Tariff setting for the development of the edible oil sector in Tanzania

Final Report

Neil Balchin, Josaphat Kweka and Maximiliano Mendez-Parra

February 2018

i i

Executive Summary

Despite strong growth in sunflower seed production, the level of edible oil processing in

Tanzania is low compared to prevailing demand, which is estimated at between 300,000 and

400,000 tons a year. Much of the demand gap is currently met by imported edible oil (mostly

palm oil), with imports accounting for 60% of all edible oils in Tanzania and between 55% and

70% in the case of sunflower oil. The Government of Tanzania (GoT) is keen to reduce

Tanzania’s dependence on imported edible oil by boosting domestic oil seed production and

downstream oil processing capacity. With this objective in mind, in 2016 the GoT implemented

a 10% tariff on imports of crude palm oil (CPO).

This study collects evidence through a literature review, survey, stakeholder consultations and

analysis of available data to assess the impact of the tariff on demand and supply dynamics,

with a particular focus on Tanzania’s sunflower sector. We also examine experiences in three

case study countries (Argentina, Indonesia and the Ukraine) that have successfully expanded

domestic production of oil seeds and/or facilitated greater levels of domestic value addition

within the edible oil sector. The ultimate objective of the study is to inform future policies to

facilitate greater investment in domestic production, processing and refining in Tanzania’s

sunflower sector.

Production and processing in Tanzania’s sunflower value chain

Recent growth in sunflower seed production in Tanzania has been significant, with especially

rapid growth since 2010 even though farmers’ productivity and yields are low. There has also

been growth downstream in sunflower oil production and Tanzania’s sunflower exports have

expanded significantly since 2005, mostly driven by rapid growth in exports of sunflower cake.

However, domestic capacity to supply sunflower seeds and oil still falls well short of prevailing

demand in the domestic market.

Previous studies have highlighted critical productivity challenges in Tanzania’s sunflower

sector due to a lack of high-yield seeds, low levels of capacity and outdated machinery and

technology for processing. Our research adds to this body of knowledge by highlighting a

number of challenges and constraints currently hampering prospects for expanding domestic

production and processing in the sunflower sector. Poor seed quality, exacerbated by low

farmer productivity, limited access to markets for farmers, a lack of warehousing facilities and

limited vertical integration in the sector, are major reasons for the shortages in quality seed

stock experienced by downstream processors. These processors, the majority of which operate

at a small scale, also face high input costs and low processing capacity, the latter owing to a

range of factors including a lack of finance for capital investment, outdated technology and

unreliable power supply. The uneven application of VAT along the sunflower value chain is

also said to disadvantage small-scale processors. This is one element of a more general lack of

policy coherence across the sector.

ii ii

Consumer preferences, demand and substitutability between sunflower oil and palm oil

The consumers we surveyed generally prefer sunflower oil over palm oil, even more so at

higher education and income levels. Our consumer survey also reveals a high price elasticity

of sunlower oil – meaning demand is highly sensitive to price changes – at lower education

and income levels. But the evidence on the extent to which sunflower oil and palm oil are

substitutable is mixed. At the producer level, most small-scale sunflower oil processors we

surveyed do not consider producers of other edible oils as direct competitors, suggesting a low

degree of substitutability. In contrast, our cross-price elasticity estimates suggest a degree of

substitutability between sunflower oil and palm oil in the lowest market segment, although

these estimates are based on a very small sample and should be interpreted with caution. The

true degree of substitutability between sunflower oil and palm oil has important implications

for the impact of the CPO tariff on production and processing in the sunflower sector.

Tariff impacts

Our survey of sunflower farmers and processors reveals a very low level of awareness of the

10% tariff on imported CPO. This suggests the presence of the tariff has had little effect on

production decisions and expectations. The precise impact of the tariff on prices in the

sunflower value chain is also not clear. Information provided by actors in the value chain

indicates average farm gate and market prices for sunflower seed increased (by 7% and 23%,

respectively) before and after the tariff was imposed, but it is not possible to disentangle the

tariff impact from other factors such as inflation. There are also contrasting views across

industry associations about the impact of the tariff on farm gate prices. For their part, we found

large-scale processors believe the tariff has not been effective in raising farm gate prices for

sunflower farmers. Similarly, the few farmers we surveyed that were aware of the tariff claimed

it has not influenced these prices.

More generally, our engagements with stakeholders and actors in the sunflower value chain

revealed a range of contrasting views on the extent to which the 10% tariff on imported CPO

can effectively support the development of the domestic sunflower sector in the face of these

challenges:

• There are diverse views across industry support organisations and private sector

foundations. TASUPA and the TCCIA in Dodoma support the tariff, mainly due to its

perceived benefits for small-scale processors. However, SUFA does not believe the

tariff has helped farmers, mainly because fundamental productivity challenges

constrain them from responding to price incentives.

• Large-scale processors believe the tariff has not been effective in expanding production

by small-scale sunflower oil processors, but has resulted in reduced profit margins and

higher prices for consumers. Many consumers (40% of those surveyed) also reported

an increase in the price of palm oil products, and regional and local government

authorities are concerned about the impact of the tariff in raising prices.

• Large-scale CPO processors (including refiners of CPO) feel the tariff has undermined

their competitiveness versus Kenyan producers, and contributed to the loss of the

Democratic Republic of Congo export market. They contend the tariff has resulted in

reduced profit margins and higher prices for consumers.

iii iii

• Downstream bar soap manufacturers in Tanzania indicated the tariff has raised the price

of raw materials significantly and generated supply shortages to the detriment of the

industry.

Experiences in other countries

The experiences of different edible oil industries in Argentina, Indonesia and Ukraine suggest

focusing on a bouquet of industry support policies is more effective than a narrow focus on

tariff policy. These countries have applied (differential) export taxes on raw materials as a

mechanism to stimulate supply of raw seeds for downstream processing, but this is unlikely to

be effective in the Tanzanian context. Instead, the Argentine experiences in soybeans and

sunflower demonstrate the value of government investments and R&D to support seed research

(for improved varieties) and multiplication. The Indonesian example (palm oil) shows the value

of government support for smallholders (plantation development, institutional support,

agricultural extension services) in the context of large anchor plantations. All three cases

highlight the importance of government action to improve capacity and productivity in

downstream crushing and processing.

Conclusions

Clearly, there remains a great deal of disagreement across different stakeholders in Tanzania,

both within the sunflower value chain and more broadly across the edible oils sector, about the

merits of maintaining the 10% tariff on imported CPO. Our analysis suggests the tariff has

been largely ineffective, primarily due to the existence of key constraints limiting productivity

(e.g. poor quality seeds, low processing capacity) and limited substitutability between

sunflower oil and palm oil. Nevertheless, there remain arguments in favour of, and against, the

tariff. On one hand, retaining the tariff on imported CPO could incentivise large-scale

investments in the production of refined sunflower to compete with palm-based products; and

it may drive increased demand and prices for small-scale sunflower producers and processors.

At the same time, the tariff has resulted in higher prices for consumers of palm oil and raised

the prices of domestically produced non-food palm-based products, thereby undermining the

competitiveness of Tanzanian manufacturers of these products.

Recommendations

Our findings show there is clearly a need for further dialogue among stakeholders in the

edible oil sector to inform the longer-term direction of Tanzania’s tariff policy for the

sector and build consensus around reform priorities.

While further dialogue takes place, the tariff on imported CPO should be retained at 10%,

but with a clear commitment to ensuring it remains in place only for a limited, time-

bound period (ideally to be phased out within three years).

A wider array of policy interventions to address critical constraints, rather than a narrow

focus on tariff policy, are necessary to spur investment in sunflower production and

processing and promote the long-term development of the sunflower sector in Tanzania:

iv iv

• Consider VAT exemption on sunflower oil and seed cake. A detailed study should

be undertaken to empirically assess the merits and revenue impacts of such a policy.

• Improve the availability of high quality, high oil bearing sunflower seed varieties

domestically. In the short term, support should be provided for importing hybrid

seeds. In slower time, the provision of incentives for seed importers to invest in local

production will help boost the availability of high quality seeds in the medium term.

The longer term focus should be on supporting seed research, breeding and

multiplication (e.g. by improving the capacity of agricultural research institutes in

Tanzania and investing more in R&D into improved seed varieties).

• Invest in warehousing infrastructure to improve access to storage facilities for oil

seeds.

• Eliminate producer cess on sunflower oil and cake. This should form part of wider

discussions to eliminate nuisance taxes, levies and fees on various agricultural value

chains.

• Develop a coherent strategy for the broader edible oil sector, drawing on Tanzania’s

existing Sunflower Sector Development Strategy 2016-2020.

• Strengthen the capacity of industry support organisations to engage more effectively

in national structures for policy dialogue.

v v

Table of contents

1. Introduction ....................................................................................................................................... 1

1.1 Research focus and objectives ....................................................................................................... 2

1.2 Methodology and approach ........................................................................................................... 4

1.3 Research scope and limitations ..................................................................................................... 6

1.4 Report structure ............................................................................................................................. 6

2. Literature review: key findings from a meta-analysis of existing studies .................................... 7

2.1 Summarising the main findings in the literature ........................................................................... 7

2.2 Information gaps and implications for this study ........................................................................ 11

3. The state of production and processing in Tanzania’s edible oil sector, tariff impacts and

challenges for future expansion ......................................................................................................... 12

3.1 Recent trends in production and processing in Tanzania’s sunflower (and other edible oil) value

chain .................................................................................................................................................. 12

3.2 Consumer preferences, demand and substitution ........................................................................ 17

3.3 Tariff impacts .............................................................................................................................. 23

3.3 Challenges and constraints to expanding production and processing ......................................... 36

4. Case study evidence of edible oil sector development in other countries ................................... 48

4.1 Export taxes in Ukraine’s sunflower sector................................................................................. 48

4.2 Argentina’s edible oil sector: A major global player in soybean and sunflower ......................... 52

4.3 Indonesia’s palm oil sector: A world leader with a prominent role for smallholder farmers ...... 57

5. Conclusions ...................................................................................................................................... 62

6. Recommendations ........................................................................................................................... 64

References ............................................................................................................................................ 67

Annex A: List of individuals and organisations consulted .............................................................. 71

Annex B: Characteristics and profile of survey respondents .......................................................... 73

1 1

1. Introduction

Demand for edible oils has grown significantly in recent years, both globally and regionally

within Africa. However, despite relatively strong growth in oil seed production in Tanzania

(especially in the case of sunflower seeds), production of edible oils in the country – and indeed

across East Africa – remains low compared to prevailing demand levels. On this basis, there

appear to be good prospects, particularly in the sunflower value chain, to ramp up production,

improve capacity utilisation in downstream processing, raise productivity and enhance quality

levels to capitalise on the opportunities presented by rising consumer demand. Moreover, as a

by-product of the crushing of oilseeds, the production of animal feed can benefit Tanzania’s

local livestock farming sector and/or aid integration into international value chains, while also

contributing to primary and secondary employment.

Government policy support can play an important role in coordinating the efforts of different

players in Tanzania’s various edible oil value chains and in creating the enabling environment

necessary to encourage an expansion of domestic production and processing. At present,

capacity utilisation levels are low in the processing stages of different edible oil value chains

and, as a result, Tanzania imports a significant amount of processed oil (crude and refined) due

to the limited local supply, which has been unable to satisfy demand in the domestic market

(Nyaki, 2013). The combined demand for different types of edible oil in Tanzania is estimated

at between 300,000 and 400,000 tons annually, and it is expected to expand at an annual growth

rate of 3% due to population growth and changing health concerns (Mgeni et al., 2017; Salisali,

2017). Palm oil accounts for the bulk of edible oil demand (64% of consumption), although

sunflower oil is the dominant edible oil produced domestically (accounting for 83% of total

production, compared to just 2% in the case of palm oil) (Dalberg, 2018). Salisali (2017)

estimates the deficit in demand for sunflower oil relative to current levels of domestic

production is typically in the range of 55% to 70% in any given year. Across the edible oil

sector, about 60% of total demand in Tanzania is currently satisfied through imports of crude,

refined and semi-refined oil (the majority of which is palm oil), mostly from Malaysia and

Indonesia for palm oil and the Ukraine, Argentina and Antigua and Barbuda in the case of

different sunflower oil varieties.

The Government of Tanzania (GoT) is eager to reduce the country’s dependence on imported

edible oil by boosting domestic production and processing capacity to ensure more oil seeds

and downstream edible oil products are produced locally. To this end, in 2016 the GoT opted

to stay the application of the East African Community’s (EAC) common external tariff (CET)

on crude palm oil (CPO) imports (HS code 151110) and apply a 10% duty rate instead of 0%

on these imports for one year up until 30 June 2017. The core intention behind the introduction

of the tariff was to incentivise refiners to utilise more domestically produced crude oil. It was

anticipated that the tariff would raise demand for domestic edible oil products and increase the

prices of domestically produced oil seeds, thereby benefiting smallholders as well as other oil

seed producers.

There are differing views across various stakeholders in the sector about the merits of this

policy. The tariff was introduced amid pressure from domestic small- and medium-scale oil

seed producers and local edible oil processors that use locally produced oil seeds as raw

materials. The Tanzania Edible Oil Seeds Association (TEOSA), led by small- and medium-

scale processors, was particularly vocal in advocating for the tariff increase. Proponents of the

2 2

tariff argue the use of imported CPO drives down the market price for other edible oil varieties

(including crude sunflower oil), to the detriment, for example of both domestic sunflower oil

seed producers and local sunflower oil processors that rely on domestically produced seeds. In

turn, they believe the tariff on CPO imports will incentivise local sunflower farmers to produce

more oil seeds and encourage downstream processors to invest in the sunflower oil value chain

(Salisali, 2017). Those in favour of the tariff contend that even with a 10% duty rate, imported

CPO is still cheaper than locally processed varieties such as crude sunflower oil (ibid.).

However, large processors in Tanzania are opposed to the tariff and want it to return to zero in

line with the EAC CET. The Tanzania Soap and Edible Oil Manufacturers Association, hosted

at the offices of Mohamed Enterprises Ltd, a major producer, have lobbied against the

imposition of the tariff. The Confederation of Tanzania Industries (CTI) has also been vocal in

highlighting the potentially adverse impacts of the tariff on domestic producers. Their position

is premised on the argument that the tariff raises both the price of CPO and by-products reliant

on the crude oil for their production (e.g. margarine, cooking fats, soaps) by 10%, which in

turn impacts negatively on sales volumes and incomes while reducing capacity utilisation in

the sector and not having any discernible impact in raising demand for domestically produced

crude palm or sunflower oil. These bodies contend the tariff has rendered domestic producers

of palm oil-based products and by-products uncompetitive, especially against their

counterparts in the EAC where the CET of 0% on crude palm oil imports is applied. The

anticipated result is a loss of Tanzania’s market share relative to other African exporters of

edible oils and products relying on CPO for their manufacture.

Within the context of the ongoing debate on the efficacy of the tariff, this study was undertaken

to augment the existing evidence base and provide recommendations, as articulated in a request

by the Ministry of Finance and Planning to the Agricultural Non-State Actors Forum (ANSAF),

on the most effective policy measures to facilitate greater investment in domestic production,

processing and refining in Tanzania’s edible oil sector. An important element of this request

was to recommend ways to incentivise value chain stakeholders to close the gap between

domestic supply and demand in Tanzania’s edible oils sector, and make the sector and the

whole value chain competitive in regional and global markets. This is an important objective.

Support to the sector, whether provided via trade policy measures such as tariffs or other

support mechanism, must be designed in a way that does not harm the long-term prospects and

competitiveness of domestic oil seed production and downstream processing industries and

does not impact adversely on consumers.

1.1 Research focus and objectives

The core objective of this study is to provide evidence-based recommendations to support the

GoT to coordinate the efforts of stakeholders in the edible oils sector with a view to creating

an enabling environment to encourage small-scale producers, processors and other players

involved in the value chain to invest in expanding production, enhancing productivity and

improving the quality of domestically produced crude edible oil and processed edible oil

products in Tanzania. We focus specifically on the sunflower value chain and the role of

smallholders,1 but do not exclude issues affecting other edible oil subsectors.

1 Borrowing from the definition used in Tanzania’s Sunflower Sector Development Strategy 2016-2020, we define smallholder farmers within

the sunflower value chain as farmers operating on small plots of less than five acres and typically without the use of modernised farming techniques.

3 3

The emphasis on the sunflower sector is motivated by its importance in the broader economic

context in Tanzania. The country is among the ten largest producers of sunflower in the world

(and the second largest in Africa after South Africa), accounting for approximately 2.4% of

total global production (MITI, 2016). The production of sunflowers and sunflower oil seeds is

dominated by small-scale farmers in Tanzania, implying the future development of the sector

is especially significant as a potential tool for improving the livelihoods and welfare of poorer

Tanzanians (Zeng, 2011).

We focus on the effects of the tariff imposed on CPO imports on demand and supply dynamics

in Tanzania’s sunflower oil (and broader edible oil) sector, including the extent to which it is

enforced in practice and its impacts on domestic prices and incentives to invest in local

production and processing capacity. But we also look beyond tariffs to consider other factors

influencing the performance of the sector, such as the availability of agricultural inputs, access

to finance and the use of insurance, access to extension services and markets, the quality of

supporting infrastructure, and linkages between large and small enterprises. We also consider

alternative trade and industrial policy measures to address binding constraints to investment

across the sunflower oil value chain and stimulate domestic production of sunflower oil seeds,

local processing of sunflower oils and the manufacture of downstream products reliant on

sunflower oil. Our intention is to provide evidence-based input that will help the GoT to devise

a policy framework that delivers genuine investment in the expansion of domestic production,

processing and refining, benefits smallholders, and minimises increases in retail prices for the

poorest consumers.

In broad terms, the study seeks to contribute to an improved understanding of the following:

• Cost and pricing structures for sunflower oil seeds, sunflower and palm oil and

consumer goods with high sunflower or palm oil content in Tanzania.

• The price elasticity of demand and substitution elasticities between imported and

domestically produced CPO and between sunflower and palm oil.

• The binding constraints to increasing domestic sunflower oil seed production and

raising domestic production of sunflower oil (and other edible oils).

• The impact of the tariff on imported CPO on production, prices and demand along the

sunflower and palm oil value chains in Tanzania.

• The extent to which the existing tariff on imported CPO is enforced and the potential

for tariffs to be evaded through smuggling.

• Experiences in other countries that have successfully expanded domestic production of

oil seeds and facilitated greater levels of domestic value addition within different edible

oil value chains.

• The most appropriate policy measures – including trade and industrial policy measures

– to boost production and downstream processing and refining of sunflower oil in

Tanzania, while also ensuring smallholder farmers, women and the poor benefit from

the development of the sector.

4 4

1.2 Methodology and approach

We have employed a multi-faceted research approach involving both qualitative and

quantitative research techniques. This encompassed the following:

i. a literature review and meta-analysis of existing studies;

ii. country case studies of experiences in developing the edible oil sector in leading

producers and exporters of edible oils globally;

iii. stakeholder consultations with government policy makers, industry associations and

private sector support associations, and key actors in the sunflower value chain

(including large-scale processors and refiners); and

iv. fieldwork to collect primary data via surveys of sunflower farmers, small-scale

processors, retailers and consumers.

In addition, where possible given significant constraints related to the availability of suitable

data, we also undertook a limited quantitative analysis of prices, demand elasticities for

sunflower oil and palm oil, and cost drivers in the sunflower oil processing sector, drawing on

primary data collected through the surveys.

Further details on the approach adopted for the literature review and meta-analysis, case

studies, stakeholder consultations and primary data collection are outlined below.

Literature review, meta-analysis and case studies

To inform the identification of data gaps to be filled through the study and insights on the status

of current debates, the study undertook a comprehensive literature review and a meta-analysis

of existing studies. The objective was to avail key findings and insights, and collate available

data on the edible oil sector globally and in Tanzania. The literature review gathered evidence

of effective policy tools and approaches to developing production and processing capacity,

while drawing lessons that can help to develop the edible oil sector in Tanzania. This was

complemented by a further review of literature on specific country experiences in the

development of edible oil sectors to produce case studies of the sunflower value chain in the

Ukraine, soybeans and sunflower in Argentina, and palm oil in Indonesia.

Our review of existing literature highlighted mixed findings on the effects of tariff liberalisation

on the edible oil sector. In light of this, we conducted a meta-analysis to ascertain the magnitude

and nature of tariff impacts in different settings. A descriptive analysis of the results from

different studies was conducted based on different factors, such as methodology, country of

analysis, publication (to test for publication bias), or type of data used. The analysis provided

preliminary knowledge on the effects of import tariffs on the growth of the edible oil sector. A

total of 64 studies were collated from different authors, 34 of which matched the requirements

of meta-analysis (i.e. having sufficient data for deriving elasticities or standard errors and

coefficients). To account for heterogeneity in findings, we grouped studies according to

approach and methodology used, publication, nature of data used, year of publication, and

geographical coverage. However, the number of studies was insufficient for meta-regression,

hence we resorted to descriptive meta-analysis.

5 5

Stakeholder consultations

Interaction with stakeholders can provide useful insights to augment the evidence base

available in existing studies on the edible oil sector in Tanzania. Consequently, a number of

stakeholders were interviewed to gather understanding of the impact of the tariff and gain views

on the policy issues to consider in moving the industry forward, focusing on the sunflower

value chain actors and large-scale processors/refiners of palm oil. Consultations were guided

by a set of specific questions, and included key government Ministries, Departments and

Agencies (MDAs), regional/local government, large-scale processors, industry associations

and private sector support organisations.

A number of sunflower industry associations were consulted, including the Sunflower Farmers

Association (SUFA), the TEOSA, and the Tanzania Sunflower Processors Association

(TASUPA). In the Singida region, interviews were conducted with the Sunflower Processors

Association (SISUPA) and Singida Sunflower Farmers Association (SISUFA). In each of the

associations, the team met with the Executive Director/Chairperson. In addition, a selection of

private sector support organisations, including the Tanzania Chamber of Commerce, Industry

and Agriculture (TCCIA) and the Tanzania Private Sector Foundation (TPSF) were

interviewed. The Regional Administrative Secretary Office (RAS) was also consulted.

The specific organisations and individuals consulted across these broad groupings were

identified in the early stages of the research (and are listed in Annex A). Interviews were

conducted by the national expert in the research team. The focus areas and specific questions

for the various consultations is available upon request.

Fieldwork for collecting primary data

The fieldwork for collecting primary data was conducted in three regions: Singida, Dodoma

and Dar-es-Salaam. Singida and Dodoma are the leading sunflower oil seed producing regions

in Tanzania, and hence the largest number of actors, farmers and producers are located there.

