Economic Analysis of Smallholder Rubber Agroforestry System Efficiency in Jambi Indonesia

BOGO

INDONESIA

THESIS

ATUHAIRE RODGERS

GRADUATE SCHOOL

BOGOR AGRICULTURAL UNIVERSITY

2008

work with the research topic entitled,

“ECONOMIC ANALYSIS OF SMALLHOLDER RUBBER AGROFORESTRY SYSTEM EFFICIENCY IN JAMBI”

Therefore I confirm that;

1. This thesis was done wholly while in candidature for Master of Science degree in Agricultural Economics at this University;

2. Where I have quoted from the work of others, the source was given. With the exception of such quotations, this thesis is entirely my own work;

3. I have acknowledged all main sources of help;

4. This work was done under the guidance of Prof. Dr.Ir. Bonar M.Sinaga, MA and Drs. Suseno Budidarsono, M.Sc (Bogor Agricultural University and World Agroforestry Center). Supervisor’s contribution to this research and thesis was consistent with normal supervisory practice

5. Wherever contributions of others are involved, every effort was made to indicate this clearly, with due reference to the literature.

6. None of this work has been published before submission.

Bogor, September, 16th, 2008 ATUHAIRE RODGERS

Atuhaire Rodgers, 2008. ECONOMIC ANALYSIS OF SMALLHOLDER

RUBBER AGROFORESTRY SYSTEM EFFICIENCY IN JAMBI (Bonar M.

Sinaga as the Chairman, and Suseno Budidarsono as the Member of the Advisory Committee).

The objectives of this research were; (1) to analyze production efficiency and profitability of smallholder rubber monoculture and smallholder rubber agroforestry systems, and (2) to find out the effects of policy distortions towards rubber production under smallholder monoculture and smallholder agroforestry. Smallholders are not only judged by yield per hectare; economic efficiency is not only a matter of returns to land and returns to labor but smallholders can adopt low input strategy, continue making profits at prices that would be economically viable. Such flexibility offers the possibility of efficient resource allocation in response to diversification of economic opportunities.

The Policy Analysis Matrix (PAM) with the Domestic Resource Cost (DRC) results under the baseline scenario indicated that the use of domestic resources in production of rubber was efficient and socially profitable under the two systems but more desirable under monoculture system given the current prices for physical inputs, outputs, technologies and policy transfer. However, even with sensitivity analysis of 10% increase in the price of rubber holding other factors constant does not make rubber agroforestry system more efficient than its counterpart and a 20% fall in price of rubber made rubber production under agroforestry system less efficient and undesirable. All measures are compared to the alternative policy indicators currently used. Therefore, recommendations made from this study relate to the need for diversification into better practices that can sustain efficient rubber production under the agroforestry system, encouraging private sector participation and reducing disincentives to rubber production.

Key words: Smallholder Rubber Agroforestry, Economic Efficiency, Policy Analysis Matrix

Based on the entire objectives of this research, the Policy Analysis Matrix (PAM), as an attractive organizational policy framework was used to analyze production efficiency and profitability using farm budget analysis comparing multi-year land use system budgets (smallholder rubber monoculture and smallholder rubber agroforestry systems) calculated at private and social prices and NPV valued at private prices as an indicator for production incentive and then Social prices that remove the impact of policy distortion (taxes, subsidy and other local levies). PAM indicators especially Domestic Resource Cost ratio (DRC) and Private Cost Ratio (PCR) were used as well. In other words, with DRC < 1 indicating that the use of domestic factors was socially profitable. PCR < 1 indicating that there was value added to the use of the domestic resources. All of these components were combined to provide a measure of total profitability, economic efficiency and effect of policy distortions from the existing divergences. Therefore, all results were consistent with the existing policies.

The study reveals that most of the labor costs were spent under rubber garden maintenance and its related activities mostly done by hired laborer. Under smallholder rubber monoculture system the total expenditures spent on rubber garden maintenance was Rp 4 344 million (23% out of the total cost) while, under smallholder rubber agroforestry system farmers spent Rp 2 630 million (31% out of total cost).

The discounted operational costs for both systems were financially ranging from Rp 24 797 million under agroforestry to Rp 48 168 million under monoculture, and economically ranging from Rp 37 977 million under smallholder rubber agroforestry system and 75 224 under smallholder rubber monoculture system. Based on farm budget calculation, the study also revealed that monoculture system in managing the plantation during rubber establishment has higher returns, employs more labor and also more profitable than smallholder rubber agroforestry traditional systems.

Based on the macroeconomic parameters of 2007, returns to land per hectare at private prices are 46 737 million for smallholder rubber monoculture system and Rp 18 254 million for smallholder rubber agroforestry system respectively. Economically (farm budget calculation valued at social prices), returns to land for those systems are Rp 119 492 million (smallholder rubber monoculture system) and

The analysis further reveals that with baseline scenario monoculture has a comparative advantage in the production of rubber than its counterpart, as indicated by DRC’s of 0.31 and 0.37 respectively. However, even with a 10% increase in the price of rubber holding other factors equal, agroforestry system remained less efficient than monoculture system i.e. with the DRC values of 0.35 and 0.28 respectively. This indicates that smallholder monoculture system has efficient use of resources especially when farmers get an increase in price of rubber with the prevailing rubber prices.

Social profits, which are an efficiency measure, are all positive under both systems, and indicate that scarce resources were well utilized by producing at social costs that did not exceed the importation costs. Monoculture system had a bigger NPV as compared to Agroforestry system and therefore, monoculture was found to be more efficient under the existing technologies and efficient policies.

Factor transfers show negative values, which imply positive incentives for smallholder rubber farmers under both systems. This can be attributed to the primary factors of production. Therefore factor transfers may include some effect of the policies and market imperfections that influence the profitability of alternative crops. With regard to output transfers, all values were found to be negative, which is a reflection of disincentives to farmers.

It was noted that not all policies distort the allocation of resources; some policies however, endorse to improve efficiency by correcting failure of the product. A negative transfer in the total revenue indicated that the smallholders were receiving less than the border parity price for the commodity. A negative transfer in the domestic factors represents a positive transfer to the producers.

The value of tradable input transfer was negative for both smallholder rubber monoculture and smallholder rubber agroforestry and the NPCI of these systems were less than one this indicated that producers in both systems were not taxed when tradable inputs were bought.

The general conclusion from this analysis is that even with the assumption and alteration in farm gate price of rubber and real interest rate (i.e. private and

Price distortions reduce production of rubber under smallholder rubber agroforestry system. A 20% decrease in parity price of rubber made the crop undesirable from the social point of view. The opposite is also true (i.e. with a depreciating RER, profits realized in excess of normal returns to domestic resources will decrease). When the price of rubber falls, smallholders under rubber agroforestry system reallocate factors of production (i.e. capital and labor) in order to meet their daily expenses and resume tapping when there is an increase in rubber prices again. Whereas; Smallholders under monoculture system maintain their tapping schedule and maintenance schedule

While more intensive rubber monoculture offers better productivity (yield and profitability), it also requires much higher capital and input that is beyond reach for smallholders under rubber agroforestry especially during the immature period.

Finally, Rubber agroforestry system has the capacity to provide smallholder farmers with diversified income and a range of non timber forest products than monoculture although for more economic benefits, monoculture stands out to be a better option.

1. No part or this entire thesis may be excerpted without inclusion or mentioning the source:

a. Excerption only for research and education use, writing for scientific papers, reporting, critical writing or reviewing of a problem.

b. Excerption does not inflict a financial loss in the proper interest of Bogor Agricultural University.

2. No part or this entire thesis may be transmitted and reproduced in any form without a written permission from Bogor Agricultural University.

INDONESIA

ATUHAIRE RODGERS

Thesis

Submitted for the partial fulfillment of

Master of Science

degree in

Agricultural Economics

GRADUATE SCHOOL

BOGOR AGRICULTURAL UNIVERSITY

2008

Name : Atuhaire Rodgers Registration Number : H353078211

Study Program : Agricultural Economics

Approved by, Advisory Committee

Prof.Dr.Ir. Bonar .M.Sinaga, MA Drs. Suseno Budidarsono, M.Sc

(Chairman) (Member)

Endorsed by,

Programme Coordinator Dean of Graduate School Agricultural Economics

Prof.Dr.Ir. Bonar M. Sinaga, MA Prof. Dr.Ir.Khairil A. Notodiputro, MS

There are several people without whom this thesis would not have been at all possible and whom I need to thank:

I am indebted to government of Indonesia for awarding me a scholarship to pursue Master of Science degree in Agricultural Economics under Bogor Agricultural University, World Agroforestry and Jessica Roy Memorial Research Scholarship that enabled me carry out this study for my thesis production. Without them, my further prospects for this research would not have been all that possible.

Prof. Dr. Ir. Bonar M. Sinaga, MA and Drs. Suseno Budidarsono, M.Sc. as my supervisors they have been remarkably patient, cooperative and constructive in all phases of the thesis work. Their encouragement and stimulating guidance has allowed us to integrate our separate ideas into a coherent thesis. They have given me a high degree of freedom in the exploration of my research, while keeping me cognizant of my research direction and more importantly; its completion, many things I have learned and gained from it.

