 e
Manufacture of Grain Mill Products 10 149 Foodstuff
Manufacture of Prepared Animal Feeds 11 150 Other commodities
Manufacture of Crude and Refined Oils from Vegetable 12 151 Oils and fats
(continued on next page)
10 Z.-M. Chen et al. / Energy 187 (2019) 115974

Table A 1 (continued )

Sector name Sector code Consumer expenditure items

Domestic Imported

Manufacture of Sugar 13 152 Sugar

Slaughtering and Processing of Meat 14 153 Meat
Processing of Aquatic Products 15 154 Aquatic products
Processing of Other Foods 16 155 Vegetables and fruits
Manufacture of Convenience Food Products 17 156 Other food
Manufacture of Milk and Dairy Products 18 157 Milk and milk products
Manufacture of Flavoring and Ferment Products 19 158 Condiment
Manufacture of Other Food Products n. e. c. 20 159 Other foods
Manufacture of Alcohol and Alcoholic Beverages 21 160 Liquor
Manufacture of Soft Drinks and Refined Tea Products 22 161 Drinks
Manufacture of Tobacco Products 23 162 Tobacco
Spinning, Weaving and Finishing of Cotton and Chemical Fibers 24 163 Clothing materials
Spinning, Weaving and Finishing of Wool 25 164 Clothing materials
Spinning, Weaving and Finishing of Bast and Silk Fibers 26 165 Clothing materials
Manufacture of Knitted and Crocheted Fabrics and Articles, Except Apparel 27 166 Household equipment and supplies
Manufacture of Made-up Textile Articles, Except Apparel 28 167 Bedding article
Manufacture of Textile Wearing Apparel 29 168 Clothes
Manufacture of Leather, Fur, Feather and Its Products 30 169 Other clothing items
Manufacture of Footwear 31 170 Footwear
Processing of Timbers and Manufacture of Products of Wood, Bamboo, Rattan, Palm and Straw 32 171 Furniture materials
Manufacture of Furniture 33 172 Furniture
Manufacture of Paper and Paper Products 34 173 Household sundries
Printing and Reproduction of Recording Media 35 174 Culture and entertainment supplies
Manufacture of Stationeries, Musical Instruments, Products of Arts and Crafts, Sports Goods, Games 36 175 Culture and entertainment supplies
and Toys
Manufacture of Coke Products 37 176 e
Manufacture of Basic Chemicals 38 177 e
Manufacture of Fertilizers 39 178 e
Manufacture of Pesticides 40 179 Household sundries
Manufacture of Paints, Printing Inks, Pigments and Similar Products 41 180 Culture and entertainment supplies
Manufacture of Synthetic Materials 42 181 e
Manufacture of Special Chemical Products 43 182 Culture and entertainment supplies
Manufacture of Daily-use Chemical Products 44 183 Household sundries
Manufacture of Pharmaceutical Products 45 184 Drug costs
Manufacture of Chemical Fibers 46 185 Other commodities
Manufacture of Rubber Products 47 186 Fuel and spare parts for vehicles
Manufacture of Plastic Products 48 187 Household sundries
Manufacture of Cement, Lime and Plaster 49 188 Housing
Manufacture of Products of Plaster and Cement and Similar Products 50 189 Housing
Manufacture of Brick, Stone and Other Building Materials 51 190 Housing
Manufacture of Glass and Glass Products 52 191 Household sundries
Manufacture of Cematic and Porcelain Products 53 192 Household sundries
Manufacture of Refractory Products 54 193 Household sundries
Manufacture of Products of Graphite and Other Nonmetallic Minerals 55 194 Housing
Manufacture and Casting of Basic Iron and Steel 56 195 e
Processing of Steel Rolling Processing 57 196 e
Manufacture of Ferroalloy 58 197 e
Manufacture and Casting of Non-Ferrous Metals and Related Alloys 59 198 e
Processing of Non-Ferrous Metals Rolling 60 199 e
Manufacture of Fabricated Metal Products, Except Machinery and Equipment 61 200 Household sundries
Manufacture of Boiler and Prime Mover 62 201 e
Manufacture of Metalworking Machinery 63 202 e
Manufacture of Lifting and Handling Equipment 64 203 e
Manufacture of Pump, Valve, Compressor and Similar Machinery 65 204 e
Manufacture of Movie, Office Machinery and Equipment, of Projector and Camera 66 205 Culture and entertainment supplies
Manufacture of Other General-Purpose Machinery 67 206 Household equipment
Manufacture of Machinery for Mining, Metallurgy and Construction 68 207 e
Manufacture of Machinery for Chemical Industry, Timber and Nonmetal Processing 69 208 e
Manufacture of Machinery for Agriculture, Forestry, Animal Production and Fishery 70 209 e
Manufacture of Other Special-Purpose Machinery 71 210 Household equipment
Manufacture of Motor Vehicles, Except Parts and Accessories for Motor Vehicles 72 211 Transport equipment
Manufacture of Parts and Accessories for Motor Vehicles 73 212 Fuel and spare parts for vehicles
Manufacture of Railway Transport Equipment 74 213 e
Manufacture of Boats and Ships and Floating Devices 75 214 Fuel and spare parts for vehicles
Manufacture of Other Transport Equipment 76 215 Transport equipment
Manufacture of Generators and Electric Motors 77 216 e
Manufacture of Equipment for Power Transmission and Distribution and Control 78 217 e
Manufacture of Wire, Cable, Optical Cable and Electrical Goods 79 218 e
Manufacture of Batteries 80 219 Household sundries
Manufacture of Household Appliances 81 220 Household equipment
Manufacture of Other Electrical Machinery and Equipment 82 221 Household equipment
Manufacture of Communication Equipment 83 222 Culture and entertainment supplies
Manufacture of Computer 84 223 Communication tools
Manufacture of Broadcasting, Television Equipment, of Radar and Related Equipment 85 224 Culture and entertainment supplies
 Z.-M. Chen et al. / Energy 187 (2019) 115974 11

