Next Article in Journal
Evaluating the Economic and Environmental Repercussions of the Price Paradox in Natural Resource Commodities: Market Drivers and Potential Challenges for Sustainable Development
Previous Article in Journal
Lithium Prospection in Portugal for E-Mobility and Solar PV Expansion
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Environmental Effects of Commodity Trade vs. Service Trade in Developing Countries

School of Accounting, Finance, Economics and Decision Sciences, Western Illinois University, 1 University Circle, Macomb, IL 61455, USA
Author to whom correspondence should be addressed.
Commodities 2022, 1(2), 115-126;
Received: 18 September 2022 / Revised: 31 October 2022 / Accepted: 2 November 2022 / Published: 7 November 2022


Increasing levels of carbon emissions have been a growing concern worldwide because of their adverse environmental effects. In that context, this paper examines the association between different categories of trade and carbon dioxide emissions. In particular, we analyze whether total trade, commodity trade, and service trade affect the environment differently. The analysis is based on panel data for 147 developing countries for the period from 1960 to 2020. Methodologically, the fixed-effects model, as suggested by the Hausman test, is used to examine the relationships. We present two main conclusions: (1) overall trade increases CO2 emissions, and (2) commodity trade contributes to higher levels of CO2 emissions than service trade. These results have important policy implications—climate change policies should target commodity trade sectors to help reduce environmental carbon emissions.

1. Introduction

Globalization has enabled countries to easily consume and exchange various goods and services beyond their borders. Trade is more open now than ever before because of the removal of barriers such as tariffs, quotas, etc., along with other liberal trade policies that countries have adopted to support their economic growth [1]. Trade liberalization allows countries to fully or partially remove certain types of international trade barriers, either in the form of tariffs or non-tariff barriers [2], enabling nations to focus on producing goods and services in which they have a comparative advantage [3]. That is, what a country trades depends on the opportunity cost, with lower opportunity costs resulting in improved efficiency [4,5]. In this process, a country could either help the environment by producing environmentally friendly products or hurt the environment by overproducing goods with negative externalities.
The impact of trade liberalization on the environment is a controversial issue. Some studies show that trade liberalization increases CO2 emissions that degrade the environment [6,7,8], while other empirical evidence states the opposite [9,10]. Trade results from the production of goods and services that increase CO2 emissions, but it is difficult to quantify the extent to which individual goods and services contribute to carbon emissions. For instance, the agriculture, manufacturing, mining, and service industries are different in nature and could affect the environment at different levels. Scant literature examines how various trade categories impact the environment at the national level [10,11,12]. A key issue with the single-country studies in the existing literature is that greenhouse gas (GHG) emissions are a global problem, i.e., an international externality. An individual country cannot do much about carbon emissions on its own, due to the free-rider problem [13]; hence, a broader analysis at the global level is necessary to understand such a global phenomenon.
Due to the ambiguities in the current literature about the trade–environment nexus and the limited scope of the analysis, this paper uses a global dataset to further examine how different trade structures increase CO2 gas emissions and negatively affect the environment. In particular, we analyze the disaggregated effects of total trade (percentage of GDP), commodity trade (using merchandise trade as a percentage of GDP as a proxy), and service trade (percentage of GDP) on CO2 emissions. In doing so, we fill the gap in the literature by providing empirical evidence that commodity trade and service trade impact the environment differently across all developing countries.
Disaggregating trade is vital to study how different sectors impact the environment in developing countries, because some developed countries outsource and/or offshore certain types of commodity production to developing countries for lower costs and less-stringent environmental policies [14]. As a result, more pollution is emitted in developing countries with lax environmental policies [15] than in developed countries. As such, developing countries are more vulnerable to environmental pollution than developed countries [16] if the pollution haven hypothesis holds [17,18], although it is possible to see the pollution halo hypothesis and improve environmental quality [19] if the use of environmentally friendly and clean technology is promoted in the production process. Existing literature on the trade–environment nexus focuses on total trade [9,12,20,21]. However, the analysis focusing on the impact of total trade on the environment underestimates the effects of trade categories with high negative impacts while overestimating the effects of other types with less harmful impacts. Hence, this study aims to investigate the environmental impacts of different trade categories in developing countries to show which trade categories adversely affect the environment.
This study contributes to the climate change literature by disaggregating total trade into commodities trade (using merchandise trade as a proxy) and service trade and estimating their role in carbon emissions—a proxy for environmental quality. Merchandise trade is defined as “goods which add or subtract from the stock of material resources of a country by entering (imports) or leaving (exports) its economic territory” [22]. Since merchandise trade includes most physical goods from agriculture, industry, manufacturing, etc., we use it as a proxy for commodity trade. In contrast, service trade includes services in information and communication, construction, distribution, educational, energy, environmental, financial, health, social, tourism, transport, business, and professional services [23]. Based on how commodity and service trade are measured, we hypothesized that commodity trade impacts the environment adversely, while service trade is less detrimental. Our results support these hypotheses. Furthermore, the results have important policy implications, as discussed by environmentalists, climate activists, and policymakers.
The rest of this paper is organized as follows: Section 2 examines related literature; Section 3 presents the data and methodology; Section 4 discusses the empirical results; and Section 5 concludes with a summary and policy implications.

