Correlation Between Energy

8/2/2019 Correlation Between Energy

http://slidepdf.com/reader/full/correlation-between-energy 1/14

March 2006 ©2006 Organization of the Petroleum Exporting Countries 41

Correlation between energy

usage and the rate of economic development

Salman Saif Ghouri

Abstract

This paper reviews the correlation between per capita GDP and percapita consumption of different sources of energy for OPEC Mem-ber Countries, the G-7 and three Asian countries, both with and

without natural logarithm. In addition, the paper estimates the ratiosfor total GDP and total energy consumption of different sources of en-

ergy and also estimates GDP energy consumption elasticities.The paper concludes that on a per capita basis most OPEC

Countries exhibit negative and weak relationships for all forms of

energy, including electricity. For the G-7 and Asian countries, thisrelationship is positive and strong, with the exception of oil for

G-7 countries, where there is a weak correlation. Surprisingly, mostOPEC Countries showed a comparatively strong and positive corre-

lation when tested for total GDP in relation to total energy consump-tion of the respective energy sources. The relationship for the rest of

the countries remains unchanged.Population might have distorted the results in OPEC Countries.

These results suggest that one should be cautious when drawingconclusions and not ignore the aggregate comparison, as this could

otherwise lead to wrong results.For G-7 countries, there has been a significant shift in the pat-

tern of energy consumption in relation to GDP when comparing1960–73 and 1973–2001. All adjusted downward in the later period.



42 ©2006 Organization of the Petroleum Exporting Countries OPEC Review

However, the greatest adjustment was associated with petroleum

consumption.The general conclusion is that wealth creation in G-7 countries

is directly associated with the efficient use of all forms of energy. Incontrast, most OPEC Countries exhibit a weaker linkage between

energy consumption and economic development on a per capita

basis, probably due to inefficient usage of resources or due to dis-proportionate distribution of wealth and thus energy usage.

Abstract — continued

Dr Ghouri is Senior Economist at the Business Environment Section

Corporate Planning of Qatar Petroleum, Doha, Qatar.




NUMBER OF studies have been carried out in the past that proved

the importance of energy usage for the rate of economic development.

A preliminary study (Ferguson et al , 1997) for the Group of Seven

(G-7) as a whole (the United States, Canada, Japan, Germany, France, Italy

and the United Kingdom) showed that there is a strong correlation between

electricity usage and wealth creation, but no relationship between total ener-

gy use and wealth. More recently, Ferguson, Wilkinson and Hill (2000) ex-

panded the scope of the study to almost all the countries of the world (more

than 100 countries, comprising over 99 per cent of global GDP). The gen-

eral conclusion of their study was that wealthy countries have a stronger

correlation between electricity usage and wealth creation than between to-

tal energy use and wealth. In addition, they concluded that in wealthy coun-

tries the increase in wealth over time correlates with an increase in the

e/E ratio (the proportion of the total primary energy supply consumed as elec-

tricity in kWh/toe).This paper expands the scope further and attempts to investigate the principal

driving energy source in economic development by assessing the relationship be-tween various forms of energy (oil, natural gas, coal and electricity) and GDP. It alsoestimates the elasticity of individual sources of energy in relation to GDP in OPEC,the G-7 and some fast-growing Asian developing countries with varying GDPs, and

draws a comparison.

The correlation between per capita consumption of oil, natural gas, electric-ity and total final energy; per capita total primary energy supply; and GDP per capita

is examined for the period 1971 to 2001, with and without natural logarithm.1 Like

earlier studies, this one also examines the correlation between the ratio of electricity

consumption to total primary energy supplies (e/E ratio, in toe) and per capita GDP.

Finally, it estimates energy elasticities for individual sources of energy in respective

countries. This measures how GDP varies following a one per cent increase or de-

crease in individual energy consumption.

The measure of GDP used in this analysis is GDPPPP, compiled using purchas-

ing power parities (PPPs). PPPs are the rates of currency conversion that eliminate the

differences in price levels between countries that occur because of fluctuations in ex-change rates. A given sum of money, when converted into different currencies at PPP

rates, buys the same basket of goods and services in all countries. The GDP figures

are given in 1995 US dollars.

