Modeling the Future of Indonesian Food Consumption, June 2018

Originally published
View original


Executive Summary

Modeling the Future of Indonesian Food Consumption

The growing food consumption demand in Indonesia has provided serious challenges for food policy which will have an impact in the years to come. Existing policies to increase the production of staple foods such as rice, maize and soybeans (Upsus Pajale) may not be adequate to meet increasing food demand. A growing population and middle class in Indonesia, and a high rate of urbanization have led to changes in the dietary patterns and food demand of the future. Higher incomes and better knowledge tend to make consumers demand healthier and more diversified food.

This study aims to develop a model of the future of Indonesia’s food demand up to 2045, using a baseline of food consumption in 2017 and projections to 2025 as milestones, and to draw policy relevance on food and related issues including the Medium-Term Development Planning (RPJM) of 2020-2024 by Bappenas. Food commodities include rice, maize, soybeans, sugar, beef, poultry, and fruit such as oranges, apples, bananas, mangoes and snake fruit, and vegetables such as shallots, garlic, red chilies, hot chilies, spinach and kangkung (swamp cabbage). The projected demand of Indonesian food consumption for 2025 and 2045 is built based on the functional relationship between income and food consumption at the baseline using three different scenarios of economic growth: baseline, moderate and optimistic. The food projections are built on some assumptions of population projections, composition of rural-urban population, income per capita and food affordability per capita. Almost Ideal Demand System was implemented to estimate estimated changes of food consumption with regard the changes of it’s ownr price, other food price and income. Susenas data from 2017 is used as the baseline of food demand model, including income elsticities estimation.

Impact of own price, cross elasticities and income elacticities were predicted using the almost ideal demand system (AIDS) model. Susenas data from 1990-2016 is used to analyze selected food consumption trends and examine the relationship between food consumption, price trends, and income in all 33 provinces of Indonesia.
The results show that future food demand in Indonesia is determined by existing demand, income, price and its composition, and various other factors that affect the behavior and trends of consumption. Average rice consumption in 2017 was recorded at 97.6 kilograms per capita per year, which was significantly lower than the official rice consumption on 114 kilograms per capita.
Average consumption of beef, as a protein source, is extremely high in the highest income group.
Average beef and poultry consumption in Quintile 5 is 6 and 14.7 kilograms per capita per year respectively, which is higher than the national average of 2.5 and 7.5 kilograms per capita per year.
A contrasting figure is found in maize consumption for humans, which averages 2 kilograms per capita per year, but this figure declines as income increases.

The per capita rice consumption projection at the baseline gradually increases by 1.5 percent to 99.08 kilograms per capita per year in 2025 and increases by 2 percent to 99.55 kilograms per capita in 2045. The projection of the demand for rice, after referring the correction factors at Food Balance Sheet (NBM) for the domestic uses of food for non-food purposes, such as industrial use for nonfood, feed, seed, food loss, also increases to 102.73 kilograms per capita per year in 2025 and 103.22 kilograms per capita per year in 2045. The demand for rice is also projected to increase to 127.09 kilograms per capita in 2025 and 127.70 kilograms per capita 2045, after considering food loss and waste in line with the FAO (2011). Rice consumption has different characteristics among different income groups and in rural and urban areas. Only in the highest income group has Indonesia experienced declining rice consumption, which is somewhat different from other Asian countries, where rice consumption declines are also found in medium and lower level income groups.

The poultry consumption projection shows the highest increase compared to other animal products, which is 22.1 percent in 2025 to 9.13 kilograms per capita per year, and 29.3 percent in 2045 to 9.66 kilograms per capita per year. The beef consumption projection increases by 10.3 percent to 2.79 kilograms per capita per year in 2025, and 20.4 percent to 3.04 kilograms per capita per year in 2045.
The fish consumption projection increases by 11 percent to 29.09 kilograms per capita per year in 2025 and 14.6 percent to 30.04 kilograms per capita per year in 2045. Beef is consumed by higher income groups in urban areas. Poultry is consumed by all income groups, including the lowest quintiles. Fish is consumed by both the urban and rural population.

In the category of fruit and vegetables, the highest food consumption demand projection per capita relates to apples, with an increase of 55 percent in 2025 to 1.49 kilograms per capita per year, and 73.5 percent in 2045 to 1.66 kilograms per capita per year. Consumers of apples are mostly part of the urban population in high and medium income groups. The projected demand for local fruit such as oranges, bananas, snake fruit and mangoes in 2025 and 2045 is not as high as apples, and this demand is dominated by imported apples.

The projected demand for sugar is 8.98 and 9.12 kilograms per capita in 2025 and 2045 respectively.
The increase in sugar consumption is not very significant, compared to other food commodity groups. The total consumption of sugar is projected to reach 25.6 million tons in 2025 and 29.1 million tons in 2045. The projection estimates of sugar consumption do not include indirect consumption of sugar in the form of cakes, drinks and other food products that use refined sugar and its derivatives, so the figure may be higher.

As rice remains a staple food, even in 2045, the policy relevance is that elements of food consumption could determine the level of food accessibility, and therefore food security in the country. Ensuring access to rice, especially for low and middle-income groups, is as important as the stability of the retail price of rice. The policy of food assistance targeting the poorest group of rice consumers remains relevant to maintain food and nutrition adequacy. As the government is planning to transform in-kind food assistance to non-cash subsidies, the implementation of such targeted subsidies could be adjusted in line with the latest development of infrastructure, data technology and preparedness of the stakeholders in the overall food system.

As the income elasticity of beef, poultry and fish remains high, the policy relevance is based on infrastructure improvements of the marketplace, including both modern retail markets and traditional markets which could shape the performance of value chains of these sources of animal protein. The value chain policies not only directly affect food accessibility among all income groups, but also affect many value chain players, such as retailers, processors, wholesalers, and collector traders that directly connect rural areas to farmers or producers of protein sources. These players could also help convey the messages of urban consumers to farmers and other actors along the value chains, including product specifications, food safety, health and hygienic requirements, and halal and other quality standards that have shaped the characteristics of the value chains of these protein sources.

For perishable fruit and vegetable products, the policy relevance is that actors in fruits and vegetable value chains need access to cold-storage facilities, including medium scale controlled atmosphere systems (CAS) which could improve the efficiency of horticulture products. The policy should also focus on the balance between demand-side management and supply-side or productivity improvement, as the majority of horticulture production centers are located in Java. As the majority of horticulture products are marketed through cooperation with the wholesale markets (pasar induk), large and medium cities should implement spatial planning and zoning policies for end-to-end waste management in these traditional horticulture markets. The policy relevance is that there is a need to iv support farmers who could meet rising quality and safety standards set by retail markets, or by consumers, through modern retail markets and supermarkets.

The food demand modeling exercise in Indonesia in this study is mostly aimed at strengthening national level policy analysis. Specific models for regional or sub-national levels might follow similar procedures, but extra care should be given to the structure and availability of Susenas data as the baseline. The government will play an important role in the response to the results of this study, and in anticipating various changes in demand for selected important foods in Indonesia.