In turn, as Tanzania’s largest city, Dar-es-Salaam is ideal for conducting a survey of consumers,

as well as being the headquarter for most institutions and large-scale manufacturing enterprises.

Table 1 summarises the distribution of the sample across the three regions. Given time and

resource limitations, the sample size is relatively small. In the case of farmers and consumers,

sampling was determined randomly once the focus location was identified. However, accessing

processors to survey was difficult as most were outside their factories or had temporarily closed

offices (due to shortages of raw materials). In such circumstances, the sampling was determined

by the availability of the respondent. The industry associations were extremely helpful in

locating the sampled farmers and processors.

Interviews were conducted using structured questionnaires. Upon completion of the fieldwork,

a data entry exercise was conducted on a template prepared in SPSS. Data cleaning and analysis

was done by conducting a descriptive analysis of the variables in the data, as well as calculating

various frequencies using SPSS.

6 6

Table 1: Sample for field work and response by location

Region Farmers Processors Consumers Total

Dodoma Target 10 10 50 70

Actual 20 12 51 83

Response rate (%) 200% 120% 102% 118%

Singida Target 10 10 50 70

Actual 12 13 54 79

Response rate (%) 120% 130% 108% 112%

Dar Target 0 0 50 50

Actual 0 3 50 52

Response rate (%) 200% 200% 200% 104%

Total Target 20 20 150 190

Actual 32 27 156 214

Response rate (%) 160% 135% 103% 112%

Source: Authors’ compilation.

1.3 Research scope and limitations

Wherever possible, we have attempted to utilise quantitative data in our analysis of the

sunflower value chain and the impacts of the tariff on imported CPO on prices, production and

processing along the value chain. However, in many cases our analysis was hamstrung by a

lack of detailed and sufficiently disaggregated data on production and prices in the sunflower

sector. For instance, a lack of detailed and disaggregated time series data on sunflower products

in Tanzania precluded a detailed empirical analysis of the tariff impacts. Instead, we have relied

primarily on information and data gathered through the abovementioned consultations and

surveys to understand the tariff impacts and other factors influencing production and

processing within the sunflower value chain.

1.4 Report structure

The remainder of the report is structured as follows. The next section provides a brief outline

of the main findings from previous studies of Tanzania’s edible oil sector and the effects of

tariffs in the sector, drawing on international evidence as well. Section 3 presents the main

findings from our study, focusing on the state of production and processing in Tanzania’s

sunflower sector, the impacts of the 10% tariff applied to imported CPO, and other challenges

affecting future expansion of the sector. Thereafter, three case studies are presented in Section

4, each documenting experiences in other countries – Argentina (sunflower and soybeans),

Indonesia (palm oil) and the Ukraine (sunflower) – that have successfully expanded domestic

production of oil seeds and/or facilitated greater levels of domestic value addition within the

edible oil sector. These cases provide insights into different policy options that could be

employed to develop the sunflower sector in Tanzania. Section 5 concludes and is followed in

Section 6 by a series of policy recommendations designed to support the long term

development of Tanzania’s sunflower sector.

7 7

2. Literature review: key findings from a meta-analysis of existing studies

A range of analytical work on the edible oil industry has been conducted in recent years, but

none of the studies seem to offer a balanced opinion on the feasible tariff rate that would be of

mutual benefit to farmers, as well as small-, medium- and large-scale processors, and

consumers. In general, the relationship between import tariffs and the edible oil sector shows

that imposing a tariff reduces imports, but the extent to which different actors in the sector gain

from the presence of a tariff is mixed. Studies on Tanzania are generally descriptive, mainly

presented as case studies of constraints and production trends. The lack of quantitative

estimates of tariff effects for Tanzania implies that arguments for or against an import tariff are

based more on perception than empirical evidence.

This chapter summarises key findings from a review of previous studies assessing the impact

of tariffs on the edible oil sector in terms of imports, consumption (demand), processing, and

effectiveness as a tool for promoting the development of the edible oil sector. The aim in

undertaking the review was to take stock of the data and insights available in existing studies

to identify knowledge gaps to be addressed through this study. We highlight the main findings

in the literature regarding tariff impacts (internationally and in Tanzania), before outlining key

areas of disagreement/conflicting findings. The chapter concludes by identifying key research

and data gaps relevant to this study.

2.1 Summarising the main findings in the literature

The existing supply of edible oil is insufficient to meet rising demand in Tanzania

Edible oil is a critical sector for Tanzania owing to its strategic importance as one of the

country’s biggest food items (sunflower oil accounts for 40% of cooking oil used in the

country), and as the second largest imported good in Tanzania (MAFC, 2008). Evidence

suggests the demand for edible oil has been increasing significantly with rapid population

growth, urbanisation and rising incomes (MRA, 2014), making edible oil one of the most

promising sectors in Tanzania (Mgeni et al., 2017; Salisali, 2017). Indeed, oil seed production

in the case of sunflower, sesame, and soy bean has been growing rapidly in recent years (see

Chapter 4 for a detailed discussion of production trends), including by smallholder farmers

(where seeds such as sunflower are also farmed for own consumption and for sale in the market)

and processors. However, despite the positive trends in production expansion, existing studies

suggest the supply of oil seeds is currently insufficient to meet rapidly increasing demand by

oil processors, and the current levels of domestic demand for oil seeds cannot be met by local

supply.

8 8

The sunflower oil sector in Tanzania is facing a critical productivity challenge

Several challenges affecting the edible oil sector can be deduced from the evidence presented

in existing studies conducted on Tanzania. It should be noted that these challenges may limit

the full potential of tariff policy for the sector to achieve its intended objectives, unless other

counter-policies are adopted (Jamet and Chaumet, 2016). Major challenges include low

production capacity, and poor tools and technology for production and processing. For

instance, the study my MRA (2014), and other studies (Sintoo, 2015; Match Makers

Associates, 2014; Mgeni et al., 2017; Iringo et al., 2014; Ugulumu and Inanga, 2013), indicate

that the following critical factors hamper the production activities of domestic producers and

processors:

• Low processing capacity – one study estimates that over 90% of the processors in

Dodoma, Kasulu and Kongwa councils use small Chinese oil extraction expellers, such

as Y 95, 105, 118, and 130 series. Other reported processing machines in use include

Rosedown from either India or the United Kingdom (UK), Alpha Willball and Roast,

which have the capacity to process small quantities at once, while others such as those

from the UK are reported to be efficient but expensive in terms of repairing and

replacing worn out components.

• A lack of seed kernels.

• Temporary closures of businesses due to machine breakdowns (affecting more than

60% of processors in Dodoma).

• The use of less advanced technology (Ruteri and Xu, 2009), which limits the ability of

small processors to grow and become formal processors (Coad and Tamvada, 2012;

Ruteri and Xu, 2009). The informal nature of these processors means their products are

mostly destined for domestic markets, usually not even far from their respective

districts of production (Kavishe, 2013).

Studies on Tanzania lack evidence on the effect of tariffs on the edible oil sector

Our review of studies conducted on the edible oil sector in Tanzania (and the sunflower sub-

sector in particular) shows that most studies are based on descriptive analysis of demand and

supply (production) of sunflower and edible oil, factors affecting the production and processing

of sunflower, value chain analysis of sunflower (MRA, 2014; Mgeni et al., 2017; FAO, 2014),

gender aspects in sunflower production (Ekama Development Foundation, 2015;

Vanderschaeghe and Okoth, 2017), and other related issues. Little has been done to assess how

macroeconomic factors such as trade policy (e.g. tariffs) affect the growth of this sector (and

wider issues such as gender concerns), and hence, derive policy recommendations that are

focused and applicable. A recent study by Salisali (2017) for the TPSF provides some insights

on the fiscal issues related to the sector, although in a rather descriptive manner. The study

followed the opposing views between importers and domestic producers concerning the issues

related to fiscal policy and its impact on the edible oil sector in Tanzania. This was done to

understand the existing fiscal situation in the edible oil sector, and how the prevailing fiscal

regime affects domestic farmers and edible oil processors, as well as importers. The study

collected primary data from small, medium, and large processors and found that the

introduction of a 10% tariff on imports of CPO did not have a significant impact on edible oil

businesses during the 2016/2017 season.

9 9

Salisali (2017) finds that the world market price for sunflower oil is US$ 850 per MT, which

climbs to a total of US$ 1,036 per MT once transport costs and other charges are factored in.

Locally produced sunflower oil is priced between US$ 1,211 per MT (in low season) and US$

1,036 per MT (in high season). This shows that imported crude sunflower oil is marginally

cheaper than the domestically produced equivalent, suggesting a moderate tariff on crude

sunflower oil – or, indeed, on CPO (as is already applied), depending on the level of

substitutability between crude sunflower and palm oil – could potentially help to equalise

domestic and international prices and possibly boost demand for domestic production. The

price for CPO in international markets is further reported by Salisali (2017) to range from US$

695 per MT to US$ 725 per MT, while semi-processed oil ranges between US$ 690 per MT

and US$ 732.5 per MT. The price difference between sunflower and palm oil is between US$

200 and US$ 300 per MT. Salisali (2017) argues that imposing a 10% tariff on CPO still makes

the price of crude oil cheaper in the domestic market. The study recommends addressing key

issues facing the sector such as overregulation, as well as the development of a national edible

oil sector strategy (a national strategy already exists for the sunflower sub-sector). It goes

further by recommending the protection of the sector due to its infancy, although it does not

provide a detailed assessment of what form of protection would be best suited to the sector.

Internationally, tariffs are widely used as a policy instrument to promote domestic edible oil sectors

Many governments have looked to develop strategies to protect domestic producers of

sunflower from foreign competition, principally involving the use of various trade policy

instruments, the most common of which are import duties on imported edible oil. This follows

the common finding that import tariffs reduce the importation of edible oil. Other instruments

in use to protect domestic producers include minimum support prices (Srinivasan, 2005; Jamet

and Chaumet, 2016). However, apart from reducing imports, import tariffs are said to impact

other variables in the economy, such as final prices, consumption and domestic production, to

varying degrees and at different magnitudes (Srinivasan, 2005).

A study by Dohlman et al. (2003) argues that liberalisation of the edible oil sector (removal of

tariffs) contributes to increased imports, in line with trade theory. However, as Srinivasan

(2005) argues, higher import volumes may harm domestic producers of edible oil who cannot

compete with cheaply imported varieties, thereby potentially transmitting impacts down the

value chain to processors and farmers.

Importantly, however, domestic production capacity and competitiveness are important factors

ultimately influencing the impact of tariffs on imports. Egwuma et al. (2016) show, for

example, that even though China and India adopted tariffs and managed to reduce imports of

edible oil, domestic production still declined. They posit that one of the reasons for the decline

in domestic production, despite the imposition of tariffs, stemmed from the fact that the edible

oil sectors in these countries faced other underlying challenges in production and processing

capacities.

Some studies argue tariffs on edible oil can play a positive role in protecting and promoting the

sunflower value chain. For instance, in certain countries seeking to protect their edible oil

sectors (e.g. India, China, Sri Lanka and the United Arab Emirates), it has been argued that

import tariffs played an important role as an instrument with the power to influence different

segments along the value chain, albeit in different ways (Chand, 2002; Chand et al., 2004;

Hashim, 2008).

10 10

In other instances, however, import tariffs may generate distortions in domestic edible oil

markets, particularly if the protection they afford to domestic producers renders them

inefficient. In some cases, other counter-measures have been used to correct for such

distortions. For example, until 2006, China imposed tariff-rate quotas (TRQ) on imports of

edible oil, although it later phased them out after accession to the World Trade Organisation

(WTO), and it left in place a 9% tariff on imports of soybean oil and rapeseed oil. TRQs on

other oils (sunflower, peanut and corn) were also removed and a fixed 10% tariff was adopted

(Cassiday, 2017). Subsequently, to protect the edible oil value chain in China, Jamet and

Chaumet (2016) note that the government of China supported soybean farmers (who would be

affected by tariffs) by setting minimum prices for their products. This move was aimed at

shielding these farmers from increased production costs and competition from imported

substitute products. Although the policy aimed to protect domestic producers, it hurt

consumers, as domestic prices for soybeans were higher than import prices. In 2013, the price

of imported soybeans was US$ 600 per ton, while the domestic price was US$ 750 (Jamet and

Chaumet, 2016). In this case, domestic soybean processors found it appealing to import beans

and process them domestically, as opposed to purchasing domestically produced beans.

International evidence suggests tariffs impact the various edible oil value chain actors differently

As noted earlier, the presence of import tariffs on edible oil is expected to reduce the level of

edible oil imported into a country. Under normal circumstances, all actors along the value chain

would be expected to gain (from farmers through to processors). However, our literature review

suggests otherwise, with some studies revealing that processors/producers of edible oil gain

more than farmers. That said, we are not aware of any study that unpacks the causes of these

differences in outcomes in the edible oil sector. Ghosh (2009) argues that the liberalisation of

tariffs is more beneficial to producers and consumers of edible oil in India, and recommends

further liberalisation of tariffs to 65% from the base case where tariffs were estimated at 80%.

Srinivasan (2005) also argues that producers gain more than farmers in cases of protectionism.

A 65% customs duty on edible oil in India was seen to cause an increase in both producer and

consumer surplus, raise government revenue, and make the edible oil sector more cost-

efficient. Another import consideration is that changes in prices generated by import tariffs are

likely to be borne by final consumers (Srinivasan, 2005).

Further evidence from China and India is summarised in Egwuma et al. (2016) and Chand et

al. (2004). In their studies, they argue that liberalisation in these countries harms local farmers

and domestic production of palm oil, and, on this basis, government support in the form of

improvements to the enabling environment (infrastructure and provision of inputs) for

production is recommended. Chand et al. (2004) argued that lowering tariffs on edible oil under

the WTO may hamper the potential of India’s domestic oilseed sector. They argued further that

since India was already importing significant quantities, there was a risk of being exposed to

global price volatility shocks. Another study by Gulati and Mullen (2003) assessed the impacts

of liberalisation of the edible oil sector in India, which had imposed a high level of tariff

protection in the sector to encourage domestic production. Their model found that

protectionism protected oil crushers rather than producers of oilseed (farmers). However, they

found that oil processors were less efficient than oil farmers, since processing in India was done

by small-scale industries.

The findings from the surveyed studies reveal that extreme cases of import tariffs for edible oil

can have clear adverse effects. These include the total removal of import tariffs or total bans

11 11

on imports of edible oil. For instance, Ghosh (2009) uses simultaneous equations to assess the

impacts associated with liberalising tariffs on edible oil and the oil seed subsector in India. The

study finds that the proposals by the Lahiri Committee to impose a tariff rate of 65% (down

from the prevailing rate of 80%) would be more welfare enhancing in India. Their study

recommends the avoidance of extreme policies in the edible oil sector, such as a no tariff

scenario or an extreme tariff scenario (which would reduce imports to zero), since both options

would not benefit the economy. However, Ghosh argues that if the two extreme cases must be

considered, then the zero-tariff case would be more appealing than the other extreme of zero

imports. He argues that both consumer surplus and processing margins are greater in the zero-

tariff case compared to the (high tariff) no import case.

Importantly, the studies reveal that the exact welfare enhancing or decreasing effects of tariffs

may vary from country to country. This emphasises the need to adopt a nuanced approach to

investigating the impact of tariffs on crude edible oil imports in the Tanzanian context.

2.2 Information gaps and implications for this study

In addition to the lack of clarity on tariff impacts in the edible oil sector internationally, our

review of studies specifically focused on Tanzania reveals various data and analytical gaps that

can be interrogated through further research, including that undertaken for this study. In

particular, there is a lack of conclusive evidence in the available literature on the effects on

domestic production and consumption in the edible oil sector of the 10% tariff imposed on

CPO imports.

Furthermore, the nature of the relationship between palm oil and sunflower oil products in

Tanzania has not been systematically established, making it difficult to assess consumer

preferences between the two products. Our study unpacks substitutability and price dynamics

between sunflower oil and palm oil, and provides new evidence on preferences, demand,

quality, price, supply and other dynamics based on surveys of farmers, processors and

consumers and consultations with key industry stakeholders.

Finally, although there is clarity in the existing literature on the constraints facing the sunflower

sub-sector in Tanzania, few attempts have been made to assess the potential for growth and

expansion in the sector. In addition, there has been little consideration of the merits of

alternative policies (other than a tariff on imports) to unlock the potential of the sector and

boost production and processing of sunflower oil and related products in Tanzania. In the next

chapter we present new evidence designed to address these information gaps.

12 12

3. The state of production and processing in Tanzania’s edible oil sector, tariff impacts and challenges for future expansion

In this chapter we present the main findings from our analysis. We begin with an assessment

of recent trends in sunflower production and sunflower oil processing (and in other edible oil

value chains where applicable) in Tanzania. We then turn to a (mostly qualitative) assessment

of consumer preferences, demand and substitution elasticities, focusing on sunflower oil and

palm oil. This is followed by a discussion of the varied impacts of the 10% tariff imposed on

CPO imports in 2016, drawing heavily on the perspectives of different actors in the sunflower

value chain. We close the analysis by highlighting a range of challenges and constraints to

expanding production and processing in Tanzania’s sunflower oil sector. The persistence of

these issues suggests a wide array of policy interventions are required to develop the sunflower

oil sector in Tanzania, rather than a narrow focus on tariff policy.

Wherever possible, we have attempted to base the analysis presented in this chapter on

quantitative evidence. However, in many cases the available data is insufficient for meaningful

empirical analysis. In these instances, we have drawn heavily on the insights and perspectives

of different stakeholders and sunflower value chain actors gathered through our surveys and

consultations.

3.1 Recent trends in production and processing in Tanzania’s sunflower (and other edible oil) value chain

Oil seed production

Table 2 compares recent trends in edible oil seed production in Tanzania, based on government

data. Production is heavily concentrated in sunflower, groundnuts and sesame seeds, which

collectively account for 99.2% of total production. By comparison, domestic production of oil

palm seeds is limited.

13 13

Table 2: Production of oil seeds (in tonnes)

2013/14 2014/15 2015/16p Average

(2013/14 -2015/16)

Percentage of

total in 2015/16

Sunflower 2,755,000 2,878,500 2,995,500 2,876,333 47.5

Groundnuts 1,635,735 1,835,933 2,025,595 1,832,421 32.1

Sesame 1,113,892 1,174,589 1,232,092 1,173,524 19.6

Palm oil 41,000 41,475 41,925 41,467 0.7

Soybean 6,025 6,030 6,035 6,030 0.1

Total 5,551,652 5,936,527 6,301,147 5,929,775 100.0

Source: Ministry of Agriculture, Livestock and Fisheries, as quoted in BOT (2017).

Focusing on sunflower oil seeds, and looking over a longer timeframe, Figure 1 shows rapid

growth in sunflower seed production in Tanzania since 2010.2 Indeed, the data suggests that

sunflower seed production has expanded almost nine times between 2008 and 2015. Although

the majority of the farms were established in the 1990s, actual growing of sunflower gained

momentum only from 2013 onwards. According to SUFA, this is because previously most

farmers were producing sunflower seeds for their own consumption and selling off any surplus.

In the view of SUFA, sunflower farming and processing has since grown significantly in

response to increased awareness among consumers of the health benefits of using sunflower

oil. As a result, farmers now invest in larger farms, and often replace their land for traditional

crops such as cassava/millet with sunflower farming.

Over 90% of the farmers we surveyed consider sunflower as their main cash crop and 93%

undertake sunflower farming as their main occupation – implying the crop is a significant

generator of income and employment. Clearly, given the surging momentum in the industry,

sunflower farming is increasingly becoming an economic opportunity which will greatly

benefit farmers, consumers and the economy at large. Tanzania’s sunflower seed production is

already significant on a regional level, accounting for 35% of total production on the African

continent (Salisali, 2017).

Importantly from a socio-economic development perspective, government estimates indicate

most production of sunflower oilseeds is undertaken by smallholder farmers, which account

for 95% of all producers operating in the subsector (MITI, 2016). These smallholders generally

operate farms spanning 2 acres on average, and rely on rain-fed farming that is only possible

for one season a year. In turn, medium-scale farmers make up 4% of the total population of

sunflower oil seed producers, and the remaining 1% are large-scale farmers. There are around

1.6 million producers of sunflower seeds in Tanzania according to MALF estimates.

2 This is based on FAO data. It is worth noting, however, that there is wide variation in sunflower oil seed production estimates across different

sources. These estimates range from 350,000 MT to 2.9 million MT. Nevertheless, most estimates suggest Tanzanian production of sunflowers has grown rapidly in recent years.

14 14

Figure 1: Production of sunflower seeds in Tanzania (in tonnes)

Source: FAOStat.

Despite the significant increase in production of sunflower seeds in recent years, productivity

among sunflower farmers in Tanzania appears to be low. Across the 31 farmers we surveyed,

they plant, on average, 15 hectares of sunflower with the maximum area planted reaching 100

hectares. Their yields are generally low. On average, these farmers collect nearly 829 kg per

hectare planted. The highest yield is 4,455 kg. Although calculated over the area harvested

(which is lower than the planted area), Tanzania’s average yield was 10,000 kg per hectare in

2016.3 This puts these farmers at the bottom of the productivity distribution. In fact, the most

productive of these farmers could not reach the average productivity for the country.

Downstream processing

Downstream, across all types of edible oil, domestic oil production is estimated at

approximately 180,000 MT per annum while demand stands at 400,000 MT according to

government data (MRA, 2014). There are only a few large-scale processors of edible oils in

Tanzania. In the case of sunflower oil, the major players are Mt Meru Millers4 (producing the

Sunola and Singida Fresh brands) and Sunshine – a Chinese factory (producing the Sunbelt

brand) in Dodoma; whereas in the case of palm oil, there are two major factories: Murzah Oil

(producing Korie and Sundrop brands) and MeTL or East Coast Ltd (producing the Safi brand),

although BIDCO – a Kenyan investment (producing the Kimbo brand) is resuming production

and could be the third major player.

Sunflower oil production in Tanzania has expanded rapidly in the last decade, although data

on production volumes varies across different sources. Government data on sunflower oil

processing volumes between 2009 and 2013 is presented in Table 3, indicating a total

processing volume of 275,932 MT in 2013, more than 3.5 times the volume of production in

2009. Data reported in Salisali (2017) suggests the volume of sunflower oil production in

Tanzania is lower, but has increased from 52,000 MT in 2005 to a current level of 163,000 MT

(Salisali, 2017). Based on these estimates, sunflower oil is the major edible oil produced in

Tanzania, representing up to 90% of the 180,000 MT of edible oil produced annually. Cotton

3 Based on FAOStat data. 4 Mt Meru is a US$ 50 million investment company, which operates three plants in Arusha, Bunda and Singida (and closed one in

Shinyanga). It is the largest sunflower oil seeds producer outside South Africa, processing half a million tons of sunflower seeds per year.