1. National analysis of energy/GDPTable 1 shows the GDP and energy profile for each country under considera-

tion at the end of 2001. Within OPEC, GDP varied widely. It ranged from $17 billion

for Qatar to as much as $561 bn for Indonesia. Six countries had GDPs over $100 bn,

while the remaining five had far less than $100 bn. In sharp contrast, the G-7 countries

have a strong GDP base. The United States’ GDP was close to $9 trillion, Japan’s was

$3 trillion, while France, Germany, Italy and the UK all had GDPs over $1,000 bn at

A



4 4

©

2 0 0 6 Or g a n i z a t i o n o f t h eP e t r o l e u mE x p or t i n g C o u n t r i e s

OP E C R e vi e w

Table 1

GDP and energy profile, 2001

Countries

GDP

$ bn

PPP

Population

millions

TPES/GDP

1995 PPP

Per capita

GDP $ PPP

Per capita

total final

energy

consumption

(toe)

Per capita

final

electricity

consumption

(toe)

P

co

Algeria 177 30.9 0.59 5,736 0.53 0.05

Indonesia 561 209 0.71 2,684 0.56 0.03

IR Iran 355 64.5 1.08 5,500 1.54 0.14

Iraq 30 23.8 0.37 1,264 0.98 0.13

Kuwait 34 2.0 0.6 16,747 6.71 0.98

SP Libyan AJ 28 5.4 0.45 5,089 1.99 0.34

Nigeria 102 129.9 2.85 786 0.66 0.01

Qatar 17 0.6 1.2 29,097 15.29 1.32

Saudi Arabia 256 21.4 0.78 11,974 3.20 0.44

UAE 63 3.0 0.63 21,169 6.46 0.94

Venezuela 130 24.6 0.67 5,285 1.50 0.22



M a r c h 2 0 0 6

©2 0 0 6 Or g a n i z a t i o n

o f t h eP e t r o l e u mE x p or t i n g C o u n t r i e s

4 5

Source: International Energy Agency

Countries

GDP

$ bn

PPP

Population

millions

TPES/GDP

1995 PPP

Per capita

GDP $ PPP

Per capita

total final

energy

consumption

(toe)

Per capita

final

electricity

consumption

(toe)

P

co

India 2,707 1032.0 1.08 2,623 0.37 0.03

Pakistan 240 141.5 0.88 1,700 0.37 0.03

Sri Lanka 57 18.7 0.48 3,029 0.39 0.02

Australia 490 19.5 0.2 25,196 3.75 0.79

Canada 832 31.1 0.3 26,787 5.95 1.32

France 1,394 60.9 0.2 22,895 2.85 0.56

Germany 1,922 82.3 0.2 23,342 2.99 0.52

Italy 1,287 57.9 0.1 22,223 2.32 0.41

Japan 3,125 127.2 0.2 24,572 2.69 0.62

UK 1,293 58.8 0.2 22,000 2.75 0.49

USA 8,977 285.9 0.3 31,400 5.39 1.01

Table 1 (cont’d)

GDP and energy profile, 2001




the end of 2001. In developing Asia, India leads with a GDP of $2,707 bn. However,

per capita GDP illustrates quite a different picture. Qatar’s GDP was lowest among

the OPEC Members, but it has the highest per capita income. The situation in Kuwait

and the UAE is similar. In the G-7, all countries have per capita incomes well above$20,000 per annum.

The ratio of total primary energy supplies to GDP (TPES/GDP) that measures

energy efficiency is shown in table 1. The ratio measures how much energy a country

uses to produce a dollar of output. For G-7 countries, it lies between 0.1 and 0.3, far

less than for OPEC and the Asian countries. Further analysis of other forms of energy

reveals that countries with higher per capita incomes generally consume more energy

per person than countries with lower per capita incomes.

2. Energy usage in industrial activityTable 2 shows, as a percentage, the share of each energy source used in the in-

dustrial sector. G-7 countries prudently utilize all forms of energy in their industrial

sectors, depending upon their resources. The contribution of electricity is well over

20 per cent, while in most OPEC Countries the use of electricity is marginal.2 In con-

trast, the use of oil and natural gas in industrial activity in OPEC Countries has been

substantial, owing to enormous natural gas and oil reserves in these countries. Sur-

prisingly, even the three Asian countries use far less electricity for industrial activity.

