Examining the pollution haven hypothesis, ‘dirty’ industries
in the world economy from 1995 until 2009.
Abstract: This study researches the trade effect of the pollution haven hypothesis. It focuses on the growth of imports of ‘dirty’ products in developed countries coming from developing countries from 1995 until 2009. Input-output analysis is used to compare these imports to the growth of the ‘dirty’ industries in the developed countries. The results show that there is a growth in ‘dirty’ products coming from developing countries in comparison the growth of the ‘dirty’ industries in the developed countries and thus accepting the hypothesis made in the literature review. Although there is a growth in ‘dirty’ imports it is still a small amount compared to the output of the ‘dirty’ industries in the developed countries.
Master: International Business and Management Student: Jelle Rijpkema
Student number: S1931172
Email: J.S.Rijpkema@student.rug.nl Supervisor: drs. A. Visscher
Co-assessors dr. van Hoorn
2 Table of Contents
1. Introduction ... 3
2. Literature review ... 5
2.1 Literature on location decisions ... 5
2.2 Literature on trade flows ... 6
3. Methodology ... 9
3.1 Input-output analysis ... 9
3.1.1 Traditional input-output model ... 9
3.1.2 The multi-regional input–output framework ... 10
3.1.3 Multi country framework ... 13
3.1.4 Value of exports ... 14
3.2 Developing and developed countries ... 14
3.3 Dirty industries ... 16
3.4 Dataset ... 17
4. Results ... 18
4.1 Value of direct imports in developed countries coming from developing countries ... 19
4.2 Value of indirect imports in developed countries coming from developing countries ... 22
4.3 Co2 emissions in the imports in developed countries coming from developing countries ... 23
4.4 Geographical Analysis... 23
4.4.1 Developed American trade bloc ... 25
4.4.2 Developed European trade bloc ... 27
4.4.3 Developed Asian trade bloc ... 29
4.5 Sectoral Analysis ... 31
5. Conclusion ... 33
5.1 Limitations ... 34
6. References: ... 35
3
1. Introduction
The last decade the environment and trade liberalization have gotten a growing interest from policy makers as well as from academics. The last thirty years can be defined by a steady decrease in global trade barriers and an increase in multiregional bilateral regional trade. As reported by the World bank, international trade as a percentage of the world’s GDP grew from 38% in 1985 to 52% in 2005. This caused the less developed countries to have an accelerated industrialization process. A number of industrial activities that before were positioned in developed countries moved to the less developed countries. The poorer and less developed countries had comparative advantage in low wages which made firms shift to them. Firms searched more intensively for these ‘low wage’ countries to reduce costs. Although wage played a major factor in the location decision of firms, other factors like quality of workforce or costs of resources (e.g. land prices) also played an important role. Another factor that has received the attention of academics is the possible influence of environmental regulations on firm location decision making. In these decades the environment experienced major changes. One of these negative changes, for the interest of this thesis, is the growth in emission of Carbon Dioxide (Co2) as can be seen in graph 1. These changes caused an increase in environmental regulations, especially in the developed countries. These new regulations cause higher costs for firms, the pollutors pay. Either to make their production process cleaner or have to pay fines for breaking the regulations. The institutional and regulatory differences between developing and developed countries could create pollution havens for heavy pollutant firms. Eskeland and Harrison (2003) state that the pollution haven hypothesis can be seen best as a result of the theory of comparative advantage. Countries can get a comparative advantage in the production of high pollutant (‘dirty’) goods because of the lower pollution control costs. The pollution haven hypothesis states that environmental regulations influence the location decision of firms. The differences in regulations will result in a specialization of the making of dirty products and a capital flow to the country with weaker regulations. This influences the structure of the industries of both countries and the trade that flows between them. The countries with a comparative advantage in ‘dirty’ products see an increase in ‘dirty’ industries. The ratio between dirty and clean industries in the country changes to a structure in which the majority of the firms are ‘dirty’. The import of ‘dirty’ goods from pollution havens to developed countries increases as well.
4 Is there a relative increase in import from ‘dirty’ industries from developing countries to developed countries visible in the years 1995 – 2009?
For classification of ‘dirty’ industries data is used that measures the pollution abatement and control expenditures. This measures the burden of tax regulations industries have. Industries that have the highest pollution abatement and control expenditures per unit of output are seen as ‘dirty’ industries. As for the classification of developing and developed countries the gross income per capita is used as a line of demarcation. Data from the World Bank is used to measure the gross income per capita. To analyse the import of ‘dirty’ goods in developed countries coming from developing countries a dataset is used that contains 40 countries with each 35 sectors over the period from 1995 until 2009. The scope of the analysis is restricted by the lack of data. Using this data, the aim will be to see if a the trade effect of the pollution haven hypothesis between developing and developed countries existed in the given time period. The remainder of the paper is organized as follows. Section 2 will give an overview of the relevant literature on the pollution haven hypothesis. This will be followed by section 3, in which the methodology, which shows the mathematical models and definitions that are used, will be presented. In section 4 the results are presented and discussed. The final section contains the conclusion of this paper.
5
2. Literature review
The issue of the pollution haven hypothesis has been researched by academics in multiple ways. The increase in costs, either because of the compliance with the stricter regulations or because of the payment of fines by ignoring those regulations, is the binding factor in research on pollution haven hypothesis.
The literature grossly disperses into two ways. As the pollution haven hypothesis looks geographically at the relocation of firms from ‘dirty’ industries, this can be observed from either looking at the location decisions of heavy pollutant multi-national firms from pollution intensive industries or examining the international trade in ‘dirty’ goods. Firstly the literature that focuses on the location decisions will be shortly review. Secondly, the academics that focused on the international trade in ‘dirty’ goods will be discussed.
2.1 Literature on location decisions
Most research on the relationship between strictness of regulations and firm location decision focuses on the jurisdiction used in the United States. In the literature about the mentioned topic, the difference in jurisdiction within the United States (i.e. between states) or between the United States and other countries is used as a unit of analysis. Domestic firms or international firms were used as the subject of analysis. This line of literature is comprehensively analysed by Jeppessen et al. in 2002. Weak support for the pollution haven hypothesis is shown in this study. Often the regression analysis taken in this line of literature show that the strictness of the environmental regulations has an insignificant effect on the location of the firm. In their research, Millimet and List (2004) search for arguments why, even though theories and conventional wisdom expect that stricter environmental regulations have an effect on the relocation of firms, no evidence can be found for this phenomena in empirical studies. They argue that the difference in local context is the reason for the insignificant results. While previous studies took for granted that the relationship between the stricter regulations and the impact it has on the region is homogeneous, this might not be the case. They reason that the relationship is of a heterogeneous nature. If this heterogeneous relationship is not taken into account when studying the pollution haven hypothesis it could influence the results of the research.
