DETERMINANTS OF COFFEE FARMGATE PRICES IN UGANDA

ABSTRACT

The purpose of the study was to examine the determinants of farmgate prices in Uganda. The specific objectives of the study were; to establish whether domestic coffee production significantly determines farmgate prices, to establish whether export prices significantly determine farmgate prices and to establish whether labour cost significantly determines farmgate prices. The study adopted cross sectional research design where secondary data was collected from Uganda coffee Development Authority and analysed using Microsoft excel and SPSS where a multiple linear regression model (MLR) was used.

The study found out that there was a weak negative relationship between farmgate price and domestic coffee production. This implied that the higher the quantity of domestic coffee production, the lower the farmgate prices. However, there was a strong positive relationship that was found between farmgate price and export prices which implied that higher export price increases farmgate prices. The study also found a weak negative relationship between farmgate price and labour cost where higher the labour cost reduced the farmgate prices. Therefore, export prices were the major determinant of farmgate prices paid to the local farmers in Uganda whereby labour cost and domestic coffee production had an insignificant effect on farmgate prices.

The study recommended that the government should construct demonstration farms and also strengthen the agricultural training institutions to train equip farmers with farm management skills so as to improve on the quality and quantity of coffee output to achieve its goal of 20million bags per year by 2020. Also The Uganda Coffee Development Authority should provide more information on the export prices on a daily basis to farmers so as to reduce information a symmetry and increase public access to price information along the marketing chain which could also reduce market power issues at the farmgate.

CHAPTER ONE

INTRODUCTION

1.0 Introduction

This chapter contains the background of the study, statement of the problem, objectives of the study, hypotheses as well as the scope and significance of the study.

1.1. Background to the study

This study is designed to analyse the determinants of coffee farmgate prices in Uganda. Between 2006 and mid-2008, the international prices of agricultural commodities increased considerably, by a factor larger than two. This upward trend in agricultural prices captured the world’s attention as a new food crisis was emerging. Several explanations for these movements in prices, ranging from demand-driven forces to supply shocks, have been provided by analysts, researchers, and development institutions which include the following;

The prices of China’s agricultural products have been unstable for about ten years. The abnormal volatilities are frequent, which have negative effect on farmers’ production decision making and people’s consumption. For the fluctuation of agricultural products prices, some researchers analyzed fluctuation causes and effects, the other researchers focused on agricultural prices prediction. Some researchers developed various methods to predict agricultural prices besides the traditional regression analysis method, including the neural networks, the grey method, the combinational model and Gray’s theory.

More so, as is the case of many agricultural commodities, price volatility is a major concern for stakeholders in the world coffee market. In exporting countries, volatility is a source of uncertainty in relation to export earnings and tax revenues, as well as instability in producer incomes. In importing countries, price volatility makes it difficult for roasters to control processing costs and affects profit margins for traders and stockholders, making their activities less attractive.

Africa is the region with the largest number of coffee producing countries: 25 as opposed to 11 in Asia & Oceania, 12 in Mexico & Central America and 8 in South America. Production in Africa has exhibited negative growth over the last 49 years. Average production was 19.4 million bags per crop year in the period between 1965/66 and 1988/89 when the coffee market was regulated under the export quota system. During the period between 1989/90 and 2014/15 under the free market, average production per crop year was 16 million bags. During those two periods, Africa’s share of world production has hence decreased from 24.9% to an average of 14%. Production in crop year 2014/15 is around 16.9 million bags, or 12% of the estimated world production of 141.7 million bags. Of this, an estimated 10.4 million bags were expected to be produced by just two countries Ethiopia and Uganda. Coffee is a major contributor to the economies of East African Community (EAC) members. However, recently, export of the crop has declined due to internal and external forces of supply and demand.

The export sector of most eastern and central African countries is dominated by coffee, which accounts for over 70 percent of foreign exchange earnings from total exports (USAID 2010). However, coffee output and quality in the sub region have declined. More recently, it was noted that coffee production decreased by 45 percent in 2011 compared to that in 2010 in Burundi alone. This was due to the decline in coffee prices that triggered poor coffee husbandry practices and crop over maturity. Agricultural prices in Uganda develop within a complex System, and can be analyzed in several different ways, potentially yielding different results. Spatial price transmission between different local equally-sized markets in Uganda, generally, is not that variable, but differs due to transportation costs. However, prices differ greatly along the supply chain, where the price of common agricultural goods can double from farm to city. Additionally, prices at local markets can differ due to local shocks or due to availability, i.e. the local harvest and how often traders operate in a local market. Due to the complexity of the value chain, smallholders, who are at the bottom of the chain for agricultural goods, might not experience rising prices following increased demand.

