1 Introduction

To reinforce an effective green growth policy in agriculture, it is important to focus on maximizing production using a given level of inputs and technology and minimizing emissions resulting from the same level of inputs. The contemporary literature highlights the impact of climate change and environmental degradation on agricultural production (Nelson et al. 2009; ADB 2013; IPCC 2014; Lipper et al. 2014; Kaur and Kaur 2016). However, the way agricultural production (process) impacts emission, environmental degradation, and subsequent climatic change is rarely examined.

This study contributes to this gap in the literature by analyzing how the agricultural production process affects the environment through emissions while focusing on how countries under a regional cooperation framework can work together to best manage emission containment to attain the goal of sustainable green growth in agriculture. Specifically, this chapter examines the countries’ current state of emission-management efficiency and subsequently derives the impact of regional cooperation in achieving green growth in agriculture by reducing emissions and environmental degradation. For empirical analysis, 16 South-through-East Asian (StEA is used throughout the paper to represent the regional bloc) are considered.

The next section explains the methodology and model specifications. Results and findings are presented in Sect. 14.3. A composite indicator for combining agricultural production and emission-management efficiencies is explained in Sect. 14.4. Concluding remarks and policy implications are given in Sect. 14.5.

2 Methodology and Data

2.1 Methodology

This study uses the well-established stochastic frontier production function analysis (SFPFA), which facilitates the calculation of production and emission-management efficiencies.Footnote 1 This study uses the technical inefficiency effects model for using the panel data proposed by Battese and Coelli (1995). The basic model is designed as shown in Eq. (14.1).

Maximizing

$$\begin{aligned} \mathrm{ln}{output}_{i,t} & ={\beta }_{0}+{\beta }_{1}\mathrm{ln}{land}_{i,t}+{\beta }_{2}\mathrm{ln}{labor}_{i,t}+{\beta }_{3}\mathrm{ln}{capital}_{i,t} \\ & \quad +{\beta }_{4}\mathrm{ln}{fertilizer}_{i,t}+{\beta }_{5}\mathrm{ln}{energy}_{i,t}+{\beta }_{6}\mathrm{ln}{FDI}_{i,t} \\ & \quad +time-{U}_{i,t}+{V}_{i,t} \end{aligned}$$
(14.1)

\(i=1, 2, 3, \dots \dots \dots \dots ., k\) (for each country)

\(t=1, 2, 3, \dots \dots \dots \dots .,T\) (for each year)

Here, output refers to the aggregated agriculture output, while land refers to a country’s total arable land. Labor and capital denote the total labor (in number) and capital (in USD) deployed in agriculture. Fertilizer and energy refer to the amount of fertilizer and energy consumed in agriculture production. FDI denotes the cumulative foreign direct investment (FDI) inflow since 2000 in the country’s agriculture sector. FDI inflow is used as a proxy for technological progress. Time is incorporated to capture the time trend in agriculture production. Subscripts i and t represent the ith country and time, respectively. \({U}_{i,t}\) denotes the single-sided non-negative error term for the combined effects of inefficiency, on which complete information is not available. \({V}_{i,t}\) refers to the normal statistical error term, which captures the effect of inadvertently omitted variables. The ratio of the actual output to the estimated frontier potential output is defined as the technical efficiency or production management efficiency for the concerned country for the particular year.

To understand sustainable green growth in agriculture, we need to comprehend the interrelationship among inputs, output, and emissions. Primarily, inputs (such as land, labor, capital, fertilizer, energy, and technology) maximize agricultural production. Nevertheless, striving for higher production may also instigate the deployment of more input resources, resulting in higher emissions. Thus, examining the emission-management efficiency resulting from the same inputs is imperative.

The basic emission-management efficiency model should target the minimization of emissions from agricultural inputs and activities. The model is designed as shown in Eq. (14.2).

