Basic Electricity Need of Households: Empirical Evidence from China Xiaoping He Xiamen University David Reiner EPRG, University of Cambridge.

Basic Electricity Need of Households:

Empirical Evidence from China

Xiaoping HeXiamen University

David ReinerEPRG, University of

Cambridge

Content

Background

Methodology

Empirical analysis

Conclusions

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

0

1,000

2,000

3,000

4,000

5,000

0

3

6

9

12

15

3.49

12.5713.16

12.50National totalResidential Proportion

TWh %

Electricity consumption of China’s households

Nationaltotal

Residential Residential per capita

TWh TWh kWh1980 301 11 112012 4976 622 459

Growth 9.2% 13.6% 12.5%

I. Introduction

per cap.Final energy Electricity

（ kce/year) （ kWh/year)

National 195. 459

Urban 225 500

Rural 163 414

Urban Rural

Household Consumption of Energy Commodities

Noncommercial energy

• Noncommercial energy plays a significant role in rural areas, with non-commercial energy consumed by rural households being nearly 50%

• Given non-commercial energies included， the energy consumption of urban family is higher than that of the rural in some regions

Electricity price in China

The prices of electricity, set by the government, has been kept steady at a low level.

Electricity consumption of households is subsidized by industry and commerce users

Residential electricity price is politically sensitive, the government is very careful about increasing it.

Coal prices have been deregulated from1992, then the increases of fuel cost in generation can not be transferred to end user.

The “price linkage  mechanism”, beginning in 2004 as a solution to the contradictions between the coal and electricity industries, has faced difficulties in execution.

Since July 2012, a new pricing regime for household electricity, the increasing block tariff, started nationwide in China.

Fixed tariffs vs IBTs

IBTs implemented nationwide from 2012, except Tibet and Xinjiang

Before 2012, electricity tariffs were fixed, identical within each province, rarely

adjusted.

　 kWh/month

Rate(yuan)

Sichuan0-260 0.522

181-280 0.622≥281 0.822

Beijing0-240 0.488

241-400 0.538≥400 0.788

Qinghai0-150 0.377

151-230 0.427≥231 0.677

Debates about the IBTs

IBT, comprising various quantities and charges, has often been promoted and adopted as a solution to address social equity, efficiency, or environmental concerns.

In IBTs, the price of electricity will be low for consumption up to a certain quantity, whereby any consumption exceeding the quantity will be charged a higher price.

In IBT schemes in China, the electricity volume of the first is said to be based on the basic need of household, covering 80% of residential demand.

The ability of IBTs to deliver equity targeting at the poor depends on setting the volume of electricity in the initial block equal to the basic need.

Question: how to determine the electricity volume of the first block ?

II. Methodology

Link between basic energy needs and energy poverty

Methods defining the basic energy needs

Concept of basic needs

Basic needs are “basically linked to the needs of ‘living’ at the most general level ” (UN, 1983).

It is universally recognized that energy service is of centrality for the provision of basic human needs.

No consensus on the amount of energy to meet basic needs, as energy needs vary significantly among countries and regions.

Concept of basic energy needs

Discussions on basic energy needs have been often found in the literatures on energy poverty.

An energy poverty line specifies a minimum level at which household can be considered non-energy poor, the energy quantity at the minimum level is regarded as the basic need.

Methods to define basic energy need

A. Physical quantity approach Ideology

Defines the energy poverty line (minimum amount of energy demand) based on a basket of energy goods and services.

Disadvantages

Difficulties in defining the basket Assumptions on the type of energy consuming appliances, their sizes, efficiencies and utilizations. Arbitrariness in choosing the threshold of energy poverty line; inconsistence in quantifying energy content of the basket

B. Expenditure approaches

　 Expenditure method　 Expenditure share methodIdeology 　　

　Examines the expenditure of household on energy.

　Examines the proportion of household expenditure on energy.

The energy expenditure of the household whose expenditure is at the known income-poverty line is regarded as to meet their basic energy needs.

A household is classified as energy-poor if the share of their energy expenditure in income is larger than a specific percentage (e.g. 10%)

characteristics 　　

　Assumes that the income poor definitely are energy poor.

Assumes that the income poor definitely are energy poorAssumes the poor spend a higher percentage of income on energy.

　 Expenditure approaches Advantages 　

No need to investigate what energy sources and how much of each are actually used by the poor, as income poverty /poverty line is well-defined in most countries.

　 The data is readily available; the measure technique is relatively simple.

Disadvantages 　

　 Assumes the energy poverty follows exactly the same pattern as the income poverty.

Ignores that the energy budget depends not only on the type of energy used and its price, but also the efficiencies and the costs of household appliances.

　 The preset expenditure or expenditure share is usually an arbitrary figure.

continued

C. Income invariant energy demand approach

　 Barnes et al.(2010; 2011)Ideology

Defines the threshold of energy poverty as the income decile where household energy consumption is significantly different from that at the previous income decile.

　The households below the threshold only consume a bare minimum level of energy for subsistence.

characteristics

　 the definition of energy poverty in concept is similar to expenditure methods.

Advantages

　 Does not specify any predefined figure as threshold, then no drawbacks of arbitrariness;

　 The definition of basic needs is based on demand function, easy for quantitative analysis.

III. Basic Electricity Need: empirical analysis based on survey data

Survey Data

Conducted in 2008 and in 2009, covering 1748 households.

Sampled population from households in Beijing, Shanghai, and Guangdong

Data limitation The survey is not been conducted especially for getting energy information. Except electricity consumption, more details on energy are not available, such as the amount of each type of energy , the expenditure on specific fuel, and the quality of energy services.