Coffee plays an important role in the economy and livelihoods of Uganda’s rural population. Based on the available value chain analysis studies, coffee market in Uganda does not include a wholesale market. However, the market includes other agents such as millers and rural traders. Most of Ugandan coffee is exported directly by coffee processors and therefore there is no active domestic wholesale market. In such cases, the point of competition is the border. The processors/exporters receive the full export price equivalent of the world prices depending on the point of delivery. Therefore, the observed price gap is zero that is no difference between the prices received by exporters and reference prices. The observed nominal rate of protection can be interpreted as the tax rate on coffee for the different market participants since quantitative restrictions are not imposed in this case. In other words, coffee exporters appear to receive prices very close to what they would have received in world market given all the currently observed market access costs.

However, the situation for coffee farmers is slightly different. Given the current profit margins which are generally low for processors and exporters, coffee growers began to receive some slight price incentives in recent years in the form of positive price gap. However, when reference prices are adjusted for excessive profit margins, the adjusted price gaps in recent years are often small and variable over time.

1.2 Problem Statement

The coffee market in Uganda does not include a wholesale market; it includes other agents such as millers and rural traders. Most of Ugandan coffee is exported directly by coffee processors and therefore there is no active domestic wholesale market, leading to low levels of competition in the market, hence exporters receive the full export price equivalent of the world’s prices depending on the point of delivery. This means export prices could have a direct effect on the farmgate prices since buyers of agricultural products in Uganda determine the price at which they buy the products rather than the famers. This implies that local harvest is also a determinant of farmgate price but however there are more determinants. It is against this background that the researcher prompted to analyse the determinants of coffee farmgate prices in Uganda.

1.3. Objectives of the study.

1.3.1 Main objective

To examine the determinants of coffee farmgate prices in Uganda.

1.3.2 Specific objectives

To establish whether domestic coffee production significantly determines farmgate prices.

To establish whether export prices significantly determine farmgate prices.

To establish whether labour cost significantly determines farmgate prices

1.4 Hypotheses

Domestic coffee production does not significantly determine farmgate price.

The export prices do not significantly determine farmgate price.

Labour cost does not significantly determines farmgate prices

1.5 Scope of the study

The study was confined to make an analysisof the determinants of coffee farmgate prices in Uganda. Specific emphasis was put on establishing whether domestic coffee production, export prices and labour cost significantly determine farmgate prices. The study gathered secondary data from Uganda Coffee Development Authority. The study was carried out from May to August 2017 in order to be able to capture and collect current and relevant data for the success of the study.

1.6 Significance

The findings of this study was helpful to management, policy makers and stake holders in establishing appropriate measures to stabilize and increase coffee farmgate prices

The findings of the study will also be helpful to farmers, buyers and all dealers in the coffee production chain to plan their production levels, forecast future price levels and predict future fluctuations so as to minimize losses and increase profits in order to experience improved standards of living.

CHAPTER TWO

LITERATURE REVIEW

2.0 Introduction

This chapter reviews the existing literature put forward by different scholars and personalities as well as critically analyzing the deviations in the explanations to find out the research gap in the study variables based on secondary sources like textbooks, internet, newspapers, reports and journals.

2.1 Effect of domestic coffee production on farmgate prices

Uganda produces two types of coffee: Arabica coffee (CoffeeArabica), which comprises About 70 per cent of the world’s coffee production and 10 percent of Uganda’s coffee Production; and Robusta coffee (Coffee canephora), which comprises about 30 per cent of the world’s production and 90 per cent of Uganda’s production (UCDA 2005).

Robusta coffee is indigenous to the central parts of the Uganda, while the British colonial Authorities introduced Arabica coffee at the turn of the twentieth century. Robusta is grown in the central part of Uganda in the Lake Victoria crescent, and across the west, South-west and east of the country. Arabica is grown at a higher altitude, in the areas of Mountain Elgon along Uganda’s eastern border with Kenya and in south-western Uganda along the Rwenzori Mountain range (UCDA, opcit).

