Energy and Growth: OECD and the Rest of the World

INTRODUCTION

This article investigates the strength of the relationship between energy consumption and economic growth over time. The import of this relationship is realized against the backdrop of discussions on the decoupling of economic growth and energy use with implications for energy and climate policy as mentioned here by the International Energy Agency (IEA) , mentioned here by 麦肯锡 , here by the International Institute for Sustainable Development (IISD). By analyzing 世界银行 data between 1990 and 2021 from 28 ORGANIZATION FOR ECONOMIC COOPERATION AND DEVELOPMENT (OECD) countries, 47 African countries, 5 GCC countries, and the 4 BRIC nations, I aim to examine how the strength of the relationship has changed over time between the OECD and non-OECD.

The importance of investigating the “energy use – economic growth” relationship is now heightened by the inclusion of the environmental dimension to the relationship. This comes even as CO2 emissions are closely related to the quantity of energy used and the country–level energy mix.

ENERGY AND GROWTH: The Connection

The relationship between energy consumption and economic growth can be established by a log–log regression model of a selection of countries’ per capita energy use and per capita GDP, in a reference year. A regression line through the data points accompanied by the r-square measuring statistical fit establishes the relationship between energy use and countries' economic growth, as seen here by Center for Strategic and International Studies (CSIS) . However, to understand the changing nature of this relationship over time, the r-square is computed for a log–log regression model and plotted over time for different world regions.

The regional groupings examined are member countries of the OECD, countries on the African continent, the Gulf Cooperative Council (GCC), and the BRIC (Brazil, Russia, India, and China). South Africa, a member of the BRICS coalition is excluded from that block and included in Africa. These groupings represent a total of 106 countries, of which 84 countries in our database have the complete matching time series data for the analysis between 1990 and 2021. Additionally, the nation–groups are at different stages of economic development – the OECD countries are considered to be the advanced economies of the world, African nations with huge, unmet energy needs are lagging in economic growth, while the GCC countries are huge oil-producing countries which have grown their economies significantly from oil wealth, and the BRIC are the group of countries with rapidly developing economies expected to dominate world economy by 2050.

The workflow for the analysis in this article is illustrated below

The steps (A) – (D) will produce four plots for the OECD, Africa, the GCC and the BRIC nations. ?I do not take cognizance of the changing composition of the blocks between 1990 and 2021 – for example, there were 20 countries at the formation of the OECD in 1961, which had increased to 24 by 1990 and further increased to 38 by 2021. Also, the BRIC alliance did not exist before 2006 and South Africa only joined the alliance in 2010.

The OECD - 28 Countries

Ten OECD countries were omitted from consideration due to incomplete data. These ten countries are Canada, Costa Rica, Estonia, Hungary, Iceland, Israel, Latvia, Lithuania, Slovakia, and Slovenia. Table 1 below captures the top ten countries of the OECD-28 by the cumulative average growth rate (CAGR) of the GDP per capita between 1990 and 2020. The table also captures the corresponding energy use per capita of those top 10 countries over that same period.

Ireland has the leading GDP per capita growth rate at 6.79%, with a corresponding per capita energy use growth rate of 0.13%. Colombia comes in at 10th place with a GDP per capita growth rate of 3.95% and a corresponding 0.96% growth rate in per capita energy use. While the minimum growth rate is ~ 4% (attributable to Colombia) for the top 10 countries by GDP per capita CAGR, the per capita energy use in this top 10 does not show a similar growth rate. In fact, a negative growth rate in per capita energy use is noted for countries such as Luxembourg (-1.26%), Norway (-0.34%), Czechoslovakia (-0.58%) and Denmark (-0.67%).

Table 2 compares the CAGR rankings of the GDP per capita, and energy use per capita

Between 1990 and 2020, the countries in the OECD with the most growth in per capita GDP do not necessarily rank similarly in per capita energy use. Denmark for example, ranks 9th in CAGR of GDP per capita (4.17%) but ranks 23rd in per capita energy use. While Luxembourg ranks 5th in GDP per capita growth (4.93%), it ranks at 28th – the bottom – in terms of per capita energy use. There are however, four countries in Table 2 that rank in the top 10 in terms of per capita energy use – South Korea (#1 with 3.36%), Turkey (#2 with 2.52%), Chile (# 3 with 2.31%), and Colombia (#4 with 0.96%).

For 28 out of the 38 OECD countries analyzed, Figure 1 plots is the plot of the r-square metric. In 1990, the r-square was at 70% and rose to 82% in 2001, where it remained at this plateau till 2005, when it began to decline. The decline in the r-square is sustained for 2-decades till it reaches the value of 58% in 2021. However, the r-square declined 7.60 percentage points over an 8-year period from 2005 to 2013, compared to a steeper decline of 13 percentage points over three years from 2013 to 2016.

