This chapter provides background information, location and structure of the sugar industry, role of the sugar industry, performance of the sugar industry in Kenya, challenges facing the Kenyan sugar industry, reforms in the sugar industry, the research problem, Objectives of the study, significance of the study and organization of the study.
Kenya’s economy is mainly dominated by the agricultural sector even though only 10% of the total land area has sufficient fertility and rainfall to sustain farming. About one half of the total agricultural output is non marketed subsistence production. Agriculture is the second largest contributor to Kenya’s Gross Domestic Product (GDP) after the service sector. Horticultural produce and tea are the main growth sectors and the most valuable of all Kenyan exports. Growth in the agricultural sector is directly correlated with the growth in the economy in general. Improving productivity in the sector can play a crucial role in Kenya’s economic growth and development (Nyoro, 2012).
Sugar cane is one of the industrial crops of Kenya. The sugar industry in Kenya has made a major contribution to the development of the nation. Despite its key importance to the economy, it has continued to perform dismally leading to persistent deficits in production and endangering the livelihoods of over 250,000 small scale farmers who depend on the sector. The poor performance of the sector is attributed to various factors including inadequate supply of sugar cane to factories; cane poaching; low levels of capacity utilization in the sector; low level of technology adoption and poor management (KSB, 2011).
The Manufacturing sector, in which the sugar industry belongs, has remained stagnant in its contribution to the GDP. The contribution has remained at an average of 10% for more than ten years (Kenya Economic Survey, 2015). The vision 2030 stipulates that the sector should account for 20% of the GDP. Achieving this goal requires that some underlying constraints that hinder faster growth are addressed. They include high input costs, decline in investment portfolio, high cost of credit and competition from imports.
Productivity is generally defined in terms of the improvement and technical change with which inputs are translated into outputs in the production process (Shih-Hsun. et al., 2003). Therefore productivity measures may be examined collectively (across the economy) or in specific industries using different measures. Productive efficiency is commonly used to measure the level of performance of a production unit in terms of its utilization of input resources in generating output, while minimizing the wastage of resources in the production process.
Productivity growth signifies increased economic growth, higher tax revenue, growth in employment opportunities and improved standards of living. Productivity enhances the capacity of firms to become viable, profitable and create sustainable jobs thus making productivity to be a major determinant of competitiveness (Magati and Muthoni, 2012). The Kenya Vision 2030 acknowledges that productivity improvement is vital in enhancing growth and economic prosperity. For an industry or firm to remain competitive it should be well informed of its level of productivity and how the productivity changes over time.
Sugarcane is an important industrial and cash crop in Kenya. It is mainly grown in the former western and Nyanza provinces. The crop is also grown in parts of Nandi, Kericho and Narok, Kwale and Tana River counties. About 90% of the total sugarcane production is contributed by small scale farmers. The remaining 10% comes from large scale farmers and factory nucleus estates (KSB 2003).
The industry has eleven operational sugar factories namely: Chemelil Sugar Company; Kibos Sugar and Allied Factories; Muhoroni Sugar Company (In receivership); Mumias Sugar Company; Nzoia Sugar Company; Soin Sugar Company; South Nyanza Sugar Company; Sukari Industries Limited; Transmara Sugar Company; West Kenya Sugar Company and Butali Sugar Company. Kwale International Sugar Company is yet to be commissioned (Kenya National Assembly: March 2015).
Sugar cane farming was introduced in Kenya in 1902. The first sugar processing factory was established at Miwani in 1922 and later Ramisi in Kwale District in 1927. Due to increase in demand for sugar, the government later got widely involved in sugar production through additional investments in sugarcane growing schemes and factories; Muhoroni (1966), Chemelil (1968), Mumias (1973), Nzoia (1978) and South Nyanza (1979). West Kenya (1979), Butali (2010), Kibos (2008), Soin (2008), Sukari (2011) and Transmara (2011) are private companies.
The Kenyan government involvement in the sugar sector was to address sugar consumption needs of the country through self-sufficiency in sugar production. Sugar production was as a vital import substitution strategy to save the country the much needed foreign exchange. It was also a tool of social development that could improve the livelihoods in the rural areas through employment and wealth creation (Sserenkuma and Kimera, 2006).