-

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

3,500,000

2008 2009 2010 2011 2012 2013 2014 2015

To

nnes

15 15

oil (8%) and palm oil (approximately 1.5%) constitute much smaller shares of overall

production in Tanzania. The country is also an important producer of sunflower oil in regional

terms, accounting for 23.1% of Africa’s total sunflower oil production (BOT, 2017).

Table 3: Trends in the Production and Processing of sunflower (Quantity, MT)

2009 2010 2011 2012 2013

Sunflower oil 77,706 88,949 200,621 286,831 275,932

Source: MRA (2014).

Looking at the regions in which we focus for this study, sunflower is a key strategic crop in

both Dodoma and Singida, and the regional governments in these two regions are keen to

facilitate more processing by medium- and large-scale processors, with backward linkages to

farmers. To this end, interventions to support the industry include allocating land for

investment in farming and promotion of out-grower schemes. About 70% of farmers in

Dodoma grow sunflower since there is a sure market, unlike other crops such as maize. The

distribution of sunflower processing industries in Dodoma is presented in Table 4, and indicates

there are only two medium-scale and two large-scale processing factories operating in the

region, compared to over 277 small-scale factories. The regional government successfully

facilitated the establishment of the Sunshine factory (through a Chinese investor) in Dodoma.

Another plant is being established in Kibaigwa (Kahama Oils), which is 100% domestically

owned, and will require a supply of about 180 tons of sunflower oil seeds per day.

Table 4: Distribution of Sunflower Processing Factories in Dodoma Region by size

Size Number of Factories District of Location

Large scale 1 Dodoma Urban

1 Kibaigwa

Medium Scale 2 Kongwa

4 Kondoa

Small Scale 95 Kondoa

46 Kongwa

51 Dodoma Urban

28 Mpwapwa

12 Chamwino

28 Chemba

9 Bahi

Total 277

Source: National expert’s consultations in Dodoma.

The rapid expansion of sunflower oil production in Tanzania has been accompanied by some

growth in exports of sunflower products (seeds, oil and cake) in certain years since 2005 (see

Table 5). This has been driven by both diversification into new products (e.g. sunflower cake)

and entry into new markets (such as India) (Salisali, 2017). The value of sunflower products

exported by Tanzania increased 70-fold between 2005 and 2014, rising from US$ 1 million in

2005 to US$ 70 million in 2014 (MITI, 2016). This was driven predominantly by exports of

sunflower cake, which registered especially rapid growth – the total value of Tanzania’s

exports of oil cakes and other solid residues from sunflower seeds to the rest of the world

increased from just US$ 60,409 in 2005 to US$ 7 million in 2015 and more than US$ 20.4

16 16

million in 2016. According to Dalberg (2018), the vast growth in oil seed cake exports has

limited local oil extraction. Perhaps linked to this, the value of Tanzania’s exports of crude

sunflower oil, after increasing significantly between 2008 and 2011, declined substantially

between 2013 and 2015, and were neglible 2016. Similarly, Tanzania’s sunflower oil (not

crude) exports have actually declined overall in comparison to the 2005 level, and totalled just

US$ 0.4 million in 2016.

Table 5: Tanzania’s sunflower seed, oil and cake exports to the rest of the world (US$ millions), 2005-2016

Product 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Sunflower

seeds

0.1 0.2 1.0 1.4 0.1 0.1 0.1 1.6 1.7 0.3 0.0 0.5

Crude

sunflower oil

0.0 0.3 0.2 4.8 2.0 12.4 12.9 12.7 6.2 7.1 6.3 0.2

Sunflower oil 1.8 0.4 0.4 1.8 1.0 2.3 1.4 0.7 0.9 1.9 0.6 0.4

Sunflower

seed oil cake

0.1 0.0 0.2 11.1 11.9 8.3 7.5 45.0 24.9 65.6 7.0 20.5

Source: UN Comtrade data.

Despite the impressive growth of Tanzania’s sunflower industry, and the expansion of exports

of sunflower products over the past decade, the available data suggests domestic production of

sunflower and other edible oils has not been able to meet even half of Tanzania’s existing edible

oil demand. Tanzania remains a net importer of edible oils. While the country has not been able

to capture fully its production and processing potential,5 the importation of edible oil has been

surging, with most imports originating from Singapore, the United Arab Emirates, India and

China (BOT, 2017). For its part, CPO is mostly imported from Malaysia (accounting for nearly

83% of Tanzania’s imports of CPO in 2016) and Indonesia (14.5% of all CPO imports).

Currently, Tanzania spends US$ 120 million annually on importing edible oil to close the gap

between supply and demand. The volume of Tanzania’s imports of crude edible oil rose from

0.3 million tons in 2012 to 0.5 million tons in 2015, with CPO accounting for 44.1% of overall

imports (BOT, 2017).

Imports of sunflower oil into Tanzania are also substantial. In 2016, the value of imports of

sunflower seed oil or safflower oil (sum of HS codes 151211 and 151219) from the rest of the

world totalled nearly US$ 5.6 million (although this was considerably lower than the US$ 8

million in imports recorded in 2015). Tanzania’s sunflower oil imports are sourced from

different countries than those supplying CPO, with the bulk of crude sunflower oil (HS 151211)

imported from Argentina (68.1% of total imports in 2016) and Antigua and Barbuda (31.9%)

and most of the sunflower oil other than crude (HS 151219) coming from the Ukraine (73.2%

of total imports in 2016), Egypt (5.1%), the United States (5.1%) or Singapore (4%).

To obtain a wholistic picture of the sunflower value chain in Tanzania, encompassing various

stages of production as well as trade, Table 6 presents the supply and demand balance for

sunflower seed, oils and cakes. Commodity balances help to track the components of supply

and demand, identifying potential gaps in supply using a uniform unit. This consistent

information is only available for 2013 for Tanzania. However, cross-references to other sources

allow us to extrapolate some conclusions about trends and the current situation.

5 MRA (2014) assumes that 1 metric ton of oil seeds produce about 300 liters of sunflower oil, with poor technology (machines) meaning

extraction rates of oil from sunflower seed are only around 30%, while an additional 12-15% can be achieved if chemical extraction is used. More discussion on Tanzania’s existing production and processing capacity is detailed in MRA (2014).

17 17

Table 6: Sunflower seed, oils and cake balance (2013) (in tonnes)

Supply

Sunflower seed Sunflower cake Sunflower oil

Production 1,083,000 203,800 152,850

Imports 8 1 1,758

Demand

Exports 24,592 93,831 7,304

Processing 509,500 -

Food/Feed supply 108,326 109,970 87,304

Other uses (seed, losses) 40,590 -

Stock variation 400,000 - 60,000 Source: FAO.

Note: In sunflower seed tonnes equivalent. Stock variation refers to the change in the stocks during the period.

In 2013, Tanzania imported nearly 1,800 tonnes of sunflower oil, which accounts for less than

1% of the total edible oil demand in Tanzania (estimated at around 400,000 tonnes) and almost

2% of the domestic production of sunflower oil. In fact, Tanzania enjoyed a surplus in the trade

of sunflower oil in 2013. Considering that international prices of sunflower oil have fallen

during the period, it suggests that the volumes of imported sunflower oil have remained fairly

constant. In fact, exports of sunflower seed in 2016 amounted to just 3,000 tonnes. Production

of sunflower oil, based on Salisali’s (2017) estimates, has not increased substantially since

2013. Given limitations in the utilisation of crushing capacity, oil processing volumes appear

to have remained fairly constant during the period. This suggests the main components of the

sunflower oil balance have remained relatively constant.

This has not been the case for sunflower seed. Production of sunflower seed has tripled during

the period, reaching almost 3 million tonnes in 2015. With a limited crushing capacity and a

very competitive international market for exports, most of the increased production has been

marketed domestically as food or feed, or exported as seed cake. This indicates that the sluggish

sunflower oil production in Tanzania is no longer explained by the lack of sunflower seeds to

crush. Assuming an extraction rate (30%), if all the sunflower oil seed produced in the country

was used to produce oil, it should easily cover the Tanzanian edible oil demand and it could

also be exported.

Rather than a lack of sunflower seed, other factors – including the quality of sunflower oil

seeds produced in Tanzania – may be limiting the domestic capacity to supply sunflower oil.

On one side, consumers may face higher domestic prices for sunflower oil and/or their

preferences may be oriented towards palm oil. It could also be the result of an unprofitable

crushing sector. In this sense, it also needs to be considered whether there is sufficient demand

for by-products (such as sunflower cake). In the sections that follow, we shed light on some of

these issues through insights gathered via consultations with key stakeholders in the sunflower

sector and our survey of farmers, small-scale processors, traders and consumers.

3.2 Consumer preferences, demand and substitution

The survey asked a couple of questions to understand consumer preferences for different types

of edible oil. Based on the results of our survey of consumers, the majority of respondents

18 18

prefer sunflower oil over other types of vegetable oil, including palm oil. The responses show

that 81% of consumers prefer sunflower oil, compared to around 19% who prefer palm oil

(Figure 2). However, two important considerations need to be highlighted when interpreting

these results. First, the survey is not national in coverage and it may present a bias in favour of

the areas where sunflower seed is produced. Second, the survey does not include industrial

downstream demand by the manufacturing and services sectors (e.g. restaurants). Therefore,

the conclusions on preferences across the two varieties of oil are focused exclusively on the

final demand.6

Figure 2: Surveyed consumer preferences between sunflower and palm oil

Source: Survey data (consumers of edible oil).

As shown in Figure 3, the main reasons for the surveyed consumers’ preference for sunflower

oil appears to be health concerns and quality, although some consumers are also influenced by

price. While health concerns are very influential in governing consumers’ purchasing

decisions, this does not mean that in the face of major variations in prices, consumers will not

substitute one type of oil for the other. However, it indicates a clear subjective preference for

sunflower oil. The consumers we surveyed appear to be less influenced by brand or packaging.

6 There is a certain risk of confusion in the analysis coming from the concurrent presence of households and retailers. An analysis of the final

demand should focus exclusively on household behaviour. Prices paid by households are expected to be different from those paid by retailers.

This may complicate the analysis. Moreover, whilst demand by manufacturing is explained primarily by production technologies, demand

from retailers and wholesalers depends primarily on the demand by households. They tend to stock and re-sell the products that their consumers

request. Demand by retailers should reflect the preferences of consumers.

19 19

Figure 3: Key reasons for surveyed consumers’ preference across different types of edible oil

Source: Survey data (consumers of edible oil).

The consumer survey results suggest that the more highly educated consumers in our sample

were more motivated by quality and health factors, compared to less educated customers who

were more influenced by price considerations (see Figure 4). Price appears to be the biggest

consideration in choosing the type of edible oil for consumers with no education, whereas the

highly educated consumers are more inclined to give greater weight to quality, availability and

health considerations when making their purchasing decision. Perhaps linked to this, Figure 5

shows that consumers with a lower level of education had higher incidences of consuming palm

oil compared to consumers with higher levels of education, most of whom consume only

sunflower oil.

Figure 4: Factors driving the edible oil preferences of surveyed consumers, by level of education

Source: Survey data (consumers of edible oil).

20 20

Figure 5: Surveyed consumers’ preferences for palm oil or sunflower oil, by level of education

Source: Survey data (consumers of edible oil).

In addition, using the data from our consumer survey, we investigated whether low income

earners consume more palm oil due to its cheaper price compared to higher income earners by

cross tabulating preferences by income levels. The results in Table 7 indicate that 21% of

consumers with annual income below TSh 5 million consumed palm oil and 79% consumed

sunflower, whereas 100% of consumers with annual income of more than TSh 5 million only

consumed sunflower oil.

Table 7: Surveyed consumers’ preferences for sunflower or palm oil, by income category

Sunflower oil Palm oil Total

Below TSh 5 million 79% 21% 100%

Tsh 5 to 10 million 100% 100%

Tsh 10 to 20 million 100% 100%

Above TSh 50 million 100% 100% Source: Survey data (consumers of edible oil).

Turning to the retailers, to establish how they assess the business aspect of the preferences of

their customers, we asked them to name the three fastest moving brands of edible oil in their

stores. The results presented in Table 8 indicate that across the sample of retailers, a palm oil

brand (Safi) was the fastest moving brand, followed by a sunflower oil brand, Singida.

However, customer preferences, at least based on the perceptions of the retailers we surveyed,

do appear to be relatively evenly distributed across different brands and types of edible oil.

21 21

Table 8: Fastest moving edible oil brands according to retailers

Brand Frequency Percent

Sunbelt 7 4.5%

Sundrop 7 4.5%

Singida 15 9.6%

Kindai 8 5.1%

Munga 6 3.8%

Singida bright 3 1.9%

Urafiki 4 2.6%

Safi 17 10.9%

Korie 6 3.8%

Halisi 2 1.3%

Mwenge 4 2.6%

Heshima ya jiko 1 0.6%

N/A 76 48.7%

Total 156 100.0%

Source: Survey data (consumers of edible oil).

Producer perspectives

In terms of competition across edible oil varieties from a producer perspective, none of the

surveyed small-scale sunflower processing firms indicated that they face relevant competition

from domestic processors of other types of edible oil and only one firm indicated it faced

competition from processors of imported crude edible oil (Table 9). Instead, most of their

competitors tend to be other processors of sunflower oil. Interestingly, the fact that most small-

scale processors do not consider producers of other edible oils as direct competitors suggests

that the degree of substitutability between sunflower oil and palm oil is relatively low from the

processors’ perspective.

Table 9: Sources of competition for small-scale sunflower oil processors

Yes No

Other domestic sunflower oil processors 25 3

Domestic processors of other types of edible oil 0 28

Processors of imported crude edible oil 1 27 Source: Survey data (small-scale sunflower oil processors).

Estimating demand elasticities

Based on the survey results (and specifically the information we can glean about quantities and

prices paid by households from the respondents), Table 10 suggests that, on average, palm oil

is 15% cheaper than sunflower oil. However, the price data in Table 10 shows wide variation

in prices – ranging from TSh 1,500 to TSh 5,000 – for sunflower oil. Importantly, these price

differences reflect wide variation in the quality of sunflower oil, which is dependent on the

extent to which the oil is refined (single, double or solvent extraction). The price of sunflower

22 22

oil that is processed by informal crushers (at between TSh 1,500 and TSh 3,000)7 is much lower

than the price charged by large-scale processors performing double refinery (around TSh 5,000

per litre) and even higher for solvent extraction or higher standardisation (TSh 7,000 per litre).

Indeed, the wide range in prices reflects the large difference in quality between the sunflower

oil crushed by small-scale informal processors (selling at the roadside at an average price of

TSh 3,000 per litre) and sunflower oil from large-scale processors or brands such as Alaska

(selling at supermarket chains as a price that can reach TSh 7,000 per litre). Substitutability

between sunflower and palm oil can be seen in the lower quality sunflower oil products, where

sunflower and palm-based oil are sold at comparable prices. Low income customers purchasing

within that segment of the market may have a higher degree of price elasticity, potentially

substituting between sunflower and palm oil purely on the basis of changes in prices.

Table 10: Prices paid by households

Obs.

Mean

(Shillings)

Std. Dev.

(Shillings)

Min

(Shillings)

Max

(Shillings)

Palm Oil 12 2912.5 697.4 1000 3500

Sunflower Oil 43 3427.1 591.2 1500 5000 Source: Survey data.

The information collected through the survey on prices paid and quantities consumed can be

used to estimate price-demand elasticities. These elasticities should indicate how demand

reacts to changes in prices. This will help to assess the effects of any policy affecting the prices

of edible oils (including tariff policies).

Elasticities need to be estimated by fitting demand functions to the observed data. This means

identifying the parameters associated with the own price (or the direct elasticity) and the cross-

price (indirect elasticity). The latter makes it possible to identify the degree of substitutability

of a product with respect to another.

In addition to price and quantity information, additional information is necessary to identify

the equation. Notably, data about household income is required to estimate the income-demand

elasticity. This makes it possible to assess by how much the demand of the product increases

against changes in income. Other variables can help in the identification and control for

different factors that may affect the estimates. For example, the location of the household may

help to explain variations in demand among households not explained by the variables of

interest. Moreover, a good estimation needs to be based on a sufficiently large number of

observations. This is necessary to reduce the incidence of outliers in the estimations and ensure

that the estimators meet the desired properties.

Table 11 presents some estimations of the demand elasticities for sunflower oil and palm oil

based on survey data. Unfortunately, a very limited number of observations were available to

perform the estimations. This limits the quality of the estimates and the controls that can be

introduced into the equations. Moreover, the survey did not collect data about household

income but only classified households according to income ranges. Therefore, the income

variable employed cannot be used in principle to estimate elasticities and it may not be

providing a good identification of the equation.

7 The prices reported at the bottom end of this range, where the minimum price is said to be as low as TSh 1,5000, may be refereeing to

prices for smaller units or to cases of own consumption whereby consumers are also farmers, and hence refer only to processing costs when reporting the price.

23 23

Table 11: Estimated demand elasticities for sunflower oil and palm oil

logqsun logqpalm

logpsun - 2.615 - 0.336

2.010 1.854

logppalm 2.273 0.305

1.711 1.914

Income 0.741 0.944

0.896 1.014

Adj R2 0.430 0.280

Observations 43 12

Source: Own calculations based on survey data.

Note: Standard errors in italics.

Taking these issues affecting the estimation into consideration, it is still possible to extract

some conclusions about how prices may affect the quantities demanded. Given the few

observations available for the estimation of the demand for palm oil, its estimates are not

trustworthy at all. In the case of sunflower oil, the estimated elasticity (-2.6), although not

statistically significant, may be indicating a relatively high sensitivity of demand to the price

level. However, it is important to distinguish the perceptions or opinions of consumers and

their true behaviour. Moreover, this high demand elasticity may reflect a high level of

substitution with respect to fat oil, for which information is not available.

The cross-price elasticity suggests a high degree of substitutability between sunflower oil and

palm oil. This is expected as both types of oil are likely to be highly substitutable (even when

consumers may indicate that their consumption decisions are not based on price). This suggest

that changes in the price of palm oil may lead to significant changes in the demand for

sunflower oil. However, it is necessary to interpret these estimates with care. First, they are not

statistically significant. Second, the cross-price elasticity needs to be validated with the cross-

price elasticity of the demand for palm oil. The limited number of observations in the other

estimation does not allow for an effective comparison or make it possible to test the similarity

of both parameters. Finally, the parameter of the income variable (which cannot be interpreted

as an elasticity) can only validate the proposition that higher income households consume more

sunflower oil and provides some minimum assurance that the price-demand elasticities are

pointing in the right direction.

3.3 Tariff impacts

Remarkably, awareness of the 10% tariff on CPO is extremely low amongst the sunflower

value chain actors we engaged with through the survey (see Table 12). Just four of the

sunflower farmers we surveyed claimed to have knowledge of the tariff. In turn, only three

small-scale processors were aware that the tariff was introduced in 2016. Instead, only the

associations and large-scale processors we engaged through consultations appear to be

conversant with such policy issues, raising doubt on the extent to which farmers and processors

are able to advocate for policy to support the industry.

24 24

Table 12: Awareness of the 10% tariff on CPO among sunflower farmers, small-scale processors and edible oil retailers

Farmers Processors Retailers

Farmers Frequency Percent Frequency Percent Frequency Percent

Yes 4 12.9% 3 10.7% 19 23.8%

No 27 87.1% 25 89.3% 61 76.2%

Total 31 100.0% 28 100.0% 80 100.0%

Source: Survey of sunflower farmers, small-scale processors and retailers of edible oil.

Importantly, three of the four farmers that were aware of the tariff claimed that it has had no

influence on farm gate prices. Just two of them indicated that the tariff led them to increase

output. The lack of knowledge about the tariff increase suggests that it has had no direct effect

on the output decisions of most farmers and processors (at least among those we surveyed).

This information (or its absence) has not led to changes in the behaviour of the farmers. Of

course, the tariffs may have affected the market price which, in turn, may have led to increases

in the output of farmers. However, the imposition of the tariff appears not to have affected

farmers’ expectations.

Similarly, among the three small-scale processors that were aware of the tariff, none of them

indicated that they increased production because of the tariff. Moreover, none of the producers

believed the price of sunflower oil has increased because of the tariff. This suggests that the

impact of the tariff has been very limited. It also constitutes additional evidence suggesting that

the elasticity of substitution between sunflower oil and palm oil may actually be very low.

Price impacts

In theory, the increase in the tariff applied on crude edible oil imports should raise the price

paid by consumers for palm oil. If sunflower oil is a substitute for palm oil, consumers will

increase demand for sunflower oil in response to a rise in the palm oil price. This should lead

to an increase in the price of sunflower oil that should boost production of sunflower oil and

cake. In turn, an increase in demand for sunflower seed will lead to an increase in the prices

paid to farmers and encourage them to boost production.

This theoretical adjustment needs to be validated empirically to assess whether the tariff had

the expected effect. In reality, the adjustment takes time. Even when prices react quickly, there

will likely be a lag in changes in production decisions. An increase in crushing investments

requires time to mature and even sunflower seed production may not be perfectly synchronised

with the increase in price. For example, if the price occurred immediately after sowing,

production of sunflower seed would not increase until at least one year later.

However, even when the different variables of this adjustment process have adjusted

accordingly, it is necessary to isolate the effect of the tariff from all the different variables that

may affect the system. For example, prices paid to farmers may rise because the effect of the

tariff, but also because of changes in world prices. Exchange rate variations may also affect the

price received in Shillings. Output may change due to changes in weather conditions.

25 25

Our analysis was hamstrung by a lack of availability of detailed and sufficiently disaggregated

price information for the sunflower sector. The available information was insufficient to enable

us to assess the effect of the tariff empirically. More importantly, there is no counterfactual in

the exercise. A control group (e.g. producers in a country not affected by the tariffs) would

need to considered. Alternatively, intervention analysis can be performed using time series

data. In this case, the counterfactual is the situation before the tariff is introduced. To do so,

however, detailed, disaggregated monthly price data for sunflower products is required. It is

necessary to have a sufficient number of observations – in the shape of a long enough time

series – after the intervention.

