Advertisement

Social life cycle assessment of first and second-generation ethanol production technologies in Brazil

  • Alexandre Souza
  • Marcos Djun Barbosa Watanabe
  • Otavio Cavalett
  • Cassia Maria Lie Ugaya
  • Antonio Bonomi
SOCIAL LCA IN PROGRESS

Abstract

Purpose

The main goal of this study is to suggest quantitative social metrics to evaluate different sugarcane biorefinery systems in Brazil by exploring a novel hybrid approach integrating social life cycle assessment and input-output analysis.

Methods

Social life cycle assessment is the main methodology for evaluating social aspects based on a life-cycle approach. Using this framework, a hybrid model integrating social life cycle assessment and input-output analysis was introduced to evaluate different social effects of biorefinery scenarios considering workers as the stakeholder category. Job creation, occupational accidents, wage profile, education profile, and gender profile were selected as the main inventory indicators. A case study of three scenarios considering variations in agricultural and industrial technologies (including sugarcane straw recovery and second-generation ethanol production, for instance) was carried out for evaluating present first-generation (1G-basic, 1G-optimized) and future first- and second-generation ethanol production (1G2G).

Results and discussion

The 1G-basic scenario leads to higher job creation levels over the supply chain mainly because of the influence of agricultural stage whose workers are mostly employed in sugarcane manual operations. On the other hand, 1G-optimized and 1G2G present supply chains are more reliant on the manufacturing, trade, and services sectors whose workers are associated with a lower level of occupational accidents, higher average wages, higher education level, and more participation of women in the work force.

Conclusions

The use of a novel hybrid approach integrating social life cycle assessment (SLCA) and input-output analysis (IOA) was useful to quantitatively distinguish the social effects over different present and future sugarcane biorefinery supply chains. As a consequence, this approach is very useful to support decision-making processes aiming to improve the sustainability of sugarcane biorefineries taking social aspects into account.

Keywords

Ethanol Input-output analysis Life cycle assessment Social assessment Sugarcane 

1 Introduction

The social life cycle assessment (SLCA) is a product-oriented method that aims at assessing social and socioeconomic aspects of products, including their potential positive and negative impacts along their life cycle (UNEP/SETAC 2009, 2011). Since the social life cycle assessment is in its initial phase of development, there is space for many improvements in the method and its application framework (Jørgensen et al. 2008; Jørgensen 2013; Macombe and Loeillet 2013). For instance, there is still a lack of tools able to anticipate the social consequences of a product system (Macombe and Loeillet 2013) because of the difficulty involved in finding correlations between a given technological change and its social consequences (Zamagni et al. 2011).

According to UNEP/SETAC (2009), SLCA can be classified in two main types considering the life cycle impact assessment (LCIA) phase. In the first type of LCIA, qualitative approaches based on scoring systems can be carried out, such as the subcategory assessment method (SAM) proposed by Ramirez et al. (2014) which assesses the social profile of the organizations involved in the processes along the product life cycle in relation to the fulfillment of a basic requirement (BR). In the second type, results related to subcategories rely on causal relationships with the impact categories, as presented by Norris (2006). This second type of SLCA approach can be explored to deal with the challenge of anticipating the social consequences of a given product system. An example is the use of the input-output analysis (IOA) to estimate social effects on the supply chain such as Rugani et al. (2012), Onat et al. (2014), and Bocoum and Macombe (2015). The input-output analysis is important because performing a study along the life cycle is data demanding and SLCA studies have been very limited in the number of unit processes considered. The development of the Social Hotspot Database (SHDB), for instance, uses the IOA to create a global database of activities or unit processes (also defined as country-specific sectors) in the supply chain that may be at risk for social issues to be present (Benoit-Norris et al. 2012). Although the SHDB is a very useful tool which disaggregates the economic sectors of countries, this database is not sensitive to capture the different technologies within a given economic sector.

In this sense, this paper aims to develop and apply a methodological framework to assess the social effects of present and future sugarcane biorefineries considering vertically integrated production systems, i.e., taking into account technological changes on agricultural and industrial phases. More precisely, the focus of this paper is on the inventory phase. In order to perform such type of assessment, the Virtual Sugarcane Biorefinery (VSB) framework is used. The VSB was developed at the Brazilian Bioethanol Science and Technology Laboratory (CTBE) and integrates economic, social, and environmental simulation platforms to assess different sugarcane biorefinery configurations (Bonomi et al. 2012; Cavalett et al. 2012; Dias et al. 2011).

2 Methods

The Social Life Cycle Inventory will be compiled by using a hybrid model which integrates LCA and input-output analysis (Watanabe et al. 2015) whose structure was developed to predict the effects of technological changes on sectors of the entire supply chain.

2.1 Goal, scope, and system boundaries

The main goal of the study is to compare different Brazilian sugarcane biorefineries considering their social consequences on workers of the entire supply chain. A group of five social effects are evaluated: job creation, occupational accidents, wage profile, education profile, and gender profile in order to comprehend the effect of different technologies on the supply chain structure.