2. Literature Review

This section presents an overview of the existing literature on how different types of trade impact the environment. We also present a summary of the literature highlighting the overall impact of trade liberalization on the environment. There is a vast body of literature on both theoretical and empirical analysis for estimating the effects of trade liberalization on the environment, using carbon dioxide as a proxy for environmental quality due to its highest share in greenhouse gas emissions [24]. Table 1 provides a summary of the representative literature.
In addition to the sectoral analyses listed in Table 1, the literature finds both positive and negative impacts of trade liberalization on the environment, mainly focusing on CO2 emissions. A study revealed that trade liberalization hurts the environment by increasing CO2 as the most-emitted and N2O as the least-emitted GHGs in the atmosphere [20]. This study also argues that the proportion of CO2 in the GHGs is 90%. Another study used CO2 emissions as the proxy for environmental quality with fixed- and random-effects methods on panel data from 2002 to 2016 [21]. The results showed a negative association between the environmental performance index and trade in the selected countries.
A similar study examined the associations between decomposed goods and services and CO2 and SO2 emissions using panel data for 179 countries segregated by the Organization for Economic Cooperation and Development (OECD) and non-OECD countries. Surprisingly, trade in goods reduced the CO2 intensity for OECD countries over 20 years of data when there was a boom in international trade. However, no significant impact of trade in services was reported. Moreover, OECD countries involved in international trade used clean technology to produce goods [9]. Another study decomposed merchandise trade into imports and exports for some developing countries between 1971 and 2013 [31]. The results showed that imports increased CO2 emissions, but exports and economic growth could improve the environment.
At the individual country level, a study examined the issues in China and found different effects of trade liberalization on the environment using a structural decomposition analysis on various tradable goods. It showed that the scale effect resulted in most carbon emissions being partially offset by the positive technique effect. However, the composition effect had no significant impact, resulting in overall environmental degradation [12]. The study also used CO2 as a proxy for environmental quality and found that trade liberalization deteriorates the environment. Similarly, another study in China used CO2 and SO2 as a proxy for environmental quality. Firm-level panel data from the period 1998–2007 were collected from the manufacturing sectors, along with data on pollution emissions and the degree of trade liberalization. Despite the positive impact of the technique effect, the combined negative impact of scale and composition offset this positive impact, resulting in overall environmental degradation [11].
In sum, the existing literature shows an inverse relationship between trade liberalization and the environment, using CO2 as a proxy for environmental quality. On the other hand, the literature also shows a positive effect of trade liberalization on the environment, but only in the short term [32]. We studied this issue further by analyzing the disaggregated effects of various components of trade, mainly focusing on commodity and service trade. In particular, we tested the following two hypotheses:
  • CO2 emissions are positively associated with trade volume.
  • Commodity trade contributes to higher levels of CO2 emissions than service trade.

3. Methodology

3.1. Empirical Model

We ran the three regression models represented by Equations (1)–(3) to test the hypotheses presented above.
l o g C O 2 i t = β 0 + β 1 T T R i t + β 2 G D P G r o w t h i t + β 3 F D I i t + β 4 l o g K i t + β 5 L F P i t + α i + ε i t
l o g C O 2 i t = β 0 + β 1 C T R r i t + β 2 G D P G r o w t h i t + β 3 F D I i t + β 4 l o g K i t + β 5 L F P i t + α i + ε i t
l o g C O 2 i t = β 0 + β 1 S T R i t + β 2 G D P G r o w t h i t + β 3 F D I i t + β 4 l o g K i t + β 5 L F P i t + α i + ε i t
The model specification follows the existing literature examining the determinants of environmental quality [20,32,33]. Our key outcome variable is the log of CO2 gas emissions—an environmental quality indicator. Total trade (TTR), commodity trade (CTR), and service trade (STR) are the key variables of interest included in Equations (1), (2), and (3), respectively. We also control for variables that the literature shows to be determinants of carbon emissions, i.e., annual GDP growth (GDPGrowth), foreign direct investment as a percentage of GDP (FDI), natural log of capital stock (logK), and the labor force participation rate for those aged 15 years and older (LFP). CO2 emissions are measured in kilotons and the capital stock (i.e., gross fixed capital formation) is measured in USD values. We take the natural log of CO2 emissions and the capital stock to linearize the regression model. Other control variables are expressed either in ratios or percentage forms, and no further transformations are performed. Table 2 provides further details of each variable.
Other notations in the regression model have their usual meaning— β i represents the parameters of interest, α i indicates the individual country-specific effects, and ε i t is the white noise for country i at time t. Consistent with our testable hypotheses, the three β 1 coefficients having positive estimates would mean that higher trade volumes increase CO2 emissions, while negative estimates would suggest the opposite.