Table A 1 (continued )

Sector name Sector code Consumer expenditure items

Domestic Imported

Manufacture of Audiovisual Apparatus 86 225 Culture and entertainment supplies

Manufacture of Electronic Components and Parts 87 226 e
Manufacture of Other Electronic Equipment 88 227 Culture and entertainment supplies
Manufacture of Measuring Instruments and Meters 89 228 Culture and entertainment supplies
Other Manufacture 90 229 Household sundries
Comprehensive Utilization of Waste Resources 91 230 e
Repair of Fabricated Metal Products, Machinery and Equipment 92 231 e
Production and Supply of Electricity and Steam 93 232 Water, electricity, fuel and others for
households
Production and Distribution of Gas 94 233 Water, electricity, fuel and others for
households
Production and Distribution of Water 95 234 Water, electricity, fuel and others for
households
Construction of Buildings 96 235 e
Civil Engineering 97 236 e
Construction Installtion Activities 98 237 e
Construction Completion and Finishing, Construction Activities 99 238 e
Wholesale and Retail 100 239 Services
Transport Via Railway 101 240 Trade
Transport Via Road 102 241 Trade
Water Transport 103 242 Trade
Air Transport 104 243 Trade
Transport Via Pipeline 105 244 Water, electricity fuel and other
Cargo Handling, Transport Agency 106 245 Trade
Storage 107 246 Communication service
Post 108 247 Communication service
Accommodation 109 248 Dwelling service fee
Food and Beverage Services 110 249 Food services
Telecommunication and Other Information Transmission Services 111 250 Communication service
Software and Information Technology Services 112 251 Cultural and entertainment services
Monetary Intermediation and Other Financial Services 113 252 Services
Capital Market Services 114 253 Services
Insurance 115 254 Services
Real Estate 116 255 Residential service fee
Renting and Leasing 117 256 Services
Business Services 118 257 Services
Research and Experimental Development 119 258 e
Professional Technique Services 120 259 Services
Technique Promotion and Application Services 121 260 e
Management of Water Conservancy 122 261 e
Ecological Protection and Environmental Control 123 262 e
Management of Public Facilities 124 263 e
Services to Households 125 264 Services
Repair of Motor Vehicles, Electronic Products and Households Goods and Other Services 126 265 Services
Education 127 266 Education
Health Care 128 267 Medical fees
Social Work Activities 129 268 Services
Journalism and Publishing 130 269 Cultural and entertainment services
Radio, Televisions; Movies and Audio-Video Recording Activities 131 270 Cultural and entertainment services
Cultural, Art and Entertainment Activities 132 271 Cultural and entertainment services
Sports Activities 133 272 Cultural and entertainment services
Amusement and Recreation Activities 134 273 Cultural and entertainment services
Social Security 135 274 e
Public Management and Social Organization 136 275 Services
Mining and Washing of Coal 137 276 Water, electricity, fuel and others for
households
Manufacture of Refined Petroleum Products, Processing of Nuclear Fuel 138 277 Fuel and spare parts for vehicles
Extraction of Crude Petroleum and Natural Gas 139 278 e