This widespread cultivation of coffee places in Uganda has made it among the top 10 coffee-producing Countries in the world and second only to Ethiopia among the Africa, Caribbean and Pacific (ACP) countries (CTA, 2006). Ugandan Robusta beans are uncharacteristically hard, giving them good roasting qualities. They have a mild, soft, sweet and neutral taste, and have high frothing properties suitable for popular drinks such as espressos. Uganda’s Arabica also has strong market qualities; it is wet processed (washed) to produce a mild coffee that is popular with most consumers.

Before the late 1980s, farmers in Uganda produced, harvested, and dried coffee, and sold to primary cooperative societies or private stores (UCDA, 2004). Primary societies sold their coffee to cooperative unions, while the private stores sold the beans to either huller operators or the cooperative unions who, after hulling, sold the coffee to the Coffee Marketing Board (CMB). The CMB in turn sorted and graded the coffee before export (NUCAFE, 2005). The prices paid at each level were pre-determined by the government and remained fixed irrespective of movements in the international coffee market.

After liberalization in the early 1990’s, nearly all exporters became vertically integrated, with the supply chain for coffee export being dominated by coffee processing and trading companies (Ssemogerere, 1990). The activities of the cooperatives included procurement of coffee from farmers and primary cooperatives, hulling and processing the coffee into the exportable grades and blends, and in many cases exporting the coffee as well.

Alongside the main supply chain were private traders and the old cooperative trading system. The increased liberalisation of the coffee sub-sector encouraged foreign direct investment (FDI) and brought in multinational coffee companies (UCDA, 2005). The price of coffee surged from 20 per cent of the world prices before liberalization to as high as 85 per cent at the peak of the competition, before falling back to about 70 per cent.

Traders and exporters competed for a larger market share, which compromised the Quality of coffee, as traders became more concerned about quantity rather than quality. This led to an inconsistent trend in the value of coffee exported indicating that the trend of coffee marketing in Uganda acts a big role in determining the farmgate prices.

The processing of coffee occurs in two stages. First is the primary processing (hulling) which involves the removal of husks from the dried coffee beans? Second is the roasting and grinding of coffee into a finer form that is ready for consumption (secondary processing). In Uganda it is the primary processing that is mostly commonly done(UCDA, 2006).

Out of the coffee that is produced, 95% is exported after primary processing. Only 5% is locally roasted and grounded into finer coffee ready for consumption. In most cases, this processing is done by traders. Of recent, some farmers have adopted coffee processing before sale and the number is on the increase. There are about 250 active hullers in Uganda and these are widely distributed in coffee growing areas. However the proximity of these facilities to farmers is still inadequate. Unless these facilities become equitably located to favour access by remote poor farmers, improvement in coffee quality through farmers’ processing may still remain a myth (Baffes, 2006).

2.2 Effect of export prices on farmgate coffee

Farmgate pricing: means that negotiating a price directly with the farmers at the farmgate. That is, without any of the confusing export and import fees. The prices that are paid to the farmers for the sale of their coffee beans are above Fair Trade minimums. The Farmgate coffees prices paid to the farmers can easily verify that the good price enables the people who do the work, and other responsibilities at the coffee farms to achieve the great cup quality of our coffee. Farmgate is a simple principle that allows coffee producers to make premium prices in reward for coffee quality, and to reinvest to improve quality even more in the future in our country Uganda.

The  guarantee that Farmgate prices are 50% over Fair Trade (FT) pricing, but often they are 100%+ more that FT minimums. Fair Trade is a co-operative certification – that is, it does not allow certification for small independent farms – it is for co-operatives only. They do support coffee co-operatives, but they are often not what consumers might think. There are many excellent co-operatives and many that are large, powerful, corrupt, and mired in bureaucracy. They avoid the bureaucracy of cooperatives that sometimes do not share premium prices with their farmer members. Fair Trade certifies that the co-operative received the FT price, but it does not guarantee that the men and women who produce your coffee were paid the FT price. Fair Trade is also not based on the quality of the product, so in many ways it has a commodity mind-set at its core, that coffee is coffee, just like corn is corn.