Nevertheless, our consultations with different stakeholders in the sunflower value chain did

enable us to establish some (albeit limited) information on price trends before and after the

introduction of the 10% tariff. These trends are discussed in the sub-sections below. It is

critically important, however, to reiterate that a host of factors other than the presence of the

tariff may have contributed to observed changes in farm gate, factory or consumer prices. In

most instances, it is not possible given the available information to disentangle the influence

of different factors on the observed prices. In certain instances, however, we could collate the

perceptions of different value chain actors regarding the influence of the tariff on pricing

structures.

Farm gate and market prices for sunflower seeds

Table 13 shows that the market price for sunflower seed increased by 23% from TSh 638 before

the introduction of the tariff to TSh 789 after the tariff was imposed. Over the same period,

farm gate prices increased by less than 7% (Table 14). In turn, the level of demand for

sunflower oil seed increased by 166% at the high estimate (Table 15). However, it is not

possible to attribute these changes unequivocally to the rise in the tariff. As mentioned earlier,

many other factors may affect prices and production.

Table 13: Average market price for sunflower seeds, 2015/16 - 2016/17

Average market price for

sunflower seeds for 2015/16

in TSh/kg

Average market price for

sunflower seeds for 2016/17

in TSh/kg

Mean 638.2727 789.3333

Minimum 350 500

Maximum 950 967 Source: Survey data (sunflower farmers).

Table 14: Average farm gate price for sunflower farmers, 2015/16 - 2016/17

Farm gate prices for

sunflower seeds for

2015/16 in TSh/kg

Farm gate prices for

sunflower seeds for 2016/17

in TSh/kg

Mean 618.2941 661.0769

Minimum 350 300

Maximum 875 967 Source: Survey data (sunflower farmers).

26 26

Table 15: Level of demand for sunflower oil seed, 2015/16 - 2016/17

Level of demand of for

sunflower oil seeds from

main customer in 2015/16

Level of demand of for

sunflower oil seeds

from main customer

in 2016/17

High 29 77.4

Low 35.5 6.5

Don’t Know 35.5 16.1

Total 100 100 Source: Survey data (Sunflower farmers).

Factory prices

Factory prices of sunflower oil – based on information supplied by small-scale processors –

are shown in Table 16. The majority (43%) of the small-scale processors we surveyed sell

sunflower oil at TSh 3,400 per litre; and seedcake at TSh 300 per kg (see Table 17).

Table 16: Factory prices of sunflower oil

Sunflower oil price (Tsh/litre) Frequency Percent

2700 1 3.6%

3000 2 7.1%

3100 2 7.1%

3200 4 14.3%

3400 3 10.7%

4000 1 3.6%

3200 2 7.1%

3400 12 42.9%

3500 1 3.6%

Total 28 100.0% Source: Survey data (Small-scale processors).

Table 17: Factory prices of sunflower seed cake

Sunflower cake price (Tsh/Kg) Frequency Percent

250 1 3.6%

290 1 3.6%

300 3 10.7%

310 1 3.6%

320 1 3.6%

330 1 3.6%

200 1 3.6%

380 1 3.6%

Don’t sell 18 64.3%

Total 28 100.0% Source: Survey data (Small-scale processors).

27 27

For the reasons alluded to earlier, it is not clear from the results of our survey whether or not

the presence of the tariff has driven factory price changes. Indeed, while 96.4% of the 28 small-

scale processors we surveyed indicated that factory prices had changed in the last five years,

these changes may have been driven by inflationary trends and other factors unrelated to the

imposition of the tariff.

Prices for final consumers

Figure 6 shows that about 40% of the consumers we surveyed reported experiencing an

increase in the price of palm oil over the past two years. However, most of consumers

(especially female consumers) indicated they were not aware of the tariff, which is not

surprising because it is not as much of an issue of concern to them as it would be for processors.

In fact, only 12% of the surveyed consumers were aware of the 10% tariff on crude edible oil.

Consumers’ insights into the influence of the tariff on prices in the edible oil sector are thus

limited.

Figure 6: Consumer perceptions of increases in prices of palm oil over the past two years

Source: Survey data (consumers of edible oil).

The survey asked a direct question to retailers to establish the extent to which the tariff on CPO

contributed to changes in the prices of palm oil and sunflower oil. The results are summarised

in Figure 7. They suggest the tariff had a marginally greater influence on palm oil prices,

although the responses are mixed and do not point conclusively to the tariff having any major

impacts on the prices of the two edible oil varieties.

28 28

Figure 7: Retailers’ perceptions of the contribution of the import tariff on CPO to changes in the prices of sunflower oil and palm oil

Source: Survey data (edible oil retailers).

The edible oil retailers we surveyed were also asked to identify the main factors driving

changes in the price of sunflower oil over the past two years. Their responses generally suggest

considerations related to profit margins and the impact of taxes and charges have played a key

role in driving prices changes for sunflower oil (see Table 18). In turn, transport costs and

operational overheads appear to have had a more limited impact on prices in the sub-sector.

Table 18: Retailers’ perceptions of the main factors driving changes in the price of sunflower oil over the past two years

Cost item Average percent of price

Transport costs 13.9

Operational overheads 21.7

Taxes and charges 39.9

Profit margins 46.0 Source: Survey data (edible oil retailers).

Stakeholder views on the impact of the tariff

It is also possible to establish more information on the impact of the tariff through the views of

different stakeholders in the sunflower value chain.

29 29

Large-scale edible oil processors and producers of CPO feel the tariff has undermined

competitiveness

The fact that Tanzania is the only EAC country which has a duty on CPO is of great concern

to the CPO companies operating in the country.8 They believe the tariff gives an added

competitive advantage to their Kenyan counterparts, which is compounded by the less

favourable business environment in Tanzania. Indeed, while BIDCO9 is the largest edible oil

manufacturer in Kenya, Tanzania has not taken advantage of her regional presence to promote

a competitive environment for supporting the growth of the industry. Instead the plant faces

additional costs operating in Tanzania. However, with recent improvements to the political

economy environment in the country, the company has resumed its operations and is optimistic

that it will significantly expand its investment in Tanzania in the hope that the prevailing

business environment will be further improved.

The case of edible oil industry is a useful demonstration of the role of trade policy in promoting

a country’s export competitiveness. Our interviews with large scale processors revealed that

prior to the introduction of the tariff Tanzania was exporting 30% of the processed CPO, with

its largest market being the Democratic Republic of Congo (DRC). However, Tanzania

currently does not export any CPO to the DRC, and the DRC market has been captured by

other countries. In addition to the adverse impact of the tariff, the large-scale processors argue

that Tanzania lost the DRC market due to the implementation of the Single Customs Territory

with the DRC, which made DRC bound imports very expensive.

The adverse impacts on export competitiveness appear to be reflected in the trade data for

Tanzania’s palm-based products. To illustrate this, Table 19 compares the total export values

(in US$) for margarine and bar soap exported from Tanzania in 2015 and 2016. Tanzania’s

margarine exports, which were already small in value terms, dropped to a negligible amount in

2016. Likewise, exports of bar soap products from Tanzania declined by nearly US$ 2 million

(or 11.8%) between 2015 and 2016. Importantly, however, it is not possible using these data to

isolate the impact of the tariff on imported CPO from the influence of other factors that may

explain the decline in the value of exports of these products. We return to the impact of the

tariff on the downstream bar soap manufacturing industry in the next section.

Table 19: Tanzania’s exports of selected palm-based products, 2015 and 2016

Export value (US$)

Product 2015 2016 % change in value

exported 2015-2016

Margarine 12,457 (2014 data) 56 -99.6%

Bar soap 16,797,791 14,818,224 -11.8% Source: UN Comtrade.

8 The sunflower processors object to the CET argument, arguing that the other EAC countries do not grow sunflower and rely mostly on

imports for domestic sunflower consumption, in contrast to the case of Tanzania who enjoys a comparative advantage in sunflower farming. 9 The company had more than 300 full time workers and over 350 casual laborers three years ago but currently has about 129 workers and 80

casual laborers. In Kenya, the firm operates within an integrated strategy dubbed “Farm to Pan” hence controlling a large part of the value

chain. It supports farmers to obtain better yields (e.g. supply of fertilizers and pesticides), thus raising productivity. The company cites Uganda as a good example of a country that has invested notably in palm plantations to support its palm oil industry (we show in Section 4 that a

similar approach has been successful in Indonesia). They refer specifically to the Kalangala Island in Uganda, which was converted into a

CPO production centre for the country.

30 30

Large-scale processors believe the tariff has not been effective in expanding production by

small-scale sunflower oil producers or raising farm gate prices for sunflower farmers

Clearly, the main objective of the tariff was to support the growth of small-scale sunflower oil

processors in Tanzania and raise farm gate prices for farmers. However, large-scale processors

of both sunflower oil and palm oil are equally concerned about the impact of the tariff. Both

camps agree that the tariff has not been effective in meeting these objectives. The large-scale

processors reported that farm gate prices are not affected by the tariff. Instead, they argue the

tariff has only resulted in higher prices for consumers. Murzah Oil estimate the final price to

have increased by a range of 3-5%, whereas the remaining tariff percentage is absorbed by the

company’s profit margin. Thus, according to the company spokesperson, the company suffers

more than the consumers by way of reduced profit margin. In addition, the company reported

a marginal reduction in imports, although exact figures were not readily shared. Furthermore,

due to reduced demand in the market, production has declined and a total of 234 workers have

been retrenched since September 2017. The large-scale processors also believe the tariff has

led to increased cross border smuggling of the product from Kenya to Tanzania (this is

discussed further below).

Industry support organisations have different views on the impact of the tariff

Although industry associations are generally aware of the tariff imposed on CPO in 2016, their

views regarding the impact of the tariff are often not consistent. Our interview with SISUPA

revealed that the association believes the 10% tariff had a positive impact on farm gate prices

– they indicated 1kg of sunflower oil seeds sold after the introduction of the tariff for between

TSh 600 and TSh 1,050, up from between TSh 460 and TSh 770 in the previous year (2015/16).

Conversely, SISUFA argued the farm gate price is determined by supply and demand for

sunflower oil seeds and not the tariff. According to SISUFA, the farm gate price during the

period was between TSh 650 and 900.

According to the Chairman of SUFA, the 10% tariff on CPO imports has not particularly helped

farmers. This is because farmers will not be able to respond to the price incentives on sunflower

oil if the challenges affecting their productivity have not been addressed.

In the case of small-scale processors, conditions for production have improved and many

processors have recorded increased sales, but although the introduction of the tariff might have

contributed, these improvements cannot be exclusively attributed to the tariff. First, as

confirmed by the survey data, the tariff coincided with an increase in the price of palm oil

products, thus reducing the wide gap between the price of sunflower and palm oil such that

consumers at the margin could opt to purchase higher quality sunflower oil instead of palm oil.

Second, the increase in the number of small-scale processors has also contributed to price

competition between the two types of cooking oil. Consumers in the sunflower producing areas

(Singida and Dodoma) buy sunflower from artisanal processors, and some farmers retain a

portion of sunflower oil seeds for their own consumption. Finally, the market for sunflower oil

is expanding owing to its higher quality and healthy reasons.

TASUPA suggests the 10% tariff is too small to realise a dramatic impact in the sector (also

considering the declining price of global prices of CPO). Indeed, cheap imports of palm oil are

considered detrimental for sustained growth of the sunflower industry. Despite its immediate

direct effects, the association is not concerned with other indirect or induced effects of the

31 31

tariff, unlike large-scale processors and refiners who are more concerned with far reaching

competitiveness effects and issues.

TCCIA Dodoma is in favour of increasing the tariff to protect small-scale processors.

According to the Chamber, the introduction of the tariff has been beneficial to these processors

as one can see a lot of competition in the industry (due to the equalisation effect of the tariff).

The Chamber argues that marginal consumers can now opt to purchase sunflower oil as prices

of palm oil have increased. Demand for sunflower oil is also high compared to the past,

although it could reflect increased health awareness rather than the impact of the tariff. The

TCCIA believes that importers need to be mindful of the needs of the country (especially

related to industrialisation and the health of the population). They believe importers do not

bother to invest in value chain development in the sector and argue the government needs to

make imports more expensive. Hence, the Chamber supports increasing the tariff.

Distributors and wholesalers do not want an increase in the tariff

Interviews with a number of distributors and wholesalers suggest that the demand for edible

oil has declined, perhaps owing to a liquidity crunch and potentially as a result of the impact

of increased prices due to the tariff. Although this group of stakeholders do not favour any

particular type of edible oil (sunflower or palm oil), they were clear in suggesting that the GoT

should not raise the tariff. This is because the resulting impact in higher prices will dampen

demand, thereby affecting their turnover and earnings.

Regional and local government authorities believe the tariff is problematic because it raises

consumer prices

According to the Assistant RAS for Dodoma, the imposition of the tariff is problematic because

it raises the price of a product that is consumed in large quantities by the majority of people in

Tanzania, and especially low-income households. It was argued that the 10% tariff has mainly

affected consumer prices, and not farmers or processors. The Assistant RAS argued the GoT

should rather consider other measures to promote productivity, including subsidising farmers

for seeds and fertilizers (a responsibility of the Ministry of Agriculture). It was argued that the

main challenge is on the production side rather than processing. In the Assistant RAS’ view,

there are too many sunflower processors operating in Tanzania, and many have been

established without due consideration of the raw materials required to feed them. Last season,

Dodoma produced only 54,000 tons of sunflower oil seeds – which is below 50% of last year’s

season, while Mt. Meru Millers alone required 200 tons per day. The main problem is said to

be a lack of availability of seeds. Hybrid seeds called Record are not being produced in the

breeding sections, and it is very expensive for farmers to purchase these seeds (at an estimated

cost of TSh 35,000 per kg).

Impact on the downstream bar soap manufacturing industry

The companies that import CPO on a large scale also produce bar soap in significant volumes,

ideally benefiting from easier access to the raw materials. Interestingly, both Murzah Oil and

BIDCO, two major producers of bar soap, confirmed that they face a shortage of raw materials.

In contrast, the standalone soap manufacturers (who are not processors of CPO) complain

bitterly of the increase in the price of the raw materials resulting from supply shortages.

Therefore, it appears that the standalone soap manufacturers face higher prices for the raw

material because of the tariff. These companies also reported that the imposition of the tariff

32 32

was applied to the HS code that also includes raw materials for soap manufacturers. However,

this information is yet to be confirmed by the TRA.

Palm oil remains one of the chief raw materials for production of bar soap in Tanzania. Other

important raw materials include caustic soda. Palm oil raw material is obtained from both the

domestic and overseas markets. Kigoma and Morogoro were mentioned as the main domestic

producers and Indonesia and Malaysia the principal overseas suppliers. To a large extent,

supply depends on the overseas market since domestic production is limited.

Prices of palm oil raw materials increased significantly between 2016 and 2017, largely due to

the introduction of the 10% tariff on the raw material. Table 20 shows that the price increased

by 30% for G&B soap, 55% for Yufesta products and 12% for Huruma Mbezi. The change in

prices of raw materials was reported to have been caused by rising costs of production

attributed to the introduction of the 10% tariff.

Table 20: Price of palm oil raw material by manufacturer (price per litre in TSh)

Year G&B Soap

Industry

YUFESTA

Products

Haruma

Mbezi

Mwasare

Investment

2014 900 350 NA

2015 900 350 NA

2016 900 400 (14%) NA

2017 25% - 30% 1400 (55%) 450 (12.5%) NA

Source: Author interviews in January 2018. Note: NA=Manufacturer did not purchase the raw material.

The inception of the 10% tariff on CPO in 2016 appears to have negatively affected the demand

and availability of the palm oil raw material and, as a result, prices of bar soap increased. This

was clearly explained in the interviews held with bar soap manufacturers in Dar es Salaam:

“I had to cut down demand of RBD raw material by 25 – 30% because I had to part with more

cash capital to import the same amount of palm oil raw material that I did not have…”

[Interview with G&B Soap manufacturer, Dar es Salaam].

On the other hand, the raw material seems to be scarce despite the demand:

“…my concern is scarcity of palm oil raw material. I would buy at a given price but is nowhere

…” [interviews with Mwasare Investment, Dar es Salaam].

The effects of the 10% tariff cut across both informed and uninformed bar soap manufacturers.

Table 21 shows that production volumes declined with the introduction of the tariff, except in

the case of Yufesta products (where production increased significantly after the introduction of

the tariff). This is partly explained by their use of domestic palm oil fat raw materials.

33 33

Table 21: Bar soap production (production per ton), selected manufacturers, 2014-2017

Year G&B Soap

Industry

YUFESTA

Products

Haruma

Mbezi

Mwesare Investment

2014 10 7 10 NA

2015 10 36 10 NA

2016 10 68 6 2

2017 6 85 6 2

Source: Author interview, 2018. Note: NA means no production.

The effect of the 10% tariff on consumer prices for bar soap in Tanzania is mixed. While the

price of family (G&B soap industry) and EDR (Yufesta Products) brands increased by about

5% between 2016 and 2017, the prices of Magadi (Huruma Mbezi) and Mimea (Mwasare

Investment) brands remained stable (see Table 22). The price of EDR has increased steadily

since 2014, however, suggesting that factors other than the 10% tariff have also contributed to

rising bar soap prices.

Table 22: Price of bar soap by manufacturer per bar (TSh), selected manufacturers, 2014-2017

Year G&B Soap

Industry

YUFESTA

Products

Haruma

Mbezi

Mwasare

Investment

2014 750 650 1,350 NA*

2015 750 700 1,350 NA*

2016 750 720 1,350 800

2017 800 750 1,350 800

Source: Author interviews, 2018. Note: NA indicates the manufacturer did not sell the product

The effectiveness of tariff enforcement and challenges related to smuggling

Our interview with a TRA Customs deputy in charge of border posts was useful in

understanding the extent of cross border trade in edible oils, and the effectiveness of customs

control. The interview highlighted the following:

• Edible oil accounts for a relatively small share of transactions through the Namanga Border

to/from Kenya. For instance, between January 2016 and 4 December 2017 there were only 13

transactions. However, a large amount of sunflower seedcake is exported from Tanzania to

Kenya by Mt. Meru Millers. The customs official estimated the volume of seedcake exported

to be in the range of 1,500 tons (50 trucks each transporting 30 tons) per week. Most edible oil

imported from Kenya is Sungold, which is a refined sunflower oil. However, Kimbo, which is

fatty oil, is a relatively dominant edible oil.

• BIDCO is the largest importer of Kimbo brand from Kenya, whereas, for the last year, the

company imported 26 tons. Apparently, BIDCO also produces Kimbo in its plant located in

Dar es Salaam, indicating possible intra-firm trade.

34 34

• The customs officials confirmed that there is a lot smuggling, especially of Kimbo brand, but

the TRA have strengthened border patrols and inspections. An average of 20 smuggled cases

are intercepted and fined per year. For example, two smuggling cases were intercepted in

October 2017 for cargo worth TSh 73 million, paying a total in fines and taxes of TSh 17.8

million. In November, two cases were intercepted worth TSh 14 million (cif) but the fine and

taxes amounted to TSh 19.5 million.

• The customs duty rates applied are the EAC CET rates. Refined sunflower oil is charged 25%

duty if it is outside the EAC, but if it is from within the EAC, is it only subjected to VAT of

18%. There is no export tax on seed cake exports to Kenya, but other charges apply.

• While the customs border post management is doing its best to control smuggling, they are

mindful of their role to facilitate trade rather than being a barrier to trade, thus avoiding

pervasive inspections. So far, the customs posts are supported by four police posts (Ngaramtoni,

Engikareti, Longido and Kimokoa) and suspicious cases are identified by surveillance police,

informal informers and the normal customs border officials.

• Challenges in border control exist, compounded by the long and porous border making it

difficult to implement effective patrols. The border infrastructure is also weak, resulting in lots

of smuggled goods across the border (panya routes). The official recommended the need for

raising awareness of sunflower oil for the industry to achieve its huge potential. Furthermore,

the GoT should seek other policy measures to support the industry rather than just focusing on

the tariff.

Revenue implications

Table 23 compares the customs values of imported palm oil and sunflower oil products in

Tanzania, based on TRA data. The values emphasise the magnitude of CPO imported by

Tanzania, which far exceeds imports of crude sunflower oil as well as imports of downstream

palm oil products in value terms. Notably, the customs value of CPO imports increased

significantly (by 43.6%) between 2015 and 2016, coinciding with the imposition of the 10%

tariff.

Table 23: Customs values for imported palm oil and sunflower oil products (TSh billions), 2015-2017

Customs value (TSh billion)

Product HS code 2015 2016 2017

Crude palm oil 15111000 253.2 363.7 354.8

Crude palm olein 15119010 60.2 195.5 35.4

Palm stearin, fractions 15119020 0.7 2.4 0.0

Palm olein 15119030 10.3 11.9 74.9

Palm stearin, RBD 15119040 5.7 14.1 12.7

Palm oil and its fractions 15119090 5.3 6.3 88.0

Crude sunflower oil 15121100 15.2 8.9 1.7

Sunflower/safflower oil 15121900 0.9 3.2 11.8 Source: TRA data.

The applied MFN tariff rates for the palm oil products listed in Table 23 are compared in Table

24. These tariffs do not apply to palm oil products imported from within the EAC Customs

Union. In addition, CPO imports from Mauritius entered Tanzania duty free in 2016; and

exports of palm crude palm olein, palm stearin and palm oil from Mauritius and South Africa

enjoyed duty free access into Tanzania.

35 35

Table 24: Tanzania’s applied MFN tariff rates, 2015-2017

Product 2015 2016 2017

Crude palm oil 0.00 10.00 10.00

Crude palm olein 10.00 10.00 10.00

Palm stearin, fractions 10.00 10.00 10.00

Palm olein 25.00 25.00 25.00

Palm stearin, RBD 10.00 10.00 10.00

Palm oil and its fractions 25.00 25.00 25.00

Source: UN Trains.

Taken together, the customs values and applied MFN and preferential tariff rates can be used

to estimate the total tariff revenue collected for each palm oil product. The revenue collections

by product (at the HS 8-digit level) are calculated by multiplying the customs values by the

tariff rate on a transaction-by-transaction basis, distinguishing between exporting countries for

which the MFN tariff is applied and those that enjoy preferential access into the Tanzanian

market. The resulting tariff revenue estimates are presented in Table 25. The introduction of

the tariff on CPO imports in 2016 is estimated to have generated TSh 36.3 billion in additional

revenue and a marginally smaller amount in 2017. These figures are substantially larger than

the equivalent tariff revenues collected from imports of other palm oil products in the period

between 2015 and 2017. Based on these estimates, the removal of the 10% tariff would result

in a significant loss in revenue (around TSh 36 billion) for the GoT.