Mrs. Hellen Ochieng moderated the production of this thesis between two institutions until its final stage between i.e. World Agroforestry Centre and Bogor Agricultural University. Her efforts are highly appreciated for they made us finish within the expected time frame of the research fellowship.

Dr. Nunung Kusunadi, my external examiner and the chairman of Agribusiness department, Bogor Agricultural University who provided me with good suggestions and comments on the overall presentation of this thesis.

contributions ranged from counseling and moral support.

My friends; Ronnie Babigumira, Joseph Obado, Arif Rahmanulloh, Mamihery Ravoniarijaona and Marty Hasu Linda provided me with insightful and challenging comments at various stages of my thesis production.

I would like to express my sincere gratitude to my mum Mrs. Perry karamuzi for her encouragement, “behind a strong lady there is a strong man”. And to my late dad Mr. Paddy Karamuzi who had been waiting for too long for my study. His sincere love and care are impossible to repay. My brothers; Atwinematsiko Andrew and Atuhaise Benjamin, Sisters; Ayebazibwe Calvin and Akankwatsa Catherine who stood by my side until this day.

Finally, to all those that managed to provide me with valuable comments and inputs during various consultations.

Bogor, September 16th, 2008 Atuhaire Rodgers

Bushenyi District in the western part of Uganda. He is the first born out of five children of Mrs. Perry Karamuzi and late Paddy Karamuzi.

He completed his primary school in 1993 at Kyeitembe primary school, Bushenyi. In 1997, he completed Uganda Certificate of Education at Muntuyera High School, Ntungamo district. In 1999 he finished Uganda Advanced Certificate of Education at Valley College, Bushenyi District. In 2000 he won His Excellency Martin Brennan and family scholarship to pursue undergraduate’s degree in Environmental Management at Makerere University finished his studies in 2003. In 2004 he won another scholarship from Capacity Building International (INWENT) to pursue an advanced training course in wind energy applications at Kassel University of Technology in Germany. He was also an intern at the World Wind Energy Association headquarters where he wrote a book entitled, “Africa’s Renewable Energy Sustainability Guidelines”. In 2005 he was appointed as the Country Representative of Africa Wind Energy Association – Uganda.

In 2006 he won a scholarship to pursue Masters Degree in Agricultural Economics at Bogor Agricultural University, in the same year he worked as an intern at the Center for International Forestry Research (CIFOR) focusing on biofuels and environment. He was also selected to take part in His Excellency Horst Koehler’s initiative, “Partnership with Africa” between Africa and Germany under the group

In 2007, he was invited for a roundtable discussion under the Civil Society and the Joint EU–AU Strategy to discuss how EU can handle Africa’s problems (http://europafrica.files.wordpress.com/2007/04/tn_liste200407fin.pdf). In the same year he was among the winners of World Agroforestry and Roy Joy Memorial Master Thesis Research Fellowship.

In 2008, he was crowned as the Best Presenter in Bogor Agricultural University International Students Conference and now expects to graduate in October 2008.

LIST OF TABLES... LIST OF FIGURES... LIST OF BOXES... LIST OF APPENDIXES...

i iii iv v I. INTRODUCTION... 1

1.1. 1.2. 1.3. 1.4.

Background... Statement of the Problem……… Objectives of the Study……….. Scope………...

1 5 7 7

II. LITERATURE REVIEW……… 8

2.1. 2.2. 2.3. 2.4. 2.5. Smallholder Rubber... 2.1.1. Rubber Agroforestry System... 2.1.2. Rubber Monoculture System... Policies Affecting Agriculture... 2.2.1. Agricultural Price Policy Instruments………... 2.2.2. Macroeconomic Policies Affecting Agriculture………... Smallholder Export Crop Production... Review on Measuring of Efficiency... 2.4.1. Smallholders Production Efficiency... 2.4.2. Adopted Model on Measuring Efficiency... Measuring Policy Distortions... 2.5.1. Distortions in Agricultural Markets... 2.5.2. Adopted Model... 2.5.3. Subsidy on Inputs ...

8 10 12 13 13 14 16 24 27 28 30 30 31 32

III. RESEARCH METHODOLOGY... 36 36 38 38 38 39 40 40 41 41 43 44 45 47 49 49 51 54 55 56 3.1. 3.2. 3.3. 3.4. 3.5. 3.6.

The Conceptual Framework... Hypothesis... Selection Criteria for Area of Study... 3.3.1. Selection Procedures... 3.3.2. Survey Locations... Sources of Data... Secondary Data Update... The Policy Analysis Matrix (PAM)... 3.6.1.The Farm Budget Analysis Approach... 3.6.1.1. Estimation of Private Costs and Revenues... 3.6.1.2. Estimation of Social Costs and Revenues... 3.6.1.3. Pricing the Costs and Returns... 3.6.2. The PAM Table Approach of Assessment... 3.6.2.1. Private Profitability... 3.6.2.2. Social Profitability... 3.6.2.3. Important Policy Parameters from the PAM... 3.6.3. Data needed for Analysis... 3.6.4. Sensitivity Analysis... 3.6.5. Strength and Weakness of PAM... IV. THE CASE STUDY AREA... 57

4.1. The Study Sites... 4.1.1. Physical Characteristics... 4.1.2. Land-use and Planted Area by Major Crops...

57 4.2. 57 59 Demographic Characteristics... 4.2.1. Population... 61 61

4.2.2. Size of the Household... 63 V. ECONOMIC PERFORMANCE OF SMALLHOLDERS... 65

5.1. Farm Budget Analysis and Profitability... 5.1.1. Cost of Rubber Establishment... 5.1.2. Returns to Land and Returns to Labor Assessment...

65 65 68 5.2. 5.3. 5.4.

Efficiency and Profitability with PAM Table Analysis... 5.2.1. Profitability... 5.2.2. Domestic Resource Cost ... Effects of Divergences or Policy Transfers... Effects of Fertilizer Subsidy...

70 72 73 74 77 VI. IMPACT OF POLICY DISTORTIONS ON RUBBER

PRODUCTION... 79 6.1.

6.2. 6.3.

Effects of an Increase in Producer Prices... Effects of an Increase in Interest Rate... Effects of Policy Distortions...

80 82 83 VII. CONCLUSIONS, POLICY IMPLICATIONS AND AREAS FOR

FURTHER RESEARCH... 87 7.1.

7.2. 7.3.

Conclusions... Policy Implications... Areas for Further Research...

87 89 90 REFERENCES... APPENDIX... 91 95

i

Number Page

1. Rubber Area and Growth in Indonesia from 1970 to 2003... 2

2. Natural Rubber Production and Growth in Indonesia from 1970 to 2003... 3

3. Capacity and Production of Crumb Rubber Factories Based on Provinces from 2002 to 2004... 19

4. Export Volume of Indonesian Natural Rubber Based Product from 1969 to 2002... 21

5. Volume, Value and Average Price of Indonesian Exported Natural Rubber in Indonesia from 1969 to 2002... 22

6. Indonesian Rubber Export Based on Destination from 1996 to 2001... 23

7. 2007 Macroeconomic Parameters Used in Data Update... 46

8. The Policy Analysis Matrix... 48

9. Planted Crop area by Important Crops and Administrative Locations in Hectares... 61

10. Land Area, Percentage to the Total Area, Population and Population Density 2005 Covered per District... 62

11. Number of Households by Occupation and Administrative Locations in 2005... 64

12. Cost Composition of Rubber Establishment in Private Prices... 66

13. Years of Positive Cash Flow and Operational Costs... 67

14. 30 Years Profitability Matrix for Smallholder Monoculture and Smallholder Agroforestry Systems... 70

ii

iii

Number Page

1. Number of Actively Producing Crumb Rubber Factories from 1969 to 2001... 18 2. Conceptual Framework for Analyzing Policy Impacts... 37 3. Geographical Location and land cover 2005... 58

iv

Number Page

v

Number Page

1. Private and Social Price Table for Smallholder Rubber Monoculture

System... 96 2. Private and Social Price Table for Smallholder Rubber Agroforestry

System... 97 3. Input-Output Table for Smallholder Rubber Monoculture

System... 98 4. Input-Output Table for Smallholder Rubber Agroforestry

System... 102 5. Private Budget Table for Smallholder Rubber Monoculture

System... 107 6. Private Budget Table for Smallholder Rubber Agroforestry

System... 112 7. Social Budget Table for Smallholder Rubber Monoculture

System... 117 8. Social Budget Table for Smallholder Rubber Agroforestry

System... 122 9. Cash Flow for Smallholder Rubber Monoculture

System... 127 10. Cash Flow for Smallholder Rubber Agroforestry

System... 130 11. Extended PAM Table for Smallholder Rubber Monoculture

System... 133 12. Extended PAM Table for Smallholder Rubber Agroforestry

System... 134 13. Labor Requirement for Smallholder Rubber Monoculture

vi

15. Labor Allocation for Smallholder Rubber Monoculture

System... 145 16. Labor Allocation by Stages for Smallholder Rubber Agroforestry

System... 146 17. Import Parity Price Calculation of Rice at Farm Gate... 147 18. Import Parity Price of Fertiliser... 150 19. Export Parity Price of Rubber at Farm Gate... 152 20. Domestic Timber Price of Merati... 154 21. Domestic Timber Price of Rubber Wood... 155 22. Paddy & Rice Price in Jambi... 156 23. Fertilizer Retail Price in Jambi... 157 24. Chemical Retail Prices in Jambi... 158 25. Domestic Price of Dry Rubber Content... 159 26. Real Interest Rate in 2007... 160 27. Wage Rate in Jambi... 161 28. Exchange Rate in IDR/US$ Year 2007... 162

1.1. Background

The uniqueness of agroforestry system has been documented in many reports. Torquebiau (1984), Mary and Michon (1987) and Michon (1993) reveal that agroforestry is like a forest like land use system invented by local people over generations living at the margin of rainforest.