3.2. Data Description

To compute the relationship between trade liberalization and environmental quality by segregating total trade into commodity versus service trade, we used CO2 emissions as a dependent variable, measured in kilotons, as used widely in the existing literature [9,31,34]. According to the World Development Indicators database, carbon dioxide emissions mostly consist of byproducts of the production and use of energy, including the burning of fossil fuels (i.e., their use in agriculture, manufacturing, industry, and other human activities). The measure excludes emissions from land use, e.g., deforestation. Estimates also exclude fuels supplied to ships and aircraft for international transport, because of the difficulty of apportioning the fuels among benefiting countries.
We had three key explanatory variables: total trade as a percentage of GDP was used for TTR; merchandise trade as a percentage of GDP was used for CTR; and service trade as a percentage of GDP was used for STR. Other control variables included in the model were FDI (net inflows as a percentage of GDP), GDP growth (annual percentage), log of capital (gross fixed capital formation in USD, 2015), and labor force participation rate (percentage of the total population aged 15+). We controlled for these variables because they affect environmental quality [21]. A study in the South Asian Association for Regional Cooperation (SAARC) countries found that capital and economic growth negatively impact environmental quality in the long term. Another study showed that FDI, capital, and economic growth positively impact the environment in the short term [32]. Moreover, the existing literature on FDI shows that it can support either the pollution haven hypothesis and increased CO2 [17,18] or the pollution halo hypothesis and decreased CO2 [19], depending on the sample. The literature shows that the population is also a key determinant of carbon emissions [35]. In particular, the population age and structure affect carbon emissions, primarily through the expansion of the labor force and subsequent overall economic growth [36]. To capture this effect, we included the labor force participation rate (i.e., the percentage of the total population aged 15+) in the model.
Our panel data for a sample of 147 developing countries were obtained from the World Development Indicators database maintained by the World Bank. The list of developing countries was obtained from the International Monetary Fund (IMF) website, which divides countries into two groups: advanced economies, and developing and emerging economies. According to the IMF, the main criteria used for country classification are (i) per capita income level, (ii) export diversification, and (iii) degree of integration into the global financial system. The IMF uses either sums or weighted averages of data for individual countries. However, the IMF’s statistical appendix for country classification explains that this is not a strict criterion, and other factors are considered in deciding the classification of countries. The World Bank [37] also designates nations and territories with a GNI of USD 12,535 or less as low-income economies, most of which are on the list of developing countries per the IMF classification. Based on this criterion and the availability of data in the World Bank database, the developing countries selected for this study comprised 147 out of the total 156, as listed in Appendix A. Table 3 shows the data summary.

3.3. Estimation Technique

We began the analysis with the standard fixed- and random-effects methods for panel data analysis presented above, as described in the literature [21,34]. To identify the appropriate estimation method for the given panel data, we used the Hausman specification test [38] after regressing each model using Equations (1)–(3). The test identifies whether the individual country-specific effects α i are correlated with the covariates (Xit).
H 0 : C o r r α i ,   X i = 0
H 1 :   C o r r α i ,   X i   0
If the null hypothesis is true, it indicates that the random-effects model would be appropriate for the data. However, if the alternative hypothesis is true, the fixed-effects model would be appropriate for the data. Table 4 presents the results from the Hausman specification test for each model represented in Equations (1)–(3).
For each model, the p-value is smaller than 0.05. Thus, we can reject the null hypothesis of random effects as an appropriate estimation method; that is, we use the results from the alternative hypothesis—the fixed effects—to examine our testable hypotheses. Intuitively, the fixed-effects model makes sense, as certain cross-country differences do not change over time, and it is necessary to control for them. These models were estimated with R software using the plm package.

4. Empirical Results and Discussion

This section presents the regression results comparing and contrasting the impacts of various trade categories on environmental quality.

4.1. Results Pertaining to Key Variables of Interest

The results obtained from estimating Equations (1)–(3) using the fixed-effects method are shown in Table 5. Each column represents the result for each model. The dependent variable is the natural log of carbon dioxide emissions, and our main explanatory variables are total trade (percentage of GDP), commodity trade (merchandise trade, percentage of GDP, as a proxy), and service trade (percentage of GDP). We can see that total trade has a positive and statistically significant association with carbon dioxide. Specifically, a 1% increase in trade increases carbon dioxide by 0.3%, which reflects a moderate increase in environmental pollution. This further illustrates that carbon dioxide emissions also increase as the total trade increases, adversely impacting environmental quality. These results are consistent with the existing literature [6,7,20,21,32]. This supports our first testable hypothesis.
We also see similar results for commodity trade, i.e., positive and significant relationship with carbon dioxide. As commodity trade increases by 1%, it increases CO2 by 0.4%, which reflects that commodity trade also results in deteriorating environmental quality at a large scale. While these results are consistent with the existing literature, one study showed a contradictory result, where trade in commodities reduced the CO2 emissions and improved the environment in OECD countries [9]. A possible reason for this is the increase in international trade due to technological improvement, especially in developing countries. Furthermore, service trade did not show a significant impact, as expected. There was a positive but statistically insignificant relationship between service trade and carbon dioxide emissions. Existing evidence shows similar results, with an insignificant impact of service trade on CO2 emissions [9]. The results indicate that most of the negative effects of trade on the environment come from commodity trade, rather than from service trade. This means that an increase in service trade does not directly increase carbon dioxide. This makes sense based on the definition of service trade as trade in services only, which are non-physical in nature. Hence, service trade does not deteriorate the environment, even if it does not contribute to improving the environmental quality. These results are consistent with the existing literature; unlike commodity trade, which is potentially harmful to the environment due to the increased output of tangible goods, trade in services is less harmful to the environment [29,30]. These results support our second testable hypothesis.

