Do clean development mechanisms promote sustainable development in Brazil? a cross-sectoral investigation

1. Introduction

Since the Kyoto Protocol in 1997, a number of instruments have been established in order to help developed countries fulfilling their goals in terms of reducing their combined greenhouse gases (GHG) emissions (Michaelowa, Shishlov, & Brescia, 2019; UNFCCC, 2011). Clean Development Mechanism (CDM) is one of these instruments, accumulating a total of over 4,000 projects worldwide. It consists of a way to allow developed nations to carry out sustainable projects in developing countries, so that the GHG emissions reduction generated by these projects is accounted toward the developed country’s emission goals – one of the so-called Flexibility Mechanisms. With this instrument, the actual reduction outcomes in developing countries obtain equivalence in carbon credits to be used by developed nations that had emission reduction targets (Lopes, 2002; Franguetto & Gazani, 2002; Pereira, 2002).

An interesting issue about CDMs is that they should go way beyond reducing GHG emissions, encouraging sustainable development (SD) in a much broader way. Each country is free to create its own specific rules. Brazil, for instance, who brought the very first CDM project registration worldwide, requires these projects to specify how they intend to contribute to income generation, job creation, technological development, regional integration and articulation with other sectors. It is worth noting that CDMs are not legally required to satisfy all these criteria, but they must mention which contributions they intend to make. Nonetheless, Brazil ranks among the top three countries in number of CDMs: 463 projects have been registered between 2004 and 2020, representing 4% of all projects being developed worldwide (Unep Risoe, 2020).

Given this scenario, in this paper we seek to verify the extent to which these projects intend to contribute to sustainable development (SD) in Brazil, considering a broader definition of SD as environmental, social and economic aspects (Brundtland, 1987; Olsen & Fenhann, 2008). Specifically, we seek to identify which activities contribute the most to SD, i.e. which sectors are more or less beneficial and what is the specific impact of the projects conducted in each sector on SD in general; also, we aim at pinpointing the extent to which CDM projects intend to go beyond GHG emission reduction aspects.

To this end, we qualitatively analyze 219 projects representing 47% of total projects registered in Brazil from 2004 to 2020. Moreover, we were cautious to select a very diversified sample, including projects in different sectors throughout time. As far as we can tell, we are the first to scrutinize this number of diversified projects in Brazil. The analyses were based on “Annex III” reports, a qualitative information document filled out by the CDM proponents. This information was standardized and classified, with the purpose of comparing projects and sectors in Brazil (MCTIC, 2021). Based on our results, we derive lessons from Brazil and propose paths for public policy toward encouraging SD.

2. CDMs across different countries: what is the previous evidence on sustainable development?

A CDM is a voluntary program aiming to help developed countries that have emission reduction target to fulfill their goals by building projects in developing countries, while promoting local SD (UNFCCC, 2011; Mehling & Mielke, 2012; Shishlov & Bellassen, 2016).

As described in Article 12 of the Kyoto Protocol, in order to obtain a CDM project approval, in addition to the GHG emission reduction target, proponents should indicate measures linked to the SD of the place where the project would be implemented. However, as SD definitions are very subjective, with different interpretations, the Kyoto Protocol asks project proponents for a formal description of future activities related to SD (Dirix, Peeters, & Sterckx, 2016; Benites-Lazaro & Mello-Théry, 2017; UNFCCC, 2011). Thus, the CDM host country determines the particular rules, defining the meaning of SD. However, as definitions differ from country to country, UNFCCC proposes ways to guide these understandings (UNFCCC. United Nation Framework Climate Change, 2020).

There are several definitions regarding the SD concept, and there is no single global standard for them in terms of CDM projects. Because of different approaches Olsen and Fenhann (2008), claim that SD definition should be the same worldwide, with an international standard for all CDMs. Nonetheless, the ideas considered by countries are based on the UN Brundtland Report, “Our Common Future” (Brundtland, 1987), which admits that a sustainable enterprise is one that simultaneously provides economic, social and environmental benefits. Thus, a country’s development cannot be limited to social aspects and economic growth but must incorporate complex relationships between future societies and the environment (Elkington, 2004; Sachs, 2005).

