The present paper aims at answering the following question: why did the U.S. refuse to ratify the. Kyoto Protocol? To this end, we proceed firstly with a ...
The USA non-ratification of the Kyoto Protocol An empirical validation through the Kuznets curve 1 2 Dorsaf JAMALI and Jaleleddine BEN REJEB
Abstract
The present paper aims at answering the following question: why did the U.S. refuse to ratify the Kyoto Protocol? To this end, we proceed firstly with a descriptive analysis for global conventional energy consumption and secondly we check whether or not the Environmental Kuznets Curve (EKC) exists in the USA case. Empirical analysis shows that USA is well aware that its industries are unable to fulfill their commitments to reduce emissions of carbon dioxide and other greenhouse gas dioxide as this hinders their profitability.
Key Words : CO2 emissions, GDP, ECK JEL Codes : 013, R11
1
PhD student (FSEG, Tunis) Professor in Quantitative Methods (ISG, Sousse). Head of Laboratory of Management of Innovation and Sustainable Development. 2
1
I – Introduction Since the seventies, there has been a certain awareness of the abuses and the negative effects of industrialization on the environment. As a matter of fact and since the circulation of the Meadows report, there has been a general agreement that the economic growth and the environment are divergent. These two concepts are contradictory and the emergence of a new concept to reconcile between them seems to be urgent. Thus appears the concept of sustainable development which entails that governments are committed to give substantial attention to the environment especially after the conduct of scientific studies which warned that we are now facing a global threat due to climate warming. All the states responsible of large emissions of greenhouse gas have agreed to ratify the Kyoto protocol and to reduce gas emissions to achieve the objectives of this protocol, except the US. The latter which was the first largest emitter of CO2, refused to sign the memorandum arguing that the protocol targets ward off its economic growth. The relationship between the economic growth and the environment has been a lively debatable topic in the economic literature. In this respect, many empirical studies attempted to highlight the environmental Kuznets curve in order to disprove the existence of such relationship. In effect, the EKC is often used to defend the idea that economic growth causes no danger to the environment. The decreasing effect of the EKC reflects the degree of compatibility between growth and environment. In addition, the existence of an EKC has important implications in terms of economic policies. The present paper aims at answering the following question: why did the U.S. refuse to ratify the Kyoto Protocol? To this end, we proceed firstly with a descriptive analysis for global conventional energy consumption and secondly we will check whether or not the EKC exists in the USA case.
II - Global warming: Global threat: According to the report of the Convention- Framework of the United Nations on the Climate Changes, the average temperature on Earth has increased by 0.6 degrees Celsius since the late 1800s, as one would expect it to continue to increase to 5.8 degrees Celsius by 2100, causing a rapid and deep climate change. The main reasons behind this temperature rise are mainly a century and a half of industrialization with the combustion of increasingly high quantities of oil, gas and coal, deforestation as well as certain inappropriate agricultural practices.
2
In the recent decades, alerts about the state of the planet have been raised non-stop. As a matter of fact, the climate warms, the freshwater keeps decreasing, thousands of living species are threatened with extinction, and the forests were suffering from deforestation which was accompanied by agricultural intensification holding many chemical entrants. This image becomes more and more common in developing countries, especially under the pressure of population growth. These activities, on the other hand, have increased the amount of "greenhouse gases" in the atmosphere, particularly the carbon dioxide, methane and nitrous oxide. All these gases are essential for life as they keep some of the sun's heat back into space without which the earth would be a cold and barren place. But in ever increasing quantities, these gases are pushing the global temperature to artificially high peaks altering the climate. The 1990s are believed to have been the warmest of the last millennium with 1998 being the hottest year. Additionally, the current warming trend is expected to cause extinctions. Many species of plants and animals, already weakened by pollution and abode loss, are likely to disappear in the next 100 years. Although they are not threatened to this extent, human beings are likely to face increasingly large difficulties. Recent storms, floods and droughts, for example, tend to show that computer models predict frequent "extreme weather events". The sea level has risen up to 20 cm during the 20th century and a further rise of 9-88 cm is expected by the year 2100. Higher temperatures cause the expansion of the oceans volumes and the melting of ice and the ice caps add more water. If the peak of the scale is reached, the sea could overflow in densely populated coastal areas of countries such as Bangladesh, causing the disappearance of entire nations (as the island state of the Maldives), polluting fresh water of billions of people and pushing to mass migration. One decade ago, many countries signed an international treaty known as the Convention- Framework of the United Nations on Climate Change, having as an aim to consider what can be done to reduce global warming and deal with any inevitable temperature increases. In 1997, governments agreed to make an addition to the treaty, called the Kyoto Protocol, which contained stricter measures (legally binding). The Protocol was applied on 16 February 2005. And since 1988, an Intergovernmental Panel on Climate Change has reviewed the scientific researches and provided governments with summaries and advice on climate issues.