Generally these countries use more of the resource that is abundant in their respec-

tive country.

3. Energy GDP correlationThe analysis of table 3 reveals interesting and thought-provoking results. With

the exception of Algeria, Indonesia and Venezuela, all OPEC Member Countries

show a negative and weak relationship between electricity consumption and econom-

ic growth. In contrast, the rest of the countries in the study exhibit a strong correla-

tion between electricity usage and economic growth. The analysis further reveals that,

with the exception of Algeria, Indonesia, the Islamic Republic of Iran, Kuwait and

Venezuela, all OPEC Member Countries have a weak and negative correlation be-

tween respective per capita energy consumption and per capita GDP. In sharp con-

trast, G-7 countries generally show a strong relationship, except in the case of oil.Meanwhile, Asian countries show a positive and strong correlation with each of the

energy sources used, except for coal in Pakistan and TPES in India. The last column of

table 3 supports the result of earlier studies that there is a strong relationship between

electricity consumption as a proportion of total primary energy supply (e/E ratio in

toe/toe) and per capita GDP.

Table 4 illustrates per capita GDP and respective per capita energy consumption

in natural logarithm. There is hardly any difference to the results of table 3, which is

based on per capita correlation.

Table 5 highlights the correlation between total GDP and respective total energy

consumption in natural logarithm. The results demonstrate how much wealth is cre-

ated in response to a one per cent increase (or decrease) in the use of the respective




energy source. The results are quite different to the results of table 4, which is based

on per capita correlation. Based on total GDP and total respective energy consump-

tion (without dividing by population), most OPEC Countries have a positive and sig-

nificant relationship between wealth creation and energy consumption. It is interest-

ing to note that the results for G-7 and Asian countries hardly change, while OPEC

Countries witness stronger results. This might be due to population, which may have

Table 2

Percentage share of respective energy source in industry

Electricity Oil Natural gas Coal

Algeria 16.3 4.8 76.2 2.1

Indonesia 13.2 42.8 26.4 17.6

IR Iran 11.1 37.2 49.2 1.8

Iraq 0.0 40.4 59.6 0.0

Kuwait 0.0 4.2 95.8 0.0

SP Libyan AJ 0.0 43.2 56.8 0.0Nigeria 2.2 7.0 12.1 0.1

Qatar 2.5 30.9 66.6 0.0

Saudi Arabia 7.7 92.3 0.0 0.0

UAE 1.9 13.0 85.1 0.0

Venezuela 15.6 22.8 59.3 0.3

India 13.9 26.5 10.7 26.5

Pakistan 9.4 11.3 49.1 11.5

Sri Lanka 11.0 22.4 0.0 0.1

Australia 28.4 16.5 31.4 14.1

Canada 26.5 24.2 31.2 4.6

France 25.4 33.3 31.0 6.5

Germany 27.0 31.6 28.3 11.1

Italy 29.0 23.7 41.5 5.0

Japan 26.1 48.3 6.8 16.9

UK 24.4 28.8 37.8 6.0

USA 22.6 25.1 34.5 7.3




Table 3

Correlation matrix — per capita GDP/per capita energy consumption

Coal Electricity

Total

primary

energy

supplies

OilNatural

gas

Total

final

energy

cons.

e/E

ratio

Algeria 0 0.28 0.47 0.78 0.23 0.34 0.39

Indonesia 0 0.94 0.98 0.96 0.94 0.97 0.97

IR Iran 0 –0.27 –0.23 0.83 0.88 –0.16 –0.03

Iraq 0 –0.76 –0.63 –0.64 –0.24 –0.61 –0.75

Kuwait 0 –0.31 0.36 –0.29 0.83 0.32 –0.49

SP Libyan AJ 0 –0.93 –0.85 –0.66 –0.65 –0.78 –0.89

Nigeria 0 –0.71 –0.6 –0.4 –0.66 –0.3 –0.69

Qatar 0 –0.72 –0.55 –0.66 –0.47 –0.58 –0.59

Saudi Arabia 0 –0.77 –0.62 –0.23 –0.87 –0.54 –0.82

UAE 0 –0.85 –0.91 –0.36 –0.79 –0.87 –0.46Venezuela 0 0.92 0.68 –0.32 –0.48 –0.58 –0.84