In order to perform such assessment, a hybrid model integrating process-based (LCA) and input-output based (IOA) approaches is used. The life cycle system includes all processes from sugarcane planting to ethanol use. Additionally, the input-output analysis (IOA) ensures the completeness of the life cycle inventory since the contribution of all upstream sectors are considered, including direct and indirect suppliers of inputs to sugarcane biorefineries. The boundaries of the product system under study are shown on Fig. 1. It shows the links of the sugarcane production and industrialization phases with the other economic sectors due the use of inputs in this production chain.
Fig. 1

Systems boundaries

2.2 Scenarios

Table 1 shows the main characteristics of the scenarios considered in this study. It is worthwhile to mention that in all the scenarios, agricultural and industrial phases are vertically integrated. As Table 1 shows, sugarcane production relies on different planting, harvesting, and straw recovery technologies depending on the scenario. For instance, sugarcane yields assumed for the 1G-basic, 1G-optimized, and 1G2G were of 80, 80, and 100 metric tons of sugarcane per hectare respectively (Bonomi et al. 2012; Cavalett et al. 2012). The assumption of higher productivity in 1G2G scenario reflects the potential sugarcane yield if better agricultural management was applied in the fields. As Table 1 shows, most of the 1G-basic agricultural operations rely on manual and semi-mechanized operations. Moreover, this scenario is characterized by sugarcane burning, and all sugarcane straw is lost due to the fire. In the 1G-optimized and 1G2G agricultural operations, however, only mechanized operations are observed, especially in planting and harvest stages. It is also important to highlight that these scenarios are associated with the sugarcane burning phase out and, therefore, about 50 % of total sugarcane straw available on the field is recovered using the bailing system. Such different agricultural technologies lead to different biomass production costs whose values were calculated using CanaSoft, which is the agricultural model embedded in the Virtual Sugarcane Biorefinery (Bonomi et al. 2012; Cardoso et al. 2013; Jonker et al. 2015; Cavalett et al. 2012).
Table 1

Main characteristics of the vertically integrated production scenarios considered in this study

Scenario

Sugarcane planting system

Sugarcane harvesting system

Straw recovery

Outputs

Context

1G-basic

Semi-mechanized

Manual, with pre-harvesting burning

No

▪ First-generation ethanol

Represents current sugarcane production with relatively outdated technologiesa

1G-optimized

Mechanized

Mechanized, without pre-harvesting burning

Yes

▪ First-generation ethanol

▪ Electricity

Represents a modern current technology

1G2G

Mechanized

Mechanized, without pre-harvesting burning

Yes

▪ First- and second-generation ethanol

▪ Electricity

Represents future technology

aSugarcane pre-harvesting burning still takes place either in some regions of Brazil such as the Brazilian Northeast region or areas where harvester operation is limited due to high slope (above 12 %). In center-south region, mechanized harvesting represents more than 88 % of the total harvested area (Nunes Junior 2014)

In the industrial phase, ethanol is the main product obtained from sugarcane processing. The 1G-basic scenario is characterized by producing only first-generation ethanol—about 85 l/metric ton of sugarcane. Crushing capacity is fixed in 2 million tons of sugarcane stalks per year (Bonomi et al. 2012; Cavalett et al. 2012). This scenario represents an average present technology available in most of Brazilian plants.

The 1G-optimized industrial scenario represents the best current technology available in center-south region of Brazil. Its crushing capacity is assumed to be of 4 million tons of sugarcane per year, and there is coproduction of first-generation ethanol (85 l/metric ton of sugarcane) and surplus electricity (185 kWh/ton of sugarcane) which is sold to the grid. Such a technological improvement is possible because of enhanced industrial plant characteristics such as energetic integration to reduce the steam demand and more efficient high-pressure cogeneration system.

The integrated first- and second-generation scenario (1G2G) represents a future biorefinery which combines an optimized first-generation distillery with an industrial process to produce cellulosic ethanol. The crushing capacity is about 4 million tons of sugarcane per year. This scenario was adapted from a previous study which describes the future technology for hydrolysis and fermentation of sugarcane lignocellulosic materials, such as bagasse and straw, to ethanol (Dias et al. 2011). Because a fraction of bagasse and straw is diverted to second-generation ethanol production, the 1G2G electricity output is lower (approximately 60 kWh/ton of sugarcane) than the 1G-optimized surplus. The ethanol production yield is 124 l of ethanol per ton of processed sugarcane.

2.3 Allocation

In this study, the social consequences of ethanol production from sugarcane are assessed. Considering that input-output analysis is used to compile the life-cycle inventory, a change in ethanol final demand of R$ 1 billion (approximately US$ 500 million in 2009) was assumed for all scenarios in order to assess the direct and indirect effects on the supply chain. In the 1G-optimized and 1G2G scenarios, electricity is a coproduct which corresponds to outputs of R$ 0.2 billion and R$ 0.05 billion, respectively. Therefore, both products are associated with the social consequences. In order to express the social effects of job creation and occupational accidents, the economic allocation criterion could be adopted, i.e., the higher the participation of ethanol in the total biorefinery output, the higher the allocation of a given social effect to ethanol. These results could also be converted into an energy basis, i.e., job creation and accidents would be expressed per joules of ethanol. In this study, however, social effects will not be allocated to a single product; on the other hand, they will be interpreted as the overall effect of different biorefinery technologies due to a change in the ethanol final demand of R$ 1 billion on the entire supply chain.