III - The Kyoto Protocol: A mechanism to prevent the threat When they adopted the Convention, most of the governments knew that their commitments are not sufficient to seriously address the climate change. In Conference 1 held in Berlin (March / April 1995), known as the Berlin Mandate decision, the different members began a round of negotiations in order to 3
come up with stricter decisions and more detailed commitments for industrialized countries. After two and a half years of intense negotiations, the Kyoto Protocol was adopted at Conference 3 in Kyoto, Japan, December 11, 1997. However, the complexity of the negotiations left a considerable number of issues unresolved, even after the adoption of the Kyoto Protocol. The latter defined the mechanisms main points of compliance with commitments, but did not substantiate such important rules to make them operational. Although 84 countries signed the Protocol thus showing their intention to ratify it, many hesitated to put the step into action before a clear overview of the treaty is settled. It is for this very reason that a new round of negotiations was initiated to outline the rules of the Kyoto Protocol, which was conducted in parallel with negotiations about the ongoing issues within the Convention framework. This round of negotiations finally found its culmination in conference 7 with the adoption of the Marrakesh Agreements, which established the detailed implementation of the Kyoto Protocol rules and took important decisions about the implementation of the Convention. The Kyoto Protocol, on the other hand, has the same goal as that of the Convention, as well as its principles and institutions, but significantly strengthens the Convention by committing Annex I parties to individual objectives that are legally binding to reduce greenhouse gas emissions. The ultimate goal is to reach a total reduction of greenhouse gas emissions by at least 5% in the commitment period of 2008-2012 compared to that of 1990.
4
Despite the commitment of the majority of countries to achieve the objective of the Convention, the United States has increased its emissions of greenhouse gas by 1% per year since 1990. Indeed, in 1997, the U.S. Senate refused to ratify the treaty. It is to be noted as well that no Democratic senator has ever voted in favor of the protocol. In 2000, Bill Clinton tried to negotiate amendments with the Europeans but with no success. In July 2005, the government of George W. Bush refused to resubmit the treaty for ratification because it would hamper the U.S. economy. They argued that the fight against climate change should be done not with a simple reduction of greenhouse gas emissions, but through better management of their issuance. According to the New York Times, The United States was classified as the largest emitter with around 23% of greenhouse gas emissions on the planet as it produces 25% of goods and services in the world. However, and according to researchers from the Universities of Berkeley and San Diego, China has surpassed the U.S. in CO2 emissions since 2006. Politically speaking, the government of George W. Bush explained its rejection of the protocol by the fact that the Republic of China, the second largest emitter of greenhouse gas, has no binding reduction target under the Protocol. It also justified its non-adherence by the fact that their industry is more energy efficient than the majority of the signatories. On the other hand, the U.S. government signed the "Climate Convention" of the United Nations, which came into force in 1994. As a member in this Convention, the United States took part in the conferences attended by the different parties involved concerning the climate and the Kyoto Protocol. It preferred to invest in new technologies and in clean energy and refused any binding multilateral agreement given that no obligation was imposed on the developing countries (especially China and India, the two large emitters of greenhouse gas).
