Transformation and Sources of Growth in Southeast Asian Agriculture.

• Author: Birthal, Pratap S.

1. Introduction

   Rapid growth in agriculture is a precursor to the process of structural transformation, which eventually leads to a fall in the primary sector's share in gross domestic product (GDP) and workforce. The path and speed of structural transformation across regions, however, may vary depending on their relative endowments of land, labour and capital, technological capabilities, infrastructure, institutions and governance (Echevarria 1997; Gollin, Parente, and Rogerson 2007; Bah 2011; Felipe, Mehta, and Rhee 2015; Sen 2016).

   Southeast Asia has experienced robust economic growth in the recent past. Driven by the manufacturing and services sectors, the region's GDP grew at a rate of more than 5 per cent during the 2000-15 period, making it one of the fastest growing clusters in the world. However, the pace of structural transformation has not been as robust as in developed countries. The agricultural sector grew slowly, at an annual rate of about 3 per cent, and experienced a decline in its share of GDP, from 15 per cent in 2000 to 11 per cent in 2015. This outcome is that, notwithstanding the continued employment pressure, agriculture's share in the workforce during this period declined from 49 per cent to 38 per cent--still a high number and confined to some rich countries. Plus, the pace of structural transformation has differed markedly across the countries in the region (Sen 2016).

   The transformation of agriculture in Southeast Asia has been shaped by advancements in bio-chemical technologies, investments in irrigation and infrastructure, and innovations in information and communication technologies and financial systems (Sen 2016). These have led to a substantial increase in agricultural productivity, food supplies and rural wages (Dawe 2015; Vos 2018).

   The main aim of this paper is to understand the paths to agricultural transformation in different countries in the region, and to identify the factors underlying the transformation in a broader context of economy-wide changes in domestic and international markets. The paper tracks trajectories of agricultural growth and identifies factors responsible for differences in the sources of growth within and across countries. Towards this, we decompose the growth to estimate the relative contributions of four of its main components:

1. technological change,

2. area expansion,

3. diversification, and

4. price increases.

   There are several reasons to study the sources of growth in agriculture in the region. First, despite the rapid increase in food production, some countries are not self-sufficient in cereals and depend heavily on imports. By 2050, the Southeast Asian population is projected to increase by 20 per cent, to over 800 million. Therefore, the need to produce more food from limited resources remains important. Second, agriculture in most countries is dominated by small landholdings and is likely to remain under significant employment pressure if the slow rate of labour absorption in non-agricultural sectors continues. Third, the region is characterized by fast-rising incomes, urbanization, an expanding middle class, changing demographic profiles, growing health consciousness, increasing participation of women in the workforce--all these apart from being confronted with globalization of agri-food markets. Rapid changes have been observed in consumer preferences in favour of high-value, nutrient-rich foods, such as fruits, vegetables, milk, meat, egg and fish (FAO 2017). The factors shaping consumer preferences are unlikely to subside in the near future, implying a faster growth in demand for high-value food commodities in the long run. Being perishable, most high-value commodities require different kinds of markets and post-harvest infrastructure than the widely grown cereals or other such crops. Yet, the markets in some regional countries are underdeveloped, fragmented, imperfect, and dominated by intermediaries, and the value chains for most high-value commodities are not well developed (Habito, Capistrano, and Saguiguit 2016; Stillman and Rillo 2015).

   Also, climate change is posing a significant threat to sustainable agriculture (FAO 2017). Most countries in the region, being located in the tropics and squeezed between the Indian Ocean and the Pacific Ocean, are highly exposed to frequent risks of cyclones, floods and droughts. The frequency of climate extremes is predicted to increase in the future, which will adversely affect the performance of agriculture, and subsequently the livelihoods of people dependent on agriculture and agro-based industries. Addressing these challenges is essential for achieving faster, efficient and sustainable growth in agriculture--a prerequisite for accelerating the pace of structural transformation, especially in low-income countries that depend heavily on agriculture and lack technological capabilities and financial resources to cope with emerging challenges.