2.4 Inventory analysis

The objective of an inventory is to gather relevant information considering the goal and scope of the study. Within the hybrid approach, the inventory modeling is divided in two main steps. First, the sugarcane and biorefinery scenarios’ inventories were modeled according to process-based (LCA) databases in the Virtual Sugarcane Biorefinery. This inventory modeling describes main agricultural and industrial processes related to sugarcane ethanol (and electricity) production systems. Second, the process-based inventory of evaluated scenarios is inserted into a commodity-by-industry Brazilian input-output table as a group of new sectors according to the method proposed by Watanabe et al. (2015). Therefore, an input-output-based inventory is compiled by considering all upstream direct and indirect sectors associated with the sugarcane ethanol production. The latest Brazilian input-output table used in this study covers 110 commodities and 56 industries, estimated according to official data published by the Brazilian Institute of Geography and Statistics (IBGE 2015; Guilhoto and Sesso Filho 2005; 2010).

Table 2 shows a summary of the stakeholder categories, subcategories, inventory indicators, inventory data, and data sources considered in this study. The Brazilian Ministry of Labor and Employment (MTE) and Ministry of Social Security (MPS) databases, as well as sustainability reports from sugarcane sector, were consulted in order to obtain sectorial data associated with job creation, occupational accidents, wage, education, and gender profiles. As shown in Table 2, the case study presented in this paper focuses on only 4 out of 31 subcategories described in the UNEP/SETAC (2009) SLCA guidelines. However, the evaluated categories are among the most relevant and discussed when comparing sugarcane technology differences.
Table 2

Stakeholder categories, subcategories, and inventory indicators considered in this study

Stakeholder categories

Subcategories

Inventory indicators

Inventory data

Data source

Local community

Local employment

Number of jobs

Manual and mechanized necessary working hours

Modeled using the VSB (Bonomi et al. 2012)

Access to material resources/access to immaterial resources/delocalization and migration/cultural heritage/safe and healthy living conditions/respect of indigenous rights/community engagement/secure living conditions

Workers

Health and safety

Number of occupational accidents

Incidence of occupational accidents per thousand workers of each Brazilian economic sector

MPS (2015)

Fair salary

Average wages

Costs of the agricultural and industrial working hours

Modeled using the VSB (Bonomi et al. 2012)

Wage value profile

Wage profile of each Brazilian economic sector

MTE (2015)

Equal opportunities/discrimination

Woman in the labor force participation rate

Woman participation in each Brazilian economic sector

MTE (2015)

 

Education degree profilea

Education profile of each Brazilian economy sector

MTE (2015)

Freedom of association and collective bargain/child labor/working hours/forced labor/social benefits/social security

Consumer

Health and safety/feedback mechanism/consumer privacy/transparency/end of life responsibility

Society

Public commitments to sustainability issues/contribution to economic development/preventions and mitigation of armed conflicts/technology development/corruption

Value chain actors

Fair competition/promoting social responsibility/supplier relationships/respect of intellectual property rights

aEducation profile is not a default inventory indicator in UNEP/SETAC (2009) guidelines. It was included in this study considering the high education inequality in Brazilian work force and data availability on the education of workers in Brazilian economic sectors

The two-step method used in the inventory analysis is summarized in Fig. 2. These two steps complement each other and allow the assessment of effects of specific technologies on the entire Brazilian economy by considering sectors which are directly and indirectly connected to both sugarcane and ethanol production systems.
Fig. 2

Inventory compilation based on process and input-output approaches

The first step (process-based approach) is more scenario-oriented. It is useful to describe specific sugarcane and biorefinery technologies because data related to such sectors are not available in the Brazilian input-output table. As Fig. 2 shows, sugarcane production and industrial conversion inventories were compiled based on Virtual Sugarcane Biorefinery models considering technical parameters such as the agricultural productivity, harvesting efficiency (manual or mechanical), agricultural inputs to sugarcane production, milling capacity, ethanol yield, and chemical inputs to ethanol production. Examples of these calculations are presented on several publications (Bonomi et al. 2012; Cavalett et al. 2012; Cardoso et al. 2013; Jonker et al. 2015). The life cycle inventory representing each new biorefinery and sugarcane sector was then translated into a list of purchased commodities (goods and services) matching the list of commodities described in the Brazilian commodity-by-sector direct requirements table for 2009 (most recent available table). The link between process-based analysis and input-output analysis is observed when such new biorefinery and sugarcane sectors (technological scenarios described by process-based analysis) are inserted into the input-output table.