As of 2021, the coupling between energy use per capita and GDP per capita for the OECD-28 had weakened from its peak in 2005.

The import of this r-square profile is that the member states of the OECD as a group have, in the last 2-decades, progressively dissociated energy use from economic growth. Furthermore, this group has improved energy efficiency, increasing its per capita GDP with lower per capita energy use, with some member states recording negative energy use for increased GDP per capita growth. This much was noted in the rankings and trends shown in Tables 1 and 2.

The BRIC Countries

The 4-countries of the BRIC group have experienced significant economic growth and account for 37% of global energy consumption. As Table 3 shows, Brazil has the lowest CAGR in GDP per capita in the group at 2.90%, while its energy-use per capita experienced a CAGR of 1.54%. China leads the group in the growth of its GDP per capita – in 1990, China’s GDP per capita was $981/person, and it rose to $17,189/person three decades later. This growth at a CAGR of 10% has a corresponding energy use per capita CAGR of 4.94%. India’s GDP per capita also grew by ~6% CAGR between 1990 and 2020, while its per capita energy use grew by 3.19% CAGR. Russia is the only country of the 4 BRIC group which GDP per capita grew by 4.66% CAGR, while per capita energy use shrank 0.41% CAGR – from ~68,000kWh/person (1990) to ~55,000kWh/person (2020).

China and India of the BRIC block share the same rankings between energy use per capita and GDP per capita. However, while Russia ranks #3 in GDP per capita CAGR, its negative CAGR of energy use per capita ranks it 4th in the BRIC group of nations.? Brazil which ranked 4th in the GDP per capita CAGR, ranks 3rd in the energy use per capita CAGR. This is captured in Table 4.

For the BRIC nations, the plot of the r-square metric shows the metric first declined from its 1990 level of 61% to 48% in 1998. However, from 1998, this metric rose from a trough of 48% to 97% in 2021.

The trend indicates two epochs. Before 1998, a de-coupling of energy use per capita and GDP per capita was witnessed. However, post–1998, the energy use per capita and GDP per capita show an increasingly strong relationship.? By 2008, the r-square was 82%; post–2008, however, the rate at which the r-square increased had slowed as judged by the gentle slope of the curve in the 15-years since 2008, relative to the 10-years before 2008. While the relationship between energy use and GDP is still held strongly, the relative increase in the strength of that relationship is in decline.

The Gulf Cooperative Council (GCC) Countries

The countries of the GCC considered here are Bahrain, Kuwait, Oman, Saudi Arabia and the United Arab Emirates. Qatar is omitted due to the absence of a complete data time series to include it in the analysis to evaluate the r-square metric. As a block, these countries supplied 17.8 MMbbls/day of oil in 2021, which was 20% of global supply. Known for their significant energy exports, these countries’ GDP and energy-use per capita growth trends are explored in Table 5

Bahrain ranks number 1 in GDP per capita growth with a CAGR of 2.03%. However, it ranks 4th out 5 in energy-use per capita growth (0.84%). This is indicative that Bahrain’s GDP per capita grew at odds with its per capita energy use. Oman, however, ranked 4th out of 5 in terms of CAGR for GDP per capita (0.72%), but ranked number 1 in CAGR energy use per capita (3.62%). This is suggestive that there is scope for Oman to improve its economic growth within its level of energy use.

The other countries’ GDP per capita growth and energy use per capita rankings- Kuwait, Saudi Arabia, and the UAE- were the same at 2, 3, and 5, respectively. The countries’ dissimilar rankings between GDP per capita and energy-use per capita reflect their different developmental stages and the structural relationship between the two variables.

The trend reveals that the relationship between energy use per capita and GDP per capita for the 5-group GCC has tended to wax and wane through time. The GCC-5's r-square metric was volatile from 1992 to 2004, though generally trending downward from 48% in 1992 to 44% in 2004. However, in the 5-years between 2004 and 2009, the r-square steeply declined 34 percentage points from 44% (2004) to 10% (2009). The r-square recovered from its 2009 low of 10% to 66% in 2017. As of 2021, however, the coupling between energy use and the GDP of the GCC weakened as indicated by the r-square of ~50% relative to its 2017 position.

The African Countries

Forty-seven countries of the African continent are captured. The countries omitted are Djibouti, Eritrea, Liberia, Libya, Sao Tome and Principe, Somalia, and South Sudan, which comprise about 2% of Africa’s $3.14 trillion GDP. ?Africa is an important continent in the energy use–economic growth conversation, especially due to its constrained economic growth, limited energy use and minuscule carbon emissions.

Table 7 captures the top 10 countries by the CAGR of their GDP per capita between 1990 and 2021.