The Kenya Sugar Directorate under the AFFA is the regulatory body of the Kenya Sugar Industry. It is charged with the responsibility of regulating, developing and promoting the Kenya Sugar Industry. KESREF is the research arm of the directorate with the principal objective of conducting sugar research and developing appropriate and suitable technologies.
The sugar industry has an indirect socio-economic impact in overall terms which is significantly larger than its direct contribution to GDP because of its backward and forward linkages. Sugar cane farming supports over 250 000 small scale farmers in Kenya. This is in addition to an estimated six million Kenyans who derive their livelihood directly or indirectly from the sugar industry (KSB, 2011).
The industry contributes revenue through tax revenues to the government. The industry has also contributed immensely towards infrastructure development through roads construction and maintenance of bridges as well as provision of social amenities such as education, health, sports and recreation facilities.
The industry is a source of raw materials for other industries such as bagasse (cane residue) for power co-generation and molasses for wide range of industrial products including ethanol. Sugar is an important food item and also a critical raw material in food beverage and pharmaceutical industries. The industry has contributed to the development of urbanization through emergence of towns near sugar factories.
Kenya has been experiencing a steady rise in the domestic demand for sugar. The gap between sugar production and consumption has continued to increase making Kenya a net importer of sugar.
Area under Cane
The area under cane grew from 131 507 hectares in 2004 to 213 920 hectares in 2013 representing an increase of 62.7% %. The increase in area under cane is due to high cane demand because of new mills and expanded capacity of most sugar factories. The increase in area under has not translated to self-sufficiency in sugar production.
The total area harvested in the nucleus estates, and out-grower farms was 85, 857 hectares in 2013 compared to 54, 191 hectares in 2004 indicating an increase of 58.4%. The mean area harvested over the period was 67540.3 hectares. The largest area harvested was recorded in 2012 when 87,340 hectares of cane was harvested. However the best industry performance during the study period was achieved in 2010 when 49.83 per cent of the area under cane was harvested (KSB, 2013).
The average cane yield in Kenya during the period is 65.4 TC/Ha. This is very low compared to other COMESA countries like Egypt 126.4 TC/Ha, Zimbabwe 90 TC/Ha and Tanzania (95 TC/Ha). Low yields are attributed to poor cane husbandry and high cost of farm inputs and low yielding cane varieties (KSB, 1999 and 2013).
Figure 1: Average Sugar cane yield, Tonnes/Ha 2004-2013
Source: generated by the author from KSB data
As can be seen in Figure 1, the average cane yield in MT per hectare for the industry declined steadily from 73.8 TC/Ha in 2004 to 51.00 TC/Ha in 2012.
Sugar production in Kenya has grown from 389, 138 MT of sugar in 1996 to 600,179 MT in 2013, while sugar consumption has increased from 570,000 MT in 1996 to 841,957 MT in 2013 (KSB, 2013). The deficit in meeting domestic sugar consumption needs from local production has grown from 180, 862 MT in 1996 to 241,778 MT in 2013. This has made Kenya to regularly import sugar to meet the domestic demand for sugar.
Figure 2: Local Sugar Production compared to Estimated Sugar Consumption in 1996 to 2013
Source: Generated by the author from KSB data 2013
According to Figure 2, the gap between sugar production and consumption has continued to increase. Since Kenya is a high cost producer of sugar, the trend is expected to continue unless the efficiency and productivity of the industry is improved.
The Quality of sugar cane crushed measured as pol % cane (sucrose content) has been steadily decreasing from 13.28 in 1996 to 11.16 in 2013 against an industry target of 13.50. This is low compared to other countries in the region like Malawi with 14.26. The low sucrose content is due to poor cane varieties, fluctuating weather patterns and lack of coordinated extension services.