IV – The causes of the US non- ratification of the Kyoto protocol The U.S. attributed the non- ratification of the protocol to its interference with its economic growth, but being aware of the global warming phenomenon that threatens the whole planet, it opted to invest in clean technologies in a try to achieve the objectives of the Kyoto Protocol. The newly adopted strategy was meant to replace the protocol alternative. In this respect, a big question may arise: why is this intense opposition to ratification despite the government considerable effort to achieve the desired goal of the Kyoto protocol? In order to answer this question, we will proceed in two steps. Firstly, we will analyze, in a descriptive way, the consumption of the global conventional energy as well as that of clean energy. Secondly, we will examine the validation of the Environmental Kuznets Curve (inverted U-shaped) for the U.S, showing the relationship between GDP per capita and pollution. 5
According to what was declared by the U.S. government regarding its energy policy and its intention to invest in the technology of clean energy, we will try to examine and check if this large emitter of greenhouse gas pollutants is actually about to implement policies that go hand in hand with the objectives of sustainable development so as to maintain the welfare of present and future generations. The graph below shows the evolution of the overall energy consumption (measured in kg of oil equivalent) and clean energy over the period from 1990 to 2005 for five countries, namely the USA, China, India, France and Russia.
2,5E+12 2E+12 usa
1,5E+12
rus china
1E+12
ind 5E+11
fra
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
0
The comparative analysis clearly shows that the U.S. government is the largest user of energy followed by China which has experienced an exponential increase in energy since 2002. This increasing use of energy, regardless of the country, is the basis of greenhouse gas emissions, which cause global warming and threaten the survival of all kinds of living species. The following graph, equivalent to the previous, highlights the trend of the evolution of energy over time.
6
2,5E+12 2E+12 usa
1,5E+12
rus china
1E+12
ind 5E+11
fra
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
Moreover, the comparative analysis of the consumption of clean energy for the five countries shows the considerable effort made by France and Russia in developing this form of energy, contrary to the U.S. that clearly did not respect its commitments. 50
45 40 35 30
usa
25
rus
20
china
15
ind
10
fra
5
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
0
As a matter of fact, not only the weight of clean energy in total energy use is low for the U.S, but also has not been strengthened over the years, unlike France which increased the weight of clean energy in the total energy consumed. Currently, the weight of clean energy is 45% for France while it is just 10% for the USA.
7
2,5E+12 2E+12 1,5E+12 total energy use 1E+12
clean energy
5E+11 0 1990 1992 1994 1996 1998 2000 2002 2004
With reference to the data about energetic consumption (total and clean) in the U.S, provided by the World Bank in 2008, it can be said that there was an indifference from the part of the American government towards environmental problems. While there was an increase in the energy use at a scale of 1012, the tendency in the consumption of clean energy is stable or almost bearish as the graph shows a scale of 1011. This indicates that little efforts are made to address the problems of climate change. According to these results, the U.S. releases more than19,73 tons of CO2 per capita, then comes Canada with emissions of 14.5 tons of CO2 per capita (2005, Energy International Agency (EIA)). Even though the U.S. is classified as the biggest polluter in terms of the release of gas of greenhouse effect, it has not ratified the Kyoto Protocol (reduction of greenhouse gas emissions) and with the "race" in economic industrialization and the increase in the energy use nothing seems to reverse this trend. In fact, on average, an American car releases two times more CO2 than any other car in another country. Worried about the economic impact, the Federal Government and Congress still refuse to act. Nevertheless, several American states are beginning to implement certain programs to reduce the emission of greenhouse gas4. Several empirical studies have suggested an inverted-U relationship between the economic activity usually measured by GDP per capita, and the quality of the environment. At the first stage of the economic growth, the environment deteriorates and the personal income increases, but this degradation starts to decrease when the rise of this income crosses a turning point. This environment trend is called the environmental Kuznets curve (EKC). It is named this way because of its similarity with the relationship between the inequality level and the per capita income considered by Kuznets. According to the latter’s