On the other side, one should bear in mind that FT is a global standard, is verified by certifiers that make regular (if infrequent) visits to the cooperatives. They don’t have a third-party certifier. Instead they substitute the direct involvement at ground level in the buying process with farms, and know what they received when they are paying them through a middle-person. In this scheme, exporters and importers have a changing role, offering a service as logistics coordinators and an important one at that rather than coffee resellers. Any coffee bought off an importer/broker list does not qualify for Farmgate, and they do still buy some coffees that way because they are good quality. Further, lots from origins where hundreds of tiny farms contribute to even the smallest importable lots, such as Sumatra, or Yemen, can’t qualify for Farmgate in many cases nor can Auction Lot Kenya’s, even though the group pays extremely high prices for all these coffees, and know from direct observation that a premium reaches the farmer.

2.3 Effect of labour cost on farmgate prices.

Rural Uganda has limited access to communication tools that might be effectively used to transmit price information. Cell phone use is constrained by limited signal availability. In addition, cell phones are no longer costly. Only 14 (4%) of the farmers who marketed coffee had access to a cell phone. Much of the price information was gathered from local sources such as buyers and friends. Buyers often benefit from information asymmetries and the information they do convey may not be objective. Friends and neighbors may also be poorly informed or simply received from buyers. Given the lack of market information, it is difficult to know whether well informed farmers were actually able to negotiate higher prices without further analysis. Evidence from other countries, such as Cameroon, suggests that the dissemination of accurate labour cost can have a positive effect on farmgate prices received by farmers (Wilcox 2006).

Coffee price Volatility for example the Ethiopian coffee producers. This looks basically at the way coffee prices vary from time to time in the different periods and seasons of the year during coffee production. The outcome and benefit of the coffee producers is related to the average farmgate price paid to producers and the coffee sector accounts for over 35% of foreign exchange earnings and about 4% of GDP. Over 4 million small farmers are engaged in coffee cultivation. And about 15 million people are directly or indirectly dependent on income from coffee production. Ethiopia is the highest consumer of coffee in Africa (also the origin of Arabica coffee).

But the motivation for commodity price stabilization arises from the view that Poor households prefer price stability (risk aversion) and the poor suffer more than the rich from price instability.

Therefore governments have often set commodity price stability as a goal of economic policy. Which gives us this, understanding of the relationship between price volatility and producers’ welfare is important. It is important to estimate potential welfare gains for producers from eliminating this price risk. Welfare gains for producers from eliminating coffee price volatility are considerable and likely to make a significant difference to the lives of poor producers. The challenge for Ethiopian coffee producers is there are insufficient risk management tools available for them to manage price risk.

Therefore government intervention to deal with price volatility e.g. providing storage facilities, buffer stock systems, and ensuring Farmers willingness  to pay for insurance to mitigate price risk, providing information to farmers e.g., information on the determinants of international and domestic prices may be useful to reduce the coffee price volatility  hence showing that coffee price volatility  is  a great determinant of coffee  farm  gate prices which government and  producers  have to take great concern about in their mind.

When a new technology is introduced, farmers experiment with it before adopting. According to Saha et al. (1994), adoption can be classified into three phases namely, Information collection, decision whether or not to adopt and how much to adopt. Adoption occurs if the perceived benefit of adoption outweighs its cost. The farmer also decides on what proportion of resources to allocate to the new technology. Previous Studies have shown that the rate of adoption varies from one location to another (Ismail and Cuma, 2004), if the allocation of the resources to the new technology is high the farmers will tend to push the coffee farmgate prices high in order to benefit and meet the desired cost of production and profit levels hence the rate of technology adoption ermines a great determinant to farmgate prices in Uganda.

It is commonly assumed that rural producers have little bargaining power and are therefore price takers. In addition they have to make a choice between selling at the Farmgate or travel to nearby or more distant markets (Fafchamps and Hill, 2005).

Selling at the market provides a higher price than selling at the farmgate. In cases where the households sell at home, they sell to itinerant traders who internalise the transaction costs in their price offers. But households also need market information so that they are not exploited by these traders. Therefore social network capital serves to channel information and therefore the information costs are directly determined by the households’ level of connectedness. Information costs are non-increasing in SN. Transportation costs are mainly determined by distance to roads, road quality and distance to the market making the social network capital and agricultural prices available in the economy play a great role in determining farmgate prices of coffee since if put in to consideration the social network capital provides benefits to participating households through its function as an avenue for sharing information about markets and other opportunities that influence socioeconomic benefits. This largely depends on the structure of the networks and their functionality.