4.2. Results Pertaining to Control Variables

The overall results presented in Table 4 show a significant result based on the values of R2 and adjusted R2 (approximately 85%). The F-statistic also shows a highly significant value for all of the models. Most of the independent variables also show a high correlation with the dependent variables, making the overall estimation robust.
GDP growth shows a negative association with carbon dioxide. This further highlights that as GDP growth increases, carbon dioxide decreases, improving the environmental quality. The results are consistent with the literature on the Kuznets curve hypothesis, which suggests that environmental pollution increases at the beginning of economic growth, but when it passes a certain level of income the economic growth helps to improve environmental quality [39].
FDI shows a negative and significant relationship with carbon dioxide emissions when we consider total trade and commodity trade, but the relationship is statistically insignificant when service trade is considered. Existing literature shows that FDI can increase CO2 emissions based on the pollution haven hypothesis [20] or decrease them based on the pollution halo hypothesis [19]. Based on our results, it is possible that the foreign investments are used to move towards the use of environmentally friendly and clean technology in the production process, which would improve the environmental quality [34]. However, these findings require a separate investigation that is outside the focus of this paper, because the effects of FDI could be different based on the stringency of the receiving countries’ environmental policies [40,41] and the income levels of the countries [42,43], while the effect could also be bidirectional [44] and depend on the time horizon [33].
Capital stock also plays an important role in determining carbon dioxide emissions. The coefficients of log-capital are consistently positive and statistically significant across all specifications, indicating a significant association with carbon emissions. A potential explanation for this observation is the positive role of capital in increasing output and economic activity [45,46], which would ultimately increase CO2 emissions in the initial stages of growth. On the other hand, the labor force participation rate negatively and significantly affects carbon emissions. However, these results are inconsistent with the existing literature [36], potentially due to differences in the samples of countries included in the analysis. There is limited literature examining the role of variables such as capital and labor, and a detailed study on this topic is necessary to understand the inconsistent results.
By pooling the results from the three models, the data provide statistical evidence that both the total trade and commodity trade have a positive and statistically significant relationship with carbon dioxide emissions. However, service trade does not have a significant relationship with CO2 emissions. This implies that higher commodity trade will increase carbon dioxide emissions and deplete environmental quality.

4.3. Study Limitations and Future Prospects

This paper analyzed the impacts of commodity versus service trade on environmental outcomes. While disaggregating total trade into commodity and service trade provides us with a way to understand the issue further, there are some limitations. It would be ideal to analyze the impacts of other disaggregated trade components, such as hard commodities and soft commodities, including agriculture, manufacturing, mines and minerals, etc. This would provide a better picture to find and compare the major sources that release the most carbon dioxide in the atmosphere. Additionally, using the firm-level data for a few developing countries would allow us to empirically analyze the decomposition of trade commodities to capture the in-depth impacts of trade. Detailed analysis of the individual commodities is beyond the scope of this paper. Furthermore, many countries’ economic structures have been changing since the COVID-19 pandemic [47,48]. How the pandemic affects their economic structures and, ultimately, their carbon emissions will be a contentious issue. Future research should focus on these issues.