3 4
See appendix 2 http://www.essperans.fr/blog/wp-content/themes/default/Mini-Kyoto/Mini-Kyoto.htm
8
hypothesis, the economic growth will be the solution for environmental problems in the future. However, the reached results and the empirical findings are ambiguous. On the one hand, several researchers confirmed the existence of an environmental Kuznets curve for different measures of environmental degradation (Panayotou (1993), Selden and Song (1994), Giles and Mosk (2003), and Brajer et al. (2007)). Other authors, on the other hand, argued that there is no evidence to support the EKC hypothesis, and asserted that it is rather an increasing or decreasing monotonic relationship between pollution and the per capita income (( Holtz- Eakin and Selden (1995), Torras and Boyce (1998) and 1999 Hettige al. (1999), de Bruyn et al. ( 1998) and Cole and Elliott ( 2003)). Most of these empirical studies focused on the use of cross- country panel data to estimate the relationship between per capita income and the various environmental indicators. However, the transition from an a background study to a study in a particular country, based on a time series is a new trend for EKC researchers as they can eliminate the problems associated with cross-country data. Lindmark (2002) asserted that the historical studies of each country offers an advantage over the cross-section approaches to get the analyses closer to the dynamics raised by the EKC model. Therefore, any potential inference from these cross-country studies can just provide a general understanding of the variables relationships and thus provides little guideline on policies formulation. Since the U.S. government opted for national solutions to fight against the phenomenon of global warming and preferred to invest in clean energy in order to limit greenhouse gas emissions, it seems to be an obligation to check the EKC for the USA by highlighting the turning point. This shows at which stage one becomes aware of the severity of the phenomenon and at which level of GDP one starts to lower pollution.
The environmental curve of Kuznets
9
V – Data In our empirical analysis we used as an environmental indicator the CO2/ capita measured in tons and as an economic indicator the GDP / capita (in a constant price 2000). The choice of this environmental indicator is driven by two main reasons: the first is the availability of data for a long- term time series, and the second is that the CO2 alone accounts for 80% of greenhouse gas emissions for the U.S5. The data used in this study covered the period 1960-2007. Both indicators namely the CO2/capita and GDP / capita were excerpted from the World Bank database in 2010.
VI - Methodological approach and empirical results: The present work focuses on examining the way how the variables are long –term related and highlighting the dynamic causal relationship between these variables. VI.1 – The Co- integration test between CO2 emissions and GDP: According to Engle and Granger (1987, Econometrica), the test proceeds in two steps: Step 1: Testing the integration order between the two variables: For the two series to be co-integrated, they must be integrated of the same order. It becomes important, therefore, to determine the type of trend (deterministic or stochastic) of each variable, then the order of integration, using the Augmented Dickey- Fuller tests. The stationarity test of Augmented Dickey-Fuller: This test is sequential. We begin by estimating a model with constant and trend. The significance of the trend is tested. In case it is not significant, it is removed and the test model is redone with constant. The latter does not appear in the final version of the model if it is not significant. The following table presents the results of the unit root test: Dickey-Fuller unit root test
GDP per capita Emission of CO2 per capita
Level variable First difference variable Critical value at 1% threshold 7.565 -3.312 -2.626 0.461 -6.097 -2.626
10
It is clearly displayed that the two variables are not stationary in terms of levels, while they are in the first differences. Therefore, it can be said that the two variables of order 1 are integrated. Thus, it becomes possible now to test the co- integration of order 1 between these two variables. Step 2: The estimation of long- term relation The variable "CO2 emissions per capita" is adjusted depending on the GDP per capita by OLS (ordinary least square method). The results of this estimation are provided below: Source | SS df MS -------------+-----------------------------Model | 6.05218023 1 6.05218023 Residual | 108.926986 46 2.36797797 -------------+-----------------------------Total | 114.979167 47 2.44636525