India –0.61 0.98 –0.99 0.99 0.97 0.92 0.95

Pakistan 0.84 0.99 0.99 0.97 0.97 0.86 0.99

Sri Lanka 0 0.99 0.93 0.92 0 0.95 0.99

Australia –0.83 0.97 0.95 0.47 0.91 0.96 0.91

Canada –0.88 0.93 0.86 –0.72 0.89 0.41 0.86

France –0.92 0.99 0.97 –0.63 0.97 0.77 0.96

Germany –0.93 0.87 0.07 –0.29 0.95 –0.11 0.96

Italy –0.65 0.99 0.95 –0.57 0.98 0.95 0.99

Japan –0.55 0.98 0.95 0.68 0.99 0.91 0.56

UK –0.88 0.98 0.58 0.15 0.89 0.77 0.97

USA –0.93 0.98 0.24 –0.54 –0.78 –0.62 0.96




Table 4

Correlation matrix in natural logarithm — per capita GDP/per capita energy

consumption

Coal ElectricityNatural

gas

Petroleum

products

Total

final

energy

cons.

Total

primary

energy

supplies

Algeria 0 0.33 0.44 0.81 0.41 0.51

Indonesia 0 0.99 0.89 0.96 0.97 0.98

IR Iran 0 –0.35 –0.14 –0.22 –0.18 –0.47

Iraq 0 –0.67 –0.20 –0.57 –0.54 –0.57

Kuwait 0 –0.37 0.78 –0.23 0.52 0.37

SP Libyan AJ 0 –0.85 –0.33 –0.72 –0.76 –0.80

Nigeria 0 –0.71 –0.76 –0.43 –0.66 –0.64

Qatar 0 –0.78 –0.47 –0.75 –0.63 –0.60

Saudi Arabia 0 –0.65 –0.91 –0.29 –0.48 –0.57

UAE 0 –0.73 –0.79 –0.35 –0.88 –0.89Venezuela 0 –0.84 –0.46 –0.33 –0.56 –0.67

India –0.56 0.97 0.94 0.98 0.95 –0.99

Pakistan 0.86 0.99 0.97 0.98 0.87 0.99

Sri Lanka 0 0.99 0 0.90 0.92 0.91

Australia –0.88 0.95 0.84 0.42 0.96 0.95

Canada –0.91 0.95 0.91 –0.72 0.44 0.88

France –0.96 0.99 0.956 –0.65 0.75 0.96

Germany –0.87 0.88 0.91 –0.29 –0.05 0.16

Italy –0.63 0.99 0.98 –0.6 0.94 0.94

Japan –0.55 0.989 0.986 0.62 0.88 0.94

UK –0.95 0.979 0.84 0.12 0.73 0.53

USA –0.93 0.985 –0.82 –0.56 –0.65 0.21




Table 5

Correlation matrix in natural logarithm — GDP/energy consumption


gas

Petroleum

products

Total

final

energy

cons.

Total

primary

energy

supplies

Algeria 0 0.96 0.94 0.99 0.98 0.98

Indonesia 0 0.99 0.91 0.31 0.99 0.99

IR Iran 0 0.83 0.88 0.83 0.87 0.72

Iraq 0 –0.52 –0.27 –0.49 –0.47 –0.50

Kuwait 0 0.79 0.30 0.34 0.69 0.38

SP Libyan AJ 0 –0.58 –0.75 –0.52 –0.51 –0.58

Nigeria 0 0.76 0.75 0.66 0.88 0.56

Qatar 0 0.81 0.84 0.83 0.85 0.86

Saudi Arabia 0 0.92 0.77 0.94 0.94 0.93

UAE 0 0.94 0.77 0.94 0.77 0.89

Venezuela 0 0.89 0.91 0.91 0.92 0.90

India 0.72 0.99 0.97 0.99 0.96 0.99

Pakistan 0.93 0.99 0.99 0.99 0.70 0.99

Sri Lanka 0 0.99 0 0.98 0.98 0.97

Australia –0.42 0.98 0.90 0.96 0.99 0.99

Canada –0.87 0.98 0.96 0.15 0.96 0.98

France –0.94 0.99 0.96 –0.29 0.89 0.97

Germany –0.90 0.91 0.91 0.016 0.32 0.48

Italy –0.536 0.99 0.97 –0.32 0.93 0.93

Japan 0.27 0.98 0.98 0.79 0.93 0.95

UK –0.94 0.97 0.90 0.47 0.84 0.74

USA –0.91 0.99 –0.19 0.71 0.79 0.94




distorted the results in our per capita analysis for OPEC Countries in particular,

probably due to the inefficient use of energy or disproportionate distribution of

wealth.