2.4 Other Determinants of coffee farmgate prices

Fafchamps and Hill (2004) concluded that the likelihood of a farmer selling off farm increases with quantity and proximity to the regional markets. Having access to transportation resources may decrease the opportunity costs of transporting cocoa to a local or regional buying center.

Farmer groups organized at the village-level may provide farmers withresources such as bargaining power, collective marketing, or credit opportunities (Myers, 2004).

The source and frequency of market information may positively affect a farmer’s ability to receive price signals and their ability to negotiate trades (Townsend, 1999; Kherallah et al. 2002). The source of the information, government, newspaper or radio sources may give the farmer a sense of downstream prices which may allow the farmer to better negotiate price. However, downstream price information may be irrelevant to local market negotiations, and therefore information from produce buyers and neighbors may give the farmer a better sense of local market prices.

A farmer’s access to capital through credit may impact quantity and quality through production, purchasing of inputs, labor, or marketing (Oxfam 2002; Townsend, 1999). Depending upon the source of credit, the effects on farmgate price may be positive or negative. Credit is often extended by the buyers as a means to secure supplies. Interest may be collected through lower farmgate prices at the time of the sale.

Social factors may also impact prices for a multitude of reasons, including some unmeasured local level phenomenon. During the conflict, fighting subsided and communities appear to have recovered and are involved in agriculture(Pay-Bayee 2005). Market access may be an issue in evacuating coffee from a given area given the local infrastructure and attractiveness to buyers.

CHAPTER THREE

METHODOLOGY

3.1 Introduction

This chapter looks at the frame work of methodology which is to be used to achieve the stated objectives. This includes collection and presentation of data, analysis, hypothesis testing and presentation of the results.

3.2 Research Design

The research design was cross sectional in nature, that is a combination of descriptive and quantitative designs was used so as to collect as much as possible about the coffee farmgate prices. The quantitative method was used to analyse the data produced during the study to show relationship between the key study variables.

3.3 Data Type and Sources

The data was got from the reports realised by Uganda coffee Development Authority.

3.4 Sample Size and Sampling Procedures.

The study considered all reports and secondary data provided by Uganda coffee Development Authority.

3.5 Data Collection

Data on monthly farmgate prices is to be compiled from monthly reports from the Uganda coffee development authority.

Data on monthly exports of coffee was used to estimate total production since majority of Uganda’s coffee production is exported. Export prices were too derived from the value of exports indicated in the monthly reports and are to be converted from US dollars to local currency using the average nominal interest rates for each month.

3.6 Data Analysis

Data was analysed using Microsoft excel and SPSS and was conducted as follows; Descriptive statistics were generated and used to describe each of the variables.

Bivariate analysis was done using spearman’s rank correlation coefficient and tested for significance at a 5% level of significance using SPSS.

The data tested for presence of heteroskedasticity, autocorrelation and multi-collinearity using the relevant tests using SPSS.

Model

P_g = α+ β₁X₁+β₂X₂+β₃X₃ + ε_i

Where;

P_g = farmgate prices

α = constant

X₁ = Domestic coffee production

X₂ = Export Prices

X₃ = Labour Cost

ε_i = Errors

β₁ = coefficient of X₁

β₂ = coefficient of X₂

β₃ = coefficient of X₃

• Assumptions about the error term

Normality of the error term. This means that the error terms are symmetrically distributed around their mean and their distribution is determined by the mean and the variance.

Zero mean. This means that the error term has a value of zero on average.

Homoscedasticity. This means that the variance of the error terms is constant in each period

Non autocorrelation. This means that the error terms of different observations are independent

The error term and the explanatory variables

• Limitations and Delimitations

Limited access to relevant reports. The researcher faced a challenge of accessing relevant reports to provide relevant information for this study. However, the researcher strived to use the available reports at UCDA for various years.

CHAPTER FOUR

PRESENTATION, INTERPRETATION AND DISCUSSION OF FINDINGS

4.0 Introduction.

In this chapter, the researcher presents analyses and interprets the findings of the study.