Natural rubber is an important export commodity for Indonesia with approximately 1.3 million farm households relying on rubber cultivation that accounts for 75% of national production (Director General of Estates, 2002). The predominant traditional rubber agroforestry system, has two characteristics of interest: firstly the crop is owned by smallholder farmers (with 2-5 ha plots on average) and secondly it is a result of local farmers adapting rubber as a cash crop into their crop fallow system rotation from the early 20th century (Van Noordwijk et al, 1995; Penot and Sunario, 1997; Joshi et al, 2002). In this system, a range of products additional to rubber can be harvested for self consumption or sale. The system provides regular income for farmers, mostly from the rubber, plus the temporary benefits of food and cash crops in the initial years and fruit, timber and other products in later years.

The yields of rubber-based agroforestry system can be classified into three product groups namely: (1) latex (2) rubber wood and (3) yields of the intercrops.

Smallholder natural rubber area in Indonesia covers 3 million ha (hectares) out of which 2 million ha are smallholder rubber agroforests (Director General of Estates,

2003). The total area growth of rubber in Indonesia was 1.27 from 1970 to 2003 (Direktorat Jenderal Bina Produksi Perkebunan (2004).

Table 1. Rubber Area and Growth in Indonesia from 1970 to 2003

Description

1970 1980 1990 2003

Growth/ Year

(000Ha) (%) (000Ha) (%) (000Ha) (%) (000Ha) (%) (%)

Smallholder

1,813 78 1 947 82 2 639 84 2 797 85 1.64

Government Owned

Estates 224 10 190 8 267 9 221 7 -0.04

Private owned

Estates 281 12 246 10 234 7 272 8 -0.10

Total 2 318 100 2 383 100 3 140 100 3 290 100 1.27

Source. Ditjen Bina Produksi Perkebunan (2004)

Smallholder’s rubber area had the highest growth of 1.64%, while the government owned and private owned estates area decreased to -0.04 and -0.10% growthrespectively(Table 1). The highest production growth on smallholder’s rubber was 4.33% annually while the government owned and private estates were 1.77% and 3.88% growth annually as illustrated in Table 2.

From the natural conservation point of view, Muara Kuamang agroforestry affords environmental benefit. The forest like structure of agroforestry allows the conservation of large part of natural forest biodiversity (de Foresta and Michon, 1994). The mature agroforestry is made up of an intimate mixture of various tree crops managed by the smallholders. The trees shade out the crops, occupy different

strata and occupy high value products such as fruits and high grade timber. As far as wild animals are concerned, agroforests harbor too many wild species and most of them are protected by the Indonesian law. These animals include; monkeys, gibbons, siamang etc.

Table 2. Natural Rubber Production and Growth in Indonesia from 1970 to 2003

Description

1970 1980 1990 2003

Growth/ Year Prod (Tons) (%) Prod (Tons) (%) Prod (Tons) (%) Prod

(Tons) (%) (%)

Smallholder

571 73 715 78 913 72 1 387 77 4.33

Government Owned

Estates 118 15 186 20 217 17 187 10 1.77

Private owned

Estates 96 12 12 1 145 11 219 12 3.88

Total 785 100 913 100 1 275 100 1 793 100 3.89

Source. Ditjen Bina Produksi Perkebunan (2004).

From the economic perspective, agroforestry land use system provides a wide range source of income to farmers, their neighborhood and actors along the trading chain (Levang, 1989; Dupain 1994; Bouamrane, 1996). Trees with about 65% of the tree community provide regular cash income from the harvesting and total sales. Fruit trees comprise almost a quarter of the tree community although not on a monthly basis, also provide additional cash income.

The interesting part of the story lies in the way farmers initiate and develop this land use. In the first year, after slash and burn, subsistence food crops (primarily

dry land paddy) are planted alongside rubber and fruit trees such as durian, duku, rambutan and other trees which have economic value for additional household income. Wherever possible, smallholder farmers plant any kind of vegetable for their own need.

Crop mixture has economic importance as it makes the basis of succession of harvestable commercial products before positive cash flow (i.e. 9th year for agroforestry system). Food crops (dry land paddy and vegetables) are the first yields that are harvested mainly used for daily consumption before other commercial crops are harvested. Farmers have additional annual income from harvesting these fruit trees.

It is clear that rubber establishment creates sources of income for the operators as well as its neighborhood in harvesting the yields alongside before positive cash flow. It is also clear that there are conservation measures involved in the agroforestry system that provide income related incentives to farmers.

However, one of the major issues amongst smallholders under rubber agroforestry is how efficient and profitable their system is as compared to smallholder monoculture rubber production considering whether it is financially viable or not. Some of the evidences on this issue were considered, by analyzing the production structure and the arguments made for smallholders production efficiency and profitability. The expected outputs of the efficient production system are good price, improved product quality and enhancement of a better welfare of farmers and communities.

1.2. Statement of the Problem

Over years, developing countries have experienced a wide variety of agricultural policy regimes (Westlake, M. J. 1987). Promotion of one objective conflict with one or both; to a considerable extent, government policies often favor particular parties at the expense of the others in the use of agricultural resources, in addition to factor market imperfections. In this situation, policy makers need to trade-off the gains in one objective against the losses in the substantial use of land for smallholder rubber cultivation which might be not tolerated if the action could not bring about significant improvement. This makes it necessary to measure the effect of policies on the smallholder’s rubber agroforestry system while comparing it with the smallholder monoculture system.

Agricultural policies are influenced by government interventions mostly through price and trade policies. Looking at agriculture sector it cannot be treated in isolation, as it is substantially influenced by macro-economic factors. Some of the problems facing the sector could be attributed to market distortions. A distortion, in this case, is a degree of divergence between a situation with a particular intervention and a situation without the intervention.

Despite the current increase in price of rubber, previously Indonesian government was encouraging the replanting of rubber areas with oil palm in Jambi and other areas in Indonesia because of low price of rubber that existed at the world market (Peyman, 2003). Effectively it would take 10 to 15 years of consistent replanting and new planting to enable supply to catch up with demand. Hence, it is

predicted that rubber could see a decade of higher prices, which is estimated at least double of the present market price. Price fluctuations affect production of rubber under agroforestry system than monoculture system looking at the production base of this study in Jambi. Farmers are facing problems like low yields caused by lack of technology know how, lack of capital, lack of certified planting material, poor selection of clones, low productive SPH (stand per hectare), high tapping interference and lack of uniform productive stand amongst smallholders which have led to low rubber productivity and reduction in smallholders income and product profitability. However, multipurpose uses are considered to attain highest productivity and land uses could be available for successions of more productive crop. It is very essential to establish other shade loving crops or creepers viz. Durian, Pertai, Duku and cassava, etc., in rubber plantation for economic and ecological benefit. Due to the existing problems scanned under rubber agroforestry system, it was compared with rubber monoculture system for better results.

In some recent literature, knowledge on quantitative effects of policy distortions on smallholder rubber agroforestry production in Jambi is still insufficient. Given this background, it is the objective of the study to analyze the direction and extent of distortions that will empirically fill the literature gap.

1. Is the production of rubber more efficient and profitable under smallholder rubber monoculture or smallholder rubber agroforestry system?

2. Do policy distortions affect rubber production efficiency under smallholder rubber monoculture system or smallholder rubber agroforestry system?

1.3. Objectives of the Study

1. To analyze production efficiency and profitability of smallholder rubber monoculture and smallholder rubber agroforestry systems in Jambi.

2. To find out the impacts of policy distortions towards rubber production efficiency under smallholder rubber monoculture and smallholder rubber agroforestry systems in Jambi.

1.4. Scope

This study attempted to analyze the production efficiency and the policy distortions affecting smallholders under rubber agroforestry while comparing them with smallholders under rubber monoculture in Muara Bungo - Jambi.

Production efficiency had to be obscured and to reveal the system’s commodities production practices, evaluation of potential gains in economic efficiency terms, policy distortions, and private profitability compared with potential social profitability. Policy analysis matrix (PAM) framework was used to analyze the impact of government's agricultural policies and other factors that might have affected the systems performance. Since PAM is a static model, a sensitivity analysis was used to explore the effect of potential changes in various factors that influence profitability and efficiency of smallholders.

Although systems efficiency can be measured in terms of biological, environmental and ecological perspectives etc, this thesis did not look at any of them. It only concentrated on the economic perspective.