Some outputs of interest—such as number of workers—were calculated and converted into the proposed social metrics considered in this study. For instance, the total working hours available in the agricultural model was converted into the number of workers required to cultivate and harvest sugarcane assuming an average number of working hours per day and an average number of working days per year.

With regard to the correlation of process-based model outputs and social effects of evaluated biorefinery scenarios, a specific approach was established in this study, especially to describe the number of occupational accidents in the sugarcane production system. First, a positive correlation between the incidence of occupational accidents (number of accidents per thousand workers) in the sugarcane production and its level of mechanization was determined based on official data of Ministry of Social Security (MPS 2015) and sectoral data on sugarcane mechanization (Nunes Junior 2012). With this correlation, it was possible to estimate the number of accidents in each biorefinery scenario considering the estimated number of workers. A similar procedure was carried out to link model outputs and social effects in the sugarcane conversion scenarios (industrial stage).

The second step (input-output based) expands the boundaries of the study to the entire sugarcane ethanol supply chain and uses a general equilibrium model to estimate both direct and indirect changes in the sector outputs. Within the Virtual Sugarcane Biorefinery framework, the input-output analysis has been alternatively applied to assess the economic and environmental impacts associated with the introduction of new biorefinery technologies in the Brazilian economy (Watanabe et al. 2015).

In this study, the social effects of technologies are assessed based on previous publications which explore LCA and input-output analysis in a broader conceptual view. Hendrickson et al. (2006) for instance translated economic activity of sectors into socioeconomic effects, such as occupational safety risks by the Economic input-output life cycle assessment (EIO-LCA). In the VSB, non-square commodity-by-industry tables are used because more disaggregated data are available on official databases. Moreover, the assessment of biorefinery technologies implies on adding a set of new industries and commodities into the original input-output table in order to represent agricultural and industrial scenarios. As described in Watanabe et al. (2015), additional sectors representing evaluated technological scenarios are inserted into the original Brazilian input-output matrix, i.e., biorefinery 1G-basic, biorefinery 1G-optimized, biorefinery 1G2G, sugarcane-1G-basic, sugarcane 1G-optimized, and sugarcane 1G2G (see Table 1). Moreover, biorefinery processes outputs were also included into the original direct requirements table because each scenario requires specific set of inputs. The additional products inserted into the original table were sugarcane 1G-basic, sugarcane 1G-optimized, sugarcane 1G2G, ethanol 1G-basic, ethanol 1G-optimized, ethanol 1G2G, electricity 1G-optimized, and electricity 1G2G. Additional details on the input-output model and the resolution of such linear system of equations have been previously described by Watanabe et al. (2015).

Figure 3 shows the application of the hybrid model aiming to expand the social effects of a given scenario over its supply chain. The first step is to assume a given change in ethanol final demand, e.g., US$ 1 billion of ethanol. This change in ethanol final demand should be exactly the same for all the evaluated scenarios to allow its comparison. The input-output model is then used to simulate the impacts of ethanol production on the Brazilian economy. Considering that a variety of sectors give direct and indirect support to the ethanol production—such as sugarcane, trade, oil and gas, and transportation—the economic activity of such supporting sectors will increase due to the increase in ethanol final demand. The economic activity of sectors is the measurement in monetary terms of the sectors outputs, as calculated in the input-output model, considering a changing in final demand of a determined product (Miller and Blair 2009). In other words, the economic activity of sectors is the purchased value (in monetary units) of each sector required to meet the final demand. In order to estimate job creation over the sectors of the entire supply chain, the sector outputs (economic activities) were multiplied by the number of jobs per R$ output of each sector using official data (IBGE 2015). With regard to occupational accidents, wage profile, gender profile, and education profile, the quantification was possible because data on the worker’s profile were available in sectoral databases of the Ministry of Social Security (MPS 2015) and Ministry of Labor and Employment (MTE 2015).
Fig. 3

Quantification of occupational accidents in the biorefinery supply chain based on the EIO-LCA framework

It is important to highlight that the methodological framework presented in this paper focuses on developing social metrics sensitive to predict the effects of incremental technological changes, at the inventory analysis phase. This approach is reasonable to compare selected social aspects of ethanol production technologies. One step forward would be to express the risk levels as it is performed, for instance, in the Social Hotspot Database (SHDB). The SHDB uses an input-output model together with characterization models in order to estimate risk levels of sectors associated with child labor, forced labor, excessive working time, indigenous rights, and so on (Benoit-Norris 2014). However, risk levels are not considered mainly due to the lack of data for the specific sectoral aggregation level adopted in this project. Future studies may include these and further LCA steps, such as social impact category assessment and evaluation of hotspots.