The GDP elasticities in relation to respective energy sources are tabulated intable 6. GDP elasticity measures how much wealth is created in response to a one

per cent increase (or decrease) in the consumption of the respective energy source.

Within OPEC, with the exception of Algeria, Indonesia, IR Iran, Iraq and Venezue-

la, all countries show a weak or insignificant relationship between respective energy

consumption and wealth effects. Asian and G-7 countries in general witness strong

elasticities except for in relation to oil, particularly in G-7 countries. That is, in G-7

countries, oil is the factor that contributes least to wealth creation because the ma-

jority of the oil is imported at a high cost and over 50 per cent is used in the trans-

port sector. In contrast, the contribution of electricity is generally higher, followed

by other types of energy.

Historically, energy was cheaper prior to the oil shocks of 1973 and 1979. The

historical data for G-7 countries is available from 1960. Therefore, the data was di-

vided into two groups, 1960–73 and 1973–2001, to analyse the implications of the oil

price shocks. Table 7 draws out the comparison of the correlation between per capita

GDP and per capita energy consumption for G-7 countries in natural logarithm form

(i.e. how much per capita GDP changes in response to a one per cent increase in per

capita energy consumption). Clearly, there has been a shift in the correlation since

the oil shocks of the 1970s. In general, all countries have adjusted downward, but the

greatest re-adjustment has been associated with per capita petroleum consumption.The higher oil prices forced these countries to use oil more efficiently to produce a

dollar’s worth of output, compared to years when oil was comparatively cheap. This

was possible by using state-of-the-art-technology that improved over time. Howev-

er, there has been a marginal shift in per capita electricity consumption in relation to

per capita GDP. That is, the main driving force behind rapid economic development

of G-7 countries is the greater and more efficient use of electricity and other types of

energy. However, one should not ignore the importance of total energy usage, rather

than focusing on a per capita basis, especially when drawing conclusions for OPEC

or other developing countries.

4. ConclusionsThis paper analysed the relationship between the use of different types of en-

ergy and economic development, both on a per capita and total usage basis, with and

without natural logarithm, for OPEC, the G-7 and three Asian developing countries.

Generally, there is a strong and positive relationship between energy consumption and

economic development for the G-7 and the three Asian countries. However, a compar-

atively weaker association is witnessed for petroleum consumption in G-7 countries.

In contrast, apart from Algeria, Indonesia, IR Iran, Kuwait and Venezuela, all OPEC

Countries witness a weak and negative correlation between per capita GDP and per

capita energy consumption. This relationship, however, becomes positive and strong-

er for almost all OPEC Countries when analyses are based on total GDP and total




Table 6

GDP elasticities in relation to respective energy consumption


gas

Petroleum

products

Total

final

energy

cons.