6 supporting the pollution haven hypothesis. They also proved that Millimet and List were correct in their statement that the relationship between stricter regulations and the impact on the region is heterogeneous, and that this has an influence on the outcomes of the pollution haven hypothesis. Still, the evidence they found for the pollution haven hypothesis was not convincing, especially when other scholars who used FDI as a tool for their analysis found results that disproved the existence of the pollution haven hypothesis. Smarzynska Javorcik and Wei (2004) express in their paper that the lacking evidence can be the result of FDI being hindered to enter the country because of bureaucratic corruption. However, when taking this into account the results do support the pollution haven hypothesis weakly and do not survive robustness checks. In their paper of 2008 Levinson and Taylor shed light on the possible cause of the weak evidence that can be found in research on the pollution haven hypothesis when this is analysed by FDI flows. They point out that most pollutant plants from ‘dirty’ industries were already relocated before the data for the research was collected. Because of the missing data, the research gave a skewed (fake) view of the relocation of pollutant plants. This biases the results and the actual effect on the pollution haven hypothesis.
To summarize the first part of the literature review, the evidence found for the existence of the pollution haven hypothesis by analysing the geographical relocation of FDI flows is still weak. In the majority of the literature about geographical relocation an insignificant result is shown. Studies that want to prove the pollution haven hypothesis by FDI flow analysis show more evidence of the existence but this evidence is still weak. For this paper the strand of literature that focuses on the international trade flows to prove the pollution haven hypothesis is more relevant.
2.2 Literature on trade flows
7 domestic industry. As a result, trade has influenced the strictness of environmental regulations, which influences the trade of ‘dirty’ goods.
In contrast to the above, Janicke et al. (1997) found no support for the impact of strict environmental regulations on trade. Tobey (1990) examined the export of more pollutant countries in the year 1975. He concluded that environmental policies had no significant influence on international trade. According to Tobey, a reason for this was that the abatement costs of the stricter regulations were not sufficient enough to have a noticeable effect. In 2004, Cole and Elliott researched Tobey’s hypothesis but now with a new and improved dataset covering more countries and industries. They did this for the year 1995. The results of their research also turned out to be insignificant.
8 Input-output analysis is in the academic world an accepted way of measuring international trade flows. Since long input-output analysis has been recognised as a useful method to attribute pollution to final demand and trade (Leontief and Ford 1970). Although the amount of literature that studies the pollution haven hypothesis through input-output analysis is slim, the environmental impacts1
of trade using input-output analysis has been researched intensively. The existing literature on input-output analysis and the environment can be classified into two groups, namely single-region input-output analysis and multiple-region input-output analysis. One of the first to measure the pollution content of trade in a single country using input-output analysis was Walter (1973). The articles that were published after Walter’s research often focus on estimating the environmental ‘footprint’ of consumption of a country, both household and governmental consumption. It is assumed that, when looking at a single country context, the imported goods or services are produced with the same technology as when they are produced domestically. In reality, this often is not the case. These differences cannot be seen in single-region models. In multi-regional input-output models, where countries and regions are better distinguished, differences in technology, resource use and pollution intensities can better be quantified. This makes the analysis more specific. There is a correlation between the amount of countries in the multi-regional model and the specificity of the analysis. Haukland (2004) and Lenzen et al. (2004) have shown that the differences in outcome when using a single-region input-output analysis or a multi-region input-output analysis could be substantial. A drawback from multi-regional input-output models containing a large number of countries is the quality of the data. As the amount of countries grows, the amount of data needed grows exponentially. This increases the possibility for corrupt data which could make the dataset less reliable. In the last decade renewed interest in the emissions in consumptions together with the release of multi-regional input-output models containing more countries have led to interest in the environmental and trade topic. This resulted in an increase in published articles studying trade and impact on environment by input-output analysis. A detailed summary of the articles prior to 2007 can be seen in Wiedmann et al. (2007) and from 2007 until 2009 in Wiedmann (2009). As concluded by Wiedmann et al. (2009), multi-region input-output analysis is a sound and relevant methodology to measure the impacts of consumption on the environment. It forms a robust basis upon which methods to analyze environmental impacts can be built.
In contrast to other methods of studying the pollution haven hypothesis, input-output analysis does not result in a statistical analysis. Where other methods (e.g. FDI flows) found either significant or non-significant results for the pollution haven hypothesis, results of input-output analysis give a broader (multi-interpretable) view of the situation. The hypothesis used in this thesis
1
9 (1) will be based on the findings of Muñoz and Steininger (2010). They study the Co2 content in the imports of commodities of Austria in the years 1997 and 2004. They divide the exporting countries into annex-I countries and non annex-I countries. This division is based on the Kyoto Protocol about the emission targets of countries. The two groups are similar to the groups used in this thesis, as the annex-I countries are similar to the developed countries and non annex-I countries are similar to the developing countries. A list of annex-I and non annex-I countries can be found in appendix I. As a result, Muñoz and Steininger (2010) found that the emission in imports from non annex-I countries grew 105% compared to a growth of 21% in imports from annex-I countries. The authors conclude that there is a significant growth in the origin of pollutant imports. Based on their findings the following hypothesis is formulated:
Hypothesis 1:
If there is an increase of imports from ‘dirty’ industries in developed countries coming from developing countries of 100% or more compared to the growth of the dirty industries in the developed countries, there is evidence for the pollution haven hypothesis.
In the next section the econometric methodology and data will be discussed, after which the results and the conclusion will be presented.
3. Methodology
3.1 Input-output analysis
The calculations used in this thesis are based on the environment input-output analysis (Miller and Blair, 1985). This is a method often used and widely accepted by the academic society as it is part of the System of National Accounts. This method is based on input-output models which show the transactions in monetary values between supply chains in a country. A comprehensive introduction into input-output analysis can be found in the paper of Miller and Blair(1985).
3.1.1 Traditional input-output model
Input-output models show which part of output in an industry is used as an intermediate input in other industries. As an equation the input output model can be seen as following:
10 (2)
(3)
(4) In which x represents the total output of the economy as column vector. Y is the final demand and Ax the intermediate input. A is the matrix (K x k) of the input coefficients. For example an input
coefficient , is the value of the input of the good per output of good .
In the input-output models goods are produced to serve the intra-industry demand and the demand of the households, government and trade (export to other countries). In the process of producing commodities for final consumption, direct inputs are used as well as inputs that go in the producing of the direct inputs (indirect inputs for the commodities).
The total production (indirect and direct inputs) needed in the final demand can be calculated using the following equation.
x = ( − ) =
Where I indicates an identity matrix and the equation ( − ) stands for the Leontief inverse. Because the equation is linear the extra gross output can be calculated by change in final demand times Leontief inverse. Vector can be expressed by the amount of output from industry required directly and indirectly to produce per unit final demand from industry .