Table 25: Tanzania’s estimated tariff revenue collections (TSh billions) by palm oil product, 2015-2017

Tariff revenue (TSh billions)

2015 2016 2017

Crude palm oil 0.0 36.3 35.5

Crude palm olein 6.0 19.5 3.5

Palm stearin, fractions 0.1 0.2 0.0

Palm olein 0.9 1.4 18.6

Palm stearin, RBD 0.5 1.4 1.3

Palm oil and its fractions 0.2 0.9 20.5 Source: Own calculations based on TRA data

Summarising the evidence on the efficacy of the 10% tariff on imported CPO

Clearly, there remains a great deal of disagreement across different stakeholders in Tanzania

about the merits of maintaining the 10% tariff on CPO. Our analysis suggests there are

arguments both in favour of, and against, the tariff. Table 26 summarises the arguments that

fall on either side of the debate.

36 36

Table 26: Summary of arguments in favour and against maintaining Tanzania’s 10% tariff on imported CPO

Arguments in favour of maintaining the

tariff

Arguments against maintaining the tariff

• In the short term, if there are certain

price points in which low quality,

unrefined sunflower oil substitutes

for palm-based edible oil (there is

some evidence in this study to

suggest this may be the case,

although it is not definitive), this

could drive increased demand and

prices for small-scale sunflower

producers and processors.

• If the tariff raises the price of palm oil

and thus reduces domestic demand

for palm-based products in favour of

sunflower oil, it could produce long-

term health benefits as sunflower oil

is a healthier alternative.

• The tariff is a significant source of

revenue for the GoT.

• The tariff has largely been

ineffective, in part because there

appears to be a low level of

substitutability between sunflower

oil and palm oil (except perhaps at

lower price points in the case of low

quality sunflower oil), but also as a

result of key constraints limiting

productivity (e.g. a lack of high

yield seeds, low processing capacity)

in the sunflower value chain (see

section 3.3). It has thus had little

influence on the output decisions of

farmers and processors.

• The tariff has resulted in higher

prices for consumers of palm-

based edible oil. For instance, it has

increased the price of refined palm oil

in the domestic market, negatively

affecting low income consumers in

particular (but there is no evidence of

cross-price effects to sunflower oil).

• Similarly, the tariff has resulted in

higher prices for domestically

produced non-food products

reliant on CPO as an input. This

means higher prices for domestic

consumers of these products.

• The tariff has adversely impacted

the competitiveness of exported

palm oil and Tanzanian

manufacturers of non-food

products (e.g. bar soap

manufacturers).

3.3 Challenges and constraints to expanding production and processing

Among the large-scale processors, owing to production challenges (including a shortage of

quality seed stock) and an unfavourable business environment, Mt Meru Millers have scaled

down production and closed two of its five plants. As a result, the firm has downsized its

37 37

workforce from 1,000 in 2015 to just 400. The Group CEO noted that if it were to operate at

full capacity, the firm has capacity to create 4,000 jobs and impact over 10 million people

across the entire value chain (from farmers to retailers and their families). Currently the firm is

operating at 10% of the installed capacity. To avoid misleading policy recommendations,

however, it is important to note that the challenges faced by companies such as Mt Meru Milers

are not only attributable to the presence of the tariff on CPO.

A host of challenges impact different stakeholders across the sunflower value chain. According

to SUFA, the major challenges facing farmers include: a lack of high yield seeds; low levels of

knowledge and awareness of better farming practices; inadequate rainfall and a lack of

irrigation farming; a lack of warehouses to store their produce; and a weak voice in policy

debates due to the absence of a common platform to unite farmers (which is being addressed

by SUFA). Many of these challenges were echoed by the farmers we surveyed (see Table 27).

The majority of farmers believe inadequate finance and poor access to capital are the most

important constraints. This was followed by a lack of good quality inputs and unreliable

weather. Whilst the lack of finance and poor inputs are generally well-recognised problems in

many developing countries, there is a recognition that the weather seems to be more unreliable

than what would be expected in this kind of activity. While there is little that can be done to

address the weather and rainfall, there are tools that could be used to minimise or smooth their

effects. For instance, insurance for weather has been developed in many countries, including

developing countries such as Argentina.10

Table 27: Main challenges limiting the production of sunflower seed

Number of farmers identifying the issue as the main

challenge they face

Bad weather/unreliable rainfall 8

Lack of improved farm inputs 8

Inadequate capital and poor access to

capital 9

Lack of reliable market 1

Lack of pesticides 2

Inadequate arable land for farming 2

Inadequate availability of farm

implements 1

Source: Survey of sunflower farmers.

Across the value chain, our interview with the Dodoma Regional Executive Officer for TCCIA

(Chamber of Commerce) highlighted three main challenges affecting the edible oil sector: (a)

tax evasion, with importers claiming to import CPO but in fact importing refined palm oil; (b)

inferior technology; and (c) low quality of palm oil. According to the Assistant RAS for

Dodoma, other challenges affecting the industry include taxes (the producer cess) charged by

local government authorities (TSh 2,00 per 20 litres of oil) and unreliable electricity supply.

In the remainder of this section, we discuss a range of challenges affecting various players in

the sunflower oil value chain in greater detail.

10 See https://www.sancorseguros.com.ar/en/campo-multirriesgos.

38 38

Poor quality seeds

It was established through our consultations with sunflower industry associations that Singida

has a deficit of about 75% of quality sunflower seeds. Farmers rely on recycled seeds and

‘quality declared seeds’ produced in very low quantities. Few farmers are able to access quality

seeds from sunflower processors and private shops. In attempting to address these issues, the

RAS offices encourage the councils to support the already available sunflower seeds

production groups and to continue to establish more groups. They also encourage investors to

invest in production of quality sunflower oil seeds. In response to the efforts, FaidaMali, in

collaboration with AMSDT, plans to support about 16,000 farmers in 2018. In addition, the

region is implanting a sensitisation program so that farmers accept hybrid sunflower oil seeds

in the future.

At present, however, the sunflower processing industry faces serious problems related to a

shortage of seed stock owing to very low productivity among farmers. As a result, according

to the SIDO Business Development Manager in the Dodoma Regional Office, over 50 small

processors have closed down in 2017. Most sunflower processing factories operate for just 3 –

4 months per year, and are closed for the remainder of the year due to a lack of seed stock.

Moreover, while this results in high production costs, prices for sunflower oil are very

competitive. Processors purchase a 70kg bag of sunflower seeds at TSh 50,000 to TSh 60,000,

which produces 20 litres of sunflower oil currently selling at TSh 60,000. This suggests the

premium is either zero or very small, meaning that processors only profit from the production

and sale of seedcakes.

These challenges related to seed quality are confirmed by the results of our survey of sunflower

farmers. Figure 8 shows that only one third of the farmers we surveyed use improved seed

varieties or certified seeds.

Figure 8: Use of improved seeds by surveyed sunflower farmers in Tanzania

Source: Survey data (Sunflower farmers).

The quality of oil seeds can be compromised if there are no proper warehousing facilities to

store harvested seeds. When asked whether there is a central place to collect seeds produced

by local farmers, 93% of the respondents said no and only 7% indicated they have access to a

central collection facility. A lack of availability of warehousing facilities appears to be a major

constraint affecting small-scale processors. As a result, most of them (85%) keep stock of

39 39

sunflower oil seeds for a very short time (less than a month), thereby increasing costs of

production (see Table 28). Furthermore, over 96% of the small-scale processors we surveyed

store the oil seeds using their own on-site warehouses, with only one processor making use of

a rented warehouse off-site. Almost all (96.4%) of the processors we surveyed do not undertake

further refining, clearly illustrating the low level of capacity in the industry.

Table 28: Time taken to stock sunflower seeds

On average, how long do you keep your stock of sunflower oil

seeds?

Frequency Percent

Less than a month 24 85.7

More than a

month 4 14.3

Total 28 100

Source: Survey data (small-scale processors).

According to the Assistant RAS for Dodoma, since smallholder sunflower farmers are not

assured of access to farm inputs and implements, a well-coordinated, structured and subsidised

supply of the inputs would benefit the poorer farmers and facilitate growth of the sunflower

value chain. It was suggested during our interview with the RAS office that the government

should help to bring greater structure to the market, including by establishing market centres,

setting prices and creating standards. In addition, SISUFA suggested the provision of training

and extension services (modern farm practices) to farmers. The Ministry of Agriculture could

be more proactive in availing foundation seed for sustainability and affordability purposes. The

best way to address the problem is for the GoT to invest in research and development into high

yield seed varieties.

Limited access to markets for sunflower seed farmers

The survey data further shows that little, if any, value addition activities are undertaken by

sunflower oil seed farmers post-harvest. Most farmers we surveyed said that the only activities

they undertake after harvesting are packaging and storage (until the seed price rises) (see Figure

9). Over 70% of the customers are individual traders, but also sell to small-scale processors

and consolidators. The most prominent modes of delivery are buyers picking directly from

farms (67%) or from offsite storage (13%).

40 40

Figure 9: Post-harvest value addition activities undertaken by sunflower oil seed farmers

Source: Survey data (sunflower farmers).

There also appears to be a lack of vertical integration in the sector. As shown in Figure 10, the

main customers for sunflower farmers are individual traders (77.4%) which include

middlemen. In a well-integrated industry, one would see a larger share of small processors as

major customers to sunflower farmers. The relationship between small-scale processors and

farmers is critical for enhancing backward integration in the industry. However, only 18% of

the small-scale processors we surveyed appear to have formal relationship with farmers (e.g.

through contractual farming).

Figure 10: Main customers for sunflower farmers

Source: Survey data (sunflower farmers).

The role of middlemen in the marketing of sunflower oil seed is a real challenge inhibiting

sunflower farmers from enjoying the full benefit of the market price. Our survey of sunflower

farmers found that 52% of farmers sell to middlemen. Middlemen often buy crops cheaply

41 41

from farmers while crops are on field unharvested (through a form of informal contract) mainly

due to the financial needs of the farmer. In turn, middlemen sell the crops at higher prices after

the harvest period, thus denying the farmer higher farm gate income. The survey probed the

respondents further to understand whether the farmers know the price middlemen sell the

produce. The responses show that the majority of farmers indeed sell to middlemen (individual

traders), but 83% of them hardly know what price middlemen charge.

Furthermore, farmers do not seek to take their produce to market owing to prohibitive costs of

transport and a lack of market information (see Figure 11). Given the high transport costs and

long distance to markets, buyers pick directly from farms – a situation in which the farmer does

not have much bargaining power.

Figure 11: Factors determining access to markets for sunflower oil seed farmers

Source: Survey data (sunflower farmers).

The above factors play a critical role in determining not only the productivity of sunflower

farmers but, more importantly, the farmers’ earnings. If farmers do not realise adequate

earnings from their farming activities, they will not be able to grow and expand production

and, hence, will be affected by the vicious cycle of low productivity and poverty.

High input costs and low processing capacity, particularly for small-scale sunflower oil processors

As noted earlier, there are only a handful of large-scale processors of sunflower in Tanzania.

Instead, there has been a mushrooming of small-scale processing enterprises, mostly operating

as sole proprietors, responding to the lucrative domestic demand for sunflower oil. Our survey

covered 28 sunflower processors, 23 of them male-owned and only 5 female-owned. The

distribution of the surveyed firms by size is shown in Figure 12. The majority of the processors

we surveyed (57.1%) operate at a small scale, processing less than 2,000 kg of oil seeds per

day.

42 42

Figure 12: Distribution of surveyed sunflower oil processors by size

Source: Survey data (small-scale processors).

A lack of adequate sunflower seed stock was mentioned by 96.4% of the surveyed processors

as constraining their ability to produce at installed capacity. Indeed, Table 29 shows that

capacity utilisation among processors is generally low. Of the 28 processors we surveyed, only

three operate at capacity greater than 81%. More than one-third of the processors we surveyed

operate at 50% of less of their total installed capacity (and 18% at below 30%). As a result,

most of them operate 3 to 4 months in a year and remain idle most of the other months.

Table 29: Capacity utilisation in the sunflower industry

What is your capacity utilisation relative to installed capacity?

Frequency Percent

Below 30% 5 17.9

30% - 50% 5 17.9

51% - 80% 15 53.6

81% - 100% 3 10.7

Total 28 100.0 Source: Survey data (small-scale processors).

All of the processors we surveyed indicated they plan to increase production in the next year.

The capacity utilisation levels outlined in Table 29 suggest that the majority could increase

production by up to 30% (taking the mid-point of the most frequent category). This suggests

that in the short run, capacity may not be an issue to increase production. However, they

indicate that among the most severe constraints they face are a lack of availability of raw seeds

for processing or insufficient quality of the available seed stock. This is very interesting

considering that sunflower seed production has increased significantly in recent years (as

shown in Figure 1) and that it is primarily marketed within Tanzania (rather than being

imported). This may suggest that small-scale processors’ access to quality seed is being

crowded out by larger processors or other users of sunflower seed (e.g. for livestock feeding).

In addition, other major factors constraining the expansion of processing include a lack of

finance, outdated technology and unreliable power supplies (Table 30). Difficulty accessing

finance to increase operating stocks and to finance other variable costs appears to be a major

issue for small-scale processors. Almost all (92%) of the small-scale processors we interviewed

indicated they rely on their own savings as their main source of finance and do not receive

financial support from the government.

43 43

Table 30: Challenges and constraints to expanding production

Challenge/constraints

1=Yes; 2=No

Severity

(1= Least…5 =

Most severe)

Availability of raw materials (seeds

stock) 3 5

Quality of raw materials (seeds) 26 3.07

Outdated technology/machinery 25 3.04

Availability of finance 26 4.07

Reliable power supply 26 3.07

Human resource challenges incl.

availability of skilled labour 0 0

Stiff competition from imported crude

oil and processors 1 2

Unfavorable business environment

(incl. regulations, taxation) 25 3.08

Source: Survey data (small-scale processors).

In the face of such an acute challenge, the survey probed to understand why the processors are

not considering importing sunflower seeds or crude sunflower oil. Emphasising the points

made above, the main reasons cited are a lack of capital and low production capacity (see

Figure 13). Some small-scale processors are satisfied with the locally obtained stock, implying

reluctance to grow.

Figure 13: Reasons for not importing sunflower oil seeds

Source: Survey data (small-scale processors).

The importance of the price of the input (sunflower seed) is crucial for processors. On average,

more than 60% of the cost of production is the cost of the sunflower seeds (see Table 31). This

is followed by the machinery. This suggests that any production decision, in addition to an

increase in the price received, is expected to be explained by the price of the sunflower seed.

None of the processors surveyed import sunflower seed and 76% of the supply of oilseeds is

bought directly from farmers. The rest is sourced through own production and traders. The high

44 44

incidence of sunflower seed and the fact that most of it is sourced in Tanzania suggest that any

policy to increase oil processing needs to take a value chain perspective.

Table 31: Average cost structure of small-scale sunflower oil producers (in %)

Item Mean Min Max

Sunflower seeds 61.47 8 90

Labour 4.47 0 20

Machinery 14.05 0 70

Electricity 8.63 1 20

Taxes 6.26 0 20

Transport 2.92 0 10

Water 1.26 0 6

Other 0.53 0 5 Note: Based on a sample of 19 survey respondents.

Unlocking new investment in sunflower solvent extraction could help to boost sunflower oil

processing volumes in Tanzania. A parallel study undertaken by Dalberg (2018) indicates as

much as 30,000 MT of oil seed cake available nationally is not currently being used for solvent

extraction. Instead, as described earlier, the bulk of the seed cake is exported. The Dalberg

(2018) study argues using solvent extraction to process refined sunflower oil would help to

bring the sunflower oil retail price in line with the price of palm oil imports. At present,

however, among the five major sunflower oil refiners operating in Tanzania, only Mt Meru has

existing capacity for solvent extraction (Dalberg, 2018). New investment in solvent extraction

is required to drive down production costs and make this form of processing more competitive

in Tanzania.

VAT-related issues

Tanzania’s VAT collections on imported sunflower oil (crude and refined) have fluctuated in

recent years (see Figure 14).11 VAT collected on imported crude sunflower oil has declined

significantly since 2015, falling from more than TSh 2.8 billion in 2015 to just TSh 340.3

million in 2017. At the same time, revenue collected from VAT on imported refined sunflower

oil has increased substantially over the past three years, from around TSh 198 million in 2015

to nearly TSh 2.7 billion in 2017. Clearly, the increase in imports of refined sunflower oil is in

response to the increasing demand. When combined, the VAT revenue collected on imported

sunflower oil in Tanzania amounted to marginally more than TSh 3 billion in 2017. In turn,

based on data supplied by TRA, VAT collected on imported sunflower seed cake totalled just

TSh 78,720 in 2017 (up from TSh 25,741 in 2015).

11 The authors submitted a request to TRA to obtain data on VAT collections for domestic sales of products along the sunflower value chain,

but unfortunately no data was forthcoming. The data presented in this subsection thus refers only to VAT on imported sunflower products.

45 45

Figure 14: Tanzania’s total VAT revenue collections (in TSh billions) for imported crude and refined sunflower oil, 2015-2017

Source: TRA data.

Differences in VAT structures between Kenya and Tanzania are affecting the competitiveness

of the Tanzanian sunflower value chain. Kenya imports edible oil but also exports to

neighbouring countries. Tanzania imports CPO (refined mixed with crude – 2%, and is 13-14%

FFA), while seedcake is exported to Kenya. Tanzania also imports animal feeds from Kenya.

In the case of sunflower, VAT is applied on seed cake but not on animal feed. In Kenya, there

is no VAT on input or output.

A key concern raised by TEOSA is that the current VAT structure is not favourable for small-

scale producers. The association recommends the VAT threshold be raised to TSh 900 million

instead of the current level of TSh 100 million.

A lack of policy coherence

While the Sunflower Sector Development Strategy 2016-2020 is in place, Tanzania does not

have a coherent strategy for the broader edible oil sector. One of the key focus areas for such a

strategy would be alignment of trade and industrial policy, balancing the interests of various

stakeholders and beneficiaries across the various edible oil value chains. Understandably, the

price effects of the tariff on consumers cannot be ignored, and the opportunity for promoting a

vibrant sunflower industry should be harnessed. Thus, balancing demand and supply conditions

requires careful, consistent policy making and continuous monitoring.

In our interviews with large-scale processors, they argued the current policy environment does

not lend support to policy consistency. For instance:

• It is not clear why local government authorities impose a crop cess (5%) for sunflower

processors. For example, Mt Meru pays producer cess in five regions where it has branches.

-

1

1

2

2

3

3

4

2015 2016 2017

TS

h b

illi

ons

Crude sunflower oil Sunflower oil

46 46

• Tanzania charges VAT on seed cake (input) while animal feed (the output) is not charged VAT.

In Kenya, there is no VAT on input or output.

• It takes 3 years in Tanzania to approve a seed variety, compared to just 6 months in France.

• The standards agenda for the sector is not clear – while the Tanzania Bureau of Standards allows

the sale of crude oil, it is not allowed by the Tanzanian Food and Drug Authority. Mt Meru

pays millions of Shillings to the radiation commission but is not aware of the benefits from

these payments.

• The Government has removed and imposed the tariff four times since 2000, which is indicative

of the absence of a long-term vision for the industry

Moreover, industry support associations in the sector generally believe past policy support has

not been sufficient. They argue that while the Ministry of Agriculture is mostly effective at

making policies (working closely with ACT and ANSAF), it is less effective at developing

programmes for implementation, except for the ASDP (with World Bank support) that

mentioned a focus on the sunflower sector. For instance, it is not clear if recommendations

often made by TASUPA for the industry have been taken on board by the Ministry of Finance.

In addition, the industry associations doubt the validity of the statistics on sunflower supply

capacity being produced by the Ministry, and recommend the need to scrutinise them.

However, according to TASUPA, the GoT has recently made key improvements in producer

cess (reducing the rate from 5% to 3% and introducing a waiver for crops below one ton). Three

years ago, the Ministry of Industry and Trade established a task force on edible oil, but progress

for implementation is yet to be seen. Sunflower was one of the most selected product under the

One District, One Product (ODOP) initiative, but the status is not well known. Finally, the

various industry associations consider the current industrialisation drive as an opportunity for

the government and other stakeholders to make significant strides in promoting the sunflower

industry.

Capacity constraints in industry support organisations

In general, all the associations and private sector organisations in the sunflower (and wider

edible oil) sector appear to be conversant with the key issues and constraints limiting growth

in the industry. However, although they have an important role12 to play in promoting the

interests of their members for the growth of the sector, the overall picture shows that these

organisations are facing serious capacity deficits. Indeed, the activities and effective

functioning of an organisation depends so much on the person championing it (usually the

Chairperson or often the Director(s)) to the extent that his/her absence has a direct bearing on

the association. At times, some have had a MoU or other frameworks to work with other

organisations but such relationships are largely ad hoc. More importantly, the associations lack

collaboration amongst them, which would be extremely critical for advocating for industry

integration across the value chain.

Each of the organisations appears to have a clear mandate relating to a particular section of the

value chain. However, no assessments have been produced to show their effectiveness, nor

their position in the national structure for policy dialogue (i.e. whether they are members of

TPSF). Our interview with TASUPA indicated that there have been several initiatives to co-

12 TCCIA is positive that industry associations play a useful role, especially TEOSA which helped the formation

of many other associations such as TASUPA, CeSOFA, MbeSOFA and SUFA. However, enterprises have not

fully realised the need and importance of associations.

47 47

opt the association in policy discussions or to provide representation for policy decisions.

However, more generally, there appears to be no strategic relationship with government

ministries (Agriculture, Industry and Trade and Finance).

Aligned to this, one of the main issues is the lack of a unified voice across the value chain. The

different players act in silos and often with conflicting support to one another. Such tendencies

have contributed to a weakening of the voice and lack of action on the part of key policy actors

to address important issues affecting the sector.

48 48

4. Case study evidence of edible oil sector development in other countries

In this chapter, we examine experiences in other countries that have successfully expanded

domestic production of oil seeds and/or facilitated greater levels of domestic value addition

within the edible oil sector. We focus on three countries: (i) the Ukraine, the world’s leading

producer and exporter of sunflower oil; (ii) Argentina, a major global player in soybean and

sunflower production and processing with a high level of export-orientation; and (iii)

Indonesia: the world’s largest producer and exporter of palm oil, and where smallholders have

a major role in domestic production. We pay particular attention to the policy instruments used

in these countries to stimulate domestic production and/or processing of edible oils. The aim is

to draw relevant lessons from these experiences that may be applicable to the Tanzanian

context.