A key question about CDMs is whether they are successfully promoting SD in host countries, and how SD can be encouraged. Previous literature indicates that the results of CDM projects in terms of SD in developing countries are not clear (Kim, 2004; Dirix et al., 2016). Moreover, prior research indicates that less than 1% of global projects contribute to SD in developing countries (Sutter & Parreño, 2007). It also states that SD activities in developing countries can be significantly improved (Olsen & Fenhann, 2008) as they do not seem to substantially contribute, for instance, in reducing poverty in CDM host countries (Dirix et al., 2016). Similarly, Siedenburg, Brown, and Hoch (2016) evaluated GHG emission reduction projects in Madagascar and Mali and concluded that the number of SD initiatives related to CDM has been low, with several obstacles to be overcome, particularly in rural areas. Perhaps the most important of these obstacles is the lack of a universal definition of SD.

Such a lack of SD definition has also dominated previous empirical literature, as each group of scholars tends not to focus on a single country, let alone Brazil, and also not emphasizing specific aspects, such as technological transfer, job, income and education generation. As each nation is free to set its own guidelines and requirements for SD, the concept becomes very subjective and ad-hoc, depending on each country’s perceptions and needs. This is evident, for example, in the comparison of Brazilian and Peruvian hydroelectric projects as stated by Cole and Roberts (2011), indicating that the former’s propositions regarding SD are more subjective than the latter’s. In this sense, this paper proposes to analyze Brazil specificities related to SD in CDM projects.

Uddin, Blommerde, Taplin, and Laurence (2015) analyzed 30 Chinese projects focusing on methane reduction in coal mines and concluded that variables related to technology transfer stood out compared to others. Nonetheless, this is not always the case, as only 39% of all global CDMs encompass technology and/or knowledge transfer, and most of those are large-scale projects (Seres & Haites, 2008), importing technology namely from countries like Japan, Germany, USA, France and England (Dechezleprêtre, Glachant, & Ménière, 2009). As stated by Michaelowa et al. (2019), technology transfer depends on the CDM project activity, with some sectors being more dependent on foreign technology than others. This is the case described in Costa-Júnior, Pasini, and Andrade (2013), whereby technology import does not frequently occur in Brazil because most of the technology required for CDM it is locally developed.

Some other research studies focus on the contribution of CDM projects in terms of its ability to generate work and income or develop educational programs. This is the case of Olsen and Fenhann (2008), who advocate that “generation of work and income” is the aspect that contributes the most to SD in CDM projects, as this had been found in 68% of their analyzed projects. Benites-Lazaro and Mello-Théry (2017) find a very similar result in their dynamic panel analyses, although their aim was to assess the relationship between SD projects and their stakeholders, and not the particularities of SD in each of the CDM projects. In the same vein, Mori-Clement and Bednar-Friedl (2019) and Mori-Clement (2019) find that CDMs encourage income generation, while only those in the hydro sector contribute to reducing poverty at the municipal level in Brazil (Mori-Clement, 2019).

This diversity of understanding of SD definitions and lenses leads to a secondary issue: the lack of incentives for project proponents to promote local SD, primarily because of monitoring and enforcement deficiencies. The rationale here is that if very little is done to ensure that SD takes place within host countries, project proponents have little incentive to promote it. This helps explaining why many CDM projects do not contribute satisfactorily to SD in host countries (Olsen & Fenhann, 2008; Olsen, Arens, & Mersmann, 2018). This is not surprising, after all, it is not easy to create an incentive system or enforcement mechanisms on something that is not unanimously well defined to begin with.

3. CDM requirements in Brazil

The very first CDM project to be registered worldwide was a Brazilian project, “Novagerar”, a landfill company which promoted local SD through energy generation for schools and surrounding communities, back in 2004. This example has been followed in a rapidly growing number of projects ever since. To date, Brazil ranks among the top three countries in number of CDMs: 344 projects have been registered there between 2004 and 2020, representing 4% of all projects being developed worldwide (Unep Risoe, 2020).