The environmental input output matrix
The environmental input output matrix is derived from coupling the model with Co2 emissions, in which is a row vector with each element
representing the Co2 emissions per unit output from
industry . Multiplying the vector with the Leontief inverse, which shows the total direct and indirect emissions intensity for each industry, gives the emission multiplier vector .
M = ( − )
Emission multiplier vector multiplied with the final demand vector(Y), gives the total emission used in final consumption (households, government and trade) from the sectors in the industry(F).
=
3.1.2 The multi-regional input–output framework
11 (5)
(6) domestic and/or foreign industries. The produced goods can also be consumed by foreign or domestic households/government. This changes the basic input output equation to:
= +
and are the domestic input coefficients, and and are the imported inputs from the
other country. Corresponding and are domestic final demand and and are final
demand for imported products. With multiple countries the Leontief inverse transforms to the following equation: =
When wanted to calculate the emission coefficients of multiple countries, import and export should be taken into account. For the product imported in country coming from country the emissions coefficient of country should be used and visa versa.
When examining the emissions in the imports from country to a problem occurs. This problem is explained in the article of Koopman, Wang and Wei(2014), who did research on the value-added of the gross exports of countries and breaking this up into various components. They found nine elements which add up to the total gross exports of a country. Two of the nine elements are revered to as ‘pure double counting items’ namely ‘double counted intermediate exports produced at home’ and the other ‘double counted intermediate exports produced abroad’.
When a good crosses the border its gross value is measured and added in the statistics, rather than the net value added between the border crossings. So if the same goods cross the border multiple times, its gross value is added multiple times to the statistics. The origin of the intermediate inputs could either be foreign or domestic but this does not matter as long as the gross value is counted double.
Every time the intermediate inputs cross a border it gets counted, so the value of the goods exported in the multiregional input output tables can grow indefinitely. As for this problem, it also arises when calculating the imports of a country, it is relevant for this thesis. According to Los, Timmer and De Vries, a solution can be found in calculating the exports while using an
unidirectional trade flow . This means transforming the data to a dataset with one directional flow of data between countries, e.g. removing exports from to 2
, and thus creating a counterfeited world economy.
2
12 (9) (8) (7) (10) (11) This will result in a multi-regional input output model in which there is no flow of goods from to . The total output of this model is the total output in a world with no exports from to . When subtracting the total output of the counterfeited economy from the actual total output in the world economy, the genuine amount of export from country to can be calculated.
The formulas for the counterfeited world economy are going to be as following.
= 0
and = 0
Matrix Z stands for the values of the trade of intermediate inputs between and within countries. Matrix Y stands for the values of the final demand from and in between countries for goods. In these matrixes the flow of intermediate goods from country to country , is set to zero. The same goes for the final demand in country for goods from country . The new input coefficient matrix A can be formed with:
=
In a multi-regional setting, emission coefficients vectors (C) of both countries are also taken into the calculation. When interested in measuring the emission in the imports coming from , the emission coefficients from country can be set to zero. The emission coefficient vector is as follows:
C = [C 0]
Using the same equation as before the emissions in the counterfeited world economy (F ) can be
calculated by
= = M
The number of the emissions exported by country to can be calculated by using the following formula.
!"## # $%!& !' % # = −
This equation combines the nine elements found by Koopman, Wang and Wei but it excludes the ‘pure double counting items’ elements. Thus improving the equation for emissions in gross exports.
13 (13) (12)
(14) 3.1.3 Multi country framework
When lifting the mathematical equations from a two country world to a multi country settings, attention should be paid to the new dimension in trade that comes with it. Consider a world existing out of three countries. If one wants to measure the imports (or exports) from country to it should be taken into account that there is a flow of goods from to going through country &. Applying the equation in the note of Los, Timmer and De Vries to a multi-country setting excludes the indirect imports that come from the producing country to the receiving country through a series of distribution countries. Since the imports of a country express its interdependence with other countries, both direct and indirect import flows should be calculated (Ang and Su, 2010). As well as staying close to the existing literature which also focuses on direct trade flows, indirect trade flows will also be analysed. Johnson and Noguera (2011) suggest a definition of gross exports which contains direct and indirect trade flows. They calculate the total value added of goods which are produced in sector in country and absorbed by country . Applying this method to the
framework of Los, Timmer and De Vries comes up with the following equations for a multi country world when calculating the gross exports from country .
= ( ) and = ( 0 0 0 )
In the new counterfeited economy the total final demand of country is set to zero. In this model there is no exporting from country and country as well as no final demand of country for goods from country &. The next equations are the same as for the two country world economy. The
equations for the input coefficients matrix and total emissions in the counterfeited economy, where the emission coefficient vector C =*C
0 0+, are:
= = = M
The equation for the amount of emissions in the goods which are absorbed in country which originated in country is as following:
!"## # $%!& $ $%!& !' % # = −
14 (15)
(16)
(17) 3.1.4 Value of exports
To measure the value of the exports of ‘dirty industries from developing to developed countries an equivalent of the equations above can be used. Instead of using the emission coefficient vector an value coefficient vector ( ) is used to calculate the values of the outputs. As the input-output
models are already expressed in terms of value (million dollars) the value coefficient vectors of the industries in the different countries can be set to one. Since the interest of this study lies in the value of exports of the ’dirty’ industries of the developing countries, the value of the other industries are set to zero.
= [1 0]
Using the value coefficient vector in the formulas above gives the value of either the direct or indirect exports.
= = M = = M
, -! . $%!& !' % # = − , -! . $%!& !' % # = −
3.2 Developing and developed countries
A developed country is seen as a state which has a highly developed economy and an advanced infrastructure. In contrast a developing also known as a less developed country is characterised by an underdeveloped industrial base, lower living standards and a lower score on the human
development index (life expectancy, education, and income). The most used criterion to evaluate the degree of economic development is income per capita, gross domestic product or level of
15 Table 1a
Income groups based on gross income per capita Income group Gross income per capita
Low income $ 1,036- $4,085
Lower middle income $ 1,036- $4,085 Upper middle income $ 4,086- $ 12,615 High income $ 12,616 or higher
Table 2. Income groups based on gross income per capita
The World Bank classifies the developed and developing countries as follows, developed countries are countries with a high income and developing countries are countries with an upper middle income. The low income group are least developed countries and the low middle income are the transition countries. Based on the GNI per Capita the 40 countries used in the dataset are divided into developing countries and developed countries(table 1b).
Table 1b
List of developing and developed countries Developing countries Developed countries
Brazil Australia Korea
Bulgaria Austria Latvia
China Belgium Lithuania
Hungary Canada Luxembourg
Mexico Cyprus Malta
Romania Czech Republic Netherlands
Turkey Denmark Portugal
India Estonia Slovak Republic
Indonesia Finland Slovenia
Poland France Spain
Russia Germany Sweden
Greece Taiwan
Ireland
United Kingdom
Italy United States
Japan
Table 3. List of developing and developed countries
16 slightly above the high income line, which makes them developing countries compared to other high income countries.