3 Results

In order to simulate the effects of different biorefinery scenarios, a change in ethanol final demand of R$ 1 billion (Brazilian Reais) was considered in the hybrid LCA model. This shock in final demand is equivalent to approximately US$ 500 million or 1.145 billion liters, considering the base year of 2009. As previously described, results related to economic activity of sectors (see Table 3) were the basis to estimate the number of workers mobilized in each sector of the Brazilian economy. As Table 3 shows, 1G-optimized scenario leads to the highest economic activity: about R$ 2.30 billion, approximately US$ 1.15 billion. The main sectors involved in the ethanol supply chain are sugarcane, oil refining and coke, other chemicals, trade, metal products, rubber, and plastic products and chemical products. The 1G-basic and 1G2G lead to similar economic activity (about R$ 1.9 billion); however, different sectors are activated in the Brazilian economy. Less participation of sugarcane sector is observed in 1G2G sector because of the higher ethanol yields obtained from the use of bagasse and straw to produce second-generation ethanol.
Table 3

Comparison of top 10 sectors for economic activity considering different sugarcane ethanol supply chains (R$ billion)

1G2G

Total

Direct

Indirect

1G-optimized

Total

Direct

Indirect

1G-basic

Total

Direct

Indirect

Sugarcane industry

1.049

1.049

Sugarcane industry

1.215

1.215

Sugarcane industry

1.000

1.000

Sugarcane

0.324

0.305

0.018

Sugarcane

0.534

0.417

0.118

Sugarcane

0.478

0.457

0.022

Oil refining and coke

0.064

0.001

0.063

Oil refining and coke

0.085

0.001

0.085

Oil refining and coke

0.068

0.003

0.065

Other chemicals

0.053

0.048

0.005

Chemical products

0.073

0.003

0.070

Chemical products

0.052

0.002

0.050

Trade

0.050

0.017

0.033

Trade

0.051

0.007

0.044

Trade

0.038

0.007

0.031

Metal products

0.048

0.033

0.015

Metal products

0.033

0.012

0.021

Transportation, storage and mail

0.026

0.007

0.019

Rubber and plastic products

0.045

0.028

0.017

Transportation, storage and mail

0.033

0.006

0.027

Oil and natural gas

0.024

0.024

Chemical products

0.042

0.002

0.040

Rubber and plastic products

0.033

0.009

0.024

Services to companies

0.023

0.012

0.011

Transportation, storage, and mail

0.029

0.007

0.023

Oil and natural gas

0.030

0.030

Metal products

0.022

0.011

0.010

Services to companies

0.026

0.012

0.015

Services to companies

0.030

0.012

0.017

Rubber and plastic products

0.019

0.008

0.010

Other sectors (55 remaining)

0.174

0.018

0.156

Other sectors (55 remaining)

0.163

0.020

0.143

Other sectors (55 remaining)

0.122

0.018

0.104

Total

1.905

1.520

0.385

Total

2.280

1.701

0.579

Total

1.871

1.525

0.346

Table 4 shows the top 10 sectors considering job creation in the evaluated scenarios. The 1G-basic scenario is related to the highest job creation level, approximately 12,400 workers in the entire supply chain. The 1G-optimized and 1G2G scenarios are associated with lower levels roughly 7700 and 6900, respectively. It is mostly because of technological changes in the sugarcane sector, such as mechanical harvesting. It is clear that the number of direct jobs in the 1G-basic scenario is higher because the sugarcane production technology is more reliant on manual operations than observed in the other scenarios with higher agricultural mechanization levels. Moreover, 1G-basic scenario is also related to higher employment in the industrial conversion phase since the ethanol production is less efficient, i.e., more workers per liter of ethanol are required. On the other hand, 1G-optimized and 1G2G scenarios are related to a higher share of indirect workers, i.e., those workers involved in the production of inputs used in the ethanol production. In this case, indirect workers are predominantly associated with trade, services to companies, transportation, storage and mail services, and other sectors whose outputs give support to the ethanol supply chain.
Table 4

Comparison of top 10 sectors for job creation considering different sugarcane ethanol supply chains

1G2G workers

Total

Direct

Indirect

1G-optimized workers

Total

Direct

Indirect

1G-basic workers

Total

Direct

Indirect

Trade

1629

553

1076

Trade

1636

212

1424

Sugarcane

5835

5572

263

Sugarcane industry

1062

1062

Sugarcane industry

1423

1423

Sugarcane industry

3002

3002

Sugarcane

725

684

41

Sugarcane

1363

1063

300

Trade

1229

218

1011

Services to companies

599

266

333

Services to companies

675

280

395

Services to companies

523

269

254

Metal products

569

393

176

Transportation, storage, and mail

483

94

389

Transportation, storage, and mail

375

102

273

Transportation, storage, and mail

428

98

330

Metal products

394

141

253

Metal products

262

135

127

Rubber and plastic products

317

196

121

Rubber and plastic products

230

60

170

Rubber and plastic products

131

58

73

Other chemicals

291

264

27

Wood products, excluding furniture

161

161

Chemical products

79

4

75

Agriculture and forestry

144

144

Agriculture and forestry

131

131

Other products from extractive industry

77

77

Wood products, excluding furniture

125

125

Chemical products

111

4

107

Repair and maintenance services

76

76

Other sectors (55 remaining)