As mentioned before, host country is free to set its own regulations toward SD implementation. In Brazil, these are described in a document named “Annex III”. It contains detailed guidelines to describe how the project intends to contribute to local SD, promoting economic, environmental and social improvements, in line with Olsen and Fenhann (2008). These standards were defined by the Inter-ministerial Commission on Global Climate Change (CIMGC), through Resolution no. 01, 2003. These documents are made public through the website of the Ministry of Science, Technology and Innovation (MCTIC, 2020).

CDM proponents are required to fill out the guide in Annex III, in a highly qualitative manner, specifying how they intend to contribute to specific criteria (Table 1).

This provides preliminary evidence in favor of CDM projects developed in Brazil fulfilling SD in a broader way, not only in terms of environmental impacts. Nonetheless, the numerous works mentioned before raise questions about the effectiveness of these projects in practice since each proponent is free to choose which sustainability aspects to focus on. Hence, in the coming sections, we assess the extent to which CDM projects carried out in multiple sectors in Brazil intend to fulfill each of these requirements and play an active role in encouraging SD in that country.

4. Research design and methodology

This research is qualitative and exploratory, and relies on a collection of Brazilian CDM project reports, whose information is provided in “Annex III”. These documents are available and dispersed throughout the Ministry of Science, Technology, Innovations and Communications website (MCTI, 2020; Unep Risoe, 2020; UNFCCC. United Nation Framework Climate Change, 2020).

Noteworthy, the information in “Annex III” is quite subjective, differing from firm to firm. Not all documents follow the same pattern and certain proponents only describe the environmental standards required by law, such as EIA/RIMA Brazilian environmental licenses. These environmental requirements do not belong in the scope of “Annex III”, which should refer to what the company proposes in addition to the environmental effects, aiming at promoting SD.

Our population comprised 463 Brazilian projects approved under the CDM program between November, 2004 and January, 2020. Out of those, 219 reports were selected to be analyzed or 47% in total (Table 2).

In order to guide our choice, we sought to gather a representative sample, with a sufficient number of projects, and containing a considerable share of total projects by sector and by year [1]. We also had discretion to involve projects with different profiles, including a few unusual projects, which present innovative ideas, like power generation out of rice husks. It is worth noting that while some sectors counted only for one project, this proportion is in line with their representativeness in the general population. Table 3 below shows the breakdown of our selection by sector.

The methodology of this study is quite different compared to previous research, like those by Olsen and Fenhann (2008) and Benites-Lazaro and Mello-Théry (2017). First of all, the information in this work refers to a different period (November, 2004 to January, 2020). Additionally, it considers some diverse characteristics and details from earlier studies. Much of the existing literature evaluates countries in general and not Brazil specifically.

To guide our analyses, we took a series of steps. First, relying on Olsen and Fenhann (2008) and Olsen et al. (2018), we defined a number of indicators to classify SD in CDMs into environmental, social and economic aspects. We also added our own set of indicators, “independent criteria”, to further dig into the impact of CDMs in Brazil, going beyond existing literature. All these indicators are shown in Table 4 below. It is worth noting that they do not intend to quantify SD, but to guide the collection of descriptive information from Annex III reports. In order to collect data, each Annex III report has been read by different authors of this paper, one by one, in order to identify the relevant information for our research. The criteria in Table 4 guided our data collection.

After collecting data according to the criteria laid out in Table 4, we summarized the information into Table 5, counting the number of projects satisfying each criterion and computing simple and weighted averages, both overall and per sector, per indicator. This will help build a general picture of our results.

Finally, we further dig into Annex III reports to gather qualitative and specific information to help explaining some of our results and identifying gaps and strengths in CDMs in Brazil. The Annex III information is qualitative, so each proponent is free to fill in the document in their own way, like PCH Cachoeirão describing job creation: “number of engineers, 25; archaeologists, 13”.

5. Results: do CDMs intend to promote SD in Brazil?

As shown in Table 5, a general result of this study is that 46% of all analyzed projects claim to contribute satisfactorily to SD (weighted average of 52%, considering the number of projects per sector), although most of them do not clarify exactly how companies intend to get it done. Methane Avoidance projects, such as swine farming (72% of the projects), are the most explicit ones in terms of details, better describing what they actually intend to do in regards to SD. Despite limitation, our general result surprisingly contradicts some previous studies, which do not find enough evidence of SD in CDM host countries (Sutter & Parreño, 2007).