3.3 Dirty industries
Stricter environmental policies are introduced to reduce the amount of pollution and therefore influences the polluting industries (‘dirty’ industries) the most. These industries are researched in the analysis. Highly polluting industries can be classified on their pollution abatement costs and control of pollution expenditures. A often cited article on the definition of ‘dirty’ industries is of Low 1992. He classifies ‘dirty’ industries as industries where the pollution abatement costs and control expenditures are approximately I % or more of their total sales. His classification can be seen in table 2a.
Table 2a
Industries ranked by pollution Ranking Industry
1 pulp and waste paper petroleum products 2 residual petroleum products
3 organic chemicals 4 inorganic chemicals 5 Fertilizers 6 chemical materials 7 veneers, plywood 8 wood manufacturers 9 paper, paperboard 10 lime, cement contr. Matrl. 11 iron and steel
12 non ferrous metals 13 Metal manufacturers
Table 2a. Pollutant industries, from most pollutant to least pollutant according to Low, 1992
17
Table 2b
Industries ranked by pollution Ranking Industry
1 Iron and steel 2 Nonferrous metals 3 Industrial chemicals 4 Petroleum refineries
5 Nonmetallic mineral production 6 Pulp and paper
7 Other chemicals 8 Rubber products 9 Leather products 10 Metal products
Table 2b. Pollutant industries according to Grether and de Melo, 1996 and 2003
The industries that are found to be pollutant- intensive through the two different way of measuring are largely similar to each other. The industries with the overall highest pollution ranking and which are present in our data are listed below.
Pulp, Paper, Printing and Publishing Coke, Refined Petroleum and Nuclear Fuel Chemicals and Chemical Products
Basic Metals and Fabricated Metal
3.4 Dataset
This thesis exploits a dataset that has not been used often in previous research on environmental input-output analysis. The data source used in this thesis are the tables from the world input output dataset (Timmer 2012) released in November 2013, and are therefore relatively recent. The dataset contains bilateral input output data from 1995 to 2010 of 40 countries, including all European countries and 13 other countries that contribute to a large extent to the world GDP. Together, the countries in the dataset cover 84%3
of the total world GDP. Countries that have been left out and are in the top 25 of countries with the highest GDP are Saudi Arabia, Switzerland, Iran and Norway. A limitation of the dataset is the lacking representation of the African continent. Although the contribution of this continent to the world GDP is relatively small, some countries (e.g. Nigeria and South Africa) are seen as developing countries. Trade flows in ‘dirty’ products from these countries to developed countries are not taken into account. In comparison to the total world trade flows, the
3
18 mentioned countries do not bring a lot to the table. A 41st ‘country’ is added to the dataset called Rest of the world(RoW), which combines the output numbers of the rest of the world. The data covers 35 industries for each country, which make comparisons and calculations with and within the data possible. The choice for this dataset is based on the size of the dataset and the reliability of it. Given the time for this research, calculations with larger datasets (e.g. the EORA dataset covering 187 countries and a total of 15,909 sectors) have not been used, since special equipment would be needed in order to do calculations of a large size. The quality and reliability of the world input-output dataset is high, since the choice of the countries in the dataset is based on the quality of the data of the countries, while countries with qualitatively poor data are left out. The data also have to be official to be taken into the dataset. The construction of the tables has been done by combining national input-output tables with bilateral international trade flows, and by following the
conventions of the system of national accounts. Together, the lineage, accuracy and the consistency increase the quality of the data according to Clarke and Goodchild, 2008. Data on the emissions of the different sectors from the countries are also obtained from the world input-output database. The emission data are organized in the same that it fits the world input output tables (40 countries, 35 industries). The national statistics of emissions have been made compatible with the national accounts used in the world input-output dataset. Where the national statistics were lacking, additional data collections and estimations have been used. A comprehensive list of sources used in creating the emission data can be found in the paper of Genty, Arto, and Neuwahl (2012). The transformation of the statistics has been based on the accounting principles of NAMEA-energy, NAMEA-air, NAMEA-materials, NAMEA-land and NAMEA-water (SEEA, 2003). These emission data contain emission coefficients of all the 40 countries and RoW from 1995 to 2009.
4. Results
Since most academics, who studied the trade effect of the pollution haven hypothesis, took direct trade flows as their unit of measurement and because the fundaments of the trade effect of the pollution haven hypothesis are based on direct trade flows, the focus of the analysis will be on the direct trade flows from developing to developed countries.
19 on the geographical location, countries from the developing and developed group will be put in trade blocs after which the trade flows of ‘dirty’ goods between them are calculated. The last group of results will contain a sectoral analysis. In the last group of results the differences within the four ‘dirty’ industries will be revealed.
It should be taken into consideration that the results are in millions of dollars, representing the value of the outputs of the industries. This does not represent the output in products per industry. In the case of a growth in the value in an industry, this does not mean the production of that industry changed with the same amount. The growth in value can also be explained for example by inflation.
4.1 Value of direct imports in developed countries coming from developing countries To the extent of measuring the direct imports in developed countries coming from developing countries, it is necessary to remove the double counted items from the dataset. This is done by calculating the total emissions of the world minus the total emissions in a world where there would be no imports in developed countries coming from developing countries. This is done by using equation 17 as suggested by Los, Timmer and De Vries (2014). The results of this procedure, which is used for the years 1995 to 2009, can be seen in table 3a.
For each year the total value of the world’s output (the goods from all the industries from all the 41 countries) (World output) and the value of imports of ‘dirty’ goods in developed coming from developing countries is given (Direct Imports). Since the growth ‘dirty’ industries in the developing countries also plays an important role in the growth of the imports from these countries (Output developing industries) this is also given in table 3a. For every variable the change percentage is given, taking 1995 as the base year.
20
Graph 2. Percentage growth of direct imports and ‘dirty’ industries in developing countries compared to world output
A feature of developing countries is rapid economic growth (Khanna et al., 2005). This leads to the industries in the developing countries to grow faster than the average (growth) of the world. Thus explaining the difference in the growth of the world output and the growth of the ‘dirty’ industries in the developing countries as seen in graph 2. The fact that industries in developing countries grow faster can also explain the growth in the imports of ‘dirty’ products coming from those countries. To see if this growth is originating from the growth in the developing countries or by the pollution haven hypothesis, a closer look at the trade effect of the pollution haven hypothesis is needed. As firms in ‘dirty’ industries move to developing countries the imports of ‘dirty’ products in developed countries increases and the output of ‘dirty’ products in developed countries decreases. Even when not all firms relocate there can still be growth in the ‘dirty’ industries in developed countries. If the pollution haven hypothesis is true, the growth of imports of ‘dirty’ products in developed countries is higher than the growth in output of the ‘dirty’ industries in developed countries. Leading to a change in the ratio what is produced at home and what is imported. This can be seen in table 3b and in graph 3.