4.1 Descriptive Analysis

Table 4.1.1 Domestic coffee production for the period 2012 – 2015

Study results show the minimum domestic coffee production in 60kgs/bags in the study period (2012-2015) was 195,723kg and the maximum was 373,927kg, giving an average of 279,062kgs and a standard deviation of 54,870.

Table 4.1.2. Farmgate prices for Robusta coffee in Uganda for the years 2012 to 2015

The results show that the minimum farmgate price for Robusta coffee in Uganda in the study period (2012-2015) was shs. 1500 and the maximum was shs. 2,275 giving an average of shs. 1926.75 and a standard deviation of 267.2356425.

Table 4.1.3. Export prices for Robusta coffee

The results show that the minimum export prices for Robusta coffee in Uganda in the study period (2012-2015) was $ 1.7525 and the maximum was $ 2.29 giving an average of $ 2.031458 and a standard deviation of 0.184556.

4.2 Bivariate Analysis

4.2.1. Relationship between Farmgate prices and domestic coffee production

Ho. There is no significant relationship between farmgate price and domestic coffee production

Table 4.2.1.1.  Correlation between farmgate price and domestic coffee production

The correlation coefficient (-0.0576) shows a weak negative relationship between farmgate price and domestic coffee production. This means that the higher the quantity of domestic coffee production, the lower the farmgate prices. However, this is statistically insignificant at 1% level of significance since the p value (0.221) > 0.01. Therefore we fail to reject the null hypothesis and conclude that there is no significant relationship between farmgate price and domestic coffee production.

4.2.2. Relationship between Farmgate prices and export price

Ho. There is no significant relationship between farmgate price and export price

Table 4.2.2.1. Correlation between Farmgate prices and export price

*. Correlation is significant at the 0.01 level (2-tailed).

The correlation coefficient (0.6934) shows a strong positive relationship between farmgate price and export prices. This is statistically significant at 1% level of significance since the p value (0.000) < 0.01.This means that the higher the export price, the higher the farmgate prices. Hence, we reject the null hypothesis and conclude that export prices have a significant impact the farmgate prices.

4.2.3. Relationship between Farmgate prices and labour cost

Ho. There is no significant relationship between farmgate price and labour cost

Table 4.2.3.1. Correlation between Farmgate prices and labour cost

The correlation coefficient (-0.0525) shows a weak negative relationship between farmgate price and labour cost. This means that the higher the labour cost, the lower the farmgate prices. However, this is statistically insignificant at 1% level of significance since the p value (0.296) > 0.01. Therefore we fail to reject the null hypothesis and conclude that there is no significant relationship between farmgate price and labour cost.

4.3 Test for heteroskedasticity, autocorrelation and multicollinearity

4.3.1 Test for heteroskedasticity

Using the Park test

Ho: There is no heteroskedasticity in the residuals.

Table 4.2.1.1. Test for heteroskedasticity

We accept the null hypothesis and conclude that there is no heteroskedasticity in the series (p>0.05).

4.3.2 Test for autocorrelation

Using the Breusch pagan test for serial correlation

Ho: There is no serial correlation in the residuals.

Table 4.3.2.1.  Test for autocorrelation

Intercept= n=48

We accept the null hypothesis and conclude that there is no serial correlation in the residuals (P> 0.05).

4.2.3 Test for multicollinearity

Ho: Export price is not correlated to domestic coffee production.

Table 4.2.3.1. Test for multicollinearity

We accept the null hypothesis and conclude that the explanatory variables are not correlated (P> 0.05).

4.4 Empirical Results and Discussion

A regression model was run to examinant the determinants of coffee farmgate prices in Uganda.

The study conducted a multiple regression analysis from the regression equation thus:

P_g = α + β₁X₁ + β₂ X₂+β₃ X₃ + ε_i

Where P_g = farmgate prices, α = constant, X₁ = Domestic coffee production, X₂ = Export Prices and X₃ = Labour Cost

Ho: farmgate prices does not depend on domestic coffee production.

Ho: farmgate prices does not depend on export prices.

Ho: farmgate prices does not depend on labour cost.