Equatorial Guinea ranks topmost in the CAGR of the GDP per capita between 1990 and 2020 at 12.28%, while its CAGR for energy use per capita is at 9.39%. The 10th-ranked country by CAGR of GDP per capita is Lesotho, with a rate of 4.35%, while its energy-use per capita CAGR is at 5.30%. As of 2020, Mauritius, with a GDP per capita of $24,681/person, had the highest GDP per capita of the top 10 African countries by GDP per capita growth rate. Correspondingly, its CAGR of energy use per capita was 4.01% - having grown from 7,287kWh/person (1990) to 22,400kWh/person (2020).

Notably, the big economies of the African continent – Egypt, Nigeria, and South Africa – are not in the top ten. Their CAGR of GDP per capita are 4.09%, 3.31%, and 2.78%, respectively, ranking them 13th, 24th, and 31st respectively. The CAGR of the energy use per capita of these big economies of the African continent are 1.29%, 0.46%, and -0.02%, respectively.

Table 8 compares the top 10 countries by the CAGR rankings of the GDP per capita, and the CAGR rankings of the energy use per capita.

A key revelation is that not all countries ranked in the top 10 by CAGR of GDP per capita are in the top 10 by CAGR of energy use per capita. Carbo Verde presents the most divergent case – ranked 2nd by CAGR of GDP per capita growth (7.06%) and 40th by CAGR of energy-use per capita growth (-0.31%). Carbo Verde's energy use per capita shrank from 1,993kWh/person (1990) to 1,821kWh/person (2020). Rwanda is another case study. It ranked 6th in terms of CAGR of GDP per capita but 37th in terms of CAGR by energy use per capita (0.07%). Ghana is the third country with a GDP per capita growth rate ranked 8th; however, it ranks 20th in the growth rate of energy use per capita at 1.79%. This highlights the structural differences in these economies. Specifically, it suggests that Carbo Verde, Rwanda, and Ghana have managed increasing energy efficiencies - growing their GDP per capita with lower per capita energy use.

Turning to Figure 4, the r-square declined from 76% in 1990 to 68% in 2010, but there is still a strong relationship. The r-square has increased from the 2010 level to 73% in 2019. Throughout the period under consideration, the r-square has ranged between 67% and 78%.

For the cohort of African countries, the coupling of energy use per capita and economic growth is evident and has remained consistent over the 30+ years between 1990 and 2021. This is even if country-specific examples exist where economic growth is not commensurate with energy use. The next section compares the r-square trends for the different country-groupings.

PUTTING IT TOGETHER

Comparing the trend of r-square for the various country groupings, the group of 47 African countries had the tightest coupling between energy use per capita and GDP per capita at 76% (the GCC-5 data commenced in 1992). However, by 2019, the BRIC countries exhibit the tightest coupling between the two variables as indicated by the r-square of 96%. Over the 3-decades from 1990’s, the GCC-5 exhibited the lowest r-square of all the country groupings, such that by 2019, with an r-square of 49%, which is the highest for the GCC, it is still the least of all the country groups.

Based on the r-square, the coupling between the OECD's energy use per capita and per capita GDP has weakened in the last 20 years. For every other economic block, however, energy use per capita has shown an increasing coupling to per capita GDP, albeit at varying intensities, since 2008 – the year of the financial crises. The weakening relationship between energy use and GDP for the OECD has been explained by the transition towards more service-oriented economies that require less energy and improvements in energy efficiency.

KEY TAKEAWAYS

1. Energy plays a significant role in economic growth, which has been amply demonstrated in several other articles. Attention has been drawn to the decoupling of energy use per capita and GDP per capita, as found here in Our World in Data .
2. However, this is only the case for the OECD countries. The relationship has been strengthening for the other regional economic blocks, especially since 2008 – with the BRIC countries showing the strongest coupling between per capita energy use and GDP per capita.
3. Since 2008, while the relationship between energy use and GDP growth has become stronger for BRIC, its pace of increase has slowed as though approaching a “peak.” If this block follows the trajectory of the OECD, a decline in the strength of the relationship should begin to occur. The choices of this group will determine how long the relationship is at a peak and what the decoupling profile will look like.
4. Even with the least energy-economy coupling amongst the blocks, the per capita carbon emissions of the GCC are at par with that of the OECD. A concern for the GCC may be improving per capita GDP (economic growth) with the least energy use per capita while trying to drive down emissions per capita.
5. Africa’s energy-economy coupling has been second after the BRIC group since 2014. However, its per capita energy use and GDP per capita are comparatively low. For Africa, the coupling between energy use per capita and GDP per capita is important and must be enhanced; however, the effort required is to seek a path that increases and translates energy use into GDP gains.