0.0

05

.01

.01

5P

roba

bilit

y D

ens

ity

0 200 400 600 800electricity per capita

Kernel density estimate

kernel = gaussian, bandwidth = 5.2304

urban areas

0.0

1.0

2.0

3.0

4.0

5P

roba

bilit

y D

ens

ity

0 50 100 150electricity per capita

Kernel density estimate

kernel = gaussian, bandwidth = 2.6669

rural areas

25 8

Density estimates of Electricity Consumption

( kWh per capita )

Electricity need varies with income and the control factors :

Xij , control variables

, income dummy variables, y splits the sample into ten categories by the quantiles of per capita income

Modelling electricity demand

Hypothesis: if there exists a basic minimum amount of electricity consumption that a household must maintain to subsist, then electricity consumption up to that amount should be unresponsive to changes in income.

Estimated parameters

Control Variables rural urbanConstant -0.9767 9.789Number of family members -0.1450 -0.1780Log of Household living area 0.1438 0.1753Age of household head -0.0055 -0.0001Gender of household head 0.0669 0.1519Education of household head (years) 0.0127 0.0351Distance to the nearest commercial center -0.0041 0.0000

Has frequent electricity outage (1= Yes,0=No) -0.1514 -0.0674

Latitude of community -0.1156 -0.0600Longitude of community 0.0478 -0.0365Area of agricultural land 0.0004  Amount of Stocks, bonds and deposits   0.0001Household has electricity consumption for production (1= Yes, 0=No) 0.4926 0.3316

Log price of electricity -6.0082 -1.1302

Log price of gas -1.7587 -1.4949

Findings about the control variables

Family size and household living area significantly influence the demand for electricity by households in either rural areas or urban areas. Family size has negative impact, while living area has positive impact.

The education level and the gender of household head significantly influence the demand for electricity by households in urban areas rather than in rural areas.

Findings (continued)

The distance from the house to commercial center affects the electricity of rural households rather than urban households.

The latitude significantly affects household electricity demand, but the longitude does not.

Rural families and urban families respond differently to energy price changes. The urban respond negatively to gas price; while the rural respond negatively to electricity price.

Estimates of income dummiesIncom

edecile

Rural Urban

2 -0.0206 0.00853 -0.0505 0.04594 0.0066 0.15045 -0.0380 0.33236 0.1668 0.25357 0.3493 0.37988 0.2923 0.40799 0.3064 0.3943

10 0.4020 0.4646

The changes in electricity consumption at lower levels of income are not as sensitive to slight changes in income as those with changes at higher levels of income. Rural families are less elastic to income than urban families; high-income families are more sensitive than low-income families.

In the case of the electricity consumption, it is far from having been reached a theoretical saturation status.

4000 4500 5000 5500 6000 6500 70000

20

40

60

ruraldecile 7 (5300-6667)

kWh

4000 5000 6000 7000 8000 9000 100000

20

40

60

80

100 urban

decile 4 (5250-6938)

Polynomial (decile 4 (5250-6938))

decile 5 (7133-9600)

kWh

The electricity demand does not respond to income changes until a specific income decile.

Income decile

Rural Urban

Electricity Income Electricity Income

1 10.6 366 38.3 6242 11.4 1068 40.2 25623 10.1 2108 37.7 44304 12.7 2849 44.8 61035 12.8 3669 47.7 82286 15.9 4764 47.8 109947 22.8 6038 54.0 137298 20.3 7884 64.9 172739 24.3 10831 61.3 2326210 27.1 34967 87.6 72943

Electricity consumption by income decile

Rural Electricity vs urban Electricity

Electricity demand of the rural family is much lower, even when its income roughly equal to that of urban family

Electricity consumption per capita at the bare minimum level is 22.8kWh in rural areas, 47.7 kWh in urban areas.

From village to city, the

energy consumption

pattern changes

much

1 2 3 4 5 6 7 8 9 100

20

40

60

80

100

10.610.410.112.712.815.922.820.324.327.1

38.340.237.744.847.747.8

54.064.961.3

87.6rural

Income decile

kWh

Why the rural electricity need is lower

a. In urban areas, heating /cooling with electricity is popular; the electricity used for entertainment and house electric appliances is greater than in rural areas.

b. In rural areas, many households still consume traditional biomass resources for cooking; electricity is mainly used for lighting and some electric appliances, and the electricity for lighting is larger than in urban areas.

Proportion of Households by Cooking Fuel

Rural UrbanFirewood 56.4% 5.0%Electricity 16.5% 9.3%Gas 21.0% 83.1%Coal 3.0% 1.7%Solar 0.0% 0.2%Biogas 2.0% 0.2%Other 1.1% 0.4%

c. Energies used in rural areas are less convenient and efficient than those used in urban areas.

d. Coal remains an important energy source for heating in rural areas, especially in north China (e.g., Beijing).

e.g. Energy use in Beijing(kce per capita)

ItemTotal Energy consumption

Coal

Residentia

l total665 114

Urban 606 34 Rural 996 568

VI. Conclusions

The basic electricity need is defined based on the concept “energy poverty” and estimated from the electricity demand function.

The basic electricity need of rural family is fewer, because the energy consumption pattern changes much from village to city.

A few of household-related factors affect household electricity consumption, such as latitude, living area, family size, energy prices.

Given other factors controlled, the electricity consumption will be income-sensitive at higher income levels.

If there is a decrease in income, electricity demand does not necessarily decrease, although the burden of expenditure on electricity may increase significantly

Given the increasing structure of prices in IBTs ， preferential policies on residential electricity should be applied targeted at low-income families.

End

Thank you very much！