Minimizing

$$\begin{aligned} \mathrm{ln}{Emission}_{i,t} & = {\beta }_{0}+{\beta }_{1}\mathrm{ln}{Land}_{i,t}+ {\beta }_{2}\mathrm{ln}{Labor}_{i,t}+{\beta }_{3}\mathrm{ln}{Capital}_{i ,t} \\ & \quad + {{\beta }_{4}\mathrm{ln}{Fertilizer}_{i,t}+\mathrm{ln}{Energy}_{i,t}+\beta }_{6} {\mathrm{ln}FDI}_{i,t} \\ & \quad +time+{U}_{i,t}+{V}_{i,t}\end{aligned}$$
(14.2)

\(i=1, 2, 3, \dots \dots \dots \dots ., k\) (for each country)

\(t=1, 2, 3, \dots \dots \dots \dots .,T\) (for each year)

Here, emission refers to the aggregated emissions (of all greenhouse gases [GHG], including CO2, methane [CH4], and nitrous oxide [N2O]) produced in the different agricultural activities. The other variables (i.e., land, labor, capital, fertilizer, energy, FDI, time, and error terms) have been defined earlier. The estimated frontier potential minimum emission ratio to the actual emission is defined as the emission-management efficiency for the concerned country for the particular year. For the most emission-efficient country, the ratio will be 1, and greater than 1 indicates emission inefficiency.

The software FRONTIER 4.1 was used to perform the estimation with the maximum likelihood method as introduced by Coelli (1996). The country-specific production and emission-management efficiencies were calculated based on the methods suggested to estimate the stochastic production and cost frontiers.

2.2 Description of the Data

The data on aggregated agriculture production (in million USD), aggregated agriculture emission (in gigagram carbon dioxide equivalent [CO2e]), arable land (in thousand hectares), capital (in million USD), fertilizer (in kilogram [kg]), energy consumption in agriculture (in terajoule), and FDI inflows in agriculture (in million USD) are extracted from the database of the Food and Agriculture Organization of the United Nations (FAO). Labor (in thousand) data was collected from the Asian Development Bank’s (ADB) Key Development Indicators (various years).

Data were collected for the 2000–2013 period for all StEA countries except Bhutan, Brunei, and the Maldives, for which a complete set of data are not available. Along with these three countries, Singapore was also removed from further analysis because of its low-scale agriculture emission. In aggregate, these four countries emit only 0.03% of the total agricultural emission of the region; hence, skipping these countries from the analysis would not have a substantial influence on the agricultural policy implication.

3 Results and Findings

3.1 Estimations of Production Efficiency of the Countries

The agricultural production efficiency of the countries for the 2000–2013 period is presented in Table 14.1. It reveals that China has the highest production efficiency at 94.7%. Japan, Vietnam, and South Korea follow China with 91.1, 90.6, and 85.1%. Thailand, on the contrary, has the lowest efficiency at 36.5%. Cambodia, Lao PDR, and Nepal are the other least-efficient countries, with efficiency levels of 36.5%, 46.1%, and 49.8%, respectively. Among the other large-scale producers, India has a production efficiency of 65.2%, and Indonesia has 58.4%.

Table 14.1 Country-wise production management efficiency in agricultural production, 2000–2013
Full size table

3.2 Regional Cooperation

The impact of regional cooperation was estimated by considering the StEA as a single bloc. Inputs and output variables of all the countries were summed up to get the aggregated level of inputs and output for the whole region. This single bloc is then put into the model, and its efficiency was calculated by the maximum likelihood estimation by comparing the entities with regard to the best-practice performer. The estimation reveals that this regional bloc’s overall production management efficiency is 83.7%. It implies that if the countries could work under a regional cooperation bloc, on average, their agriculture production efficiency would be 83.7%. Since most of the large agriculture producers, such as China, Japan, and South Korea, have much higher efficiencies in production, the overall weighted efficiency of the StEA region remains relatively higher. It also means that the regional cooperation bloc can work together to further increase its production toward the untapped frontier potential production of 16.3% without deploying any additional resources (i.e., inputs).