4.1 Export taxes in Ukraine’s sunflower sector

Sunflower is a major part of Ukrainian agriculture and the key component of the country’s

edible oil sector. It is the most profitable crop in the Ukraine and accounts for 70% of all oil

seed crops grown in the country. The Ukraine is a major global producer of sunflower seed and

sunflower oil. The country is the world’s largest sunflower seed producer, contributing around

25% of global sunflower seed output (Polevoy et al., 2013). Further downstream, the Ukraine

produces approximately 5 million metric tons of sunflower oil (based on 2015/16 data) and

accounts for more than half (in the region of 55%) of all global sunflower oil exports

(UKRDELYC Agricultural Group, 2017; USDA, 2017).

Substantial growth in the land area under cultivation for sunflower in the Ukraine since the late

1990s – which doubled in size from 2.5 million hectares in 1995 to 5.1 million hectares in 2012

(Polevoy et al., 2013) – has helped consolidate the country’s position as a major global player

in the sunflower sector. This expansion has been backed by recent increases in productivity,

particularly since 2008.

Most sunflower cultivation in the country is undertaken by large agricultural enterprises. These

enterprises benefit from sizeable financial resources which enable them to make investments

to improve technology, a factor that has contributed to recent productivity improvements

(Polevoy et al., 2013). Other factors, such as significant increases in the use of mineral

fertilisers since the early 2000s have also helped to raise sunflower yields (ibid.).

Nevertheless, while yields in the sector exceed those of other seed crops in the Ukraine, they

remain below potential (Denisenko, 2013). The lower than optimal yields are due to a variety

of factors including limited use of fertilisers, a lack of crop rotation, insufficient farm

technology, low quality seeds and the persistent spread of diseases affecting sunflower crops

(Polevoy et al., 2013).

49 49

Despite these productivity related challenges, the domestic sector does benefit from substantial

capacity for downstream refining of sunflower oil. There are 40 large sunflower oil refineries

operating in the Ukraine, with a combined average capacity to refine between 5.5 and 6 million

tons of processed sunflower oil seeds (UKRDELYC Agricultural Group, 2017). Overall

capacity in the country is further enhanced by the presence of a number of small, independently

operated refineries.

The introduction of export duties on raw sunflower seed exports has played a major role in

encouraging downstream domestic processing activity in the Ukraine. The presence of these

duties has meant the vast majority (estimated at as much as 95%) of domestically produced

sunflower seeds are used in downstream processing plants within the Ukraine. Consequently,

the Ukraine is only a marginal exporter of sunflower seeds.

Stimulating downstream processing through export duties, but at what cost?

These export duties represent an important explanatory factor in the success of Ukraine’s

sunflower oil processing industry. Prior to the introduction of the duties, it was more lucrative

for Ukrainian producers to export their sunflower seeds because domestic processors were

unable to match the prices paid by foreign buyers (Kuhn and Nivievskyi, 2005). Many domestic

processors were using outdated technology, faced high energy costs and were operating at low

levels of efficiency. As a result, the domestic price for sunflower oil was higher than the

prevailing world market price (Kuhn and Nivievskyi, 2005). This rendered domestic processors

uncompetitive on the world stage and production of sunflower oil in Ukraine declined in the

1990s. Instead, much of the sunflower seed produced in the Ukraine was exported in raw form

and crushed by foreign processors outside the country. In the 1998/99 season, which

immediately preceding the imposition of the export tax, around 35% of the sunflower seeds

harvested in the country were exported (FAO, 2002).

An export duty on sunflower seeds was introduced in 1999 and set at 23%. No duties were

imposed on exports of sunflower oil or meal. The tax on raw seed exports was designed to

encourage downstream processing activity within the country by guaranteeing the availability

of raw seed inputs at lower prices. However, loopholes in the administration of the export tax

– which still allowed sunflower seed to be exported duty free under tolling contracts with

foreign crushers – initially undermined its effectiveness (Fry, 2004). The existence of the

loophole meant around 80% of sunflower seeds produced in the Ukraine in the 2000/01 season

were exported duty free under the tolling arrangements (FAO, 2002). In response, the

Ukrainian Government reduced the magnitude of the export duty to 17% in July 2001, but

closed the loophole by removing the special exemption for overseas tolling. This meant the tax

was more effective in limiting exports of raw sunflower seeds. The Ukraine’s commitments

upon accession to the World Trade Organisation in 2008 necessitated a reduction in the size of

the export tax. This saw the duty gradually reduced to its current level of 10%.

The introduction of the export tax was followed by a dramatic decline in exports of sunflower

oil seeds. In 2015, the value of sunflower seeds exported by the Ukraine totalled US$ 20.8

million, accounting for just 0.7% of global sunflower seed exports (UN Comtrade data). This

represented a substantial drop from the more than US$ 201.1 million in sunflower seeds

exported from the Ukraine in 1998, the year immediately preceding the imposition of the export

tax.

50 50

Initially, the export duties were combined with high import duties on oil seeds and significant

protection for downstream processed products. Imported sunflower oil faced a duty of €800

per tonne in the domestic market (FAO, 2002). The Ukraine’s applied MFN duties, which are

levied on both crude sunflower oil and processed sunflower oil, have remained relatively high,

particularly for processed sunflower oil products and sunflower meal (see Table 32). This

contrasts notably with the current tariff regime for sunflower oil imports into Tanzania, where

a 35% duty is applied on processed sunflower oil imports whereas there no tariff is levied on

imports of crude sunflower oil. The high level of protection on crude and processed sunflower

oil has had a major effect in limiting sunflower oil imports into the Ukraine. Notably, however,

palm oil, which is not considered to be in direct competition with sunflower oil in the Ukrainian

market, can be imported duty free. Again, this contrasts with the current tariff regime in

Tanzania, where a 10% tariff is applied to crude palm oil imports.

Table 32: Ukraine’s applied MFN tariffs on sunflower products, selected years

Applied MFN tariff

Year Sunflower oil seed

(HS 120600)

Crude sunflower oil

(HS 151211)

Sunflower oil

(HS 151219)

Sunflower oil-cake

(HS 230630)

2002 0 10 10

2006 10 15 15 20

2007 10 15 15 20

2008 10 15 15 20

2009 6.67 16 20 20

2010 6.67 16 20 20

2011 6.67 16 20 20

2012 6.67 16 20 20

2013 6.67 16 20 20

2014 6.67 16 20 20

2015 6.67 16 20 20

2016 6.67 16 20 20

Source: UN Trains.

Note: Years selected based on the availability of tariff data in the Trains database.

The imposition of the export duties has been credited with a major role in stimulating

significant increases in the crushing capacity of Ukrainian plants. The introduction of the duties

has been accompanied by substantial increases in crushing and sunflower oil production (see

Figure 15) as well as in the production of sunflower meal (Kuhn and Nivievskyi, 2005). The

tax generated a price differential between domestic and international prices for sunflower

seeds. This enabled crushers to acquire seeds domestically at lower prices. At the same time, it

boosted the availability of domestically produced sunflower seed for processing by local

crushers. These factors helped to raise foreign and domestic investment in the production of

sunflower oil and other refined products (e.g. margarine, mayonnaise) (Fry, 2004).

51 51

Figure 15: Ukraine’s sunflower oil production levels (in tonnes), 1998-2014

Source: Own elaboration using FAOStat data.

The result was a decline in exports of sunflower oil seeds, and an increase in exports of

sunflower oil and refined products. Figure 16 illustrates the dramatic growth in the value of

Ukraine’s exports of crude sunflower oil (HS 151211) and other sunflower oil (HS 151219)

after the introduction of the export duty on raw sunflower oil seeds in 1999.

Figure 16: Ukraine’s total exports of sunflower oil (crude and other) to the world, US$ billions, 1998-2015

Source: Own elaboration using UN Comtrade data.

The increased investment facilitated significant modernisation of Ukrainian crushing plants

from 2000 onwards (Polevoy et al., 2013). This helped crushing firms to become more

competitive, with crushing costs falling from the end of the 1990s. Ironically, however, the

improved competitiveness also reduced the rationale for retaining export taxes, particularly

given that crushing costs fell appreciably only a few years after the introduction of the export

duty.

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

To

nnes

(m

illi

ons)

0

1

1

2

2

3

3

4

4

5

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

US

$ b

illi

ons

Crude sunflower oil (HS 151211) Sunflower oil, other (HS 151219)

52 52

At the same time, however, the investment in crushing facilities also resulted in significant

levels of under-utilised capacity. This has been mitigated, in part, by recent moves to construct

multi-functional facilities that can process different types of edible oil (e.g. sunflower, soybean

and rapeseed oil). The flexibility afforded by these facilities enables producers to alternate

production between different varieties depending on market trends (USDA, 2017).

It is important to bear in mind when considering the efficacy of Ukraine’s export tax policy,

particularly in relation to its relevance to the Tanzanian context, that there are contrasting

objectives for different stakeholders at the various levels of the sunflower value chain. While

primary producers want higher prices for domestic sunflower oil seeds, crushers (and other

downstream processors) benefit from lower prices and a steady supply of raw seeds inputs.

Ukrainian sunflower farmers, the primary producers in the value chain, have been adversely

affected by the export tax. This is because the policy resulted in lower farm gate prices. The

initial reduction in farm gate prices after the introduction of the tax was relatively small (around

3%) due to internal competition between Ukrainian producers downstream; but the gap

between these prices and those on the international market for sunflower oil seeds quickly

widened (for instance, farm gate prices in the Ukraine were 25% lower in

September/November 2002). In effect, the presence of the export tax has meant Ukrainian

sunflower farmers have been cross-subsidising the investments of downstream processors in

crushing capacity through the lower prices they receive for their sunflower seeds.

As a result, the export tax has had a significant impact on the structure of the sunflower sector

in the Ukraine. Some smaller seed exporters were forced out of the market (Fry, 2004). In turn,

there has been a consolidation of the sector, with larger crushers taking on increasingly

prominent market shares (ibid.).

4.2 Argentina’s edible oil sector: A major global player in soybean and sunflower

Production of edible oils is a key manufacturing activity in Argentina. The sector accounts for

around 11% of the value added generated by Argentina’s food industry and approximately 2%

of the country’s total manufacturing value added (Romero, 2016). Much of the activity within

the sector is focused within the soybean and sunflower value chains. Both sectors exhibit high

levels of efficiency from primary production through to downstream crushing and processing,

and Argentina ranks among the most competitive oil seed processors in the world (Cetrángolo

et al., 2002).

Argentina is a leading global player in soy bean and sunflower production and processing and

the oil seed sector constitutes a key component of external trade. The country is a major

exporter of sunflower and soybean oil and these segments of the value chain are highly export

oriented, with relatively low levels of domestic consumption. Table 33 shows that Argentina

was the world’s largest exporter of crude soybean oil in 2016, accounting for more than half of

total global exports, and was also the largest exporter of soybean oil-cake (contributing more

than 40% of exports across the world). This is driven by significant crushing capacity within

the country and a strong orientation towards processing and value addition. Argentina also

ranks among the top two global exporters of soybean seed, crude sunflower/safflower oil and

sunflower seed oil-cake, and the third largest exporter of soybeans and soybean oil.

53 53

Table 33: Argentina’s exports of soybean, sunflower and associated downstream processed products in global perspective, 2016

Product Total value of

exports

(USD billion)

Share (%) and rank in

global exports

Soy beans (HS 120100) 3.23 6.3% (3rd)

Soy bean seed (HS 120110) 0.02 13.3% (2nd)

Crude soy bean oil (HS150710) 3.96 51.6% (1st)

Soy bean oil (HS 150790) 0.14 9.9% (3rd)

Soy bean oil cake (HS 230400) 9.97 40.6% (1st)

Sunflower oil seeds (HS 120600) 0.15 4.6% (8th)

Crude sunflower/safflower oil (HS 151211) 0.45 15.5% (2nd)

Sunflower/safflower oil (HS 151219) 0.06 2.1% (11th)

Sunflower seed oil cake (HS 230630) 0.11 13.3% (2nd) Source: Own calculations using UN Comtrade data.

Sunflower has traditionally been a key crop in Argentina and the commercial value of

sunflower oil seeds has been recognised by the Ministry of Agriculture since the early 1940s.

Since then, the development of effective research institutions and mechanisms within the

country, complemented by a conducive regulatory environment and enhancements to the

protection and enforcement of intellectual property rights, has facilitated the invention of

improved seed varieties and hybrids. Alongside the adoption of new technologies, this helped

to raise productivity and spearhead growth in yields and sunflower production in Argentina.

That said, the sunflower sector in Argentina has suffered in recent years, owing in part to the

rising popularity of soybean as well as the adaptability of the sunflower plant to different soils

and growing conditions. Soybean, bolstered by high levels of demand and low production

costs, is increasingly regarded as more competitive than sunflower in Argentina (Castaño,

2017). The production of soy beans, and related processing, now rank among the largest and

most dynamic agricultural and agro-processing activities in the country, while also constituting

an important source of fiscal revenue (Regunaga, 2010). As a result, in recent years many

domestic farmers have replaced sunflower with soy bean crops, relegating sunflower to

marginal areas with less nutrient-rich soil. Nevertheless, recent state interventions, which

included the elimination of export duties on sunflower (discussed further below), have boosted

sunflower acreage and the production of sunflower oil seeds. On the back of these

developments, the sunflower production area in Argentina increased by 36% to 1.7 million

hectares in the 2016/17 season (USDA, 2016). In addition to oil seed production, activity in

Argentina spans the length of the sunflower value chain, from local raw oil production (for

both domestic consumption and export), refined oil production (mostly for domestic

consumption, with some exported to Chile and Paraguay), preparation of meal and hulls, and

the production of biodiesel (six industrial companies and a number of other independent

producers are currently engaged in biodiesel production in Argentina) (Castaño, 2017).

In the remainder of this case study we document some of the key policy drivers, and other

important developments, contributing to the success of the soybean and sunflower value chains

in Argentina.

54 54

Policy factors and other key drivers of development in Argentina’s edible oil sector

In the case of both the sunflower and soybean value chains, significant investment in research

and development (R&D) for the creation of new seed varieties and hybrids has been an

important driver of improvements in yields, production levels and competitiveness. At 2.96

tons per hectare, average seed yields for soybean in Argentina are on par with those achieved

in other leading producer countries (Castaño, 2017). At the same time, the registration of new

sunflower seed cultivars has continued at a rapid rate even amid reduced interest from farmers

in the sunflower crop. The creation of the INTA, a publicly funded agricultural research and

development organisation, in the late 1950s helped drive research into crop genetics and the

development of better technologies. In addition, the development of a seeds biotech industry

in Argentina helped spur R&D into seed inputs adapted to varied agro-ecological conditions

across the country. Expanded private sector involvement in R&D focused on improving seed

inputs has also helped offset gradually waning investment of public resources into these

pursuits.

Notable improvements in intellectual property legislation and protection helped create a

conducive environment for investment in R&D to support the invention of new and improved

seed varieties. This included the development of Seed and Patent Laws to protect intellectual

property rights. The Argentine Seed Association has played a prominent role in promoting the

protection of these rights. Aligned with this, Argentina was also an early adopter of genetically

modified (GM) crops in the mid-1990s, and the introduction of transgenic soybean, for

instance, led to significant expansion of the soybean planted area and major savings on

production costs (Trigo, 2016). This is in sharp contrast with Indonesia, for example, which

historically resisted the use of GM seed for staple foods.

Similar technological and organisational innovations also helped develop dynamic competitive

advantages in the soybean and sunflower value chains. Argentina has pioneered no-tilling

farming, which allows for better moisture retention, less soil erosion, better organic soil content

and lower equipment, manpower and fuel costs, all of which facilitate productivity

improvements in primary production (Bard, 2013). At the crushing and processing stages of

soybean and sunflower oil production, investment from multinational firms focused on

exporting facilitated increases in firm productivity and growth in average plant sizes

(Cetrángolo et al., 2002). There are currently as many as 48 factories actively involved in oil

seed crushing in Argentina, collectively boasting an estimated crushing capacity of 202,831

tons per day (Calzada, 2017). Crushing costs in Argentina are among the lowest in the world

owing to high levels of technological efficiency, marking the crushing industry as one of the

most competitive globally (Regunaga, 2010; Hatum, 2011).

Strategic investments to support oil processing have also enhanced the competitiveness of the

sector. Locating processing facilities close to upstream production zones and ports (crucial for

exporting processed oils to overseas markets) means transaction costs are reduced. Most

soybean in Argentina is produced, for example, within 200-300 kilometres of ports (Regunaga,

2010). Many crushing plants are also located at, or in close proximity to, these ports. Large

investments have been undertaken to expand milling and storage capacity.

Consolidation and vertical integration at the farm and firm level has been instrumental in

achieving economies of scale and reducing oil processing costs. Vertical integration has

enabled strong linkages to producers operating in a modern and competitive input industry

(Regunaga, 2010). Many of the large crushing firms in the country are vertically integrated, in

some cases producing seeds, chemicals and fertilizers. This has led to reductions in costs along

the value chain and improved competitiveness. For soybeans, the use of larger equipment,

55 55

computerised controlling systems, and innovative solutions (such as using plastic bags as

transitory silos) has further enhanced efficiency and competitiveness.

AGD, a producer of both soybean and sunflower oil as well as peanut oil, is a clear example of

a highly vertically integrated Argentinean firm operating within the edible oil sector. The firm

has targeted backward integration by setting up grain collecting plants and purchasing offices

across the country in order to purchase raw oil seeds directly from farmers (Hatum, 2011).

AGD has also actively engaged in the development of supporting transportation and logistics

infrastructure to better link its industrial plants to storage facilities and overseas markets. This

includes, for instance, a 40-year concession to operate the NCA railway, which links the

company’s storage facilities to its industrial plants and port infrastructure.

At the macroeconomic level, key elements of the institutional environment in Argentina aided

the development of the soybean and sunflower sectors from the early 1990s onwards. Currency

free convertibility and sound macro and fiscal policies played an influential role in stabilising

domestic prices after 1991 (Regunaga, 2010). An improved climate for domestic and foreign

investment – spurred by free movement of capital, non-discrimination and the signing of

several bilateral investment treaties with key developed countries – helped attract major

investments in seed production, fertiliser and chemical plants, and in crushing, storing and port

facilities (Regunaga, 2010).

The role of export taxes and import duties in Argentina’s soybean and sunflower value chains

Argentina’s trade policy regime has also influenced dynamics in the soybean and sunflower

sectors, albeit not always in a positive way for all segments of the value chain. Argentina is a

prominent user of export taxes, which it applies on a variety of different products. Typically,

these taxes serve to reduce the domestic price of inputs used in production by raising their

supply in the domestic market (Mendez-Parra et al., 2016). For many years, Argentina’s trade

policy makers have applied differential export taxes in the soybean, sunflower, peanut and

cottonseed value chains with the aim of protecting domestic processors against tariff escalation

(which involves progressively increasing import duties processed or higher value-added

products) in key export markets in developed countries. Specifically, Argentina has applied

higher export taxes on raw oil seeds materials compared to processed oil products in order to

promote domestic processing. The differential export tax compensates downstream producers

facing higher import duties (due to tariff escalation) by taxing inputs (in this case the raw

sunflower oil seeds) relatively more than processed and final products (ibid.). The use of

differential export taxes has been very important in stimulating industrialisation and

downstream value addition within the sector in Argentina.

In the case of soybean, the export taxes are set at 35% on soybeans, 32% on soybean by-

products (oil and meal) and 20% on soy biodiesel. The 3% differential in the export tax enables

crushing firms to purchase soybean domestically at a price cheaper than the Free on Board

export price (Regunaga, 2010). This reduction in the domestic price for inputs has helped

promote domestic processing by indirectly subsidising higher-value-added manufacturing

(Bouet et al., 2014). Indeed, the differential export tax has reduced production costs and

improved the competitiveness of Argentine producers of soybean oil and meal (Costa et al.,

2009). On the other hand, however, it has also affected primary producers of raw soybean (who

receive a lower price for their products in the domestic market and are taxed on products they

wish to export) and resulted in a transfer of income from producers to consumers and the

government (through greater tax revenue collections).

56 56

Differential export taxes have also been applied in Argentina’s sunflower sector in the past –

in the shape of a 32% tax on sunflower oil seed compared to a 30% tax for exported oil.

However, in an important policy change, the export duties on sunflower were eliminated on 16

December 2015, while a 30% export tax on soybeans was retained. This generated incentives

to engage in sunflower production and contributed to an increase in the sunflower area planted

in the 2016/17 season. At the same time, sunflower producers in Argentina have increasingly

focused on niche markets to counter the growing dominance of Ukrainian producers (see

section 3.1). This has involved a strategic shift towards growing sunflower seed varieties

producing high oleic sunflower oil, which has a longer shelf life and is more stable when used

in cooking and for manufacturing foods. In the 2015/16 season, high oleic sunflower accounted

for one-quarter of the total sunflower production area (USDA, 2016).

On the import side, the Argentine Government applies tariffs of varied magnitudes on imports

of primary and processed soybean and sunflower products, with the exception of soybean seed

which can be imported duty free. For comparative purposes, the applied MFN tariffs at the HS

6-digit level in 2000 and 2017 (current) are outlined in Table 34. While tariffs have generally

fallen marginally since 2000 on both raw seed and processed soybean and sunflower oil and

meal products, domestic producers of these products are still protected by tariffs on imported

products. These tariffs are higher on downstream oil products compared to oil seeds, thus

complementing the effect of the export taxes in promoting the development of the domestic

edible oil processing industry. Interestingly, the MFN tariffs currently applied on crude and

non-crude soybean oil and sunflower oil are similar in magnitude to the tariff Tanzania applies

on CPO imports.

Table 34: Argentina’s applied MFN tariffs on imports of primary and processed soybean and sunflower products, 2000 and 2017

Applied MFN tariff

Product 2017 2000

Soy beans (HS 120100) 4.0 5.5

Soy bean seed (HS 120110) 0.0 0.0

Crude soy bean oil (HS150710) 10.0 13.0

Soy bean oil (HS 150790) 11.3 14.33

Soy bean oil cake (HS 230400) 6.0 9.0

Sunflower oil seeds (HS 120600) 4.0 5.5

Crude sunflower/safflower oil (HS 151211) 10.0 13.0

Sunflower/safflower oil (HS 151219) 11.3 14.33

Sunflower seed oil cake (HS 230630) 6.0 9.0 Source: WTO Integrated Database notifications.