Number of obs F( 1, 46) Prob > F R-squared Adj R-squared Root MSE
= = = = = =
48 2.56 0.1167 0.0526 0.0320 1.5388
-----------------------------------------------------------------------------c02pc | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------gdppc | .049112 .00307 10.60 0.000 -.0127241 .1109481 _cons | 18.4084 .8052632 22.86 0.000 16.78749 20.02931 ------------------------------------------------------------------------------
There is a positive correlation (coefficient equals 0.049) that is highly significant (significance threshold oriented towards zero) between CO2 emissions per capita and GDP per capita. Stationary test of residues Residues were picked up from the previous regression and assayed for stationarity. Calculated variable -2.154 Theoretical variable at a threshold of 5% -1.950 Theoretical variable at a threshold of 1% -2.625
The value of the statistics found is -2154 which is significant at the threshold of 5%, but not at that of 1%. Accordingly, the residues are stationary at the 5% threshold. We can then estimate the error correction model. VI.2 – The estimation of the Error Correction Model (ECM): In this respect, we recall that the CO2 and GDP are not stationary but co-integrated of order 1. Thus, we can adjust the emission of CO2 in relation to the GDP using an error correction model. The estimation results of the variation in CO2 emissions per capita depending on that in the GDP per capita and the residues postponed for a period are displayed below: 11
Source | SS df MS -------------+-----------------------------Model | 3.66668904 2 1.83334452 Residual | 17.1418216 44 .389586855 -------------+-----------------------------Total | 20.8085106 46 .452358927
Number of obs F( 2, 44) Prob > F R-squared Adj R-squared Root MSE
= = = = = =
47 4.71 0.0141 0.1762 0.1388 .62417
-----------------------------------------------------------------------------dco2 | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------dgdppc | .410173 .1949669 2.10 0.041 .017243 .8031031 residu(-1) | -.118012 .0608056 -1.94 0.059 -.2405577 .0045337 _cons | -.1476568 .1369628 -1.08 0.287 -.4236872 .1283736 ------------------------------------------------------------------------------
The coefficient associated with the delayed residues for a period (co-integration towards equilibrium) is negative and statistically significant at 5.9%. Representation of error correction is validated at a threshold of 5.6%.
VII – Adjustment of the emission of the CO2 per capita (CO2C) depending on the GDP per capita (GDP2) In order to check the existence of the EKC, the following three models were estimated: Simple linear form: CO 2 H t 0 1 PIBH t t
(1)
Quadratic form: CO 2 H t 0 1 PIBH t 2 PIBH t2 t
(2)
Cubic form:
CO 2 H t 0 1 PIBH t 2 PIBH t2 PIBH t3 t
(3)
The estimation results are displayed below: Simple linear regression: Source | SS df MS -------------+-----------------------------Model | 6.05218023 1 6.05218023 Residual | 108.926986 46 2.36797797 -------------+-----------------------------Total | 114.979167 47 2.44636525
Number of obs F( 1, 46) Prob > F R-squared Adj R-squared Root MSE
= = = = = =
48 2.56 0.1167 0.0526 0.0320 1.5388
-----------------------------------------------------------------------------c02pc | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------gdppc | .049112 .03072 1.60 0.117 -.0127241 .1109481 _cons | 18.4084 .8052632 22.86 0.000 16.78749 20.02931 ------------------------------------------------------------------------------
Quadratic regression: Source | SS df MS -------------+-----------------------------Model | 32.5445847 2 16.2722924 Residual | 82.4345819 45 1.8318796 -------------+-----------------------------Total | 114.979167 47 2.44636525
Number of obs F( 2, 45) Prob > F R-squared Adj R-squared Root MSE
= = = = = =
48 8.88 0.0006 0.2830 0.2512 1.3535
-----------------------------------------------------------------------------c02pc | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+----------------------------------------------------------------
12
gdppc | .8549361 .2136143 4.00 0.000 .4246948 1.285177 gdppc2 | -.0153969 .0040487 -3.80 0.000 -.0235514 -.0072423 _cons | 8.684242 2.65333 3.27 0.002 3.340161 14.02832 ------------------------------------------------------------------------------
Cubic regression: Source | SS df MS -------------+-----------------------------Model | 62.5718795 3 20.8572932 Residual | 52.4072872 44 1.19107471 -------------+-----------------------------Total | 114.979167 47 2.44636525