2.1. Smallholder Rubber

Rubber cultivation fits farmer’s basis resource satiation reasonably since it only requires minimal cash expenditure for relatively inexpensive items as seeds and hand tools (Barlow, 1990). Furthermore, commercial rubber production on a small scale was cleverly adapted to the indigenous institution of traditional land systems, under which chief output of land could be acquired almost without cost. In due course, farmers are able to exploit the growing demand for rubber to an extent permitted by their basic resources and market links. This attracted many other small farmers who are convinced by significance of continuous cash income from rubber in supplementing their subsistence.

Introduction of rubber to the local people opened up opportunities for utilizing land more economically instead of leaving the land fallow after one or two years of food crop farming small farmers intercropped with rubber. After 8 to 10 years, with negligible labor input for up keeping young trees, the land became valuable rubber garden and generated continuous income as supplement to their subsistence. Profits secured from rubber during the price boom encouraged many farmers to shift to rubber cultivation as source of livelihood. But to varying degrees, they continued to produce subsistence crops, which later proved valuable when rubber markets were so depressed. According to Knorr (1945), in the 1920’s Indonesia smallholders were less dependent on rubber than their Malaysia counterparts. Their diversified farming made them less vulnerable to the dramatic price decline during the 1930’s of great

depression. Judging from the continuously increasing planted area of smallholder rubber, it may be inferred that the relative importance of rubber in smallholders must be increased too.

Rubber is one of the Indonesia’s major resources: a major source of household income for over eight million people, and among the country’s list, it’s the largest agricultural generator of foreign exchange (Barlow, 1990). This makes Indonesia to be among the world’s largest rubber producers. Rubber that is produced on large plantations or estates employs modern agricultural technologies, heavy capital investments, and a wage labor force. But the bulk of Indonesia’s rubber, 75% at the most count in 1990 by Colin Barlow, produced in tiny gardens of hectare by smallholder agroforesters Barlow and Muharminto (1982) who produce rubber with household labor to meet certain part typically not the major part of their household’s income requirements. Rubber is ideally suited to this purpose, in the words of Barlow (1990), it offers “flexible management… limited reliance on skill… easy disposal of output … and a good cash potential”.

Colin Barlow in 1996 classified the period of rubber growth in three phases; (1) Subsistent economic phase indicated by traditional farming i.e. selling and buying process, shifting cultivation and little involvement of the government; (2) early transformation of farming indicated by the dominancy of commercialized agriculture in rubber marketing including international marketing, the beginning of the implementation of rubber cultivation technology, and (3) the development of agriculture in which the agricultural sector was dominant but both industrial and service sectors developed. Villages market improved, labor and land value also

increased, transportation costs decreased, innovation of planting materials improved and government roles started to increase as well. In the current state, Indonesian rubber sector is in the third phase.

2.1.1. Rubber Agroforestry System

Agroforestry is a new name for a very old farming system and to get a picture of its concept, ICRAF (World Agroforestry Centre) defines agroforestry as a collective name for land use systems and practices in which woody perennials are deliberately integrated with crops and or animals on the same land management unit. The integration can be either in a spatial mixture or in a temporal sequence with both ecological and economic interactions between woody and non-woody components.

In 1994, the World Agroforestry Centre (ICRAF), in collaboration with CIRAD-France and the Indonesian Rubber Research Institute, established a network of improved Rubber Agroforestry Systems (RAS) on-farm demonstration plots in Jambi, West Sumatra and West Kalimantan. These RAS (Box 1) are less intensive than intensive monoculture systems but more appropriate for smallholder farmers.

In Jambi agroforestry is very important for the farmers and ranges from 20 to 60% of the cultivated land (Kabai and Dässman, 1997). Looking at rubber agroforestry, it is particularly interesting for two reasons. Firstly, it is relatively profitable and secondly, it is entirely smallholder-oriented. One of the best ideas with rubber agroforestry is that the smallholders are able to generate income from different crops or species and during the whole year. The Ministry of Forestry is encouraging

farmers to plant trees on production forest land and this is to create important buffer zones around protected areas.

In many cases, where communities have an economic incentive to implement agroforestry systems, many of which are environmentally benign, these systems frequently appear to be financially more attractive (as well as environmentally sound, compared with alternative uses of forests). For example, research by IPB in 2002

Box 1: Rubber Agroforestry Systems (RAS) best for smallholder farmers The first system, RAS-1, is similar to the traditional jungle rubber system, but recommended clones are used instead of unselected rubber seedlings. The clones must be able to compete with the natural secondary forest growth. Various planting densities (550 and 750 trees/ha) and weeding protocols were tested to ascertain the minimum management necessary for optimum production. Intensive weeding is limited to the two-meter strip of rubber rows; the space between rubber rows is less intensively weeded. This is important for smallholder farmers who need to maintain or increase labour productivity. The system is very much in line with the fallow enrichment concept and suits a large number of smallholders because of its simplicity.

The second system, RAS-2, is a more complex agroforestry system. Rubber trees are planted at normal density (550 stems/ha) along with perennial timber and fruit trees (92 to 270 trees/ha) after slashing and burning. Annual crops, mainly upland rice, are intercropped for the first two or three years, under various rates of fertilization. Planting densities of selected species were tested according to pre-established tree typology. Tree species such as rambutan, durian and petai were included. Natural regeneration is allowed in between rubber rows and farmers decide which naturally regenerating species to maintain.

In RAS-3, annual crops, mainly rice, are grown in the first year only, with legume crops such as Mucuna, Pueraria and Flemingia planted immediately after the rice harvest. Fast growing multipurpose trees (such as Paraserianthes falcataria, Acacia mangium and Gmelina arborea) can also be used. These trees can shade Imperata in the early years of rubber establishment while after seven to eight years; these can be harvested and sold to the pulp industry to provide farmers with extra income.

West Lampung shows a substantial financial advantage of the complex agroforestry systems practiced by customary communities over rubber or oil palm monocultures. 2.1.2. Rubber Monoculture System

This is comprises of a single crop i.e. rubber rather than multiple of crop species. In most cases monoculture is a response to the intensification of farming and a drive to increase yields.

Many projects have been implemented in Indonesia in recent decades to improve rubber productivity by introducing more intensive monoculture systems. These include the Nucleus Estate and Smallholder project; the Rehabilitation and Replanting for Export Commodities project; the Smallholder Rubber Development Project (SRDP), the Tree Crops Smallholder Development project and the Tree Crops Smallholder Sector project. Outside government project areas, most smallholders cannot use recommended technologies because they are limited because they are both financially and resourcefully incapacitated.

Between 1980’s and 1990’s the Indonesian government policy for the development of clonal rubber plantations for smallholders at the national scale resulted in the introduction of “clones”1, mainly in the form of rubber monoculture. This particular cropping system, which is quite different from the traditional one, was based on a technological package involving a high level of inputs (fertilizers, pesticides and herbicides, cover crops etc.). The project started with 15 % of smallholder rubber producers in Indonesia (350 000 ha) with clonal planting material which allows threefold multiplication of rubber yields (i.e. 1500 kg/ha/year with 1

clones). The disadvantage of monoculture lies in the fact that it is not easily reproducible by smallholders due to lack of capital, credit, availability of planting material and technical information.

Because rubber does not yield a return on investments for 6 to 7 years, which is a serious disadvantage, monoculture requires a relatively high starting capital to purchase the necessary inputs for the first three years in order to get quick returns. The cost of clonal planting material is high as well as the quality and availability leaving a lot to be desired and emerging network of private nurseries is not yet sufficiently advanced.

2.2. Policies Affecting Agriculture

Policies influencing the agricultural sector in Indonesia fall into of two categories; i.e. agricultural price policies and macro-economic policies. Agricultural price policies are commodity specific. Each price policy targets only one commodity (e.g., rubber) at a time. Price policies also can influence agricultural inputs. Macro-economic policies are nation-wide in coverage. Macro policies thus affect all commodities simultaneously.

2.2.1. Agricultural Price Policy Instruments

All agricultural price policy instruments create transfers either to or from the producers or consumers of the affected commodity and the government budget. Some price policies affect only two of these three groups, whereas other instruments affect all three groups. In all instances, at least one group loses and at least one other group benefits. Policy analysts need to consider three categories of agricultural policy

instruments e.g. taxes and subsidies, international trade restrictions, and direct controls.

Taxes and subsidies on agricultural commodities result in transfers between the public budget and producers and consumers. Taxes transfer resources to the government, whereas subsidies transfer resources away from the government. For example, a direct production subsidy transfers resources from the government budget to agricultural producers.

International trade restrictions are taxes or quotas that limit either imports or exports. By restricting trade, these price policy instruments change domestic price levels. Import restrictions raise domestic prices above comparable world prices, whereas export restrictions lower domestic prices beneath comparable world prices.

Direct controls are government regulations of prices, marketing margins or cropping choices. Typically, direct controls must be accompanied by trade restrictions or taxes/subsidies to be effective. Otherwise, “black markets” of illegal trade render the direct controls ineffective. Occasionally, some governments have sufficient police power to enforce direct controls in the absence of accompanying trade regulations. Direct controls of cropping choices can be enforced, for example, if the government allocates irrigation water or purchased inputs.