5. Summary, Conclusions, and Policy Implications

This paper’s main objective was to find and distinguish the impacts of international trade on the environmental quality of developing countries, which are more susceptible to climate change. The goal was to see whether segregating trade into commodity and service trade would affect the environmental quality differently. We postulated two testable hypotheses: (1) overall trade increases CO2 emissions, and (2) commodity trade contributes to higher levels of CO2 emissions than service trade. We tested these hypotheses by examining the effects of total trade (as a percentage of GDP), commodity trade (using merchandise trade as a percentage of GDP as a proxy), and service trade (as a percentage of GDP) on CO2 emissions (a proxy for environmental quality). We used panel data from 147 developing countries for the period 1960–2020 and analyzed the data using the fixed-effects model as suggested by the Hausman test. We estimated three models separately to compare and contrast the effects of different trade categories. The results show that an increase in total trade negatively impacts the environment, mainly contributed by commodity trade rather than service trade. Thus, it can be inferred from the results that commodities release a higher share of carbon dioxide emissions in contrast to trade in services. Commodity trade involves tangible goods such as manufacturing, agriculture, mines and minerals, stones and precious elements, etc., which emit several harmful gasses as byproducts during their production processes. On the other hand, service trade is trade in services such as human resources and skills in information technology, engineering, etc., showing no statistically significant effect on the environment. Hence, our results support both of our testable hypotheses.
From a policy perspective, appropriate tools are necessary to control harmful gas emissions from the manufacturing and industrial sectors, which make up most of the trade in commodities. To do so, it is imperative to first identify the commodities produced by developing countries. One way to curb carbon emissions is to use environmentally friendly technology to produce hard commodities such as metals and iron and to extract elements such as coal, chromite, and minerals. Governments could also play an appropriate role in providing subsidies to invest more in environmentally friendly commodities through environment and climate change ministries in collaboration with international NGOs. Moreover, countries could devise a carbon pricing policy to either increase the tax on each unit of carbon emission generated or promote the production of commodities with lower carbon emissions. According to world trade figures, countries such as China, India, Russia, etc., trade the highest amounts of commodities. Therefore, they should act immediately and pass on the appropriate strategies to other countries to help them reduce carbon emissions and improve environmental quality.

Author Contributions

Conceptualization, S.G. and M.Z.S.; methodology, S.G. and M.Z.S.; software, M.Z.S.; validation, S.G. and M.Z.S.; formal analysis, S.G. and M.Z.S.; investigation, S.G. and M.Z.S..; resources, S.G. and M.Z.S.; data curation, M.Z.S.; writing—original draft preparation, S.G. and M.Z.S.; writing—review and editing, S.G. and M.Z.S. All authors have read and agreed to the published version of the manuscript.


This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data used in this study are publicly available at the World Bank Database.


The authors are grateful to the four anonymous referees and the journal editor for their helpful comments that helped improve the quality of this paper.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Table A1. List of 147 developing countries.
Table A1. List of 147 developing countries.
AfghanistanSouth SudanKazakhstanRomania
AlbaniaChileKenyaRussian Federation
AngolaColombiaSri LankaSamoa
Antigua and BarbudaComorosLiberiaSao Tome and Principe
ArgentinaCongo, Dem. Rep.LibyaSaudi Arabia
ArmeniaCosta RicaMadagascarSenegal
BahamasDominican RepublicMaldivesSierra Leone
BahrainEcuadorMaliSolomon Islands
BangladeshEgyptMarshall IslandsSomalia
BarbadosEl SalvadorMauritaniaSouth Africa
BelarusEquatorial GuineaMauritiusSt. Kitts and Nevis
BelizeEritreaMexicoSt. Lucia
BeninEswatiniMoldovaSt. Vincent and the Grenadines
Bosnia and HerzegovinaGabonMoroccoSyria
BotswanaThe GambiaMozambiqueTajikistan
BrazilNorth MacedoniaNamibiaTanzania
Brunei DarussalamGeorgiaNauruThailand
Burkina FasoGrenadaNicaraguaTonga
BurundiGuatemalaNigerTrinidad and Tobago
Cabo VerdeGuinea-BissauOmanTurkey
Central African RepublicHondurasPalauTunisia
ChadIndiaPapua New GuineaUganda
LebanonJordan NigeriaVietnam