Number of obs F( 3, 44) Prob > F R-squared Adj R-squared Root MSE
= = = = = =
48 17.51 0.0000 0.5442 0.5131 1.0914
-----------------------------------------------------------------------------c02pc | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------gdppc | 5.907668 1.020958 5.79 0.000 3.850061 7.965274 gdppc2 | -.2180231 .0404877 -5.38 0.000 -.2996208 -.1364254 gdppc3 | .0025798 .0005138 5.02 0.000 .0015443 .0036153 _cons | -31.1186 8.210944 -3.79 0.000 -47.66667 -14.57053 ------------------------------------------------------------------------------
Given that, the endogenous variable is the same between the three models (CO2 emissions per capita in year t: CO2 H t ), we used the adjusted coefficient of determination (coefficient of global significance) to choose between the three estimations. When comparing the results of the estimations, we opted for the most appropriate form, i.e. Model III which has the highest coefficient (0.513 versus 0.251 for model II and 0.032 for model I). According to this third form, the relationship pollution-GDP follows a bell curve with a resumption of an increase in pollution. Accordingly, the pollution-GDP curve is not symmetric as shown by model II, but has an inflection point, as presented in the following graph:
16
18
20
22
24
Emission du CO2 en Fonction du PIB (USA)
15
20
25
30 GDPPC
C02PC
13
Fitted values
35
40
VIII- Conclusion The current work aims to empirically validate the Environmental Kuznets Curve (EKC) for the USA. To achieve this, we began by testing the existence of a co-integration relationship between pollution and GDP. Then we proceeded to the adjustment of pollution in relation to GDP. It was found out that the environmental situation in the U.S. is associated with the economic growth. However and despite this convergence, the Environmental Kuznets Curve (EKC) is not verified. Indeed, the most appropriate model to adjust pollution in terms of GDP is the cubic model (the N form) not the quadratic one which underlies the inverted U-shape. This somehow explains the US non-ratification of the Kyoto Protocol as it lies on “the wrong side” of the N curve. In a word, the USA is well aware that its industries are unable to fulfill their commitments to reduce emissions of carbon dioxide and other greenhouse gas dioxide as this hinders their profitability.
14
References
Brajer, V., Mead, R.W., Xiao, F., 2007. Health benefits of tunneling through the chinese environmental Kuznets curve. Ecological Economics 66 (4), 674—686. Cole, M.A., Elliott, R.J.R., 2003. Determining the trade—environment composition effect: the role of capital, labor and environmental regulations. Journal of Environmental Economics and Management 46 (3), 363—383. Bruyn, S.M., Van den Bergh, J.C.J.M., 0pschoor, J.B., 1998. Economic growth and emissions: reconsidering the empirical basis of environmental Kuznets curves. Ecological Economics 25, 161—175. Dickey, D., Fuller, W., 1979. Distribution of the estimators for autoregressive time series with a unit root. Journal of American Statistical Association 74, 427—431. Giles, D.E., Mosk, C., 2003. Ruminant eructation and a long-run environmental Kuznets curve for enteric methane in New Zealand: Conventional and fuzzy regression. Econometrics Working Paper, vol. 0306. Canada. Department of Economics, University of Victoria. Grossman, G.M., Krueger, A.B., 1991. Environmental Impacts of a North American free trade agreement. NBER Working Paper 3914. NBER, Cambridge. Hettige, H., Mani, M., Wheeler, D., 1999. Industrial pollution in economic development: Kuznets revisited. In: Fredriksson, P. (Ed.), Trade, Global Policy and the Environment. World Bank Discussion Papers, vol. 402. World Bank, Washington DC Holtz-Eakin, D., Selden, T.M., 1995. Stoking the fires? C02 emissions and economic growth. Journal of Public Economics 57, 85—101. Lindmark, M., 2002. An EKC-pattern in historical perspective: carbon dioxide emissions, technology, fuel prices and growth in Sweden, 1870—1997. Ecological Economics 42, 333—347. Panayotou, T., 1993. Empirical tests and policy analysis of environmental degradation at different stages of economic development. Working Paper WP238 technology and Employment Programme, Geneva: international Labor Office. Selden, T.M., Song, D., 1994. Environmental quality and development: is there a Kuznets curve for air pollution? Journal of Environmental Economics and Environmental Management 27, 147—162. Torras, M., Boyce, J.K., 1998. Income, inequality and pollution: a reassessment of the environmental Kuznets curve. Ecological Economics 25, 147—160. 15