1027

148

878

Other sectors (55 remaining)

1096

193

902

Other sectors (55 remaining)

808

122

686

Total

6915

3664

3251

Total

7703

3471

4232

Total

12397

9482

2916

Figure 4 illustrates that the supply chain configurations for 1G-basic scenario, 1G-optimized, and 1G2G are quite different. In the 1G-basic scenario, for instance, 47 % of workers of the entire supply chain are employed in the sugarcane sector. The relative frequencies of workers in 1G-basic biorefinery and trade sectors, for instance, are about 24 and 10 %, respectively. On the other hand, 1G-optimized scenario is mainly characterized by workers from trade (21 %), whereas sugarcane sector corresponds to only 18 %. The participation of other economic sectors such as services and transportation is also higher in 1G-optimized scenario, about 9 and 6 %, respectively. In the 1G2G supply chain, a higher relative frequency of workers from trade is observed (24 %) as well as from industrial conversion phase (15 %) and sugarcane production (10 %). Also, the relative frequency of other sectors is also higher compared to 1G-basic and 1G-optimized scenarios.
Fig. 4

Relative frequency of workers and their respective sectors for a 1G-basic, b 1G-optimized, and c 1G2G supply chains

According to Fig. 5, the projection of occupational accidents in 1G-basic scenario is higher, about 550 accidents per year. The main sectors involved are sugarcane production (293 accidents), sugarcane industry (168), trade (18), transportation, storage and mail services (14), metal products (11), and other sectors. The sugarcane sector is the main contributor to accidents in the 1G-basic supply chain mainly because of two factors. Firstly, the predominance of manual operations leads to a higher number of workers involved in sugarcane cultivation and harvesting. Second, manual harvesting has a positive correlation with occupational accidents; the higher the exposition of workers to the environment, the higher the likelihood of an accident. These two combined effects contribute to increase the incidence of occupational accidents in 1G-basic scenario. On the other hand, 1G-optimized and 1G2G scenarios are expected to have lower levels of accidents because their supply chains are comparatively less reliant on the sugarcane production sector due to higher industrial conversion efficiencies. Additionally, mechanical harvesting in the 1G-optimized and 1G2G scenarios decrease the incidence of accidents rate per number of workers. As a result, 1G-optimized and 1G2G have lower number of direct accidents. In the 1G-basic supply chain, about 85 % of accidents were direct (sugarcane production and sugarcane industry, mostly). In contrast, direct accidents in 1G-optimized and 1G2G were about 55 and 40 %, respectively, because other sectors related to lower level of occupational accidents are activated.
Fig. 5

Number of total occupational accidents per year over the supply chain considering a R$ 1 billion change in ethanol final demand

Figure 6 shows the wage profile of different ethanol supply chain scenarios. The 1G-basic supply chain has highest percentage of workers associated with the range of 1.01 to 1.5 Brazilian minimum wages (about 235 US$ per month in 2009), about 70 % of them have salaries within this range. Considering a relatively higher participation of workers from sugarcane sector, this result was expected because unskilled workers performing manual operations (mainly for planting and harvesting operations) at agricultural stage are historically related to low salaries. Although other sectors involved in the 1G-basic supply chain have better wage profiles, the predominance of sugarcane workers leads to a higher frequency of low salaries. The 1G-optimized and 1G2G supply chains are also predominantly characterized by a salary range of 1.01 to 1.5 minimum wages; however, it is clear that the average wage profile was improved due to the lower participation of sugarcane workers and higher contribution of labor from other sectors related to industry and services. Moreover, 1G-optimized and 1G2G sugarcane workers are better paid due to the increasing share of work on mechanical operations in the sugarcane production system, especially those associated with sugarcane planting and harvest. As a consequence, a higher percentage of workers of the supply chain were placed in the 1.51 to 2.0 and 2.01 to 3.0 minimum wages position.
Fig. 6

Wage profiles of different sugarcane ethanol supply chains: 1G-basic, 1G-optimized, and 1G2G

As Fig. 7 shows, the 1G-basic, 1G-optimized, and 1G2G lead to different results in terms of relative frequency of female workers: 15, 24, and 26 %, respectively. These results were expected because 1G-basic presents a predominance of male workers employed in sugarcane production (whose average gender distribution is approximately 92 % male and 8 % female). The 1G-optimized and 1G2G supply chains have higher female participation due to trade, industry, and other services whose relative frequency of female workers is higher compared to the sugarcane agricultural sector. In trade sector, for instance, about 45 % of workers are women. Although the participation of female workers is still low in all scenarios, 1G-optimized and 1G2G supply chains may lead to more opportunities for female workers.
Fig. 7