  1. Kowalski, P. Impact of Changes in Tariffs on Developing Countries’ Government Revenue; OECD Trade Policy Papers; OECD: Paris, France, 2005. [Google Scholar]
  2. Lee, E. Trade Liberalization and Employment; Working Paper No. 5; United Nations, Department of Economic and Social Affairs: New York, NY, USA, 2005; Available online: (accessed on 31 October 2022).
  3. Lopez, J. Comparative advantage, economic growth and free trade. Rev. Econ. Contemp. 2005, 9, 313–335. [Google Scholar]
  4. Goldin, I. Comparative Advantage: Theory and Application to Developing Country Agriculture; OECD Development Centre Working Papers, No. 16; OECD Publishing: Paris, France, 1990. [Google Scholar]
  5. Costinot, A.; Donaldson, D.; Vogel, J.; Werning, I. Comparative Advantage and Optimal Trade Policy. Q. J. Econ. 2015, 130, 659–702. [Google Scholar] [CrossRef]
  6. Duan, K.; Cao, M.; Malim, N.A.K. The Relationship between Trade Liberalization, Financial Development and Carbon Dioxide Emission—An Empirical Analysis. Sustainability 2022, 14, 10308. [Google Scholar] [CrossRef]
  7. Kim, D.-H.; Suen, Y.-B.; Lin, S.-C. Carbon dioxide emissions and trade: Evidence from disaggregate trade data. Energy Econ. 2018, 78, 13–28. [Google Scholar] [CrossRef]
  8. Shahbaz, M.; Nasreen, S.; Ahmed, K.; Hammoudeh, S. Trade openness–carbon emissions nexus: The importance of turning points of trade openness for country panels. Energy Econ. 2017, 61, 221–232. [Google Scholar] [CrossRef][Green Version]
  9. Ma, T.; Wang, Y. Globalization and environment: Effects of international trade on emission intensity reduction of pollutants causing global and local concerns. J. Environ. Manag. 2021, 297, 113249. [Google Scholar] [CrossRef]
  10. Shuquan, H. Impact of Trade on Environmental Quality: A Business Ethics Perspective and Evidence from China. Bus. Ethics Leadersh. 2019, 3, 43–48. [Google Scholar]
  11. He, L.-Y.; Lin, X.; Liu, Q. How did free trade reshape the transitional China? Evidence from heterogeneous exporters and firm-level pollution emissions. Emerg. Mark. Financ. Trade 2019, 56, 1651–1676. [Google Scholar] [CrossRef]
  12. Zhang, Y. Scale, Technique and Composition Effects in Trade-Related Carbon Emissions in China. Environ. Resour. Econ. 2011, 51, 371–389. [Google Scholar] [CrossRef]
  13. Jeffrey, F.A. Environmental Effects of International Trade; HKS Faculty Research Working Paper Series; John F. Kennedy School of Government, Harvard University: Cambridge, MA, USA, 2009. [Google Scholar]
  14. Walter, I.; Ugelow, J.L. Environmental Policies in Developing Countries. Technol. Dev. Environ. Impact 1979, 8, 102–109. [Google Scholar]
  15. Neumayer, E. Pollution Havens: An Analysis of Policy Options for Dealing with an Elusive Phenomenon. J. Environ. Dev. 2001, 10, 147–177. [Google Scholar] [CrossRef]
  16. UNDP; UNFCCC. Nationally Determined Contributions Global Outlook Report 2019; NDC: 2019. Available online: (accessed on 31 October 2022).
  17. Singhania, M.; Saini, N. Demystifying pollution haven hypothesis: Role of FDI. J. Bus. Res. 2020, 123, 516–528. [Google Scholar] [CrossRef] [PubMed]
  18. Gyamfi, B.A.; Bein, M.A.; Udemba, N.E.; Bekun, V.F. Investigating the pollution haven hypothesis in oil and non-oil sub-Saharan Africa countries: Evidence from quantile regression technique. Resour. Policy 2021, 73, 102119. [Google Scholar] [CrossRef]
  19. Zhu, H.; Duan, L.; Guo, Y.; Yu, K. The effects of FDI, economic growth and energy consumption on carbon emissions in ASEAN-5: Evidence from panel quantile regression. Econ. Model. 2016, 58, 237–248. [Google Scholar] [CrossRef][Green Version]
  20. Onwachukwu, C.I.; Yan, K.-M.I.; Tu, K. The causal effect of trade liberalization on the environment. J. Clean. Prod. 2021, 318, 128615. [Google Scholar] [CrossRef]
  21. Raza, A.; Sui, H.; Jermsittiparsert, K.; Żukiewicz-Sobczak, W.; Sobczak, P. Trade Liberalization and Environmental Performance Index: Mediation Role of Climate Change Performance and Greenfield Investment. Sustainability 2021, 13, 9734. [Google Scholar] [CrossRef]
  22. United Nations. International Merchandise Trade Statistics—Concepts and Definitions; United Nations, Department of Economic and Social Affairs: New York, NY, USA, 2004. [Google Scholar]
  23. World Trade Organization. World Trade Report; World Trade Organization. 2015. Available online: (accessed on 31 October 2022).
  24. USEPA. Report on the Environment: Greenhouse Gases; United States Environmental Protection Agency: Washington, DC, USA, 2020. [Google Scholar]
  25. Lankoski, J. Environmental Effects of Agriculture Trade Liberalization and Domestic Agricultural Policy Reforms; UNCTAD: Geneva, Switzerland, 1997; Available online: (accessed on 31 October 2022).
  26. Balogh, J.M.; Jámbor, A. The Environmental Impacts of Agricultural Trade: A Systematic Literature Review. Sustainability 2020, 12, 1152. [Google Scholar] [CrossRef][Green Version]
  27. United States Environmental Protection Agency. Considering the Full Array of Impacts on Human Health and Environment by Products, Services and Supply Chain; United States Environmental Protection Agency: Washington, DC, USA, 2021. [Google Scholar]