Sector-wise, as shown in Table 5, those accumulating the largest number of projects claiming to promote SD in Brazil are Biomass, Wind, Hydro, Methane Avoidance and Landfill Gas. It is worth mentioning that sectors such as the EE industry, NO₂ and Reforestation presented even greater averages (e.g. 82% for the EE industry), but their representativeness in terms of the absolute number of projects is very low, with some of these presenting only a single project. Our general and sector-wise figures are partially in line with Lazaro and Gremaud (2017), who state that it is not clear which sector contributes the most to SD. After all, the general picture we provide above indicates that while some sectors concentrate a larger number of projects than others, their respective impacts tend to be much lower than those with fewer projects. But what matters most to SD? The number of projects promoting it or their relative impact? What are the actions taken by each sector to promote SD?

In order to answer these questions, we must consider the specificities underlying the general figures presented above. In this respect, the very first aspect that deserves to be scrutinized is the proponent being sustainable before the CDM proposal. This indicates whether CDMs encouraged firms toward SD or merely leveraged their previous SD efforts. In this respect, the sectors of Methane Avoidance, Hydro and Landfill Gas presented the lowest proportion of previously sustainable firms. At the same time, these sectors accumulated some of the largest number of CDM projects (roughly 50% of our sample), nearing 50% on average in terms of SD promotion. Preliminarily, this shows that CDMs have been very successful in promoting SD within these industries. In the same vein, the EE industry and EE supply side had no sustainable firms before CDMs, but ranked among the best ones in SD after the CDM proposals. Nonetheless, as mentioned before, these sectors accumulated a very small number of projects.

On the other hand, while Biomass projects added up to 20% of our sample, 41% of all firms in this sector had already developed sustainable actions prior to their CDMs. This implies that the average SD incidence (67% of CDMs) in this sector is not as representative as the cases of aforementioned sectors.

This result does not imply that Biomass sector do not promote SD, but that almost half of the projects were already prone to developing SD. Perhaps this explains why this industry is the second one contributing the most to “sustainability on top of GHG emission reduction”. While 50% of all our sample claimed to promote “sustainability on top of GHG emission reduction”, the Biomass sector alone accounts for 22% out of those 50%. Among the projects meeting this criterion, 60% come from Methane Avoidance, Landfill Gas and Hydro sectors, while 10% come from Fossil Fuel Switch, NO₂, Reforestation and Wind. These results preliminarily indicate that CDMs are somehow helping to promote SD in Brazil, in addition to reducing GHG emissions – fulfilling the goals for which they have been conceived on the first place. Nonetheless, in order to have a better picture of the extent to which these projects contribute to SD in Brazil, and how they do it, all three dimensions of SD will be scrutinized. This analysis is shown in the coming sections.

6. Conclusion

CDMs go way further intending to reduce GHG emissions in Brazil. As described in previous sections, they actually intend to promote SD in all three spheres (environmental, social and economic). Beyond GHG emission reduction, 61% of all CDMs seek to promote other environmental impacts. After all, CDMs were conceived with environmental aspects in mind. However, what was actually surprising about our findings was that 91% of our sample claimed to seek for social improvements in working conditions and job creation. These results are consistent with Olsen and Fenhann (2008), but even more sensible.

These findings not only surpass by far the environmental contribution of these projects, but also contradict previous studies which found mixed results (Kim, 2004) or no evidence of SD in CDM host countries (Sutter & Parreño, 2007).

In terms of CDM social impacts in Brazil, we could see that large-scale projects contribute the most to “income generation and articulation with other sectors”, the strongest of all SD dimensions in the country (Seres & Haites, 2008; Costa-Júnior et al., 2013). Said large-scale projects’ activities are subjected to high level of regulation, and must comply with strict and well-established environmental standards, positively affecting local communities. Biomass and Methane Avoidance stand out in terms of courses dissemination, energy distribution at reduced cost and schools construction, and other infrastructure activities. These activities for sure can help foster local development and positively contribute to improve local population’s income and standard of living. These are ultimate goals in terms of local impacts. Developing nations in general should critically assess and scrutinize the nature of CDM activities beyond GHG decrease, in order to favor those with greater potential to play an active role in increasing employment and income. As shown before, in Brazil, those projects are the ones requiring more skilled labor (Godoy, 2013; Godoy & Saes, 2016).