Percentage of female workers in different ethanol production supply chains

As Fig. 8 shows, there are also different education profiles associated with the 1G-basic, 1G-optimized, and 1G2G supply chains. In the 1G-basic, a high fraction of workers did not complete elementary school, corresponding to a relative frequency of 23 %. Again, it is because of the contribution from unskilled workers related to manual operations in the sugarcane sector (29 % of workers correspond to such education level in the sugarcane production phase). On the other hand, 1G-optimized and 1G2G are related to different education profiles whose main education level corresponds to complete high school—33 and 38 % relative frequencies, respectively. Considering that more technology-intensive scenarios have a lower participation of workers from the agricultural sector, this result is explained by higher participation of workers from sectors related to manufacturing and other services whose workers have higher education levels compared to agriculture. Trade and transportation sectors, for instance, have a higher participation of workers with complete high school education, with relative frequencies of 56 and 42 %, respectively.
Fig. 8

Education profile of different sugarcane ethanol supply chains

4 Discussion and conclusions

The use of a hybrid method based on social life cycle assessment principles was useful to differentiate biorefinery scenarios representing present and future ethanol production technologies as well as to quantify their social effects on the sectors of the Brazilian economy. It is clear that more modern scenarios (1G-optimized and 1G2G) are associated with lower job creation levels when compared to the outdated technology (1G-basic) which is more reliant on manual human labor, especially those dedicated to sugarcane planting and harvesting operations. On the other hand, new agricultural and industrial technologies are related to better social effects because they have potential to decrease the incidence of occupational accidents as well as to improve education levels, average wages, and participation of female workers on the work force considering the entire supply chain. This effect is observed because there is an increase of the relative frequency of workers from other sectors with more balanced gender equality, such as trade sector.

It is important to highlight that, in general, quantitative aspects calculated using econometric and process models—such as that considered in this paper—very unlikely will be capable of dealing with all the important aspects of social sustainability. In this context, finding appropriate social metrics of a given product system or quantify risk levels of a specific social issue still remains a challenge. In this study, a model-based (or desktop-research-based) approach was used to focus on assessing main social issues related to sugarcane supply chain’s working conditions. As results shown, the proposed hybrid SLCA model is important to support decision-making processes related to novel technologies for sugarcane and ethanol production systems. Such a quantitative assessment and the possibility of anticipating the social effects of future technologies may also be useful to decision-making processes associated with different sectors or other product systems. Moreover, this ex-ante study based on both process- and input-output-based inventories ensured the inventory completeness, and as a consequence, social effects were assessed considering the entire supply chain.

Although the results in this paper are focused on workers as a stakeholder category, future studies may use the same approach to expand the boundaries of SLCA to assess new stakeholder categories, subcategories, and inventory indicators. Moreover, considering that the hybrid model contains a comprehensive database on social effects of diverse Brazilian sectors, this framework could be further explored for quantifying social effects of other product systems in Brazil. As mentioned before, another possibility is to explore the overlaps of the hybrid SLCA model presented in this article with other input-output databases, such the Social Hotspot Database (SHDB), in order to expand the scope of the assessment to provide social risk data on sector and country level.

Notes

Acknowledgments

The authors are grateful to Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Capes, Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq (project 453921/2014-0), and Fundação de Amparo à Pesquisa do Estado de São Paulo - FAPESP (projects 2012/15.359-1 and 2010/17139-3) for the financial support.