To understand the link between enhanced production management efficiency and emission reduction potential, this study uses the calculation as shown in Table 14.2.

Table 14.2 Estimation of the link between optimal production efficiency and emission reduction
Full size table

With x amount of aggregated input, the regional bloc can produce y output. Since emission or environmental degradation results from using inputs in the agriculture production process, let’s assume that x amount of aggregated input leads toward an aggregated emission of m. With a further increase of efficiency by 16.3%, x amount of aggregated input will produce 116.3y amount of aggregated output, but the amount of aggregated emission will stay at m. Hence, to produce the agricultural output at the current level with this higher level of production management efficiency, the emission level may be reduced to m/1.163 or 86.0% of its current emission level. Therefore, the production management efficiency increase of the production process may save up to 14.0% of its current emission level, considering that the production level remains at its current level.

3.2.1 Synergy Effect for Potential Gap Reduction

The synergy effect claims that the combined action of a group of countries (i.e., regional cooperation) should bring added benefits over the sum of an individual country’s actions. Hence, it is important to calculate the synergy effect in any regional cooperation framework to understand the impact of combined action.

For this purpose, gaps in frontier potential agricultural productions (i.e., frontier potential production minus actual production) are calculated separately for all 16 countries. Then the model considered the whole regional bloc (StEA) as a single entity and calculated the gap for that whole bloc. As Table 14.3 shows, the gap in the potential agricultural production of the StEA (as a single bloc) is smaller than the sum of all 16 countries’ gaps. The calculation, thus, supports the synergy effect phenomenon. To quantify the impact of regional cooperation, the differences between those gaps from the potential (the sum of individual country’s gaps minus the whole bloc’s gap) are measured as percentages of the gap without cooperation. Hence, the synergy effect of the regional bloc is calculated as:

Table 14.3 Synergy effect measure (i.e., the impact of regional cooperation) to close the potential gaps
Full size table
$$\begin{aligned} & Impact\,of\,regional\,bloc\,(synergy\,effect) \\ & = \frac{\left(Gap\,without\,forming\,regional\,bloc-Gap\,as\,a\,regional\,bloc\right)} {Gap\,without\,forming\,regional\,bloc} = \\ & = \frac{\left(sum\,of\,gaps\,from\,potential\,of\,each\,country-gap\,from\,potential\,for\,StEA\right)}{sum\,of\,gaps\,from\,potential\,of\,each\,country}\end{aligned}$$

Table 14.3 reveals that the synergy effect, on average, is 34%

3.3 Estimations of the Emission-Management Efficiency of the Countries

The emission-management efficiency of the countries for the 2000–2013 period is shown in Table 14.4. It depicts that China has the highest emission-management efficiency at 70.4%. Thailand, Malaysia, and Sri Lanka follow China with 60.3%, 58.7%, and 54.3% efficiency. India and Indonesia have 42.8% and 53.3% emission-management efficiencies, respectively. Myanmar, Lao, and Cambodia are the three least-efficient countries, with efficiency levels of 30.5%, 32.2%, and 34.0%, respectively.

Table 14.4 Country-wise emission-management efficiency in agricultural production (2000–2013)
Full size table

3.3.1 Impact of Regional Cooperation

A similar approach as described earlier is adopted to measure the impact of regional cooperation on the aggregated emission-management efficiency, which considers the StEA as a single bloc. Inputs and output variables of all the countries are summed up to get the aggregated level of inputs and output for the whole region. This single bloc is then put into the model, and its efficiency was calculated using the maximum likelihood estimation by comparing the entities with regard to the best-practice performer. The estimation reveals that the overall technical efficiency of this regional bloc for emission containment is 52.6%. It implies that if the countries could work under a regional cooperation bloc, on average, their agriculture emission-management efficiency would be 52.6%. It also refers that if the regional cooperation bloc can work together with the given set of inputs, the region can further contain (i.e., decrease) the agricultural emission by 47.4% from the current level.

This study uses the following calculation to understand the link between the potential emission-management efficiency and the potential production level, as shown in Table 14.5.