   We identify crops or crop groups that can potentially influence the pace of agricultural transformation in Southeast Asia via their effects on agricultural growth. A comprehensive understanding of sources of growth is important for prioritizing efforts and investments to cope with challenges to the future growth of agriculture. We believe that such an analysis can provide evidence-based feedback to policymakers, research managers, development organizations and financial institutions to design and implement technically feasible, economically viable and environmentally sound strategies for faster, efficient, sustainable and inclusive growth of agriculture.

   The rest of the paper is organized as follows. The next section briefly discusses data sources and analytical approach to assess the relative contributions of area expansion, price increases, crop diversification and technological change to agricultural growth. The subsequent discusses nature, extent and speed of agricultural transformation in regional countries. Relative contributions of different sources of growth are analysed and discussed in the fourth section. The fifth section discusses the factors that can explain differences in their contributions within and across the countries in the region. Conclusions and implications make up the last section.

2. Data and Analytical Approach

   2.1 Data Sources

   We use data on GDP by sectors from FAOSTAT, (1) and on workforce from the Asian Productivity Organization (APO). (2) Agricultural growth is an outcome of the change in cropped area or crop yields or real producer prices or land reallocation among crops or a combination of these. Information on these variables has also been sourced from FAOSTAT. However, the FAOSTAT database lacks time-series on producer prices for all commodities and for all countries. Yet, it contains time-series on value of output of most crops. We divide the value of output of each crop by its physical output to generate a series of producer prices as unit values. We then deflate the value of output by the GDP deflator (US dollar at 2010 constant prices). The time-series on agricultural data, especially commodity prices, is often noisy or shows a random walk that can potentially bias estimates of growth and its components. Therefore, we smoothen the data series on area, production and value of output applying the Hodrick--Prescott (HP) filter and use the transformed series for analysing the sources of growth. (3)

   2.2 Decomposition of Growth

   We decompose the overall growth in agriculture following a simple "growth accounting" approach as employed by Minot et al. (2006) and Joshi, Birthal, and Minot (2006). The growth or change in revenue from a single crop at two points in time or over a period can be decomposed into approximate effects of area expansion, yield improvements and price increases.

   Let [A.sub.i] be the area of crop i, [Y.sub.i] its yield and [P.sub.i] real producer price, then the revenue [R.sub.i] from crop i can be expressed as:

   [R.sub.i] = [A.sub.i][Y.sub.i][P.sub.i] (1)

   Summing up revenues of n crops provides total revenue:

   [Please download the PDF to view the mathematical expression] (2)

   In the decomposition of the total revenue from n crops, there is one more source of change, viz., area re-allocation across crops or crop diversification. To account for this, we express area of crop i as share in the total cropped area, that is, [Please download the PDF to view the mathematical expression], and substitute this in equation (2):

   [Please download the PDF to view the mathematical expression] (3)

   Total derivative of equation (3) provides us absolute contribution of each component to the change in total revenue:

   [Please download the PDF to view the mathematical expression] (4)

   The second term on the right-hand side of equation (4) can be further decomposed from a change in sums to the sum of changes, and can be written as:

   [Please download the PDF to view the mathematical expression] (5)

   Further expansion of the second term in equation (5) yields the following expression:

   [Please download the PDF to view the mathematical expression] (6)

   Equation (6) provides the change in total revenue due to a change in (i) total cropped area, (ii) crop yields or technological change, (iii) real produce prices, and (iv) land reallocation or diversification. The first term provides the change in gross revenue due to a change in the total cropped area. The second term captures the effect of a change in real prices of agricultural commodities on gross revenue. The third term measures the effect of a change in gross revenue due to a change in crop yields or technological change. The fourth term captures the change in gross revenue due to re-allocation of land among crops or a change in crop composition. A positive value on the fourth term suggests a re-allocation of land from lower-value to higher-value crops. On dividing both sides of equation (6) by the overall change in gross revenue, we get the proportionate contribution of the...