Note: In order to distinguish endogenous price change and exogenous price shock clearly as well as to easy calculation, the sector order of the original table is rearranged.
‘dd’stands for sectors which households do not consume.

References [6] Callan T, Lyons S, Scott S, Tol RS, Verde S. The distributional implications of a
carbon tax in Ireland. Energy Policy 2009;37(2):407e12.
[7] Sanchez MV. Welfare effects of rising oil prices in oil-importing developing
[1] BP. The statistical review of world energy 2018. 2018.
countries. Develop Econ 2011;49(3):321e46.
[2] Gupta P, Goyal A. Impact of oil price fluctuations on Indian economy. OPEC
[8] Olayungbo D, Hassan W. Effects of oil price on food prices in developing oil
Energy Review 2015;39(2):141e61.
exporting countries: a panel autoregressive distributed lag analysis. OPEC
[3] Kilian L, Vigfusson RJ. The role of oil price shocks in causing US recessions.
Energy Review 2016;40(4):397e411.
J Money Credit Bank 2017;49(8):1747e76.
[9] Jimenez Rodriguez R, Sanchez M. Oil price shocks and Japanese macroeco-
[4] Jimenez Rodriguez R, Sanchez M. Oil price shocks and real GDP growth:
nomic developments. Asian Pac Econ Lit 2012;26(1):69e83.
empirical evidence for some OECD countries. Appl Econ 2005;37(2):201e28.
[10] Al-mulali U, Che Sab CNB. The impact of oil shocks on China's GDP: a time
[5] Lescaroux F, Mignon V. Measuring the effects of oil prices on China's economy:
series analysis. OPEC Energy Review 2013;37(1):20e9.
a factor-augmented vector autoregressive approach. Pac Econ Rev 2009;14(3):
[11] Berument H, Taşçı H. Inflationary effect of crude oil prices in Turkey. Phys A
410e25.
12 Z.-M. Chen et al. / Energy 187 (2019) 115974