Total

primary

energy

supplies

Algeria 0 0.29 0.01* 0.52 0.37 0.39

Indonesia 0 0.41 0.036* 0.33 0.97 0.84

Iran 0 1.49 0.17 0.75 0.86 0.24

IR Iraq 0 1.43 0.51 0.91 1.21 1.3

Kuwait 0 (0.07)* (0.01)* –0.13 (0.04)* (0.05)*

SP Libyan AJ 0 0.10* 0.13 (0.18)* 0.07* (0.06)*

Nigeria 0 0.016* (0.02)* 0.02* (0.009)* 0.012*

Qatar 0 (0.15)* (0.02)* 0.37 (0.02)* (0.03)*

Saudi Arabia 0 0.04* 0.001* 0.14 0.15 0.15

UAE 0 0.19 0.05* –0.4 0.054* 0.22

Venezuela 0 0.23 0.042 0.62 0.56 (0.077)*

India 0.35* 0.77 0.02 0.83 0.035* 1.39

Pakistan 0.03 0.63 0.81 1.0 2.95 1.1

Sri Lanka 0 (0.002)* 0 (0.006)* 0.006 0.77

Australia –0.02 0.79 1.35 0.019 1.35 0.1

Canada 0.07 0.51 0.41 0.1 1.68 0.34

France 0.015 0.32 0.22 0.04 0.19 0.15

Germany 0.002 0.13 0.11 0.042 0.11 0.18

Italy 0.02 0.66 0.34 0.1 0.39 0.23

Japan 0.059 0.47 0.46 0.049 0.36 0.23

UK 0.54 0.57 0.43 0.015 0.39 0.38

USA 0.058 1.107 0.52 0.17 0.59 0.46

Note: Figures in parentheses reflect the correlation matrix for 1960–73, while the others reflect the 1973–

2001 period.




consumption of the respective energy source. One should, therefore, be cautious when

drawing conclusions based on per capita figures for developing countries.The sample period is also an important factor in analysis. For example, when

analysis is carried out for G-7 countries, we see a significant downward adjustment

of energy consumption after the oil price shocks of the 1970s and the events that fol-

lowed (1973–2001), compared to periods when energy was cheaper (1960–73).

The general conclusion of this paper is more or less in line with other earlier

studies, especially in relation to electricity consumption. However, this study further

extended the scope and analysed the relationships with other forms of energy as well

to find the missing link. Developed and emerging economies of Asia are all heavily

reliant on all forms of energy and utilize these resources efficiently in their rapid eco-

nomic development. In contrast, most OPEC Countries generally failed to use the giv-

en abundant resources efficiently to accelerate the pace of economic development.

Table 7

Correlation matrix: comparison of correlation between per capita GDP

and per capita energy consumption during

1960–73 and 1973–2001 (natural logarithm)

Coal

cons.

Electricity

cons.

Natural

gas

cons.

Petroleum

products

cons.

Total

final

energy

cons.

Total

primary

energy

supplies

Australia –0.86

(–0.94)

0.96

(0.99)

0.88

(0.76)

0.36

(0.97)

0.96

(0.99)

0.95

(0.99)

Canada –0.87

(–0.97)

0.92

(0.99)

0.86

(0.99)

–0.72

(0.99)

0.11

(0.988)

0.80

(0.99)

France –0.94

(–0.96)

0.98

(0.99)

0.95

(0.99)

–0.63

(0.99)

0.70

(0.99)

0.95

(0.99)

Germany –0.87

(–0.29)

0.86

(0.98)

0.94

(0.96)

–0.37

(0.98)

–0.39

(0.97)

0.16

(0.98)

Italy –0.58

(–0.90)

0.99

(0.99)

0.97

(0.95)

–0.54

(0.98)

0.91

(0.99)

0.91

(0.99)

Japan –0.42

(–0.86)

0.92

(0.99)

0.98

(0.98)

0.58

(0.98)

0.86

(0.99)

0.92

(0.99)

UK –0.94(–0.97) 0.97(0.98) 0.89(0.96) 0.29(0.96) 0.76(0.97) 0.63(0.98)

USA –0.91

(–0.64)

0.98

(0.98)

–0.75

(0.97)

–0.54

(0.98)

–0.58

(0.98)

0.17

(0.98)

Note: Figures in parentheses reflect the correlation matrix for 1960–73, while the others reflect the 1973–

2001 period




Footnotes

1. Unless otherwise stated, we have used annual data series for each country for the period

1971–2001.

2. It appears there might be some error in the reporting of data in countries where electric-

ity usage is zero or marginal. However, our analysis is based on reported data.

References

Ferguson, R., et al (1997), “Benefits of electricity generation”, IEE Engineering Scienceand Education Journal 6(6), 225–59.

Ferguson, Ross, Wilkinson, William, and Robert Hill (2000), “Electricity use and eco-

nomic development”, Energy Policy 28, 923–34, Elsevier.

International Energy Agency (2002), Energy Statistics and Balances for OECD Coun-

tries 1960–2001 and Non-OECD Countries 1971–2001, Paris.

Correlation Between Energy

Documents

Transcript of Correlation Between Energy