References

  1. Benoit-Norris C (2014) Data for social LCA. Int J Life Cycle Assess 19(2):261–265CrossRefGoogle Scholar
  2. Benoit-Norris C, Cavan DA, Norris G (2012) Identifying social impacts in product supply chains: overview and application on the Social Hotspot Database. Sustainability 4:1946–1965CrossRefGoogle Scholar
  3. Bocoum I, Macombe C (2015) Anticipating impacts on health based on changes in income inequality caused by life cycles. J Life Cycle Assess 20:405–417CrossRefGoogle Scholar
  4. Bonomi A, Mariano AP, Jesus CDF et al. (2012) The Virtual Sugarcane Biorefinery (VSB): 2011 Report. Campinas, São Paulo: Brazilian Bioethanol Science and Technology Laboratory (CTBE), Technological Assessment Program (PAT), Internal Report, 2012. Available at: http://goo.gl/x1Ach. Accessed 07.16.2014
  5. Cardoso TF, Cavalett O, Chagas MF et al (2013) Technical and economic assessment of trash recovery in the sugarcane bioenergy production system. Sci Agric 70(5):353–360CrossRefGoogle Scholar
  6. Cavalett O, Junqueira TL, Dias MOS et al (2012) Environmental and economic assessment of sugarcane first generation biorefineries in Brazil. Clean Technol Environ 14(3):399–410CrossRefGoogle Scholar
  7. Dias MOS, Cunha MP, Jesus CDF et al (2011) Second generation ethanol in Brazil: can it compete with electricity production? Bioresour Technol 102(19):8964–8971CrossRefGoogle Scholar
  8. Guilhoto JJM, Sesso Filho UA (2010) Estimação da matriz insumo-produto utilizando dados preliminares das contas nacionais: aplicação e análise de indicadores econômicos para o Brasil em 2005 (input-output table estimation from preliminary data on the system of national accounts: application and assessment of economic indicators for Brazil in 2005). Economia e Tecnologia 6(23):53–62Google Scholar
  9. Guilhoto JJM, Sesso Filho UA (2005) Estimação da matriz insumo-produto a partir de dados preliminares das contas nacionais (input-output table estimation from preliminary data on the system of national accounts). Economia Aplicada 9(2):277–299Google Scholar
  10. Hendrickson C, Lave L, Matthews HS (2006) Environmental life cycle of goods and services: an input-output approach. RFF Press, Washington DCGoogle Scholar
  11. IBGE (2015) System of National Accounts. Brazilian Institute of Geography and Statistics (IBGE). http://www.ibge.gov.br/home/estatistica/economia/matrizinsumo_produto/default.shtm. Accessed 14 May 2015
  12. Jonker J, Van der Hilst F, Junginger HM et al (2015) Outlook for ethanol production costs in Brazil up to 2030, for different biomass crops and industrial technologies. Appl Energ 147:593–610CrossRefGoogle Scholar
  13. Jørgensen A, Le Bocq A, Nazarkina L et al (2008) Methodologies for social life cycle assessment. Int J Life Cycle Assess 13(2):96–103CrossRefGoogle Scholar
  14. Jørgensen A (2013) Social LCA - a way ahead? Int J Life Cycle Assess 18:296–299CrossRefGoogle Scholar
  15. Macombe C, Loeillet D (2013) Social life cycle assessment, for who and why? In: Macombe C (ed) Social LCAs: socio-economic effects in value chains. Cirad, Montpellier, pp 35–52Google Scholar
  16. Miller RE, Blair PD (2009) Input-output analysis: foundations and extensions. Cambridge University Press, New YorkCrossRefGoogle Scholar
  17. MPS (2015) Anuário Estatístico da Previdência Social – AEAT. Ministry of Social Security. http://www.previdencia.gov.br/estatisticas/. Accessed 20 Jan 2015
  18. MTE (2015) Base de dados on-line. Programa de Disseminação de Estatísticas do trabalho—PDET: Ministry of Labor and Employment (MTE). http://bi.mte.gov.br/bgcaged/login.php. Accessed 20 Jan 2015
  19. Norris GA (2006) Social impacts in product life cycles: towards life cycle attribute assessment. Int J Life Cycle Assess 11(1):97–104CrossRefGoogle Scholar
  20. Nunes Junior D (2012) Indicadores de desempenho da agroindústria canavieira: Safras 2010/2011 e2011/2012 (Performance Indicators of Sugarcane Agroindustry: Seasons 2010/2011 and 2011/2012). Grupo IDEA, Ribeirão Preto, BrasilGoogle Scholar
  21. Nunes Junior D (2014) Indicadores de desempenho da agroindústria canavieira: Safras 2012/2013 e 2013/2014 (Performance Indicators of Sugarcane Agroindustry: Seasons 2012/2013 and 2013/2014). Grupo IDEA, Ribeirão Preto, BrasilGoogle Scholar
  22. Onat NC, Kucukvar M, Tatari O (2014) Integrating triple bottom line input–output analysis into life cycle sustainability assessment framework: the case for US buildings. Int J Life Cycle Assess 19:1488–1505CrossRefGoogle Scholar
  23. Ramirez PKS, Petti L, Haberland NT et al (2014) Subcategory assessment method for social life cycle assessment. Part 1: methodological framework. Int J Life Cycle Assess 19(8):1515–1523CrossRefGoogle Scholar
  24. Rugani B, Panasiuk D, Benetto E (2012) An input-output based framework to evaluate human labor in life cycle assessment. Int J Life Cycle Assess 18:172–184Google Scholar
  25. UNEP/SETAC (2009) Guidelines for social life cycle assessment of products. United Nations Environment Programme, ParisGoogle Scholar
  26. UNEP/SETAC (2011) The methodological sheets for sub-categories in social life cycle assessment (SLCA). United Nations Environment Programme, ParisGoogle Scholar
  27. Watanabe MDB, Chagas MF, Cavalett O et al (2015) Hybrid input-output life cycle assessment of first and second generation ethanol production technologies in Brazil. J Ind Ecol. doi: 10.1111/jiec.12325 Google Scholar
  28. Zamagni A, Amerighi O, Buttol P (2011) Strengths or bias in social LCA? Int J Life Cycle Assess 16(7):596–598CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  • Alexandre Souza
    • 1
  • Marcos Djun Barbosa Watanabe
    • 1
  • Otavio Cavalett
    • 1
  • Cassia Maria Lie Ugaya
    • 2
    • 3
  • Antonio Bonomi
    • 1
  1. 1.Laboratório Nacional de Ciência e Tecnologia do Bioetanol (CTBE), Centro Nacional de Pesquisa em Energia e Materiais (CNPEM)CampinasBrasil
  2. 2.Universidade Tecnológica Federal do Paraná (UTFPR)CuritibaBrasil
  3. 3.CNPq fellowCuritibaBrazil

Personalised recommendations