Stat Mech Appl 2002;316(1e4):568e80. [29] Wen F, Min F, Zhang YJ, Yang C. Crude oil price shocks, monetary policy, and
[12] Cunado J, De Gracia FP. Do oil price shocks matter? Evidence for some Eu- China's economy. Int J Financ Econ 2018:11.
ropean countries. Energy Econ 2003;25(2):137e54. [30] Leontief W. Quantitative input and output relations in the economic systems
[13] Cunado J, De Gracia FP. Oil prices, economic activity and inflation: evidence of the United States. Rev Econ Stat 1936;18(3):105e25.
for some Asian countries. Q Rev Econ Financ 2005;45(1):65e83. [31] Leontief W. Environmental repercussions and the economic structure: an
[14] Lin B, Mou D. The impact of energy price increases on macro-economy: an input-output approach. Rev Econ Stat 1970;52(3):262e71.
analysis based on CGE method. Econ Res 2008;11:88e101 [in Chinese]. [32] Leontief W, Leontief WW, Duchin F, Faye D. The future impact of automation
[15] Li Z, Xing H. Influence of international oil price fluctuation to China's inflation: on workers. New York: Oxford University Press; 1986.
analysis based on new Keynesian Phillips Curve. Journal of International Trade [33] Miller RE, Blair PD. Input-output analysis: foundations and extensions. Cam-
2011;11:18e30 [in Chinese]. bridge University Press; 2009.
[16] Chen Z-M. Inflationary effect of coal price change on the Chinese economy. [34] National Bureau of Statistics. Input-output tables of China 2012. China Sta-
Appl Energy 2014;114:301e9. tistics Press; 2015 [in Chinese].
[17] Guo J, Zheng X, Chen Z-M. How does coal price drive up inflation? Reex- [35] Wind Database (in Chinese), https://www.wind.com.cn/NewSite/edb.html.
amining the relationship between coal price and general price level in China. [Accessed 8 April 2019].
Energy Econ 2016;57:265e76. [36] SINOPEC. Annual report of Shengli oilfield Branch of China. Petrochemical Co.,
[18] Morana C. Macroeconomic and financial effects of oil price shocks: evidence Ltd.; 2013 (in Chinese), http://quotes.money.163.com/f10/ggmx_600688_
for the euro area. Econ Modell 2017;64:82e96. 1335762.html. [Accessed 26 December 2018].
[19] Ravazzolo F, Rothman P. Oil and US GDP: a real-time out-of-sample exami- [37] SINOPEC. Annual report of Shengli oilfield Branch of China. Petrochemical Co.,
nation. J Money Credit Bank 2013;45(2-3):449e63. Ltd.; 2016 (in Chinese), http://guba.eastmoney.com/news,600028,621219193.
[20] Kerkhof AC, Moll HC, Drissen E, Wilting HC. Taxation of multiple greenhouse html. [Accessed 26 December 2018].
gases and the effects on income distribution: a case study of The Netherlands. [38] SINOPEC. Annual report of Shengli oilfield Branch of China. Petrochemical Co.,
Ecol Econ 2008;67(2):318e26. Ltd.; 2017 (in Chinese), http://guba.eastmoney.com/news,600028,750792752.
[21] Kpodar K, Djiofack C. The distributional effects of oil price changes on html. [Accessed 26 December 2018].
household income: evidence from Mali. J Afr Econ 2009;19(2):205e36. [39] China National Radio. Daqing Oilfield completed the 2017 oil and gas pro-
[22] Jiang Z, Ouyang X, Huang G. The distributional impacts of removing energy duction task. 2017 (in Chinese), http://hlj.cnr.cn/jdxw/20180101/t20180101_
subsidies in China. China Econ Rev 2015;33:111e22. 524083121.shtml. [Accessed 26 December 2018].
[23] Vecchi G, Andriamihaja NA. An evaluation of the welfare impact of higher [40] Yicai. Daqing and Shengli Oilfields reduced total output of 5 million tons of
energy prices in Madagascar. Africa Region Working Paper Series 2007;106. crude oil in 2016. 2017 (in Chinese), https://www.yicai.com/news/5203016.
[24] Pashardes P, Pashourtidou N, Zachariadis T. Estimating welfare aspects of html. [Accessed 8 April 2019].
changes in energy prices from preference heterogeneity. Energy Econ [41] General Administration of Customs of the PR China (in Chinese), http://www.
2014;42:58e66. customs.gov.cn/. [Accessed 8 April 2019].
[25] Rao ND. Kerosene subsidies in India: when energy policy fails as social policy. [42] Infobank. China's 2012 household income statistics of urban households by
Energy for Sustainable Development 2012;16(1):35e43. income level. 2012 (in Chinese), http://www.infobank.cn/IrisBin/Text.dll?
[26] Chen Z-M. Inventory and distribution of energy subsidies of China. Energy J db¼TJ&no¼609782&cs¼14207700&str¼2012þ%B0%B4%CA%D5%C8%EB%B5%
2017;38:48e61. C8%BC%B6%B7%D6. [Accessed 15 October 2018].
[27] Zhang H-Y, Ji Q, Fan Y. An evaluation framework for oil import security based [43] China national petroleum corporation Yearbook 2016. Petroleum Industry
on the supply chain with a case study focused on China. Energy Econ 2013;38: Press; 2018 [in Chinese].
87e95. [44] Coady D, Parry IW, Sears L, Shang B. How large are global energy subsidies?
[28] Sun M, Gao C, Shen B. Quantifying China's oil import risks and the impact on International Monetary Fund; 2015.
the national economy. Energy Policy 2014;67:605e11.