0 100 200 300 400 500 600 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 P e rc e n ta g e Year
Growth of direct imports
from 'dirty' industries
Direct imports of 'dirty' goods in developed countries coming from developing countries
Output of the 'dirty' industries in developing countries
21
Graph 3. Percentage growth of direct imports and output op ‘dirty’ industries in developed countries
While in 1995 2% of the ‘dirty’ products were imported, in 2008 this was 5,5%. The growth in the ratio can be seen in graph 3. Although the percentage of imported ‘dirty’ goods has doubled in absolute values, it is still a small amount of the total ‘dirty’ output in the developed world(graph 4).
Graph 4. Direct imports compared to the output op ‘dirty’ industries in developed countries
0 100 200 300 400 500 600 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n ta g e Year
Growth of direct imports
from 'dirty' industries
Direct imports of 'dirty' goods in developed countries coming from developing countries Output of the 'dirty' industries in developed countries
Growt ratio direct imports and output 'dirty' industries in developed countries 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 Va lu e i n b il li o n s o f d o ll a rs Year
Value of the direct imports
from 'dirty' industries
22 4.2 Value of indirect imports in developed countries coming from developing countries Most researches on the pollution haven hypothesis only look at direct trade flows i.e from country A directly to country B. In a multi-country setting, such as the real world, a part of the trade originating from country A goes through other countries before it arrives in country B also known as the indirect trade flows. The effect on the imports of ‘dirty’ industries caused by indirect trade flows should be given a look when studying the trade effect of the pollution haven hypothesis.
As mentioned in the methodology section, the method used to calculate the indirect trade flows are from the note of Los, Timmer and de Vries (2014) based on the paper of Johnson and Nogeura (2008). In their paper Johnson and Nogeura look at the total domestic value added that is absorbed in the final demand of other countries and thus including direct and indirect flows. In the calculations they exclude the domestic value added in the exports of intermediate goods. This results in outcomes that are smaller than the outcomes when the equations of Koopman Wang and Wei (2014) are used (table 4). Graph 5 shows the different growth rates of the different calculations. As the growth is following the same pattern there are no major changes in the indirect trade flows of ‘dirty’ goods in developed countries coming from developing countries compared to direct trade flows. The ratio of indirect/direct trade flows remain largely the same.
Graph 5. Percentage growth of direct and indirect imports
0 100 200 300 400 500 600 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n ta g e Years
Growth of indirect and direct imports
from 'dirty' industries
23 60 70 80 90 100 110 120 130 140 150 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 P e rc e n ta g e Year
Emissions in indirect and direct imports
from 'dirty' industries
Emissions in direct imports of 'dirty' goods in developed countries coming from developing countries Emissions in indirect imports of 'dirty' goods in developed countries coming from developing countries
Emissions in world output, all the industries in all the countries
4.3 Co2 emissions in the imports in developed countries coming from developing countries With respect to the emission in the imports of ‘dirty’ products a different pattern is shown (graph 6 & table 5). Although the value of the imports of ‘dirty’ goods has risen with more than 500%, the amount of emission in these imports have decreased in the last two years in comparison with 1995. In the first 11 years the growth in emissions follows the emissions in the total world output but with a more fluctuating pattern. In the last 4 years there is a steep decline in the amount of emission in the imports of ‘dirty’ products. Growth in the emission output, or even decline, can be explained by technological progress which make the industries cleaner, by regulations in the measurement of Co2, or by changes in composition of the imports from the four different ‘dirty’ industries. The last
possible reason for the decline stated will be analyzed in the sectoral analysis.
Graph 6. Percentage growth of emissions in direct and indirect imports compared to the total world emissions
4.4 Geographical Analysis
24 distinction has been made on basis of geographical regions. This lowers the differences within each subgroup and creates an analysis on a more national level. The developed countries are grouped in a North American trade bloc, European trade bloc and Asian trade bloc. For the developing countries, the trade blocs are an American trade bloc containing North and South American countries, an Eastern European trade bloc containing countries in eastern Europe and Russia and an Asian trade bloc. Appendix II presents a list of the countries in each trade bloc. The value of the ‘dirty’ goods going to the three different developed trade blocs can be seen in graph 7 or table 6a.
Graph 7. Direct imports per developed trade bloc
This shows that most of the value of ‘dirty’ goods over the years is imported by Europe. In absolute value the imports from the Asian trade bloc are the lowest compared to the European or American trade bloc. The value of the imports are relatively equal between the three developed regions and follow the same growth pattern. Almost no growth in the first 7 years and an increase in the last 7 years can be seen.
A closer look at from which developing trade bloc the imports are origination from is given in the following section. The results for the developed American trade bloc, European trade bloc and Asian trade bloc are given respectively in tables 6b, 6d and 6f. These table shows the total amount of ‘dirty’ imports (in billions of US dollars), from the different developing trade blocs to three developed trade blocs. 0 50.000 100.000 150.000 200.000 250.000 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 Va lu e i n m il li o n s o f d o ll a rs Year
Direct 'dirty' imports in the developed trade
blocs
Value of the 'dirty' imports in the American developed trade bloc
Value of the 'dirty' imports in the European developed trade bloc
25 4.4.1 Developed American trade bloc
Concerning the flow of goods to the American developed trade bloc (United States and Canada) it can be observed that most of the goods are coming from the developing countries in Asia(table 6b). Graph 8 shows the growth from the different developing trade blocs to the developed American trade bloc trade bloc and graph 9 shows the absolute growth in the value of the imports.
Graph 8. Percentage growth of the imports in the American developed trade bloc coming from the developing trade blocs
Graph 9. Direct imports in the American developed trade bloc coming from the developing trade blocs
This shows clearly that in absolute and in relative values most of the growth is coming from the imports from Asia. While the imports from the American or European countries have almost the
0 100 200 300 400 500 600 700 800 900 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 P e rc e n g a te Year
Growth of 'dirty' imports in the American developed
trade bloc
'Dirty' imports coming from the American developing trade bloc 'Dirty' imports coming from the European developing trade bloc 'Dirty' imports coming from the Asian developing trade bloc
0 20.000 40.000 60.000 80.000 100.000 1995 1997 1999 2001 2003 2005 2007 2009 Va lu e i n m il li o n s o f d o ll a rs Year
Direct
'dirty' imports in the American
developed trade bloc
26 same growth rate with the highest point in 2008 with 320%, Asia in 2008 has a growth of more than 800%. Most of the growth in imports happens from 2003 and onwards. While the first part has a steady growth, the second part shows a more exponential growth.
Graph 10 and 11 show the growth of the imports of ‘dirty’ goods versus the total growth of the dirty sector in America. The imports of ‘dirty’ products as a percentage of the total output of the American ‘dirty’ industries can be seen in graph 10 and the percentage growth can be seen in graph 11(table 6c).