Importantly, however, the tariff protection for domestic producers appears to have only had a

marginal influence on the development of the soybean and sunflower value chains in

Argentina. Instead, the range of policy and other factors discussed above – including the

industry consolidation to reach economies of scale; vertical integration, particularly backward

integration from crushing to primary input production; the increasing use of advanced

technologies; R&D into improved seed varieties facilitated by clear protection of intellectual

property rights; and a generally conducive macroeconomic environment – have been central to

the success of the soybean and sunflower sectors in Argentina.

57 57

4.3 Indonesia’s palm oil sector: A world leader with a prominent role for smallholder farmers

Global palm oil consumption has grown rapidly over the past two decades, averaging 7% per

annum. Much of this has been fuelled by strong growth in consumption in Asian countries,

especially China and India. Further growth in global consumption is expected as demand for

palm oil in the commercial processed food industry and in oleo-chemical industries, where it

is used to produce products such as cosmetics, soaps and pharmaceuticals, continues to expand

(Budidarsono et al., 2013). Indonesia, as the world leader in palm oil production, is well-placed

to benefit from this growth.

Indonesia is the world’s largest producer of both CPO and palm kernel oil, a distinction it has

held since 2008. In 2016, Indonesia produced 36 million metric tons of palm oil, contributing

more than 61% of total global palm oil production (Indonesia Investments, 2017). Palm oil

production in Indonesia is supported by a vast network of plantations, covering close to 11

million hectares, or around 40% of Indonesia’s available arable land (Harman Agribusiness,

2016). The planted area in the country has grown rapidly since the 1970s at an average rate of

around 245,000 hectares each year (ibid.).

Unsurprisingly given Indonesia’s stature as a global producer, the country is also the world’s

foremost exporter of palm oil. In 2016, Indonesia’s CPO (HS 151110) exports totalled more

than US$ 3.3 billion, accounting for 47.6% of global exports.13 Indonesia’s exports of other

types of palm oil (HS 151190) were even larger, amounting to US$ 11.1 billion or 54.8% of

global exports.14 In both cases, Indonesia is the leading exporter globally. Figure 17 compares

the annual volumes of Indonesia’s palm oil production and exports since 2008, illustrating the

rapid growth of the sector in recent years.

Figure 17: Indonesia’s annual palm oil production and export levels, 2008-2016

Source: Own elaboration using data from Indonesian Palm Oil Producers Association & Ministry of Agriculture.

The palm oil sector occupies an important role from a socio-economic development perspective

in Indonesia and has long been seen by the government as a key vehicle for rural development.

13 Own calculations based on UN Comtrade data. 14 Ibid.

0

5

10

15

20

25

30

35

2008 2009 2010 2011 2012 2013 2014 2015 2016

To

ns

(mil

lio

n)

Production Exports

58 58

The sector is a major source of jobs and income in rural areas. Estimates suggest the sector

either directly or indirectly supports between 4-6 million economically active people in rural

parts of Indonesia (Dradjat, 2012). A study by Jelsma et al. (2017), based on a sample of

farmers in specific regions in Indonesia, found palm oil was responsible for the majority of

income earned by most farmers in their sample. Notwithstanding palm oil’s specific

contribution to rural development, the sector also makes a major contribution to the national

economy more generally. The sector contributes between 1.5-2.5% of Indonesia’s total gross

domestic product (Indonesia Investments, 2017). It is also an important source of foreign

exchange and employment as well as tax revenue for national and regional governments

(Feintrenie et al., 2010; Dradjat, 2012).

Smallholder farmers play a major role in palm oil production in Indonesia. Smallholders own

or manage as much as 44% of the total oil palm planted area across the country, a share that

has increased steadily over the past two decades (up from 28% in 2000 and 38% in 2010)

(Harman Agribusiness, 2016). In turn, they contribute approximately 35% of Indonesia’s total

CPO production and 34% of overall palm kernel oil production (Daemeter, 2015). Many of

these smallholders derive significant returns from palm oil cultivation (Bronkhorst et al., 2017).

Indeed, the palm oil sector is recognised as an important driver of improvements to rural

livelihoods in many communities in Sumatra and Kalimantan (Rist et al., 2010). Understanding

the dynamics behind the prominent role of smallholders in Indonesia’s palm oil sector is of

great relevance to this study given its emphasis on ensuring smallholders benefit alongside the

broader development of Tanzania’s edible oil sector.

Boosting the involvement of smallholders in Indonesia’s palm oil sector

Oil palms have been cultivated commercially in Indonesia for more than 100 years, dating back

to 1911. The early development of the sector occurred on the back of an existing culture of

plantation development that had been inculcated through the cultivation and processing of latex

rubber. Large-scale plantations were thus already commonplace in Indonesia; and a strong

export-oriented plantation culture already existed on the east coast of Sumatra. Moreover,

plantation agriculture represented a key element of the Indonesian Government’s development

policy, and investment in plantations in areas such as Sumatra’s east coast was regarded as an

important driver of local economic development (Budidarsono et al., 2013).

Strong government emphasis on stimulating oil palm expansion, particularly through support

for the development of plantations, is a key feature of the historical development of the palm

oil sector in Indonesia. In broad terms, the government was able to stimulate investments in oil

palm plantations through direct interventions to provide institutional support, agricultural

extension services and access to land and capital (Dradjat, 2012).

The development of oil palm plantations in Indonesia has occurred in three main stages since

the 1970s. In the first period, spanning the late 1970s until 1994, the New Order Government

favoured government subsidisation and direct state intervention in the plantation system via

direct investments through state-owned enterprises. These investments were largely focused

on facilitating access to land and capital in remote rural areas (McCarthy, 2010). A Nucleus

Estate Smallholder scheme (known as the PIR, the Indonesian acronym for Perkebunan Inti

Rakyat) was introduced as a key operational vehicle for facilitating the inclusion of

smallholders and driving rural socio-economic development during this period. The scheme,

which was derived from contract farming models, involved joint ventures between oil palm

companies and smallholders wherein outgrower arrangements were used to link smallholders

59 59

to state-owned plantation companies. Local land was availed by the state for this purpose (with

smallholders ceding some of their land to an oil palm company) and concessional loans were

provided to plantation companies on the condition that they develop 20% of the plantation area

for smallholders (with the area referred to as ‘plasma’). Importantly, in addition to developing

the plantation area, the state-owned plantation companies supported smallholders by providing

inputs (seedlings), technical assistance and finance (Budidarsono, 2013).

The PIR scheme was implemented in tandem with the government’s transmigration

programme, which involved moving people from densely populated to scarcely populated areas

of the country. This process helped to provide a ready source of labour for new plantations

(especially for workers to engage in crop care and harvesting) (Budidarsono, 2013).

Consequently, the PIR and transmigration schemes collectively enabled large palm oil

companies in Indonesia to access land and low-cost labour. In turn, smallholder farmers

benefited from access to improved seeds and technical assistance from palm oil company

technicians (Feintrenie et al., 2010). Smallholders also received financial support from the

government, including to cover living expenses.

In the late 1980s, the Indonesian Government gradually began to take on a less active role in

the palm oil sector. The gradual withdrawal of state support enabled the emergence of more

commercially-oriented investment in the sector (Bronkhorst et al., 2017). A transitional period

between 1994 and 1998 saw a change in the government’s policy approach to encourage more

private sector initiatives and attract foreign direct investment into the sector. Specific actions

undertaken by the government to encourage private sector involvement included the provision

of credit at concessionary rates for developing oil palm estates, planting new crops or erecting

crushing facilities (backed by rate subsidies for executing banks) (Dradjat, 2012). As a result,

the private sector increasingly took on the responsibility of developing plantations for

smallholders (Jelsma et al., 2017). For instance, under the Primary Cooperative Credit for

Members (or KKPA) scheme, the responsibility for resolving land issues, providing training

and extension services to plasma cooperatives and establishing infrastructure gradually shifted

to the large plantation companies without direct state involvement (McCarthy, 2010). The

decentralised governance system grouped farmers into cooperatives that coordinated

smallholder farmers and were responsible for credit and infrastructure management and

occasionally also the management of plantations (Daemeter, 2015; Jelsma et al., 2017).

The various incarnations of the PIR and KKPA schemes are generally regarded as being

successful in reducing poverty in Indonesia. In support of this argument, some analysts point

to the significantly lower poverty rates in PIR project areas (2-7%) compared with the national

poverty level (14%) (Drajat, 2010).

The schemes have also been credited with aiding the development of management capacity

within smallholder palm oil producers (Jelsma et al., 2017). They also helped integrate

smallholders into large oil palm plantation areas. Moreover, the schemes played a critical role

in creating a cohort of small-scale farmers operating with improved technology and knowledge

for oil palm cultivation and who were more effectively integrated into local commodity markets

(McCarthy, 2010). The schemes also provided a pathway to ease smallholders’ access to

finance and investment capital. Smallholders could obtain land certificates after paying off their

loans, and were able to use these certificates as collateral when seeking to borrow money from

local banks (Budidarsono, 2013).

That said, the development of Indonesia’s smallholder palm oil farmers did occur in an uneven

manner, with varied impacts from the schemes observed from village to village. Some farmers

sold their land to oil palm companies rather than developing their smallholdings, thereby

60 60

losing, or significantly reducing, an important source of agricultural income (Rist et al., 2010).

Others were overly focused on the short term, planting oil palm and then selling their

plantations before reaching the production stage (ibid.). High planting and management costs

also meant some farmers faced mounting levels of debt, particularly early on and during times

of declining CPO prices (ibid.).

The most recent stage of palm oil development in Indonesia (in operation since 1998 and

frequently described as the laissez faire period) has seen a shift to a market-driven model. This

has been characterised by further decentralisation and the development of public-private

partnerships. A feature of the new model has been partnerships between existing estates and

large, capital-intensive companies who are willing to invest in labour-intensive palm oil

projects (Zen et al., 2006). The role of state-owned plantation companies – at least when

measured in terms of the percentage of the total planted area they hold – has declined

considerably as the presence of the private sector and smallholders has grown (Hawkins et al.,

2016). Even so, the emphasis on community development has been maintained. Companies

applying for a plantation business license are required to provide 20% of the total plantation

area to the local community. These ‘plasma’ plantations are transferred to small community

landholders, who operate the plantations under their supervision (ibid.).

Despite the influence of these schemes, which have had a notable impact in boosting the

presence of small-scale farmers in the sector, many smallholders still face numerous

challenges. High costs to secure land and for planting have constrained many smallholders

from expanding their acreage under production (Jelsma et al., 2017). Productivity levels among

Indonesian smallholders are also relatively low. In the case of CPO, past estimates of yields

for smallholders suggest they have been around 35-40% lower than those achieved by private-

or government-owned plantations (Teoh, 2010). More recent estimates suggest that

productivity levels of smallholder plantations are currently between one third and 50% of those

of large corporations (Jakarta Post, 2017).

Experimenting with export taxes to boost downstream value addition

Indonesia’s experience with the use of export taxes in the palm oil sector is also of interest to

the Tanzanian context. Rather than focusing on tariffs on imported CPO, the Indonesian

Government has emphasised the use of export taxes on domestically produced CPO.

Indonesia’s application of a differential export tax on palm oil was intended to encourage

exports of refined rather than crude palm oil (Deese and Reeder, 2007). These tariffs were

structured with a view to encouraging Indonesian producers to add value to locally produced

CPO, and thereby develop the downstream palm oil industry (Hawkins et al., 2016).

Importantly, the taxes also aim to guarantee the availability of domestic CPO, a key raw

material in the production of cooking oil (which is a key staple in Indonesia) (Rifin, 2014). To

this end, Indonesia has imposed an export tax on CPO and its derivatives since September

1994.

This has been implemented as a progressive export tax since September 2007. The magnitude

of the tax is determined by a reference price set by the Ministry of Trade based on the prevailing

international price of CPO in Rotterdam (Rifin, 2014). It is set at zero when the calculated

reference price is below US$ 750 per metric ton. This rises to US$ 3 per metric ton when the

reference price falls within the range of US$ 750-800 per metric ton, and to US$ 18 per metric

ton applied when it ranges between US$ 800-850 per metric ton. The tax rate increases

progressively at higher thresholds (e.g. US$ 52 per metric ton when the reference price falls

61 61

between US$ 900-950 or US$ 74 at US$ 950-1000). It is anticipated that the higher export tax

will encourage planters and domestic CPO producers to sell greater volumes domestically

rather than in the export market. This policy is in line with a shift in government approach

towards greater prioritisation of downstream, refined products.

To compensate for a loss of revenue when the tax is set at zero, the Indonesian Government

imposes additional export levies. These are currently set at US$ 50 per metric ton for CPO

exports and US$ 30 per metric ton for exports of processed palm oil products (Indonesian

Investments, 2017). A portion of the proceeds from these levies are used to fund the

government’s biodiesel subsidy programme as well as replanting operations and R&D in the

palm oil industry (ibid.).

The available evidence suggests the introduction of the export tax has led to a surge in refining

capacity. According to Indonesian Investments (2017), Indonesia’s total palm oil refining

capacity increased from 23.1 million tons in 2012 to 30.7 million tons in 2013 and 45 million

tons in 2015. This has been spurred, in part, by sizeable investments from large Indonesian

companies. As a result, Indonesia now boasts capacity for refining palm oil that is well in

excess of existing annual CPO production levels.

In other areas, the export tax has harmed local producers in the upstream segments of the palm

oil value chain. Previous analysis indicates the export tax reduced the domestic CPO price,

thereby transferring income from oil palm growers to consumers (Larson, 1996). Other studies

have shown that the export tax reduced investment, production and exports (Susila, 2004;

Obado et al., 2009), and undermined competitiveness (Rifin, 2010 and 2014).

62 62

5. Conclusions

The existing level of domestic edible oil production in Tanzania falls well short of prevailing

demand, with the gap currently met by imported varieties. Mindful of this, the GoT is eager to

reduce Tanzania’s dependence on imported edible oil by boosting domestic production and

processing capacity to ensure more oil seeds and downstream edible oil products are produced

locally. To support this objective, in 2016 the GoT opted to stay the application of the EAC’s

CET on crude edible oil imports and instead apply a 10% duty rate. It was anticipated that the

tariff would incentivise refiners to utilise more domestically produced crude oil, raising

demand for domestic edible oil products and increasing the prices of domestically produced oil

seeds, and thereby benefiting smallholders as well as other oil seed producers. In this study, we

have attempted to investigate the implications of this decision on prices, production and

processing in Tanzania’s edible oil sector – focusing on the sunflower value chain – and

examine how it has impacted different actors within the value chain.

A lack of detailed and disaggregated time series data on sunflower products in Tanzania has

precluded a detailed empirical analysis of the tariff impacts. Instead, we have drawn primarily

on information and data gathered through consultations with government policy makers,

industry associations and private sector support associations, and key actors in the sunflower

value chain (including large-scale processors and refiners), together with surveys of sunflower

farmers, small-scale processors, retailers and consumers, to understand the tariff impacts and

other factors influencing production and processing within the sunflower value chain.

At the macro-level, production of sunflower in Tanzania has grown substantially in the last five

years. However, despite considerable recent expansion in sunflower oil production and some

growth in exports of sunflower products (mostly seed cake), domestic oil seed crushing

capacity remains low. There are few large-scale processors operating in the sector, and the

many small-scale processors generally face a range of capacity constraints owing to a reliance

on poor quality or outdated technology and machinery and constraints in accessing finance to

boost investment, among other challenges. As a result, domestic production of sunflower and

other edible oils has not been able to meet even half of Tanzania’s existing edible oil demand.

Instead, in the absence of sizeable exports, much of the surplus sunflower seed appears to be

consumed domestically without undergoing further processing.

Our research suggests a number of challenges and constraints currently hamper prospects for

expanding domestic production and processing in the sunflower sector, which will not be

addressed by a tariff on CPO. Poor seed quality, exacerbated by low farmer productivity,

limited access to markets for farmers, a lack of warehousing facilities and limited vertical

integration in the sector, are major reasons for the shortages in quality seed stock experienced

by downstream processors. These processors, the majority of which operate at a small scale,

also face high input costs and low processing capacity, the latter owing to a range of factors

including a lack of finance for capital investment, outdated technology and unreliable power

supply. The uneven application of VAT along the sunflower value chain is also said to

disadvantage small-scale processors. This is one element of a more general lack of policy

coherence across the sector.

Our engagements with stakeholders and actors in the sunflower value chain revealed a range

of contrasting views on the extent to which the 10% tariff on imported CPO can effectively

63 63

support the development of the domestic sunflower sector in the face of these challenges. While

certain industry associations and private sector organisations such as TASUPA and the TCCIA

in Dodoma support the tariff, mainly due to its perceived benefits for small-scale processors,

large-scale processors as well as distributors and wholesalers want the tariff reduced or

eliminated.

Remarkably, the majority of the sunflower farmers and small-scale processors we surveyed

were not even aware of the tariff. Of the few farmers that did know of the tariff’s existence,

most claimed that it has had no influence on farm gate prices and it appears to have had little

effect on farmers’ expectations and output decisions. Similarly, none of the very low number

of small-scale processors that were aware of the tariff believed that the price of sunflower oil

had increased as a result of its introduction. This suggests that the impact of the tariff has been

very limited on these segments of the value chain.

Furthermore, while the evidence we gathered on farm gate, market, factory and consumer

prices for sunflower oil seeds, sunflower oil and other sunflower products suggests these prices

increased between the periods immediately before and after the tariff was introduced, there is

little clear evidence to suggest the presence of the tariff was a major factor driving these price

changes. There is thus no definitive evidence, for example, to suggest the tariff has helped to

increase the prices of domestically produced oil seeds, one of its intended objectives.

On balance, the available evidence on the positive impacts of the tariff for sunflower farmers

and small-scale sunflower oil processors appears to be limited, largely because of the existence

of key constraints limiting productivity and the apparent limits to the extent to which sunflower

oil and palm oil are substitutable. Uncertainty about the level of substitutability between

sunflower oil and palm oil raises doubts about whether continuing to impose a tariff on

imported CPO will have the intended effects on production and processing in the sunflower

sector. While our cross-price elasticity estimates suggest a high degree of substitutability

between sunflower oil and palm oil, the very small sample upon which these estimates are

based implies the result should be interpreted with caution. Moreover, our survey of consumers

generally revealed a preference for sunflower oil over palm oil, although income levels and

education appear to be important determinants of consumers’ preferences. The more highly

educated consumers in our sample were motivated more by quality and health factors,

compared to less educated customers who were influenced more by price considerations.

Consumers with a lower level of education had higher incidences of consuming palm oil

compared to consumers with higher levels of education, most of whom consume only

sunflower oil. These factors reduce the extent to which sunflower oil and palm oil are

substitutable by consumers. At the same time, most small-scale sunflower processors do not

consider producers of other edible oils as direct competitors, also pointing to a lower degree of

substitutability between sunflower oil and palm oil than perhaps has traditionally been

envisaged.

At the same time, the tariff appears to have had neutral, or even negative, impacts at a number

of different levels in the sunflower value chain. Large-scale processors believe the tariff has

not been effective in expanding production by small-scale sunflower oil producers or raising

farm gate prices for sunflower farmers. Certain industry associations, such as SUFA, also argue

the tariff has not particularly helped farmers because they are unable to adequately respond to

price advantage it affords to sunflower oil due to the myriad productivity challenges they face.

In addition, large-scale sunflower and CPO processors both feel the tariff has undermined

Tanzania’s competitiveness in downstream segments of the value chain. Our study finds

evidence that the tariff on CPO led to an increase in the price of refined palm oil, thereby

64 64

adversely affecting low income consumers. For this reason, regional and local government

authorities believe the tariff is problematic because it raises consumer prices. Furthermore, our

engagements with Tanzanian bar soap manufacturers revealed the tariff has raised the price of

raw materials significantly and generated supply shortages to the detriment of the industry.

Clearly, there remains a great deal of disagreement across different stakeholders in Tanzania,

both within the sunflower value chain and more broadly across the edible oils sector, about the

merits of maintaining the 10% tariff on CPO. Our analysis suggests there are arguments both

in favour of, and against, the tariff. On one hand, the tariff generates a significant amount of

revenue for the GoT (we estimate total revenue collected through the duty on CPO at around

TSh 36 million each year). Moreover, parallel research by Dalberg (2018) suggests investment

in solvent extraction for processing refined sunflower oil could help bring the sunflower oil

retail price down to a similar level to the price of imported palm oil. In this context, the presence

of the tariff could, in theory at least, help provide incentives for large-scale investments in

solvent extraction for the production of refined sunflower oil, which would help drive down

production costs and potentially enable domestically produced sunflower oil to compete on

price with palm-based products. In the short-term, the tariff could also drive increased demand

and prices for small-scale sunflower producers and processors at certain price points in which

unrefined sunflower oil is a closer substitute for palm-based edible oil. However, the evidence

collected through our study is inconclusive, but does indicate there is quite wide variation in

sunflower oil prices and, at lower points in the price distribution, sunflower oil may substitute

more closely for palm oil. Also in relation to prices, by raising the price of refined palm oil in

the domestic market, the tariff on CPO may reduce domestic demand for palm-based oil in

favour of sunflower oil, thereby producing long-term health benefits as sunflower oil is a

healthier alternative. We have not, however, found convincing evidence of these demand, price

and substitution effects in this study, although the extent to which we are able to investigate

them is hampered by the availability of suitably disaggregated quantitative data on pricing.

Instead, as indicated earlier, the available evidence suggests the tariff on imported CPO has

largely been ineffective in achieving its intended outcomes. At the same time, the tariff has

also had some clearly negative impacts. It has resulted in higher prices for consumers of palm-

based edible oil, with the higher price for refined palm oil in the domestic market particularly

negative for low income consumers, who are more likely to consume palm oil. It has also raised

the prices of domestically produced non-food products reliant on CPO as an input, which again

impacts negatively on Tanzanian consumers and also undermines the competitiveness of

Tanzanian manufacturers of these products (e.g. bar soap manufacturers) as well as Tanzania’s

exports of palm oil and palm-based products.

6. Recommendations

There is clearly a need for further dialogue among stakeholders in the edible oil sector to

inform the longer-term direction of Tanzania’s tariff policy for the sector and build

consensus around reform priorities. Even with the evidence presented in this study, and

considering the broader arguments for and against the tariff, it is difficult to settle on a clear

view given the information at hand. A dialogue process among key stakeholders could help to

build consensus around reform priorities and their sequencing.