The combined installed capacity of all sugar factories in the country is 30,866.4 TCD. This could produce approximately 1,187,910.08, MT of sugar per year leading to surplus sugar production of over 300, 000 MT. However during the period, the average capacity utilized was 19 239.33 TCD (59.535%). The low capacity is attributed to unscheduled factory stoppages, factory breakdowns and lack of cane (KSB, 1999 and 2006).
The Kenyan Sugar Industry has the potential to generate 120 MW of electricity. It is only Mumias which is generating 38MW of which 26 MW is exported to the National Grid. In addition to electricity, the miller produces 22 million litres of ethanol and 15 million litres of bottled water (Kenya National Assembly; 2015). Unlike Kenyan firms, sugar firms in the COMESA region have diversified their operations into co-generation to reduce costs on electricity and earn revenue. Challenges to product diversification have been uncompetitive price mechanism; limited technology and factory capacity; weak legal and regulatory framework (KSI, 2010- 2014).
Kenya’s sugar exports decreased from 24,478 MT in 1996 to 11,580MT in 2004. From 2004 sugar exports decreased further to 104 MT in 2013. In contrast, sugar imports have steadily increased from 65,816 MT in 1996 to 238,046 MT in 2013.The decrease in sugar exports is mainly due to relatively higher domestic ex-factory prices. The increase in imports is as a result of increase in sugar demand.
Table1: Sugar Production and Trade in Kenya (2008 – 2013)
|Self-sufficiency ratio (%)||70.31||74.81||66.94||77.90||67.43||71.60||71.50|
|Import dependency ratio (%)||29.67||25.18||33.05||22.10||32.57||28.40||28.50|
Source: Author’s compilation from KSB data
Kenya is a net importer of sugar with an import dependency ratio ranging from 25.18% to 33.05% and a self-sufficiency ratio ranging from 66.94% to 74.81% during the period 2008 – 2013. The mean import dependency ratio during the period was 28.50 per cent against a self-sufficiency ratio of 71.5 per cent, an indication that local sugar production cannot sustain the domestic sugar consumption.
Kenyan sugar factories are high cost producers of sugar. This has reduced the competitiveness of the industry (KSB 2007). The cost of sugar production in Kenya is currently estimated at USD 870 per MT which is twice the cost of production in other COMESA competing countries. This is very high compared to Zimbabwe (USD 300), Malawi (USD 350), Swaziland (USD340), Sudan (USD 340), and Zambia (USD 400), (Kenya National Assembly, 2015).
The sugar industry is constrained by low production capacities, lack of clear harvesting schedules, huge debts, managerial inefficiency, cane poaching, unreliable and fluctuating weather conditions, outdated technology equipment and machinery.
Kenya sugar factories have a combined installed capacity of 30,866.4 TCD. The factories continue to operate at low capacities because of significant technical and management limitations (KSI, 2009 and KSB, 2010). The main determinant of a sugar factory’s production efficiency is its conversion ratio, which measures the amount of sugar cane needed to produce one MT of sugar. A comparison of conversion ratios between private and government owned factories reveals a significant difference. In 2008, the conversion rate for Mumias was 9.65 while Muhoroni was 12.62 (KSB, 2010). This means Muhoroni required an additional three MT of cane to produce the same amount of sugar as Mumias.
Kenya is a member of COMESA free trade agreement and is therefore bound by the provision of the free trade protocol that allow duty and quota free access of sugar from COMESA FTA countries into the Kenyan market. This has resulted in an influx of imported sugar at lower prices thus rendering locally produced sugar non-competitive. Most of Kenya’s exports end up in the COMESA region from which Kenya earns a lot of benefits through foreign exchange. It is therefore difficult for Kenya to restrict imports from COMESA countries because it is bound by the COMESA FTA protocol (KSB, 2006).