  28. United Nations Conference on Trade and Development. The Role of Exports in Manufacturing Pollution in Sub-Saharan Africa, South Asia: Towards a Better Trade-Environment Governance; UNCTAD: Geneva, Switzerland, 2021; Available online: (accessed on 31 October 2022).
  29. Sun, C.; Ma, T.; Ouyang, X.; Wang, R. Does Service Trade Globalization Promote Trade and Low-Carbon Globalization? Evidence from 30 Countries. Emerg. Mark. Financ. Trade 2019, 57, 1455–1473. [Google Scholar] [CrossRef]
  30. Feng, R.; Shen, C.; Huang, L.; Tang, X. Does trade in services improve carbon efficiency?—Analysis based on international panel data. Technol. Forecast. Soc. Chang. 2021, 174, 121298. [Google Scholar] [CrossRef]
  31. Appiah, K.; Worae, T.A.; Yeboah, B.; Yeboah, M. The causal nexus betwen trade openness and environmental pollution in selected emerging economies. Ecol. Indic. 2022, 138, 108872. [Google Scholar] [CrossRef]
  32. Sun, H.-P.; Tariq, G.; Haris, M.; Mohsin, M. Evaluating the environmental effects of economic openness: Evidence from SAARC countries. Environ. Sci. Pollut. Res. 2019, 26, 24542–24551. [Google Scholar] [CrossRef]
  33. Sharma, S.S. Determinants of carbon dioxide emissions: Empirical evidence from 69 countries. Appl. Energy 2011, 88, 376–382. [Google Scholar] [CrossRef]
  34. Acheampong, A.O.; Adams, S.; Boateng, E. Do globalization and renewable energy contribute to carbon emissions mitigation in Sub-Saharan Africa? Sci. Total. Environ. 2019, 677, 436–446. [Google Scholar] [CrossRef] [PubMed]
  35. Alam, M.M.; Murad, W.M.; Noman, A.H.M.; Ozturk, I. Relationships among Carbon Emissions, Economic Growth, Energy Cnsumption and Population Growth: Testing Environmental Kuznets Curve Hypothesis for Brazil, China, India and Indonesia. Ecol. Indic. 2016, 70, 477–479. [Google Scholar] [CrossRef]
  36. Zhu, Q.; Peng, X. The impacts of population change on carbon emissions in China during 1978–2008. Environ. Impact Assess. Rev. 2012, 36, 1–8. [Google Scholar] [CrossRef]
  37. Serajuddin, U.; Hamadeh, N. New World Bank Country Classifications by Income Level: 2020–2021; World Bank: Washington, DC, USA, 2021. [Google Scholar]
  38. Hausman, A.J. Specification Tests in Econometics. Econometrica 1978, 46, 1251–1271. [Google Scholar]
  39. Dinda, S. Environmental Kuznets Curve Hypothesis: A Survey. Ecol. Econ. 2004, 49, 431–455. [Google Scholar] [CrossRef][Green Version]
  40. Cole, M.A.; Fredriksson, P.G. Institutionalized pollution havens. Ecol. Econ. 2009, 68, 1239–1256. [Google Scholar] [CrossRef][Green Version]
  41. Mert, M.; Bölük, G. Do foreign direct investment and renewable energy consumption affect the CO2 emissions? New evidence from a panel ARDL approach to Kyoto Annex countries. Environ. Sci. Pollut. Res. 2016, 23, 21669–21681. [Google Scholar] [CrossRef]
  42. Pao, H.-T.; Tsai, C.-M. Income and CO2 emissions: Evidence from panel unit root and cointegration tests. Energy Policy 2009, 37, 413–423. [Google Scholar]
  43. Jaunky, V.C. The CO2 emissions-income nexus: Evidence from rich countries. Energy Policy 2011, 39, 1228–1240. [Google Scholar] [CrossRef]
  44. Pao, H.-T.; Tsai, C.-M. Multivariate Granger causality between CO2 emissions, energy consumption, FDI (foreign direct investment) and GDP (gross domestic product): Evidence from a panel of BRIC (Brazil, Russian Federation, India, and China) countries. Energy 2011, 36, 685–693. [Google Scholar] [CrossRef]
  45. Chow, G.C. Capital Formation and Economic Growth in China. Chin. Econ. Hist. Since 2017, 1949, 1186–1221. [Google Scholar]
  46. Solow, R.M. Technical Progress, Capital Formation, and Economic Growth. Am. Econ. Rev. 1962, 52, 76–86. [Google Scholar]
  47. Vidyaa, C.T.; Prabheesh, K.P. Implications of COVID-19 Pandemic on the Global Trade Networks. Emerg. Mark. Financ. Trade 2020, 56, 2408–2421. [Google Scholar] [CrossRef]
  48. Duan, W.; Zhu, S.; Lai, M. The impact of COVID-19 on China’s trade and outward FDI and related countermeasures. J. Chin. Econ. Bus. Stud. 2020, 18, 355–364. [Google Scholar] [CrossRef]
Table 1. Review of literature for agriculture, manufacturing, and service trade.
Table 1. Review of literature for agriculture, manufacturing, and service trade.
Title and Author(s)ObjectivesResults and Findings
Trade in Agriculture
Environmental effects of agricultural trade liberalization [25].Examined environmental impacts due to agricultural trade liberalization and domestic policy reforms through existing literature.Indirect impacts were caused due to changes in location, intensity, product mix, and agricultural technology, while direct impacts included pollution due to the transportation of agricultural products, as well as the potential migration of harmful species of plants, animals, and insects that could disrupt the food chain.