2.2.2. Macro-economic Policies Affecting Agriculture.

Agricultural producers and consumers are heavily influenced by macro-economic policies even though they often have little influence over the setting of these nation-wide policies. Three categories of macroeconomic policies monetary and

fiscal policies, foreign exchange rate policies, and factor price, natural resource, and land use policies affect agriculture Walter P. Falcon, and Scott R. Pearson (1995).

Monetary and fiscal policies are the core of macro-economic policy because together they influence the level of economic activity and the rate of price inflation in the national economy, as measured by increases in indexes of consumer or producer prices. Monetary policies refer to controls over the rate of increase in the country’s supply of money and hence the aggregate demand in the economy. If the supply of money is increased faster than the growth of aggregate goods and services, inflationary pressure ensues. Fiscal policies refer to the balance between the government taxing policies that raise government revenue and the public expenditure policies that use that revenue. When government spending exceeds revenue, the government runs a fiscal deficit. That result creates inflation if the government covers the deficit by expanding the money supply.

Foreign exchange rate policies directly affect agricultural prices and costs. The foreign exchange rate is the conversion ratio at which domestic currency exchanges for foreign currency. Most agricultural commodities are traded internationally, and most countries either import or export a portion of their agricultural demand or supply. For internationally tradable commodities, the world price sets the domestic price in the absence of trade restrictions. The exchange rate thus directly influences the price of an agricultural commodity because the domestic price (in local currency) of a tradable commodity is equal to the world price (in foreign currency) times the exchange rate (the ratio of domestic to foreign currency).

Factor price policies directly affect agricultural costs of production. The primary factors of production are land, labor, and capital. Land and labor costs typically make up a substantial portion of the costs of producing most agricultural commodities in developing countries. Governments often enact macro policies that affect land rental rates, wage rates, or interest rates throughout the economy. Other factor price policies, such as minimum wage floors or interest rate ceilings, influence some sectors more than others. Some governments introduce special policies to attempt to control land uses or to govern the exploitation of natural resources, such as minerals or water. These macro policies can also influence the costs of agricultural production.

2.3. Smallholder Export Crop Production

The rural economy of many developing countries is based on export crop production. Export crops are grown for international markets. Most of them share some common characteristics.

Firstly, they are high value commodities handled through reasonably concentrated marketing systems. Secondly, their production relies on the use of some purchased inputs. Finally, they are also internationally tradable, such that their domestic price is closely linked to a world market price mediated through the domestic exchange rate (Dorwardet. al., 1998).

The World Bank (2005) recognized a number of contributions that export crops could make to agricultural development at the household level. Export crops are the nucleus around which extension services, input supply and marketing are built. Food crops often benefit from residual fertilizers in the soil when they follow export crops

in rotation. Export crops also allow the purchase of productivity enhancing equipment and accumulation of capital for other investments.

It should be noted that critical issues facing export crop production under market liberalization are somewhat different from those facing food crops. Many staple food crops in Indonesia are essentially non-tradable internationally, as low value-to-weight ratios preclude profitable trade and/or because they are not widely consumed outside the country. Even where international trade might exist (as for example with rice and maize), intra-country transport costs might inflate prices within producing regions, thus hindering goods movements into the world market. Some of the core requirements of the reform policies, including real exchange rate depreciation coupled with the removal of subsidies on purchased inputs, have dramatically reduced profitability of input use of most major food crops. Indeed, such cases of reduced profitability have been reported in Jambi (Hawassi et al 2003). Changes in relative input-output price ratios have been less adverse for export crops than for food crops. In general therefore, the use of purchased inputs remains profitable for export crops as compared to food crops.

Crumb rubber factories were introduced by the government; most of them were private and depended only on the raw material from smallholder rubber growers. State owned crumb rubber factories were integrated with the private owned estates but the number was limited (Gapkindo 2004).

In 1968, crumb rubber processing technology was introduced and commercially applied in Indonesia. Success and popularity of this crumb rubber industry was shown by the fast growth of crumb rubber factories from 65 in 1969 to

135 in 1974 (Gapkind again significantly in 1

Source: Gapkindo, 200 Figure 1. Number of A

2001

In 1984 the num capacity of raw rubber get raw rubber and si their operation.

The sustainability raw rubber material fr capacity and the raw geographical positions N u m b e r o f fa c to r ie s

pkindo, 2002). This number decreased to 87 in 1 tly in 1994 to 115 factories.

2002.

er of Actively Producing Crumb Rubber Factor

the number of crumb rubber factories decreased du rubber raw material. There was tight competition a and since the factories could not compete, they e

ability of crumb rubber industry depends strongly rial from smallholder rubber farmers. The balanc

raw rubber material production should be ons cause macro inefficiency especially in transp

0 20 40 60 80 100 120 140

69 71 73 72 74 75 76 82 83 84 85 86 90 94 95 97 99

Year

7 in 1984 but increased

actories from 1969 to

sed due to the decreased tition among factories to they ended up stopping

trongly on the supply of balance between factory ld be maintained. Poor

transportation.

Table 3. Capacity and Production of Crumb Rubber Factories Based on Provinces from 2002 to 2004

NO Province

Capacity1 Production2 Deficit

Ton / year

1. Nangroe Aceh

Darussalam 0 47 097 47 097

2. North Sumatera 422 600 197 087 225 513

3. Riau 127 600 197 424 69 824

4. West Sumatera 163 000 64 288 98 712

5. Jambi 162 200 183 610 21 410

6. Bengkulu 44 000 33 563 10 437

7. South Sumatera 629 400 291 666 337 734

8. Lampung 24 000 25 097 1 097

9. West Kalimantan 160 000 173 165 13 165

10. Central Kalimantan 55 800 137 921 82 121

11. South Kalimantan 128 000 54 181 73 819

12. East Kalimantan 0 14 617 14 617

13. Others 16 800 21 751 4 951

T o t a l 1 933 400 1 441 467 491 933

Source: 1Gapkindo, 2004 2Ditjenbun 2004

Totally there was a difference of 491 933 tons (Table 3). This number reflects a deficit in raw rubber material. The largest number was in the largest raw rubber producing provinces i.e. North and South Sumatera. This condition is caused by; a relatively tight competition in the raw rubber material marketing system. Another influence caused by the shortage of raw rubber material is the irrational traffic of raw rubber viewed from the transportation side.

According to Table 3, the largest rubber producing provinces are North Sumatera and South Sumatera. The reasons could be due to relatively tight competition in the raw rubber material and the irrational traffic of raw rubber viewed from the transportation side. Indonesian natural rubber export destination countries (Table 6) are United States, Western European Countries, Japan, China, Singapore and South Korea. The export to the U.S decreased to -3.5% from 1996 to 2001 due to the slowdown in economics and automobile industries in the U.S. Export to the Asian countries increased by 23.3% annually, China increased by 33.4% annually due to the fast growth of economics at an average rate of 10% annually. Export to western European countries also increased to 8.5% annually (Table 6).

According to the Indonesian Rubber Association (Gabungan Perusahaan Karet Indonesia, Gapkindo in 2002 there were 91 crumb rubber Standard Indonesian Rubber (SIR) factories, with 21 560 workers and 89 Ribbed Smoked Sheet (RSS) factories all around Indonesia, in Java, Sumatera, and Kalimantan Islands and concentrated latex or crepe factories. Most of the crumb rubber factories belonged to private companies, smoked sheet factories to government-owned estates. In 2003 total capacity of the factories that belonged to Gapkindo members were 2.2 million tons, 20% more than the raw rubber material produced, which meant that factories should be more aggressive in buying raw rubber material. The Gapkindo members were government-owned estates, private companies, processors, exporters, traders, and buyers.

Table 4. Export Volume of Indonesian Natural Rubber Based Product from 1969 to 2002

Product Type

1969 1980 1990 2002

Volume

(000 tons) (%)

Volume

(000 ton) (%)

Volume

(000 ton) (%)

Volume

(000 ton) (%)

RSS 388 59 192 20 124 12 44 3

SIR 4 1 658 67 915 85 1 435 96

Crepe 79 12 81 8 4 0 0 0

Latex 34 5 44 4 32 3 86 1

Others 153 23 1 0 2 0 78 1

Total 657 100 976 100 1 077 100 1 496 100

Source: International Rubber Study Group (2002).

In 1969 Indonesian Natural Rubber (NR) export was dominated by Ribbed Smoked Sheet (RSS), but ten years later by technically specified rubber (Standard Indonesian Rubber SIR) and in 2002 SIR dominated, 96% of the total export. The growth of Indonesian NR export based on the type of product during 1969-2002 is shown in Table 4. Based on its value, the Indonesian NR export fluctuated (Table 5), export price was very much influenced by the international rubber price, supply and demand, and competitiveness of Indonesian NR export compared to export from other exporting countries such as Malaysia.

Since the economic crisis which was triggered by the monetary crisis in middle of 1997 in several South East Asian countries, the economic growth was negative and the inflation rate also increased. This condition influenced the world rubber demand

as well as the supply. This was shown by the negative growth and a drastic decrease in exchange rate of the main natural rubber producing countries.