Graph 10. Direct imports coming from developing trade blocs as a percentage of the output of the American developed trade bloc
Graph 11. Percentage growth of the imports coming from the developing trade blocs compared to the output of the American developed trade bloc
0 0,5 1 1,5 2 2,5 3 3,5 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 P e rc e n t Year
'Dirty' imports as a percentage of the output of the
'dirty' industries in the American developed trade
bloc
'Dirty' imports coming from the American developing trade bloc 'Dirty' imports coming from the European developing trade bloc 'Dirty' imports coming from the Asian developing trade bloc
0 0,5 1 1,5 2 2,5 3 3,5 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 P e rc e n t Year
'Dirty' imports as a percentage of the output of
the 'dirty' industries in the American developed
trade bloc
27 Through all years most of the ‘dirty’ goods are coming from the developing countries in North and South America leading up to 5% of the total output in 2008. The percentage of ‘dirty’ goods originating from the European developing trade bloc remained constant over the year while the other two trade blocs grew. ‘Dirty’ imports coming from the Asian developing trade bloc grew the most with 350% in 2008. While in 1995 it was just 0,7%, it grew to 3% of the total output in the industry in 2008.
4.4.2 Developed European trade bloc
Taking a closer look at the ‘dirty’ trade flows to the European developed trade bloc, a similar
situation can be seen (graph 12 and 13; table 6d). Percentage wise the imports coming from the Asian developing trade bloc grew the most. Especially the last 7 years show the most growth (graph 12). In absolute values most ‘dirty’ goods are coming from the European developing trade bloc. This can be explained by the geographical distance/transportation costs. Only a small part of the imports of ‘dirty’ goods are coming from the American trade bloc and this trade bloc shows the lowest growth as well.
Graph 12. Percentage growth of the imports in the European trade bloc coming from the developing trade blocs
0 100 200 300 400 500 600 700 800 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n ta g e Year
Growth of 'dirty' imports in the European
developed trade bloc
28
Graph 13. Direct imports in the European developed trade bloc coming from the developing trade blocs
Graph 14 and 15 show the growth in the ‘dirty’ goods compared to the output of the ‘dirty’ industries of the European developed trade bloc(table 6e).
Graph 14. Direct imports coming from developing trade blocs as a percentage of the output of the European developed trade bloc 0 20.000 40.000 60.000 80.000 100.000 120.000 Va lu e i n m il li o n s o f d o ll a rs Year
Direct 'dirty' imports to the European
developed trade bloc
'Dirty' imports coming from the American developing trade bloc 'Dirty' imports coming from the European developing trade bloc 'Dirty' imports coming from the Asian developing trade bloc
0 0,5 1 1,5 2 2,5 3 3,5 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n t Year
'Dirty' imports as a percentage of the output of
the 'dirty' industries in the European developed
trade bloc
29
Graph 15. Percentage growth of the imports coming from the developing trade blocs compared to the output of the American developed trade bloc
These graphs confirm the results stated above about European developed trade bloc (graph 14). Most trade is coming from the European developing trade bloc, while the Asian developing trade bloc shows the most growth (graph 14), 0,6% in 1995 to 2,3% in 2008.
4.4.3 Developed Asian trade bloc
With respect to the imports of ‘dirty’ goods to developed Asian trade bloc (Japan, Korea and Australia) a slightly different pattern can be seen (graph 16 & 17 (table 6f)). While the American developed trade bloc imported a proportion of their ‘dirty’ goods from the European developing trade bloc, the European developed trade bloc imported from the American developing trade bloc. The Asian developed trade bloc does neither. Their imports coming from America or European trade blocs decrease in the first 8 years and after 14 years they were the same as they were in 1995. In contrast to this, the goods coming from the Asian developing trade bloc grew with almost 600%. Putting it in perspective, in 2008 the ‘dirty’ goods coming from the Asian developing countries was almost 10 times the amount coming from European and American developing countries combined.
0 100 200 300 400 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n ta g e Year
Growth in imports compared to the growth
'dirty' sector of the European developed trade
bloc
30
Graph 16. Percentage growth of the imports in the Asian developed trade bloc coming from the developing trade blocs
Graph 17. Direct imports of the American developed trade bloc coming from the developing trade blocs
Graph 18 shows the same story. The imports coming from European or American developing
countries stay beneath 0,5% of the total output of the ‘dirty’ industries in the developed countries in Asia. The goods coming from the developing countries in Asia grew from 1,2% to 4,8% in 2008, an increase of 393% (graph 19 & table 6g).
0 100 200 300 400 500 600 700 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n ta g e Year
Growth of 'dirty' imports in the Asian
developed trade bloc
'Dirty' imports coming from the American developing trade bloc 'Dirty' imports coming from the European developing trade bloc 'Dirty' imports coming from the Asian developing trade bloc
0 20.000 40.000 60.000 80.000 100.000 120.000 1995 1997 1999 2001 2003 2005 2007 2009 Va lu e i n m il li o n s o f d o ll a rs Year
Direct 'dirty' imports in the Asian developed
trade bloc
31
Graph 18. Direct imports coming from developing trade blocs as a percentage of the output of the Asian developed trade bloc
Graph 19. Percentage growth of the imports coming from the developing trade blocs compared to the output of the American developed trade bloc
4.5 Sectoral Analysis
Multi-regional input-output analysis has the ability to estimate the international trade between industries. Table 7 presents the imports of the four different ‘dirty’ industries over the years 1995 until 2009 from developing to developed countries. An image of the growth rate of the four
industries can be seen in graph 20. The four industries have relatively the same growth pattern. Only the Pulp, Paper, Printing and Publishing has a lower growth rate compared to the other three. Again most of the growth is in the last years, 2003 and onwards. Graph 21 shows the value of ‘dirty’
0 1 2 3 4 5 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n t Year
'Dirty' imports as a percentage of the output of
the 'dirty' industries in the Asian developed
trade bloc
'Dirty' imports coming from the American developing trade bloc 'Dirty' imports coming from the European developing trade bloc 'Dirty' imports coming from the Asian developing trade bloc
0 50 100 150 200 250 300 350 400 450 1995 1997 1999 2001 2003 2005 2007 2009 P e rc e n ta g e Year
Growth in imports compared to the growth
'dirty' sector of the Asian developed trade bloc
32 imports from the 4 different industries. The imports of Basic Metals and Fabricated Metal covers more than 66% of the total value of the imports. All the industries except Pulp, Paper, Printing and Publishing show the same growth pattern in absolute values as well. Because the pattern is the same over the years the changes in total emissions in the industries, section 4.3 of the analysis, cannot be explained by changes in composition of the four industries.