The environmental impacts of agricultural trade: A systematic literature review [26].Investigating the relationships between agricultural trade and the environment based on existing literature in international economics.Focused on local (land use) rather than global emissions (GHGs). A negative relationship between agricultural trade and the environment (pollution) was found, but some studies show a positive relationship, while very few studies could not find any relationship.
Considering the full array of impacts on human health and the environment by products, services and supply chain [27]: US Environmental Protection Agency Report.Manufacturing trade-offs in the environment, health, and resource use.Selection of bioproducts by the producers reduces fossil fuel extraction and inputs as well as GHGs, but the growing and harvesting of bio-feedstocks may also pollute water sources and degrade soil quality.
The role of exports in manufacturing pollution in sub-Saharan Africa, South Asia: toward better trade-environment governance [28]:UNCTAD Report (2021).To assess the export-associated pollution by the manufacturing sector in sub-Saharan Africa and South Asia.Identified the public and private environmental governance opportunities to achieve more sustainable production linked to trade. Should follow sustainable practices despite adding value in the products in the manufacturing sector.
Trade in Services
Does Service Trade Globalization Promote Trade and Low-Carbon Globalization? Evidence from 30 Countries [29].Investigated the effects of service trade globalization on low-carbon globalization through the Unified Efficiency Index and Energy-Environmental Performance Index in 30 countries from 1980 to 2013 using a tobit model.Service trade openness showed positive effects on both energy and CO2 emission efficiency that have further improved with time, which is good for the environment. Emerging service sectors promoted the improvement of energy and CO2 emission efficiency, while the traditional sectors hindered the efficiency improvement. There existed a “catch-up” effect between less-developed countries and developed countries in terms of energy and CO2 emission efficiency.
Does trade in services improve carbon efficiency?—Analysis based on international panel data [30].Finding the theoretical and empirical impacts of service trade on carbon efficiency using panel data for 55 countries from 2001 to 2015 using a slacks-based measure model with a global Malmquist–Luenberger index. Carbon efficiency was improved due to service trade, which was greater in exports than imports. Each country showed a different effect for exports and imports of services.
Table 2. Description of variables.
Table 2. Description of variables.
VariableDescriptionUnit of Measurement
logCO2Log of carbon dioxide emissions Kilotons
TTRTotal tradeAs a percentage of GDP
CTRCommodity tradeAs a percentage of GDP
STRService tradeAs a percentage of GDP
GDPGDP growthAnnual percentage
FDIForeign direct investmentNet inflows as percentage of GDP
logKLog of capitalGross fixed capital formation
Labor force participation rate
Individual country-specific effects
Percentage of total population aged 15+
εError term
itCountry and time period
Table 3. Descriptive statistics.
Table 3. Descriptive statistics.
Log of CO23.52391.15090.56437.01347821
Total trade69.325037.33520.0210245.36946063
Commodity trade55.783132.35712.7226311.51147056
Service trade21.354318.83020.6250143.98055121
GDP growth3.77426.9498−64.0471149.97306870
Foreign direct investment3.01665.8095−55.2341161.82386172
Labor force participation rate62.026812.111031.440090.32004185
Log of capital9.63310.88795.748111.91314037
Notes: (1) there are many missing observations, which is why the total count for each variable is different; (2) we removed the outliers before running the final regression analysis.
Table 4. Hausman test results.
Table 4. Hausman test results.
ModelChi-Squared Statisticp-Value
Total trade as a percentage of GDP (TTR)34.26<0.0001
Commodity trade as a percentage of GDP (CTR)62.722<0.0001
Service trade as a percentage of GDP (STR)63.973<0.0001
Table 5. Results obtained from the fixed-effects estimation.
Table 5. Results obtained from the fixed-effects estimation.
Dependent Variable:
Total trade (% of GDP)0.003 ***
Commodity trade (% of GDP) 0.004 ***
Service trade (% of GDP) 0.001
GDP growth rate−0.010 ***−0.011 ***−0.010 ***
FDI (% of GDP)−0.012 *** −0.009 ***−0.002
Log capital1.024 *** 1.011 ***0.984 ***
Labor force participation rate−0.008 ***−0.008 ***−0.009 ***
Adjusted R20.8550.8540.842
F-statistic3057.205 ***3051.155 ***2560.033 ***
(df = 5; 2546)(df = 5; 2568)(df = 5; 2363)
Note: *** p < 0.01.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Saeed, M.Z.; Ghimire, S. Environmental Effects of Commodity Trade vs. Service Trade in Developing Countries. Commodities 2022, 1, 115-126.

AMA Style

Saeed MZ, Ghimire S. Environmental Effects of Commodity Trade vs. Service Trade in Developing Countries. Commodities. 2022; 1(2):115-126.

Chicago/Turabian Style

Saeed, Mohammad Zohaib, and Shankar Ghimire. 2022. "Environmental Effects of Commodity Trade vs. Service Trade in Developing Countries" Commodities 1, no. 2: 115-126.

Article Metrics

Back to TopTop