Table 5. Volume, Value and Average Price of Indonesian Exported Natural Rubber from 1969 to 2002

Year Volume

(000 US $)

Export value (000 US$)

Average price (US $/kg)

1969

657 171 750 0.26

1980

976 1 165 321 1.19

1990

1 077 846 876 0.79

2000

1 380 888 623 0.64

2002

1 496 1 037 562 0.69

Source. Ditjen Bina Produksi Perkebunan, 2004.

According to Burger and Smit (2001), between 1997 and 2000 the world rubber market experienced a turbulence development, as well as serious shock from 11th September 2001 tragedy (The World Trade Center attack). In this respect, the price for natural rubber went down to its lowest point.

Table 6. Indonesian Rubber Export Based on Destination from 1996 – 2001

Destination

1996 2001

Growth / year Volume

(000 ton) (%)

Volume

(000 ton) (%) (%)

Asia & Africa 433 30 488 34 2.5

Japan 106 7 151 10 8.6

Korea 48 3 60 4 4.9

China 51 4 137 9 33.4

Singapore 130 9 78 5 -8.0

Others 97 7 62 4 -7.3

America 783 55 690 47 -2.4

United States of America 628 44 517 36 -3.5

Others

155 11 173 12 2.3

Europe 198 14 258 18 6.1

West Europe 151 11 215 15 8.5

East Europe 47 3 425 3 -1.8

Australia & New Zealand 199 1 161 1 -3.8

Total 1 434 100 1 453 100 0.3

2.4. Review on Measuring of Efficiency

It thus appears that the amount of output from a currently known way of production is determined not only by levels of inputs but also the existing organizational arrangements i.e. institutions. An institutional factor is entered into the production calculus in terms of its influence on the efficiency resource use. The conventional perception is that institutions may have great influence on the ability and willingness of a producer to correctly allocate resources at hand subject to his objective e.g. output maximization, profit or revenue maximization, cost minimization etc.

In deriving testable hypotheses of producer behavior, efficient production is usually taken for granted. The standard assumption is that markets and information are perfect, allowing equilibrium prices to clear any excess demand for and supply inputs and output. This leads to a production situation with zero profit, which ensures that inefficient producers may exist in the short run only and will eventually be driven out of the industry during the transition towards long run equilibrium. Such equilibrium is simply a conceptualization of a pareto-optimal state of production and consumption with a somewhat dynamic property. In reality, an economy or industry may experience a continuous transition from one state of equilibrium to another, during which some inefficient producers may survive regardless of competitive pressures. More importantly, inefficiency often perpetuates over a long period, mostly as a result of imperfect information concerning the methods of input application as well as prices. In addition to this the ability to discern available information may

differ significantly from one producer to another, leading to a production situation characterized by markedly different efficiency.

In its early application, the concept of efficiency deals with output variation by considering one input at the time. In this way, input use efficiency is shown in terms of productivity of an input under consideration. Later development of the concept makes it possible to measure the efficiency if input application for more than one by some index of input use. However there is a problem in providing a satisfactory index of input use, as different inputs may only be measured in different units Farrel, (1957); Griliches, (1987). Some economists have tried to get round this problem by suggesting cost comparisons as a measure of production efficiency but such a measure can only be meaningful if all producers face the same input and output prices.

Farrell (1957) suggested one method for measuring efficiency of input use through estimation of ‘production frontiers’, from which a deviation of observed output, at a given level of inputs technology is measured to give the degree of system efficiency of that particular production. When factor prices are taken into account, measurements of efficiency at observed level of output give ‘price efficiency’ or, as it is often called, allocative efficiency. The combination of system and price inefficiencies then gives ‘overall efficiency’, which is also called economic efficiency.

Farrell’s approach is considered a sensitive way to simultaneously measure the efficiency of all relevant inputs applied in a production process. But more satisfactory applications of this approach had been delayed for some time due to

absence of computational techniques for estimating a production frontier. Understandably, the applications of Farrell’s approach are simply confined to efficiency measurements, which are obtained by comparing output of individual producers to an average output of the industry. Lau and Yotopoulos (1971), for example used the approach in comparing efficiency of a production unit relative to another. By fitting an average production function for each production separately, they then compared the constant terms to give relative efficiency measurements. Although their method (i.e. unit-output-price profit function) can get round the problem of model specification for physical input – output relationships, it does not measure the efficiency of a particular unit (e.g. farm or firm) from the maximum possible output under actual circumstances of production. In addition, this method becomes impractical when comparing the efficiency of more than two production units. Further, this methods fails to specifically distinguish the source of differences in the performance of production units considered, in a sense, it does not show the relative importance of two components of ‘efficiency parameter’ (i.e. efficiency related error terms and white noise). In other words, a difference in the efficiency of two production units may be statistically significant, but it cannot be ascertained whether the difference is due to efficiency –related error terms (i.e. methods of input application) or random errors. Unless these two error terms can be decomposed, there is a possibility for making an incorrect conclusion about efficiency differentials. For example, production units under investigation may be equally (in) efficient, but considered as being different because their efficiency parameters are statistically different due to white noise.

With the advent of more sophisticated econometric techniques, production possibility frontiers with decomposed error terms can be estimated with great ease. A parametric representation of production with decomposed error terms was first and independently suggested by Aigner, Lovell and Schmidt (1977), Meeusen and Broeck (1977), Battesse and Cora, (1977). As noted by Bauer (1990), Econometric techniques for the estimation of production function basically remain unchanged. The only advancement is found in the specification of error terms, which consists of two parts. There is statistical noise with a normal distribution and inefficiency with a one-sided distribution.

2.4.1. Smallholders Production Efficiency

A fair amount of attention has been directed at assessing the relative efficiency of smallholders. However as Carr (2003) points out, many smallholders are not trying to maximize production but to take a risk-free strategy given the policy constraints. Higher smallholder productivity would be possible if these constraints were removed. He argues that unlike food crops, which are grown everywhere, export crops are only grown in areas well suited for their growth. There are few ecological constraints and many well developed technologies available for the smallholders. However, unlike food production export production is very sensitive to government policies on agriculture, exchange rates and retailing. Technical matters require further research but to a great extent production increases will be achieved more by policies which remove the constraints on smallholders using technology.

The accumulated evidence suggests that smallholders are not as efficient if judged only by yield per hectare. However economic efficiency is not only a matter

of returns to land. Smallholders can adopt a low-input and low-output strategy and continue making a profit at prices that would not be economically viable for estates. ‘Such flexibility offers the possibility of efficient resource allocation in response to the diversification of economic opportunities in the developing economies, as well as being a form of insurance against the uncertainties inherent in world markets (Tiffen and Mortimore, 1990). Furthermore, the political case for smallholder rubber agroforestry farming rests not only on efficiency, but also on equity considerations in the distribution of land and in the regional knock-on effects that smallholdings would generate.

2.4.2. Adopted Model on Measuring Efficiency

The Policy Analysis Matrix (PAM), developed by Monke and Person (1989), provides an organizational framework, which identifies patterns of incentives for economic sectors at each level of the commodity chain, and analyses the impact of direct policy on these patterns at each level as well as their effects on different production technologies, Staal and Shapiro, (1994). The impact of specific commodity and macro-economic policies is gauged by comparing results in the presence and absence of the policy Scarborough and Kydd, (1992). According to Shapiro and Staal (1995), policy in the context of marketing of agricultural commodities is defined as those government decisions (market interventions) which alter the prices economic agents (such as farmers, traders, processors, wholesalers, retailers and consumers) face and which affect their incomes and welfare.

The PAM is a matrix of information about agricultural and natural resource policies and market imperfections that is created by comparing multi-year land use

system budget calculated at private and social prices (Monke and Pearson, 1995). Private prices are the prices that farm households are facing (local or domestic market price of input and output). Therefore, profitability or NPV valued at private prices, so called private profitability, is an indicator for production incentive Tomich et al, (1998). Social prices are the economic prices that remove the impact of policy distortion (taxes, subsidy and other local levies) and market imperfections. Usually it is derived from export or import parity prices of particular inputs or outputs. Profitability measured at social prices, so called social profitability, is an indicator of potential profitability.

In brief, the Domestic Resource Cost (DRC) approach uses social profitability to measure efficiency; the greater the profitability, the stronger the efficiency (Monke and Pearson, 1989). Specifically, systems efficiency can be measured by a DRC ratio: DRC = Value of domestic factors at economic prices per unit of output

Value added at economic prices per unit of output

In addition to computing profitability, the PAM is used to compute protection coefficients such as DRC, EPC, ERP, NPC and NRP of which DRC is the most important. For this study DRC is computed to determine the production efficiency of the rubber smallholders under agroforestry and monoculture systems respectively.

DRC is a summary measure of the efficiency of domestic production and is interpreted as the costs required for earning a unit of foreign exchange. DRC is also used in comparing relative efficiency among agricultural commodities and is defined as the shadow value of non-tradable factor input used in an activity per unit of

tradable value added. In other words DRC indicates whether the use of domestic factors is socially profitable or not.