Graph 20. Percentage growth of the imports of ‘dirty’ products per industry in developed countries coming from developing countries
Graph 21. Imports of ‘dirty’ products per industry in developed countries coming from developing countries
0 100 200 300 400 500 600 700 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 P e rc e n ta g e Year
Growth of the imports per 'dirty' industry
to developed countries coming from developing countries
Pulp, Paper, Printing and Publishing
Coke, Refined Petroleum and Nuclear Fuel
Chemicals and Chemical Products
Basic Metals and Fabricated Metal 0 50.000 100.000 150.000 200.000 250.000 300.000 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 Va lu e i n m il li o n s o f d o ll a rs Year
Value of imports per 'dirty' industry
to developed countries coming from developing countries
Pulp, Paper, Printing and Publishing
Coke, Refined Petroleum and Nuclear Fuel Chemicals and Chemical Products
33
5. Conclusion
The objective of this thesis was to find evidence for the existence of the pollution haven hypothesis. This was done by examining the trade impact of the pollution haven hypothesis, which states that the differences in the level of strictness of environmental regulations will lead to ‘dirty’ industries moving to countries with less strict environmental regulations, in order to save costs. As a result, countries with a high level of environmental regulation strictness will import ‘dirty’ products from countries with lower strictness of environmental regulations. To measure this, by using multi-regional input-output analysis, this thesis studied the trade flows in ‘dirty’ goods coming from high pollutant industries from developing countries (low strictness of environmental regulations) to developed countries (high strictness of environmental regulations).
The results have shown that the value of the imports of ‘dirty’ goods in developed countries coming from developing countries increases with more than 100% compared to the output of the ‘dirty’ industries in the developed countries. Thus finding support for the hypothesis made in the literature review. Although there is a strong growth in imports of ‘dirty’ goods, a doubling in ten years, it remains only a few percentages compared to the total output of the industries. The trade effect of pollution haven hypothesis is growing but is still very small and has a relatively small impact on the ‘dirty’ sector in developed countries, with a maximum of 5,5% of the total output in the developed industries.
Dividing the developing and developed countries into trade blocs showed a more indebt overview of the growth in the flow of ‘dirty’ products between regions. This shows that most of the imports are coming from the developing countries that are close to the developed countries (e.g. the European developed trade bloc got most imports from east Europe and Russia, while American and Canadian flows of ‘dirty’ goods originated mostly from Brazil and Mexico and the Asian developed trade bloc imported most ‘dirty’ goods from the Asian developing trade bloc). This can easily be explained by transportation costs. The environmental regulatory costs that are saved by importing ‘dirty’ goods from developing countries are lost when importing them from distant developing countries, an increasing the costs of transportation.
The geographical analysis also highlights the importance of the Asian developing trade bloc (China, India and Indonesia). In all the three developed trade blocs ‘dirty’ goods coming from these three countries grew the most. While in 1995 40% of the ‘dirty’ goods originated in the Asian developing trade bloc, in 2007 this grew to almost 60% of all ‘dirty’ goods being imported by developed countries. If this growth continues, in 30 to 40 years the three countries will be
34 this effect is growing it remains small in comparison to the output of the ‘dirty’ industries in the developed countries(only a few percent).
In conclusion, the hypothesis stated in this thesis has been confirmed. The ‘dirty’ imports in developed countries coming from developing countries more than doubled in comparison to the growth of the ‘dirty’ industries in developed countries. Thus finding evidence for a trade effect on imports of ‘dirty’ goods, although in absolute terms this trade effect is still small. The existence of this trade effect is a necessity for proving the pollution haven hypothesis. As the pollution haven hypothesis leads to an increase in imports of ‘dirty’ products. Further research can investigate if the trade effect found is caused by relocating firms from ‘dirty’ industries seeking to reduce costs and so proving the pollution haven hypothesis.
5.1 Limitations
35 6. References:
Ang, B. W., & Su, B. (2010). Input–output analysis of CO2emissions embodied in trade: The effects of spatial aggregation. Ecological Economics, 70
Antweiler, W., Copeland, B. R., & Taylor, M. S. (2001). is free trade trade good for the environment? American economic review, 91(4)
Busse, M. (2004). Trade, Environmental Regulations and the World Trade Organization. World Bank Policy Research Working Paper
Clarke, K. C., & Goodchild, M. F. (2002). ’Data quality in massive data sets. Handbook of Massive Data Sets, massive computing, 4(18)
Cole, M. A., & Elliott, R. J. R. (2003). ’Do Environmental regulations influence trade patterns? Testing old and new trade theories. the world economy, 28(8)
Copeland, B. R., & Taylor, M. S. (2004). TRADE, GROWTH AND THE ENVIRONMENT. journal of economic literature, 41
Dean, J. M., Lovely, M. E., & Wang, H. (2005). ’Are foreign investors attracted to weak environmental regulations? Journal of Development Economics, 90
Ederington, J., & Minier, J. (2001). Is environmental policy a secondary trade barrier? And empirical analysis. Canadian journal of economcis, 36(1)
Eskeland, G. S., & Harrison, A. E. (2003). Moving to greener pastures? Multinationals and the pollution haven hypothesis. Journal of Development Economics, 70
Genty, A., Arto, I., & Neuwahl, F. (2012). Final database of environmental satellite accounts: technical report on their compilation. WIOD, deliverable 4.6
Grether, J., & De Melo, J. (2003). Globalization and dirty industries: do pollution havens matter? NBER Working Paper No. 9776
Grether, J. M., & De Melo, J. (1996). Commerce, Environnement et Relations Nord-Sud: les enjeux et quelques tendances récentes. Revue d’économie du développement, 4
Grossman, G., & Kreuger, A. (1995). Economic growth and the environment. Quarterly Journal of Economics, 110
Hanna, R. (2004). US environmental regulation and FDI: evidence from a panel of US based multinational firms. Massachusetts Institute of Technology, Cambridge, MA
Haukland, E. (2004). Trade and environment: emissions intensity of Norway's imports and exports. Industrial Ecology Programme, 1(1)