Table 4.4.1. Model Summary

Adjusted R squared is coefficient of determination which tells us the variation in the dependent variable due to changes in the independent variable. From the findings in the table above the value of adjusted R squared (0.593) shows that 59.3% of the variations in the farmgate pries can be well explained by domestic coffee production, export prices and labour costs. Hence a moderate fit.

From the results in the above table the established regression equation was;

P_g= -134.05-641X₁+ 0.424 X₂ – 0.739X₃:

A unit increase in domestic coffee production would on average lead to (641) decreases in farmgate prices however, this is statistically insignificant at 5% level of significance since the p value 0.221 > 0.05. Hence we fail to reject the null hypothesis and conclude that farmgate prices do not depend on domestic coffee production. However, the researcher’s observation it is not true because the higher the quantity of coffee produced that means farmgate price have to lower down and also the lower the quantity produced, the higher the farmgate price. This finding concurs with that of (Baffes, 2006) who argued that out of the coffee that is produced, 95% is exported after primary processing. Only 5% is locally roasted and grounded into finer coffee ready for consumption. In most cases, this processing is done by traders. Of recent, some farmers have adopted coffee processing before sale and the number is on the increase. This led to an inconsistent trend in the value of coffee exported indicating that the trend of coffee marketing in Uganda acts a big role in determining the farmgate prices.

A unit increase in export prices by 1$ would on average lead to (0.42) increases in farmgate prices and this is statistically significant at 5% level of significance since the p value (0.000) < 0.05. Hence we reject the null hypothesis and conclude that farmgate prices depend on export prices. Therefore, this is very true because the higher the price for our coffee exports means we will also see an in increase in the farmgate prices while the lower the export price for our coffee exports, the lower the farmgate prices. Evidently, the coffee market in Uganda does not include a wholesale market. However, the market includes other agents such as millers and rural traders. Most of Ugandan coffee is exported directly by coffee processors and therefore there is no active domestic wholesale market. In such cases, the point of competition is the border. The exporters receive the full export price equivalent of the world prices depending on the point of delivery. This means export prices have a direct effect on the farm gate prices since buyers of agricultural products in Uganda determine the price at which they buy the products rather than the famers. Also, spatial price transmission between different local equally-sized markets in Uganda, generally, is not that variable, but differs due to transportation costs. This agrees with (Fafchamps and Hill, 2005) who argued that it is commonly assumed that rural producers have little bargaining power and are therefore price takers. In addition they have to make a choice between selling at the Farmgate or travel to nearby or more distant markets.

A unit increase in labour cost by shs. 1 would on average lead to (0.739) decrease in farmgate prices. However this is statistically insignificant at 5% level of significance since the p value 0.296> 0.05. Hence we fail to reject the null hypothesis and conclude that farmgate price does not depend on labour costs. Prices at local markets (no matter the size) can differ due to local shocks or due to availability, i.e. the local harvest and how often traders operate in a local market. This implies that local labour input is also a determinant of farm gate price.

Therefore from the results above, export prices is a major determinant of farmgate prices paid to the local farmers in Uganda whereby labour cost and domestic coffee production have an insignificant effect on farmgate prices.

CHAPTER FIVE

SUMMARY, CONCLUSIONS AND POLICY RECOMMENDATIONS

5.0 Introduction

This chapter presents the summary, conclusion and recommendation of the findings presented   in the previous chapter.

5.1 Summary of findings

While the model has two explanatory variables, the farmgate price which accounts for about 57% of the variation in the farmgates. The relationship between export prices and farmgate prices is brought about by the fact that most of Uganda’s coffee is exported and the nature of the supply chain which does not include a whole sale market which leads to direct interaction between exporters and producers of the coffee.

The relationship between farmgate prices and export prices is positive which implies that high export prices will translate into high farmgate prices and the converse is true. The level of production has no significant effect on the level of farmgate prices. This is because most of Uganda’s coffee is exported.

The local market is too small to be affected by variations in production since quantities supplied will always be able to meet the demand hence unlike other agricultural products, there will not be a scarcity arising from low production or surplus resulting from over production to drive prices high or low respectively.

Furthermore, domestic production has no significant effect on export prices because demand in the world’s largest consumer and largest importer of Uganda’s coffee (76.09% as of February 2016), European Union, has stagnated slightly at an estimated 42 million bags, averaging growth of 0.8% per year since 2012.