Commodities are ranked according to DRC value, and this ranking can then be taken as an indication of competitive advantage or disadvantage.

It should be noted that social profitability needs to be gauged under “shadow” instead of market prices. As opposed to observable market prices, shadow prices are “social” prices reflecting the value of social benefits or costs. For example, agroforestry high profitability in rubber may not reflect efficient resource utilization, but could result from direct or indirect government interventions artificially lowering the production costs or raising the output prices. Therefore, using distorted market prices to measure profitability tends to result in a “false” indication of efficiency; and shadow prices, which measure the true or social value of production costs and revenues, should be used in calculating DRC ratios for efficiency assessment. Empirical DRC analyses are often conducted based on the “Policy Analysis Matrix” (PAM), which is a standard apparatus for policy decision-makings (Monke and Pearson, 1989) and used widely in efficiency analysis of related to agriculture commodities.

2.5. Measuring Distortions

2.5.1. Distortions in Agricultural Commodity Markets

More efforts have been put in place recently in measuring distortions in the agricultural sector. Several studies have sought to evaluate the costs of such trade distortions. In the late 1970’s and early 1980’s several less developed countries raised

producer prices for cereals relative to other competing opportunities, thus increasing incentives for food production. A good example is shown in the study in Kenya by Jabara (1985). The study revealed that producer prices (output prices deflated by input prices) for food and other crops increased substantially from 1979 to the early 1980s and these increases in prices were associated with the increase in marketed agricultural production. In another major study on pricing policy in developing countries, the United Nations Food and Agriculture Organization (FAO) (1985) noted that, while international cereal prices fell to 18 per cent between years 1978 and 1982, developing countries domestic producer prices rose on average by 5 per cent over the same period.

The study by Byerlee and Sain (1986) found no consistent evidence of price disincentives for agricultural commodity producers in less developed countries. Although the countries kept prices low to consumers, according to the Byerlee and Sain study, the countries did so using policies that did not directly tax producers. It is clear that government taxation of the exportable commodities, which is embraced in the controlled prices set by the government, is a policy considered unfavorable. Westlake (1987) draws attention to the policy distortions that existed in developing countries major agricultural subsectors and tried to visualize the empirical method to estimate distortions on national levels.

2.5.2. Adopted Model

Standard ratios reflecting the degree of price divergences or distortions are normally calculated to compare profitability and efficiency of different crops. These ratios facilitate comparisons among activities, particularly when the production

process and outputs are dissimilar. A number of protection coefficients are calculated in a standard PAM. The most commonly used protection coefficients are Nominal Protection Coefficients (NPC) and Effective Protection Coefficient (EPC).

Many studies have quantified the distortions of input prices to be caused by policies (Repetto, 1985; Tjornhom, 1995; Lee and Espinosa, 1997), and there are a few techniques to evaluate the degree of protection or subsidy. Most of the techniques that have used methods in which the distortion is observed by the comparison of world price, or border price, and the prevailing market price, which allows the use of the conceptual framework constructed.

2.5.3. Subsidy on Inputs

The method used by Repetto (1995), Lee and Espinosa (1997) compared the subsidy on pesticides and fertilizers against that on other inputs. Conceptually, their subsidized prices were the same as the private prices in most other analyses. However, a difference was that the subsidized prices were estimated by adding up costs of importation, transportation, and sale with current subsidies.

The unsubsidized prices set as benchmark prices are different from the social costs in several studies. These are calculated as if fertilizers and pesticides are treated as any other production input. “(t) his approach compares the specific sources of fertilizer and pesticide subsidies (currency overvaluation, favorable tariff treatment, tax exemptions, etc.) against a counterfactual which measures the “standard” social costs of inputs and goods (particularly imported goods) in an economy.” Lee, Espinosa (2000).

The series of studies summarized by Krueger, Schiff, and Valdes (2001), and applied to a pesticide case by Tjornhom (2005), use the nominal and effective rates of protection for their analysis. The nominal protection coefficient (NPC) and the nominal protection rate (NPR) are measures of the protection of domestic producers or consumers from imports or exports. NPC is calculated as the ratio between the domestic price and the border price, and NPR is the percent difference between the domestic and border prices. A positive NPR represents a protection on the producer and dis-protection on the consumer, while a negative one means dis-protection of the producer and protection of the consumer.

Strength of these measures is that they may be applied to measure price distortion at any level in the production-consumption chain Tsakok, (2000). However, the drawback of that strength is that these indexes only consider output or input prices, so their scope tends to be narrow when compared to the effective coefficient and rate of protection.

The effective protection coefficient (EPC) and the effective protection rates (EPR) are different from their corresponding nominal measures, because they indicate the degree of protection on production structures. The EPC, for example, calculates the value added in domestic prices over the value added in border prices for a commodity. The value added is defined as the difference between the output value and the tradable input value. Because of that definition, the value added can also be interpreted as the return on non-tradable domestic factors of production such as land, labor and capital (Tsakok, 2000). The EPR is the percent difference between the value added in domestic prices and border prices.

EPC greater than 1 means that, producers receive a greater return with the policy intervention. Any positive EPC implies incentive to producers, but Tsakok, (2000) explains that the incentive is only potential and not actual. This happens because the difference in price levels does not result in concrete resource reallocation. A negative EPC indicates that there is a flaw in the decision to undergo the production process under existing productivity and cost conditions.

2.5.4. Market Failures

Price formation through market mechanisms in decentralized economies frequently fail to arrive at equilibrium levels. The answer lies in non-competitive behavior, asymmetric information, severely segmented input and output markets, and the presence of externalities are among the causes of market failure, signaling wrong input and output prices. This is a situation that is commonly found in developing economies Myint (2005).

Distortions express themselves to markets through a complex web of economic activity interactions. One possible net outcome takes the form of wrong input prices. For example as a result of market distortions the capital price may appear low relative to wages despite abundant labor. This may lead to development of capital using technology. The implications of such technical change are particularly important to smallholder producers whose cash situation does not permit them to adopt new technology. Their problem of technology adoption is also exacerbated by an imbalance access to formal capital markets. As a result, these involuntary non-adopters are excluded from the partition of income generated.

2.5.5. Government Interventions

Institutional changes induced by economic forces may not always progress along socially optimal path of economic development. The presence of social-private divergences in resource use and income distribution resulted from such changes in a legitimate justification for “deliberate exogenous efforts” undertaken by governments, which are specifically aimed at narrowing the gap by planning and stimulating changes in relevant institutions.

It is interesting to note that many economists support ‘intervention approach’. Johnston and Kilby (2003), for example maintained that Laissez-faire is certainly not the best alternative for government policies in product, factor and money markets in late developing countries, even though interventions may impede healthy developments of some sectors. This is particularly true in backward economies such as those of the 19th century e.g. Germany and Russia, where necessary prerequisites for economic growth have not developed (Gerschenkron, 2002). Therefore, a more powerful stimulus than pecuniary incentives is needed to stimulate economic growth.

In preventing distortions governments often intervene. Policy measures for such intervention may range from modest discouragement to complete banning of the development and adoption of undesirable technology. Policy measures are usually coupled with fuller support to reach activities in labor using technology development. Encouragement usually comes in form of patent and license granted to private interventions. Along with those, budget subsidies may be allocated to public education and research institutes advancing natural science and engineering. Such encouragement is basically justified for research that has substantial payback.

3. 1. The Conceptual Framework

Distortions in the context of production of agricultural commodities can be defined as those interventions which lead to price alterations that farmers face affecting their incomes and welfare and hence reductions in rubber production. These interventions alter economic incentives to the producers. The effects of divergence as a result of interventions can be measured at the market levels. Profitability of production systems (defined as the difference between revenues and costs) is a key issue affected by changes in commodity, domestic factor and input markets. Distortions in the input/output markets can result into a net taxation of the production system hence lowering of farmer’s profit. This could lead to changes in cropping patterns and resource allocation among the alternative crop enterprises (shifting towards a more desirable enterprise). Institutional market failure could also result into a situation in which markets do not function efficiently because of inadequate development or lack of appropriate regulations. Therefore, identifying institutional forms of market failure and their effect is also important in evaluating agricultural commodity markets. Changes in profitability as a result of policy intervention can be measured within the production systems.

Distortions come from an understanding of how they affect profits. Distortions create incentives or disincentives for production systems and the long-term environment for the development and sustainability of agricultural production, hence directly having the impacts on rubber production initiatives. The aim of this

study is to analyze production profitability and efficiency of rubber under smallholder monoculture and smallholder agroforestry systems, distortions and the patterns of incentives for rubber agroforestry system under the current policies.

Figure 2. Conceptual Framework for Analyzing Policy Impacts

PRICING POLICY (E.g. Removal of input subsidies

setting of produce indicative price, exchange rate, taxes etc INSTITUTIONAL

ARRANGEMENTS (e.g. Crop buying systems, market regulations

etc

DISTORTIONS IN THE INPUT/OUTPUT MARKETS (Produce, tradable and

non-tradable input prices

SMALLHOLDER PRODUCERS

PROFITABILITY (Income and Welfare)

CHANGES IN CROPPING PATTERN/REALLOCATION OF

FACTORS OF PRODUCTION AGRICULTURAL

3.2. Hypothesis

The objectives set above are sought against the background of these hypotheses which states that;

1. Production of rubber is more efficient and profitable under smallholder rubber agroforestry system than smallholder rubber monoculture system

2. Policy distortions affect production efficiency under smallholder rubber agroforestry system than smallholder monoculture system.