Table 14.5 Estimation of the link between optimal emission-management efficiency and potential production level
Full size table

With x amount of aggregated input, the regional bloc can produce y output and m level of emission. Now for the potential efficiency in emission containment, with the given set of inputs, emissions can be further reduced by 47.4% from the current level. Therefore, the optimal emission level with the given input (x) and given production level (y) would be (1 − 0.474)m or 0.526 m.

Now, assume that the StEA region adopts the potential emission-management efficiency practices. If at that efficiency level, it wants to remain at the current emission level (m), it would, in turn, allow the aggregated input level to (1/0.526) x (i.e., 1.90x). Since x level of input produces y level of output, the new level of input (i.e., 1.90x) would produce a 1.90y level of output. Therefore, the potential production management efficiency in the emission-management process may further enhance the production level by 90% from its current level, considering it allows emissions to remain at the current level.

4 Green Growth Index in Agriculture (GGIA): A Composite Indicator

For an effective green growth policy in agriculture, a country should simultaneously focus on growing its production and lowering production-related emissions. This becomes plausible by attaining higher efficiency both in production and emissions management. In reality, some countries may have higher production management efficiency owing to better use of inputs to maximize production, while some countries may have high efficiency in managing emissions. Hence, a combined index needs to be prepared to know the resultant efficiency toward attaining a green growth policy. Saltelli (2007) mentions that a composite indicator is easier for general interpretation and more useful in evaluating performance than following several separate indicators’ trends. It is rational to argue that a composite indicator (or combined index) always helps initiate the discussion and motivate the common interest.

Though the formulation of a composite indicator seems to follow the mathematical or computational models, the essence and justification for it depend on the intended purpose, peer acceptance, and the craftsmanship of the modeler (Rosen 1991). The Organisation for Economic Co-operation and Development (OECD 2008) handbook identifies two key steps (i.e., multivariate analysis and normalization) as essential steps for aggregation. Multivariate analysis requires that the structure of the dataset is investigated and its suitability for combination be assessed. Since this study attempts to combine two factors, both efficiency terms, and for the same set of 16 countries, it is plausible to combine. Normalization is also attained since both efficiencies are estimated based on best practices as the benchmark. Yet, the key discussion for an aggregation technique remains toward the arbitrary weighting process (Sharpe 2004). This study reveals that both agriculture production management efficiency and emission-management efficiency are equally significant for attaining sustainable green growth in agriculture. It also shows strong linkages between the optimal production efficiency with emission reduction and optimal emission-management efficiency with increasing production. Hence, equal weights were considered for the aggregation technique to concurrently emphasize the countries’ production and emission-management efficiencies.

The next question is whether the additive (linear aggregation) or multiplicative (geometric aggregation) technique is suitable for this study. In fact, a country with lower scores would prefer an additive aggregation technique over the multiplicative. However, it also means that multiplicative aggregation would set greater incentives to address the limiting factors more intensely for the low-score country as it would give it a better chance of improving its position in the ranking (Munda and Nardo 2005). The multiplicative aggregation technique was considered to emphasize improving the countries’ performance. The GGIA is thus proposed as follows:

$$\mathrm{GGIA}= \sqrt{\mathrm{Production\,efficiency }\times \mathrm{Emission}-\mathrm{management\,efficiency}}$$

Based on 2000–2013 average efficiencies, GGIA was calculated (Table 14.6). China (82%), Japan (67%), and South Korea (66%) have the highest GGIA. Cambodia (40%), Lao PDR (40%), and Thailand (47%), on the contrary, have the lowest GGIAs. The whole of the StEA region has a GGIA of 66%.

Table 14.6 Countries’ overall performance regarding production efficiency, emission-management efficiency, and GGIA
Full size table

5 Concluding Remarks and Policy Implications

5.1 Concluding Remarks

An ever-growing population and the climate change phenomenon have compelled producers to exploit the resources in agriculture production even faster. This study examined regional cooperation’s prospective roles in attaining sustainable green growth by enhancing countries’ production management and emission efficiencies.