36 Holtz-eakin, D., & Selder, T. M. (1995). Stoking the fires? CO2emissions and economic growth.
Journal of Public Economics, 57
Janicke, M., Binder, M., & Monch, H. (1997). ’Dirty industries’’: Patterns of change in industrial countries. environmental and resource economics, 9
Jeppessen, T., List, J. A., & Folmer, H. (2002). Environmental Regulations and New Plant Location Decisions: Evidence from a Meta-Analysis. Journal of Regional Science, 42(1)
Johnson, R., & Noguera, G. (2012). Accounting for Intermediates: Production Sharing and Trade in Value Added. Journal of International Economics, 82(2)
Kalt, J. (1988). The Impact of Domestic Environmental regulatory Policies on U.S. International Competitiveness, in Spence, M. and H. Hazard . International Competitiveness, Cambridge, MA Keller, W., & Levinson, A. (2002). Pollution abatement costs and foreign direct investment inflows to US states. The Review of Economics and Statistics,, 84(4)
Khanna, T., Palepu, K. G., & Sinha, J. (2005). ’Strategies that fit emerging markets. Risk and reward in world markets, Harvard business review, 1(1)
Koopman, R., Wang, Z., & Wei, S. (2014). Tracing Value-Added and Double Counting in Gross Exports. American Economic Review, 104
Lenzen, M., Pade, L. L., & Munksgaard, J. (2004). CO2 multipliers in multi-region input–output models. Economic Systems Research , 16(4)
Leontief, W. (1953). ’’Domestic production and foreign trade; the American capital position re-examined. Proceedings of the American philosophical society, 97(5)
Leontief, W., & Ford, D. (1970). Environmental repercussions and the economic structure: an input– output approach. The Review of Economics and Statistics, 52(3)
Levinson, A., & Taylor, S. M. (2008). Unmasking the pollution haven effect. International Economic Review, 49(1)
Los, B., Timmer, T., & De Vries, G. (2014). ’Tracing value-added and double counting in gross exports: comment
Low, P. (1992). ’Trade measures and environmental quality: the implication for Mexican exports. international trade and environment, Washington D.C.: World Bank Discussion papers, 159
Mani, M., & Wheeler, D. (1999). In Search of Pollution Havens? Dirty Industry in the World Economy, 1960-95. in Fredriksson (ed).
Milimet, D. L., & List, J. A. (2004). ’The case of the missing pollution haven hypothesis. Journal of regulatory economics, 26(3)
37 Muñoz, P., & Steininger, K. W. (2010). ’austria’s CO2 responsibility and carbon content of its
international trade. ecological economics, 69
Ratnayake, R. (1998). Do Stringent Environmental Regulations Reduce International
Competitiveness? Evidence from an Inter-industry Analysis. International Journal Economics of Business, 5(1)
SEEA (2003). Handbook of National Accounting: Integrated Environmental Accounting 2003.United Nations, European Commission, International Monetary Fund, Organisation for Economic
Cooperation and Development. UN, New York
Selden, T. M., & Song, D. (1994). Environmental quality and development: is there a Kuznets Curve for air pollution emissions? Journal of Environmental Economics and Management , 27(2)
Smarzynska Javorcik , B. K., & Wei, S. J. (2004). Pollution havens and foreign direct investment: dirty secret or popular myth? Contributions to Economic Analysis & Policy, 3(2)
Timmer, M. P. (2012). The World Input-Output Database (WIOD): Contents, Sources and Methods. WIOD Working Paper
Tobey, J. A. (1990). ’The effects of domestic environmental policies on patterns of world trade: an empirical test. kyklos, 43(2)
Van Beers, C., & Van den Bergh, J. C. J. M. (1997). An empirical Multi-country analysis of the impact of envirometnal regulations on foreign trade flows. kyklos, 50(1)
Walter, I. (1973). The pollution content of American trade. Western Economic Journal , 9(1) Wheeler, D., & Mody, A. (1992). International investment location decisions—the case of US firms. Journal of International Economics, 33
Wiedmann, T., Lenzen, M., Turner, K., & Barrett, J. (2007). Examining the global environmental impact of regional consumption activities—part 2: review of input–output models for the assessment of environmental impacts embodied in trade. ecological economics, 61(1)
Wiedmann, T., Wilting, H., Lutter, S., Palm, V., Giljum, S., Wadeskog, A., & Nijdam, D. (2009). Development of a Methodology for the Assessment of Global Environmental Impacts of Traded Goods and Services. ERA-NET SKEP Project
World Bank, 2008, World Development Indicators, Washington, DC: World Bank
38 7. Tables
Table 3a
Outputs developing countries and direct imports (in billions of dollars)
Year World output
Gowth world ouput Output developing industries Growth output developing industries Direct imports Growth direct imports 1995 55182 100 818 100 97 100 1996 56556 102 900 110 96 112 1997 56599 103 963 118 117 117 1998 56061 102 901 110 112 110 1999 58306 106 891 109 103 109 2000 60817 110 1024 125 131 124 2001 60203 109 1036 127 122 127 2002 62264 113 1091 133 128 134 2003 70209 127 1336 163 158 163 2004 79624 144 1708 209 225 206 2005 87526 159 2130 260 257 260 2006 95929 174 2643 323 337 320 2007 109398 198 3432 420 396 421 2008 122789 223 4277 523 465 529 2009 114158 207 4048 495 316 518 World output
= Value of output of all the industries in all the countries
Growth world
output = the percentage change in world output Output developing
industries = total output of 'dirty' industries of the developing countries Growth output
developing industries
= the percentage change in total output of 'dirty' industries of the developing countries
Direct imports = Value of the direct imports from developing to developed countries Growth direct
imports
39 Table 3b
Outputs developed countries and direct imports (in billions of dollars)
Year Output developed industries Growth output developed industries Direct imports Growth direct imports Direct percentage of output Growth direct percentage of output 1995 4446 100 97 100 2,2% 100 1996 4319 97 96 112 2,2% 102 1997 4239 95 117 117 2,8% 127 1998 4037 91 112 110 2,8% 127 1999 4192 94 103 109 2,5% 112 2000 4486 101 131 124 2,9% 134 2001 4208 95 122 127 2,9% 132 2002 4261 96 128 134 3,0% 137 2003 4760 107 158 163 3,3% 152 2004 5583 126 225 206 4,0% 185 2005 6224 140 257 260 4,1% 189 2006 6793 153 337 320 5,0% 227 2007 7603 171 396 421 5,2% 239 2008 8446 190 465 529 5,5% 252 2009 6722 151 316 518 4,7% 215
Output developed industries = Total value of the output of 'dirty' industries of the developed countries
Growth output developed industries = The percentage change in the total output of 'dirty' industries of the developed countries with 1995 as the base year
Direct imports = Value of the direct imports from developing to developed countries
Growth direct imports = The percentage change of the value of the direct imports from developing to developed countries with 1995 as the base year Direct percentage of ouput = The percentage of direct imports in the output of the
developed industries Growth direct percentage of output
40 Table 4
Indirect and direct imports (in billions of dollars)
Year Direct imports Growth direct imports Indirect imports Growth indirect imports 1995 97 100 28 100 1996 96 112 29 104 1997 117 117 34 122 1998 112 110 32 115 1999 103 109 30 108 2000 131 124 39 137 2001 122 127 37 132 2002 128 134 38 137 2003 158 163 50 177 2004 225 206 67 238 2005 257 260 67 238 2006 337 320 101 360 2007 396 421 125 444 2008 465 529 149 529 2009 316 518 111 395
Direct imports = Value of the direct imports from developing to developed countries
Growth direct imports = The percentage change of the value of the direct imports from developing to developed countries with 1995 as the base year
Indirect imports = Value of the indirect imports from developing to developed countries