3.3 . Selection Criteria for Area of Study

3.3.1. Selection Procedures

The first task was to select the survey province. According to the preliminary investigation, it was found out that Jambi is among the rubber producing provinces in Indonesia. According to Table 3 based on the results of the capacity and production of crumb rubber factories based on Provinces, Jambi was among the provinces with large capacity. Therefore this was the reason for selecting Jambi province as a survey location, whilst no other crucial considerations were taken in favor of this choice over the four provinces.

The next task included selecting a district with two rubber producing villages in the province as the representative of rubber cultivation under smallholder monoculture and smallholder agroforestry systems respectively. The selection procedure involved two criteria:

1. The extent of rubber cultivation by smallholders, was traced from the district level and further down to the village level; and

2. The ‘probability’ of finding monoculture and agroforestry systems in adjacent places.

Based on criteria 1, Muara Bungo district was selected as the possible candidate. After consultation with institutions (e.g. ICRAF Jambi province office) based in Muara Bungo and local officials the decision was made based on the expectations of a better chance of finding two neighboring village’s one representing rubber monoculture system and another representing rubber agroforestry system typically under smallholdings.

The representative smallholdings were based on homogeneity in terms of input and output factors which would allow profitability and efficiency analysis. All the participating smallholdings in Senamat village satisfied the criterion (i.e. smallholdings consisted of trees with the same age composition and the same inputs).

The situation was quite different in the case of traditional smallholders under rubber agroforestry in Muara Kuamang village whose smallholdings consisted of unselected rubber trees with a wide range of age composition. It was necessary to minimize possible confounding effects of age differential and for this reason; only smallholdings with rubber trees replanted in 1990’s were selected. Everything here was predetermined based on the availability of time and budget for conducting this study.

3.3.2. Survey Locations

It was desirable to cover a wide area and include as many smallholders as possible in order to allow the generalization of the findings of the study. However due to limited time, the coverage of the present study was strictly on two villages

Appendix 24. Chemical Retail Prices in Jambi

Year CPI Jambi 2007=100 Furadan (Rp/kg) Round-up (Rp/lt)

Current Real Current Real

1994 22 15 397 71 272

1995 23 14 290 62 310

1996 24 16 473 67 764

1997 26 28 900 110 526

1998 39 35 900 91 770

1999 49 34 000 69 753

2000 52 40 000 77 332

2001 55 40 000 72 220

2002 61 40 000 65 800

2003 65 40 000 61 985

2004 70 43 000 61 094

2005 84 50 000 59 433

2006 93 52 000 55 857

2007 100 12 000 12 000 60 000 60 000

Average from 1997 to 2007 12 000 71 434

Sources:

Appendix 25. Domestic Price of Dry Rubber Content

Year

Nominal Price

CPI Jambi 2007=100 Real price (Rp/kg)

Rp/kg

1989 1 229 14 8 592

1990 1 193 16 7 496

1991 1 264 17 7 393

1992 1 260 18 6 974

1993 1 424 20 7 253

1994 2 072 22 9 590

1995 3 027 23 13 198

1996 2 682 24 11 032

1997 2 369 26 9 059

1998 4 878 39 12 469

1999 3 953 49 8 110

2000 3 400 52 6 573

2001 4 269 55 7 708

2002 5 001 61 8 227

2003 5 867 65 9 092

2004 8 538 70 12 131

2005 9 651 84 11 472

2006 13 529 93 14 532

2007 13 000 100 13 000

Annual average from 1997 to 2007 10 216

Source:

GAPKINDO, Jambi for 1986 - 1999

Market price monitoring by ICRAF Muara Bungo, 2006 2001-2006 Price: Directorate General of Estate Crops, 2006 field data

Appendix 26. Real Interest Rate in 2007

Real interest rate calculation 1997 1999 2000 2004 2005 2006 2007

Real interest rate on project loan :

((1-Nominal Interest rate)/(1-Inflation)) - 1 12.25% 14.05% 13.96% 12.02% 11.03% 10.57% 10.02%

Nominal interest rate on project loan 19.84% 20. 84% 20.75% 19.01% 18.94% 18.53% 17.59% average annual inflation rates 8.65% 7.91% 7.89% 7.95% 8.89% 8.91% 8.41%

private 12.02% 11.03% 10.57% 10.02%

Social 7.02% 6.03% 5.57% 5.02%

Period Rupiah credit interest

rate by state banks Inflations (Working capital)

1990 19.26% 9.53% 1991 23.25% 9.52% 1992 22.16% 4.94% 1993 19.37% 9.77% 1994 16.77% 9.24% 1995 16.86% 8.64% 1996 19.04% 6.47% 1997 21.98% 11.05% 1998

1999 28.89% 2.01% 2000 18.43% 9.35% 2001 19.19% 12.55% 2002 18.25% 10.03% 2003 15.07% 5.06% 2004 13.41% 6.40% 2005 16.23% 17.1% 2006 15.35% 6.6% 2007 13.47% 6.59% Average until 1997 = 19.84% 8.65%

Average until 1999 = 20.84% 7.91%

Average until 2000 = 20.75% 7.89%

Average until 2004 = 19.01% 7.95% 11.06% Average until 2005 = 18.94% 8.89%

Average until 2006 = 18.53% 8.91%

Average until 2007 = 17.59% 8.41%

Source

Rupiah credit interest rate by state banks (source BPS Monthly Statistical Bulletin Dec 2007) Bank Indonesia

Appendix 27. Wage Rate in Jambi

Cost of Labor 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 AVG

Private price (Rp/person

day) 3 500 4 000 6 000 8 000 10 000 15 000 15 000 15 000 15 000 17 500 30.000 30.000 14.083 Social price (Rp/person

day) 3 500 4 000 6.000 8.000 10.000 15.000 15.000 15.000 15.000 17.500 30.000 30.000 14.083

2007=100 24.31 26.15 39.12 48.74 51.73 55.39 60.79 64.53 70.38 84.13 93.09 100

Real Labor Wage 13.404 14.241 14.278 15.279 17.998 25.212 22.971 21.639 19.840 19.365 30.000 27.928 20.796 13.404 14.241 14.278 15.279 17.998 25.212 22.971 21.639 19.840 19.365 30.000 27.928 20.796

Appendix 28. Exchange Rate in IDR/US$ Year 2007

Year Exchange Rate (IDR/US$) 1967 150 1968 296 1969 326 1970 363 1971 392 1972 415 1973 415 1974 415 1975 415 1976 415 1977 415 1978 442 1979 623 1980 627 1981 632 1982 661 1983 909

1984 1 026

1985 1 111

1986 1 283

1987 1 644

1988 1 686

1989 1 770

1990 1 843

1991 1 950

1992 2 030

1993 2 087

1994 2 161

1995 2 249

1996 2 342

1997 2 909

1998 10 014

1999 7 855

2000 8 422

2001 10 261

2002 9 311

2003 8 577

2004 8 939

2005 9 751

2006 9 141

2007 9 164

Source: http://fx.sauder.ubc.ca/

http://fx.sauder.ubc.ca/etc/USDpages.pdf Bank Indonesia

Atuhaire Rodgers, 2008

. ECONOMIC ANALYSIS OF SMALLHOLDER

RUBBER AGROFORESTRY SYSTEM EFFICIENCY IN JAMBI (

Bonar M.

Sinaga as the Chairman, and Suseno Budidarsono as the Member of the Advisory

Committee).

The objectives of this research were; (1) to analyze production efficiency and

profitability of smallholder rubber monoculture and smallholder rubber agroforestry

systems, and (2) to find out the effects of policy distortions towards rubber

production

under

smallholder

monoculture

and

smallholder

agroforestry.

Smallholders are not only judged by yield per hectare; economic efficiency is not

only a matter of returns to land and returns to labor but smallholders can adopt low

input strategy, continue making profits at prices that would be economically viable.

Such flexibility offers the possibility of efficient resource allocation in response to

diversification of economic opportunities.

The Policy Analysis Matrix (PAM) with the Domestic Resource Cost (DRC)

results under the baseline scenario indicated that the use of domestic resources in

production of rubber was efficient and socially profitable under the two systems but

more desirable under monoculture system given the current prices for physical inputs,

outputs, technologies and policy transfer. However, even with sensitivity analysis of

10% increase in the price of rubber holding other factors constant does not make

rubber agroforestry system more efficient than its counterpart and a 20% fall in price

of rubber made rubber production under agroforestry system less efficient and

undesirable. All measures are compared to the alternative policy indicators currently

used. Therefore, recommendations made from this study relate to the need for

diversification into better practices that can sustain efficient rubber production under

the agroforestry system, encouraging private sector participation and reducing

disincentives to rubber production.

Key words: Smallholder Rubber Agroforestry, Economic Efficiency, Policy Analysis

Matrix