This study adopted the stochastic frontier model to estimate the production management efficiency and emission efficiency levels of 16 countries. The empirical results reveal that China has the highest production management efficiency. Japan, Vietnam, and South Korea follow China, while Thailand, Cambodia, Lao PDR, and Nepal are the least-efficient countries in terms of agriculture production management among the StEA countries. The estimation reveals that this chosen regional bloc’s overall production management efficiency is 83.7%, implying that it can work together to further increase the region’s agriculture production by 16.3% toward untapped potential production without deploying any additional resources. Estimating the link between potential production efficiency and emission reduction reveals that under the fully-efficient scenario, an increase in the production management efficiency may reduce 14.0% of emissions from its current level, considering the production level remains the same. The synergy effect calculation also revealed that the StEA countries could have improved their production closer to the potential had they worked under a common regional cooperation bloc than working separately.

This study also examined the countries’ current state of emission-management efficiency and subsequently derived the impact of regional cooperation in achieving green growth in agriculture by reducing emissions and environmental degradation. Empirical results revealed China had the highest emission-management efficiency (70.4%), while Myanmar had the lowest (30.5%). The results further show that the overall emission efficiency in controlling emissions from agriculture is 52.6%. It also emphasizes that if the regional cooperation bloc can work together with the given set of inputs, the region can further decrease emissions from agriculture by 47.4% from the current level. The calculation also shows that regional cooperation has a positive synergy effect on emissions management for the StEA countries.

A combined index had to be prepared from each country’s production management and emission-management efficiency to understand the resultant efficiency toward attaining green growth. In this regard, the multiplicative (geometric aggregation) technique was used to generate the GGIA. The calculation showed that China, Japan, and South Korea have the highest GGIA while Cambodia, Lao PDR, and Thailand have the lowest. The GGIA of the StEA region as one bloc was 66%.

5.2 Policy Implications

The analysis presented in this study has several policy implications for strengthening sustainable green growth in the agriculture sector of StEA countries. Information on the production and emission-management efficiencies can help countries have a comprehensive idea about their respective strengths and challenges, which can help them improve their efficiencies. From a regional cooperation perspective, a policy framework based on the analysis would provide wide-ranging tools to manage intra-regional demand and supply of foods more efficiently while complying with the measures necessary for the transition toward low-emission agriculture systems. Obviously, under a regional cooperation framework, countries with higher efficiency in production management should produce more so that resources within the regional bloc are efficiently managed. Higher production under the regional cooperation framework is feasible by accelerating technology transfer, knowledge sharing, and capacity building between high-efficient and low-efficient countries. Institutional settings at the regional level should be strengthened to constantly monitor the level of progress and disseminate adequate policy, rules, and technical support to all member countries. Creating a common fund to finance agricultural green growth projects may also play an important role. In this context, the role of multinational financial institutions like the Asian Development Bank (ADB), Asian Infrastructure Investment Bank, and the World Bank is crucial. Easing trade restrictions on agricultural production inputs may also facilitate efficient production among the countries (Kalirajan and Anbumozhi 2014).

Recollections of Professor Keijiro Otsuka

Keijiro Otsuka, fondly known as ‘Kei’ among his friends, has always been very keen to introduce ‘innovation’ in terms of methodology and empirical discussions in his research. He believes in primary data collection and face-to-face interviews with stakeholders. He enumerates clearly the basic sources of economic issues through this approach, which helped him arrive at appropriate models to provide feasible solutions.

I have never seen him with a ‘long face’ when I worked with him from 2001 to 2009 in FASID/GRIPS. He was always cheerful and encouraged his students and colleagues to make a significant contribution to the literature. In this context, Kei has been a ‘nitty–gritty’ researcher and a ‘tactful taskmaster,’ which I learned when we coedited three books and two special editions of Developing Economies (2006) and the Journal of Agricultural and Development Economics (2005).