However, there is a tradeoff in terms of the number of jobs created and their relative impact on improving income. We demonstrated that in Brazil most CDMs were small-scale projects, employing less than 400 workers. While they help creating a bulky number of jobs, these projects require less skilled workers and hence, contribute the least to income generation. Therefore, in terms of policy implications, local programs should balance income generation with number of jobs, to assure that a significant number of people are employed, while a considerable share of them manages to improve their income. This is by no means a trivial endeavor, but one that should be taken on board and encouraged by developing countries.

From an economic perspective, the results are in line with previous work focusing on technological transfer and skilled labor (Michaelowa et al., 2019). There are indications that Brazilian CDMs projects mostly rely on local technology. These results provide further evidence in favor of the local impact of CDMs in Brazil. Depending on the sector, for instance NO₂ and the EE industry, firms might rely almost exclusively in foreign technology. This means that depending on the sector and the nature of the activities they develop, nations might not be able to replicate Brazilian results. One way of overcoming this is to encourage projects in those sectors relying least in foreign technology, such as Fugitive, Methane Avoidance and Wind, despite this not always being a trivial endeavor. Brazil, for instance, has great potential in renewable energy, such as wind, solar and hydro, and public policies should foster emission reduction projects in these sectors.

As far as the number of projects is concerned, we could see that those sectors accumulating the largest number of CDM projects are Biomass (67%), Hydro (50%), Methane Avoidance (50%) and Landfill Gas (46%). Among these, the swine farming projects are the ones that most act in several activities at once, such as organic fertilizer production, development of courses for the local population, odor reduction in the farms, incentive to plant different cultures, among other measures. Conversely, we also found that sectors accumulating a very small number of projects – sometimes a single project - demonstrated more SD impact than their counterparts: this is the case, for instance, of the EE industry and Reforestation sectors. It is also worth mentioning that in some sectors – like Biomass, for instance – more than others there was a prevalence of projects that were already sustainable prior to CDMs.

As for implications, there are indications that CDM is a feasible way to foster SD goals. Since a project development promotes local SD, public policies could take advantage of this mechanism. A CDM is directly linked to Agenda goal 13, affirming “take urgent measures to combat climate change and its impacts”. But, through its complementary SD actions beyond emission reduction, such as job creation or other financial support, CDM is also linked to another Agenda goal 2030: the first one, “end poverty in all its forms everywhere”. In this sense, GHG emission reduction projects can be great allies of public policies, in terms of encouraging measures to comply with the 2030 Agenda.

Regarding future studies, we suggest that other countries could have their projects analyzed as we have done for Brazil. Also, it is recommended that upcoming work focuses on indicators as taxation, state incentives or sector policies. Qualitative studies could be developed to analyze the effects of projects in terms of SD.

As limitations of our study, we highlight that our analysis is focused on ex ante CDM and SD proposals, and not on ex-post measures. This means that we could not assess the extent to which and how CDM proponents actually implemented every action declared in Annex III. Even though assessing this angle and gathering data about it is challenging, relying solely on our ex ante sample already made it possible to conclude that there is great potential for measures to be adopted aiming at SD in CDM host countries – not only in Brazil, but in countries like India and China.

We stress, however, that there are currently no regulations, mandatory targets or strict rules related to SD in Brazil related to CDM. There are no monitoring devices to both check and enforce the proposed actions. It is the key to bear in mind that if stricter standards with mandatory patterns and different norms are implemented, new transaction costs might emerge. Consequently, expenses may be higher due to monitoring, gathering new information in order to implement new activities and other kinds of costs arising from stricter rules.

Thus, it is necessary to take into consideration that GHG emission reduction programs linked to stricter SD norms could increase cost. There lies a paradox resulting from the need to expand measures related to SD in developing countries, without, however, bringing further obstacles to the implementation of new GHG emission reduction projects.