65 65

While further dialogue takes place, the tariff on imported CPO should be retained at

10%, but with a clear commitment to ensuring it remains in place only for a limited, time-

bound period (ideally to be phased out within three years). This would create some certainly

about the GoT’s short-term tariff policy approach for the sector and allow more time to see if

the tariff on CPO imports can spur large-scale investment in domestic refined sunflower

production (e.g. through solvent extraction) to compete with palm-based products. Such

investment could ultimately boost investment and growth further upstream, including for

sunflower farmers and small-scale processors. The wider dialogue process could also be

accompanied by a detailed analysis of whether an import tariff on crude sunflower oil, again

in place for a time-bound period, may be more effective in supporting investment in

downstream processing in Tanzania’s sunflower value chain.

Encouragingly, there appears to be a strong level of willingness within the domestic sunflower

value chain to expand production. Our consultations with large-scale processors such as

Murzah Oil and Mt Meru indicate they are willing to substantially expand investment along

the value chain and to promote backward integration. Murzah Oil is keen to support sunflower

farmers. Mt Meru aims to reach a capacity to process 3 million tons of sunflower per year.

Similarly, there appears to be an appetite among small-scale sunflower oil processors in

Tanzania to expand production. Across the group of sampled small-scale processors, 96% are

planning to increase production. Importantly, over 80% of the small-scale processors we

surveyed plan to increase production in the immediate future, within a timeframe of one year.

These processors are mostly operating at low levels of capacity utilisation and appear to have

significant excess capacity that could be used to raise processing volumes.

A wider array of policy interventions to address critical constraints, rather than a narrow

focus on tariff policy, are necessary to spur investment in sunflower production and

processing. The case studies of different edible oil subsectors in Argentina, Indonesia and the

Ukraine illustrate how a bouquet of policies aimed at developing these sectors, rather than a

narrow focus on tariff policy, has been effective. While some of the policy measures used in

these countries, such as export taxes on raw materials (seeds) as a mechanism to stimulate the

supply of raw edible oil seeds for downstream processing are unlikely to be effective in

Tanzania given that the challenge appears to be more related to the quality rather than the

quantity of sunflower seeds, and given that very small volumes of sunflower seeds are currently

exported, others such as the approach to improving seed varieties in Argentina’s soybean and

sunflower sectors or the support given to smallholders in Indonesia’s palm oil sector (which

included explicit support for the development of plantations and direct interventions to provide

institutional support, agricultural extension services and access to land and capital) provide

pertinent lessons for the Tanzanian context.

Bearing these lessons in mind, and on the basis of our findings, the following interventions

would help to address critical constraints to the expansion of sunflower production and

processing volumes and promote the long-term development of the sunflower sector in

Tanzania:

• Consider VAT exemption on sunflower oil and seed cake. A detailed study should

be undertaken to assess the merits and revenue impacts of such a policy. On the face of

it, this would likely have only limited impacts on domestic VAT revenue collection,

since VAT revenues constitute less than 10% of the corporation tax of large-scale oil

processors (although no data was provided by TRA to assess the impact of VAT

exemption on domestic sales of sunflower oil and seed cake). However, the revenue

collected through VAT on sunflower oil imports is substantial. Based on 2017 VAT

66 66

collection data supplied by TRA, VAT exemption on crude and refined sunflower oil

would result in a loss of around TSh 3 billion. At the same time, however, VAT

exemption at the oil and seed cake stages of the value chain would benefit small-scale

processors and provide incentives for large-scale manufacturing.

• Improve the availability of high quality, high oil bearing sunflower seed varieties

domestically. Clearly, The Ministry of Agriculture could be more proactive in availing

foundation seed for sustainability and affordability purposes. Estimates reported in a

parallel study by Dalberg (2018) indicate new, high yield seeds could increase yields

by three times the current production volumes. This will help to raise the productivity

of sunflower farmers and hence increase the supply of better quality raw materials to

downstream processors. In the short term, support should be provided for

importing hybrid seeds. In slower time, the provision of incentives for seed

importers to invest in local production will help boost the availability of high quality

seeds in the medium term. The longer-term focus should be on supporting seed

research, breeding and multiplication, for instance by improving the capacity of

agricultural research institutes in Tanzania and encouraging higher levels of investment

in R&D into improved seed varieties adapted for local conditions.

• Invest in warehousing infrastructure to improve access to storage facilities for oil

seeds. These investments could be supported through public-private partnerships or

capital from development finance institutions.

• Unlock access to finance and investment capital to boost domestic crushing

capacity (e.g. by improving access to finance for investment in improved technology

and machinery).

• Eliminate producer cess on sunflower oil and cake. This should form part of wider

discussions to eliminate nuisance taxes, levies and fees on various agricultural value

chains.

• Develop a coherent strategy for the broader edible oil sector, drawing on

Tanzania’s existing Sunflower Sector Development Strategy 2016-2020. The

strategy should take a wholistic value chain approach to the development of the sector.

• Strengthen the capacity of industry support organisations to engage more

effectively in national structures for policy dialogue. This should occur alongside

efforts to strengthen the strategic relationship between industry support organisations

and relevant GoT ministries, and improve links between specific edible oil industry

support associations and Tanzania’s broader private sector associations.

67 67

References

Ahmad, T., Ariff, T.M. and Tawang, A. (1999) Effects of trade liberalization on agriculture in

Malaysia: commodity aspects (No. 32694). United Nations Centre for Alleviation of Poverty Through

Secondary Crops' Development in Asia and the Pacific (CAPSA).

Bard, D. (2013) ‘Sustainable Practices in Argentina’s Confection Sunflower Production’, International

Food Trader, 22 May 2013.

Bouet, A., Estrades, C. and Laborde, D. (2014) ‘Differential Export Taxes along the Oilseeds Value

Chain: A Partial Equilibrium Analysis’, American Journal of Agricultural Economics, 96 (3): 924-938.

Bronkhorst, E., Cavallo, E., van Dorth tot Medler, M., Klinghammer, S., Smit, H.H., Gijsenbergh, A.

and van der Laan, C. (2017) ‘Current practices and innovations in smallholder palm oil finance in

Indonesia and Malaysia’, CIFOR Infobrief No. 186, August 2017.

Budidarsono, S., Susanti, A. and Zoomers, A. (2013) ‘Oil Palm Plantations in Indonesia: The

Implications for Migration, Settlement/Resettlement and Local Economic Development’, Chapter 6 in

Z. Fang (ed.) Biofuels – Economy, Environment and Sustainability.

Calzada, J. (2017) ‘6th in the world, Argentina produces more edible oil than Brazil’, Noticias Agricolas

(Agricultural News), 6 June 2017, Available from: http://noticiasagricolas.com.ar/?p=8529.

Cassiday, L. (2017) ‘China’s evolving edible oils industry’, Available from: https://www.aocs.org/stay-

informed/read-inform/featured-articles/chinas-evolving-edible-oils-industry-april-2017.

Castaño, F.D. (2017) ‘The sunflower crop in Argentina: past, present and potential future’, Oilseeds

and fats, Crops and Lipids, Topical issue on ‘The Oil and Protein Crop Supply Chain in South America.

Cetrángolo, H., Carlevaro, M. and Fernández, S. (2002) ‘Limitations for competitiveness in Argentinian

sunflower agrifood chain’, New Medit no. 2/2002.

Chand, R. (2002) ‘Trade Liberalization, WTO and Indian Agriculture’, New Delhi: Mittal Publications.

Chand, R., Jha, D. and Mittal, S. (2004) ‘WTO and Oilseeds Sector: Challenges of Trade Liberalization,

Economic and Political Weekly, February 7, 2004: 533-537.

Coad, A. and Tamvada, J.P. (2012) ‘Firm growth and barriers to growth among small firms in India’,

Small Business Economics, 39(2): 383–400.

Costa, R., Xia, Y., Susanto, D., Rosson, P. and Adcock, F. (2009) ‘Analyzing the Impact of Changes in

Trade and Domestic Policies: The Case of the Soybean Complex’, Paper prepared for presentation at

the Southern Agricultural Economics Association Annual Meeting, Atlanta, Georgia, 31 January – 3

February 2009.

Daemeter (2015) ‘Overview of Indonesian Oil Palm Smallholder Farmers: A Typology of

Organizational Models, Needs, and Investment Opportunities’, November 2015.

Dalberg (2018) ‘Industrialisation and Edible Oils’, Working Draft Presentation for the AAPC Special

Session – Edible Oils Sector in Tanzania, 14 February 2018.

Deese, W. and Reeder, J. (2007) ‘Export Taxes on Agricultural Products: Recent History and Economic

Modeling of Soybean Export Taxes in Argentina’, United States International Trade Commission

Journal of International Commerce and Economics, September 2007.

68 68

Denisenko, L. (2013) ‘The Analysis of World Dynamics of Sunflower Market’, Kharkov Institute of

Physics and Technology, The Journal of National Scientific Centre, 137.

Dohlman, E., Persaud, S.C. and Landes, M.R. (2003) ‘India’ Edible Oil Sector: Imports Fill Rising

Demand’, OCS090301, United States Department of Agriculture, Economic Research Service.

Dradjat, B. (2012) ‘Structure, Roles, Challenges and Opportunities of the Oil Palm Industry in

Indonesia: The Significance of Oil Palm Smallholders’, Oil Palm Bulletin, 64 (May 2012): 1-12.

Egwuma, H., Shamsudin, M.N., Mohamed, Z., Kamarulzaman, N.H. and kai Seng Wong, K. (2016)

‘Investigating the effects of import tariff liberalization on Nigerian Palm Oil Industry’, ARPN Journal

of Agricultural and Biological Science.

Ekama Development Foundation (2015). ‘Gender analysis of Irish Aid Value chain projects in

Tanzania’.

Elbehri, A., Hoffman, L., Ash, M. and Dohlman, E. (2001) ‘Global impacts of zero-for-zero trade policy

in the world oilseed market: a quantitative assessment’, In Proceedings of the 4th Conference on Global

Economic Analysis, Purdue University, West Lafayette, Indiana.

FAO (Food and Agricultural Organisation of the United Nations) (2002) ‘Ukraine: Review of the

Sunflower Oil Sector’, Report No. 02/074 EBRD-UKR.

Feintrenie, L., Chong, W.K. and Levang, P. (2010) ‘Why do farmers prefer oil palm? Lessons learnt

from Bungo district, Indonesia’, Small-scale Forestry, 9 (3): 379-396.

Fry, J. (2004) ‘The Sunflowerseed Export Tax in Ukraine’, Presentation, LMC International, Oxford,

UK, June 2004.

Ghosh, N. (2009) ‘Effects of tariff liberalization on oilseed and edible oil sector in India: Who wins and

who loses?’.

Gulati, A. and Mullen, K. (2003) ‘Responding to Policy Reform: Indian Agriculture in the 1990s and

After’, Stanford Centre for International Development, Working Paper No.189, Stanford University,

September.

Hashim, D.A (2008) ‘Indian Oilseed Products Industry: Impact of Tariff Reforms on Sources of Output

Growth’, Commodity Vision, 1(4): 86-96.

Hatum, A. (2011) ‘The transformation process of AGD, Argentina’, January 2011.

Hawkins, D., Chen, Y. and Wigglesworth, T. (2016) ‘Indonesian Palm Oil Production Sector: A Wave

of Consolidation To Come’, Hardman Agribusiness, 25 October 2016.

Indonesia Investments (2017) ‘Palm Oil’, 26 June 2017, Available from: https://www.indonesia-

investments.com/business/commodities/palm-oil/item166.

Iringo, R., Elias, P. and Majid, S. (2014) ‘CEZOSOPA –Processors Study Report 2014’, Accessed at

http://lic.or.tz/wp/.

Jakarta Post (2017) ‘Indonesia to increase palm oil production to 42 million tons by 2020’, 8 September

2017, Available from: http://www.thejakartapost.com/news/2017/09/08/indonesia-to-increase-palm-

oil-production-to-42-million-tons-by-2020.html.

Jamet, J-P. and Chaumet, J-M. (2016) ‘Soybean in China: adapting to the liberalization’, OCL 23, D604.

http://dx.doi.org/10.1051/ocl/2016044.

69 69

Jelsma, I., Schoneveld, G.C., Zoomers, A. and van Westen, A.C.M. (2017) ‘Unpacking Indonesia’s

independent oil palm smallholders: An actor-disaggregated approach to identifying environmental and

social performance challenges’, Land Use Policy, 67: 281-297.

Kavishe, C.B. (2013) ‘The Food Chain in Babati’, PowerPoint presentation.

Kuhn, A. and Nivievskyi, O. (2005) ‘Evaluating the Ukrainian Oilseed Export Tax’, Unpublished

mimeo.

Larson, D.F. (1996) ‘Indonesia’s palm oil subsector’, World Bank Policy Research Working Paper No.

1654.

Match Maker Associates (2014) ‘Sustainable growth potential of Tanzania’s agriculture’, Available

from: http://www.smeimpactfund.com/news/sustainable-potential-of-tanzanias-

agriculture.aspx#.WcOSKIrasdU.

McCarthy, J.F. (2010) ‘Processes of inclusion and adverse incorporation: oil palm and agrarian change

in Sumatra, Indonesia’, Journal of Peasant Studies, 37 (4): 821-850.

Meilke, K., Wensley, M. and Cluff, M. (2001) ‘The impact of trade liberalization on the international

oilseed complex’, Review of Agricultural Economics, 23(1): 2-17.

Mendez-Parra, M., Schubert, S.R. and Brutschin, E. (2016) ‘Export taxes and other restrictions on raw

materials and their limitation through free trade agreements: Impact on developing countries’, Paper for

the European Parliament’s Committee on Development, April 2016.

Mgeni, C.P., Sieber, S., Amjath Babu, T.S. and Mpenda, Z.T. (2017) ‘Can Sub-Saharan Africa Become

Food Self-Sufficient? Supply Response Analysis of Sunflower Oil Producers in Tanzania’. Paper

prepared for Tropentag 2017: Future Agriculture: Socio-ecological transitions and bio-cultural shifts,

20-22 September 2017, Bonn, Germany.

MITI (Ministry of Industry, Trade and Investment, United Republic of Tanzania) (2016) Sunflower

Sector Development Strategy 2016-2020. Geneva: International Trade Centre.

Moe, S.A., Than, S.M. and San, C.C. (2016) ‘The impact of world palm oil price changes on edible oil

industry of Myanmar: Case study on groundnut oil’, Journal of Agricultural Research, 3(1): 98-104.

MRA Management Associates (2014) ‘Sunflower value chain analysis report for Dodoma Urban,

Kongwa, and Kasulu districts’.

Nyaki, S. (2013) ‘Tanzania spends $ 120 million to import edible oil every year’, Corporate Digest, 22

August 2013, Available from: http://www.corporate-digest.com/index.php/tanzania spends--120-

million-to-import-edible-oil-every-year.

Obado, J., Syaukat, Y. and Siregar, H. (2009) ‘The impacts of export tax policy on the Indonesian crude

palm oil industry’, Journal of International Society for Southeast Asian Agricultural Sciences, 15 (2):

107-119.

Polevoy, V., Lukashchuk, L. and Peskovski, G. (2013) ‘Sunflower Cultivation in Ukraine: Role of

Fertilizers in Sunflower Seed Production’, International Potash Institute, No. 36, December 2013.

Regunaga, M. (2010) ‘Implications of the Organization of the Commodity Production and Processing

Industry: The Soybean Chain in Argentina’, LCSSD occasional paper series on food prices. World

Bank, Washington, DC.

Rifin, A. (2010) ‘The Effect of Export Tax on Indonesia’s Crude Palm Oil (CPO) Export

Competitiveness’, ASEAN Economic Bulletin, 27 (2): 173-184.

70 70

Rifin, A. (2014) ‘The Effect of Progressive Export Tax on Indonesian Palm Oil Industry’, Oil Palm

Industry Economic Journal, 14 (1), March 2014.

Rist, L., Feintrenie, L. and Levang, P. (2010) ‘The livelihood impact of oil palm: smallholders in

Indonesia’, Biodiversity and Conservation, April 2010.

Romero, C.A. (2016) ‘Competition Conditions in the Argentinean Oleaginous Sector Summary’,

September 2016.

Ruteri, J. M. and Xu, Q. (2009). ‘Supply Chain Management and Challenges Facing the Food Industry

Sector in Tanzania’. International Journal of Business and Management, 4(12): 70–80.

Salisali, B.M. (2017) ‘Fiscal Policy Study on Edible Oil Sector in Tanzania’, Draft consultancy report

prepared for the Tanzania Private Sector Foundation, June 2017.

Shivakumar, K.M., Kombairaju, S. and Chandrasekaran, M. (2007) ‘Trade Policy Analysis of Edible

Oil Imports of India under CGE Framework’, Department of Agricultural Economics, Tamil Nadu

Agricultural University, Coimbatore, India.

Sintoo, Y.E. (2015) ‘Factors affecting income of smallholder sunflower farmers in Singida region,

Tanzania: A case study of sunflower farmers in Singida Rural District’, Mzumbe University.

Srinivasan, P.V. (2005) ‘Impact of Trade Liberalisation on India’s Oilseed and Edible oils Sector’,

Report prepared for IGIDR-ERS/USDA Project: Indian Agricultural Markets and Policy, Indira Gandhi

Institute of Development Research, February 2, Mumbai.

Susila, W.R. (2004) ‘Contribution of oil palm industry to economic growth and poverty alleviation in

Indonesia’, Jurnal Litbank Pertanian, 23 (3): 107-114.

Teoh, C.H. (2010) Key Sustainability Issues in the Oil Palm Sector. International Finance Corporation,

World Bank Group.

Trigo, E.J. (2016) ‘Twenty Years of Genetically Modified Crops in Argentine Agriculture’, ArgenBio,

November 2016.

USDA (United States Department of Agriculture) (2016) ‘Argentina Oilseeds and Product Annual:

2016/17 Soybean Area Forecast Down as Area for Competing Crops Expands Dramatically; Sunflower

Area Expands by 36 percent in 2016/2017’, GAIN Report, 4 April 2016.

USDA (2017) ‘Ukraine Oilseeds and Product Annual: Forecast for less sunflowers, more soybeans and

rapeseed’, USDA Foreign Agricultural Service GAIN Report No. UP1706, 15 March 2017.

Zen, Z., Barlow, C. and Gondowarsito, R. (2006) ‘Oil Palm in Indonesian Socio-Economic

Improvement – A Review of Options’, Oil Palm Industry Economic Journal, 6 (1), Malaysian Palm Oil

Board.

Zeng, D.Z. (2011) ‘The Sunflower Sector in Tanzania – A Great Potential for Industrial

Competitiveness (A Sectoral Policy Study)’, Trade and Competitiveness Global Practice, The World

Bank Group.

Ugulumu, E. and Inanga, E. (2013) ‘Tanzania’s small scale sunflower farmers: Upgrading the value

chain’, International Journal of sciences: Basic and applied research.

Vanderschaeghe, M. and Okoth, H. (2017) ‘Gender analysis of the sunflower value chain. Inclusive

business in the sunflower value chain’, June, 2017. World Bank.

71 71

Annex A: List of individuals and organisations consulted

S/

N

Organisation/Company Individual(s) consulted Position

1. Bank of Tanzania Motto Lugobi Senior Economist

2. Mt Meru Millers Atul Mittal Group Executive Director

3. ALASKA Ltd. Jeniffer Bashe Chief Executive Officer

4 Murzah Oil Ltd. Nicetas Lyamuya Corporate Affairs and HRA

Manager

5 BIDCO Oil and Soap Ltd. Ruchir Shah Director of Operations

6 MeTL/East Coast Co. Ltd Hussein Kamote Advisor

7 Sunflower Farmers

Association (SUFA)

Stephen Henry Chairman

8 Tanzania Edible Oil Seeds

Association (TEOSA)

Rashid A. Mamu Chairman

Tanzania Sunflower

Processors Association

(TASUPA)

Enock A. Ndondole National Vice Chairman

9 Small Industries

Development

Organisation (SIDO) –

Dodoma

Nyangusi Mehtalam Technical Manager

10 Dodoma Regional

Government (RC/RAS

Office)

Aziza Mumba Assistant RAS – Economics

services

11 Tanzania Chamber of

Commerce, Industry and

Agriculture (TCCIA)

Idd Senge Regional Executive Officer

Dodoma

12 TRA Customs Namanga

Border Posts

Faraji Izungo

Ag. Customs Incharge

11 Singida Regional

Secretary Office

(RAS Office)

Choaji S. Beatus

Munyi Daniel

Mashaka Mrangi

Economist

Crops market officer

Seeds Officer

12 Singida Sunflower

Processors Association

(SISOPA)

Mohamed Alfan

Secretary

13 Singida Sunflower

Farmers Association

(SISUFA)

Athuman Ramadhan Sima

Mr. Ititi

Chairman

Deputy chairman

14 Mwasare Investment

(Bar soap manufacturer)

Wilfred Jumanne Mazini

Managing Director

15 YUFESTA PRODUCTS

(Bar Soap Manufacturer)

Felichism Massawe -

16 Huruma Mbezi

(Bar Soap Manufacturer)

Huruma Mbezi -

72 72

S/

N

Organisation/Company Individual(s) consulted Position

17 G&B Soap Industry Mr. Makundi

Managing Director

73 73

Annex B: Characteristics and profile of survey respondents

Farmers Processors Consumers

Sex Frequency Percent Frequency Percent Frequency Percent

Male 11 35.5 23 82.1 86 55.1

Female 20 64.5 5 17.9 70 44.9

Total 31 100 28 100 156 100

Age Frequency Percent Frequency Percent Frequency Percent

15 - 24 0 0 1 3.6 13 8.3

25 - 35 6 19.4 11 39.3 75 48.1

36 - 55 21 67.7 12 42.9 59 37.8

56+ 4 12.9 4 14.3 9 5.8

Total 31 100 28 100 156 100

Education Frequency Percent Frequency Percent Frequency Percent

No education 1 3.2 0 0 2 1.3

Primary education 25 80.6 15 53.6 81 51.9

Secondary education 5 16.1 6 21.4 51 32.7

Certificate/diploma 0 0 5 17.9 12 7.7

Degree and above 0 0 2 7.1 9 5.8

Total 31 100 28 100 156 100

Occupational status Frequency Percent

Employed 32 20.5

Trader 37 23.7

Business owner 76 48.7

Unemployed 7 4.5

Farmer 28 90.3 1 0.6

Farm owner/farmer and

employed 3 9.7

Total 31 100 156 100

Position Frequency Percent

Business owner 6 21.4

Business owner & manager 9 32.1

Manager 9 32.1

Deputy/other manager 3 10.7

Other staff, not manager 1 3.6

Total 28 100