5.2 Conclusions

The study found a weak negative relationship between farmgate price and domestic coffee production. This implied that the higher the quantity of domestic coffee production, the lower the farmgate prices. However, there was a strong positive relationship that was found between farmgate price and export prices which implied that higher export price increases farmgate prices.

The study also found a weak negative relationship between farmgate price and labour cost where higher the labour cost reduced the farmgate prices.

Therefore, export prices were the major determinant of farmgate prices paid to the local farmers in Uganda whereby labour cost and domestic coffee production had an insignificant effect on farmgate prices.

5.3 Policy Recommendations

The government should construct demonstration farms and also strengthen the agricultural training institutions to train equip farmers with farm management skills so as to improve on the quality and quantity of coffee output to achieve its goal of 20million bags per year by 2020.

The Uganda Coffee Development Authority should provide more information on the export prices on a daily basis to farmers so as to reduce information a symmetry and increase public access to price information along the marketing chain which could also reduce market power issues at the farmgate.

The government should provide subsidies to coffee farmers so as to reduce on the labour costs and operation costs incurred by the farmers.

REFERENCES

Barrett, C. B., and. J. R Li. 2002. “Distinguishing between Equilibrium and Integration in Spatial Price Analysis.” American Journal of Agricultural Economics. 84: 292-307.

Barry Callebaut. (2008). “Barry Callebut.” Internet Site: http://www.barry-callebaut.com/

Belsley, D.A., E. Kuh, and R.E. Welsch. (1980). Regression Diagnostics. New York: John Wiley & Sons, Inc.

Bryce Cooke & Miguel Robles. “Recent Food Price movements. A time series analysis” IFPRI, 2009

Byekwaso J. “Output Manager of Output 3 (Farmers Organizations), of ABDC of ASPS of Danida”, Interview with authors. 2008

Cleaver, K. M., and W.G. Donovan. (1995). Agriculture, poverty, and policy reform in Sub-Saharan Africa. Washington, D.C.: World Bank

Fafchamps, M. and R.V. Hill. (2005). “Selling at the Farm-Gate or Traveling to Market.” American Journal of Agricultural Economics 87(3):717-734

FAO- MAFAP “Analysis of incentives and disincentives for coffee in Uganda” Internet.

Foyn, T.H.Y &L.L. Kaspersen. “Price transmission for agricultural commodities in Uganda: An empirical vector autoregressive analysis

Gibbon, P., and S. Ponte. 2005. Trading down Africa, value chains, and the global economy. Philadelphia: Temple University Press.

ICO “World coffee trade (1963 – 2013): A review of the markets, challenges and opportunities facing the sector” Internet: www.ico.org/news/icc-111-5-r1e-world-coffee-outlook.pdf, Feb2014 [Jun. 8, 2016]

ICO, “Sustainability of the coffee sector in Africa” Internet www.ico.org/documents/…/icc-114-5e-overview-coffee-sector-africa.pdf, 2015 [Jun. 8, 2016]

Kennedy, J. (2005). “Rapid Assessment of Cocoa and Coffee Production Base in Liberia.”

Michael J. Boskin. (2008). Consumer Price Indexes.” The Concise Encyclopedia of Economics.

Oredein, O. 2007. “Nigeria’s Ondo State Passes Law Prohibiting Cocoa Tampering.” Dow Jones Newswire. 14 July 2007.

Unpublished, Monrovia, Liberia.

W.M. Ndayitwayeko et al. “Comparative advantage of the eastern and central Africa in the coffee export sector: the case of Burundi.” African Crop Science Journal, Vol. 22, Issue Supplement s4, pp. 987 – 995, 2014

Yi Wang &Xin Su. “Fractal Analysis of the Agricultural Products Prices Time Series” International Journal of u- and e- Service, Science and Technology Vol.8, 2015

APPENDICES

Appendix I:Coffee Exports for Calendar Years 2012-2015

AVERAGE PRICES PAID TO FARMERS AND PROCESSORS FROM JULY 2012 UP TO DEC 2015 IN UGANDA SHILLINGS PER KILOGRAM (U SHS/KG & $/KG)   

DOMESTIC COFFEE PRODUCTION FOR THE PERIOD OF (2012-2015)

Export prices for 2012-2015