In Kenya, sugar is not classified as a basic food and therefore currently attracts a VAT of 16 per cent. Cess is also levied to facilitate the development of infrastructure. All firm inputs including; fertilizers, herbicides and machinery such as boilers and tractors are taxed. Contrary to practices in countries like India, Sudan and Egypt, Kenyan sugarcane farmers do not receive subsidies. This leads to high cost of production and high prices of domestically produced sugar. There have been claims of double taxation in which sugar inputs are taxed and VAT is levied on the final product. The double taxation has been identified as the cause of high prices for local sugar. Suggestions have been made to classify sugar as a food item like maize and other food crops for it to be zero rated (MAFAP, 2013)
Corruption remains a major challenge in the management of sugar firms. According to KACC (2010), there is blatant corruption and mismanagement of most institutions connected to the sugar sub sector. Incidences of corruption are cited in the appointment of CEOs of sugar firms, employment and promotion of staff in the mills, theft of sugar from the factories and the process of accessing loans from KSB, issuance of licenses to new sugar factories in contravention to guidelines set in the Sugar Act (2001).
1.1.5 Reforms in the Industry
According to KACC (2010), Kenya suffered the biggest crises to its sugar sector between 1998 and 2001. Most sugar mills suffered serious financial crises which almost resulted in collapse of the industry. The main causes of the crises were; managerial inefficiency and unregulated importation of sugar due to liberalization. The government initiated policy reforms to save the industry from collapse. This led to the enactment of the Sugar Act 2001 and a new regulator, the Kenya Sugar Board (KSB). Administrative reforms backed by tough trade measures immediately followed to control the behavior of players in the trade and those in the distribution chain (Ssrenkuma and Kimera, 2006).
Beginning 2001, the Kenya government has renegotiated COMESA safeguards on five different occasions to give the industry sufficient time to improve its productivity and efficiency. In recent case, Kenya was allowed one year extension effective March 1st 2015 to improve the efficiency and productivity of its sugar industry (Kenya National Assembly, 2015).
The operations of the sugar industry are funded through the Sugar Development Fund (SDF). SDF was established in 1992 to extend loans to the industry for factory rehabilitation and cane development. It also provides grants for operations of the Sugar Directorate, KESREF and development of roads infrastructure in the cane growing areas. The Kenya Sugar Directorate, Out-grower institutions, Millers, Transporters, Farmers and the Kenya Sugar Research Foundation are all eligible for the fund.
Following the expiry of COMESA safeguards in February 2015, Kenya was granted a one year extension effective March 1st 2015 to improve the productivity and efficiency of its sugar industry. Kenya has been successfully negotiating the extension of COMESA safeguards since 2001.The COMESA safeguards allow Kenya to limit the entry of sugar imports to 350,000 MT to plug the annual production deficit. However, World Trade Organization (WTO) rules permit a maximum of ten years for such special trade protection measures. Therefore, there is likely to be no future extensions of COMESA safeguards.
For many years, sugar consumption has exceeded sugar production. Most sugar firms have accumulated high debts as they continue to operate with inefficiency leading to low levels of productivity. Therefore, there is an urgent need to address the productivity growth of sugar production in Kenya to improve the competitiveness of the industry. Policy makers, industry stakeholders and the sugar factory need empirical information on the productivity of the industry.
The study aims to answer the following questions: (i). How has the productivity of the sugar industry in Kenya changed during the period? (ii) Have there been productivity growth in the industry? (iii)What are the factors that determine productivity in the sugar industry in Kenya?
The main objective of the study is to assess the performance of the sugar industry in Kenya in terms of its productivity. The specific objectives are,
(i) To analyze the performance of the sugar industry in Kenya.
(ii) To investigate the factors that explains the productivity in the sugar industry.
(iii) To measure the productivity changes of the sugar industry in Kenya.
(iv) Suggest recommendations to improve productivity in the industry.
Most of the previous studies in the sugar industry have mainly focused on the Technical efficiency of Sugar factories in Kenya. Studies on production efficiency have mainly been done at the grower (farm) level. There are no reported productivity studies at the firm level using financial data for the sugar industry in Kenya.
This study attempts to fill this gap by estimating the firm level total factor productivity and its components, for the sugar firms in the industry and to assess the variations in TFP growth between firms and over time. This study therefore would provide a fresh perspective on the productivity in sugar sector for use in developing appropriate policy responses towards the industry. The study aims to decompose TFP growth into technical change, technical efficiency change and scale efficiency change for understanding the source of productivity of sugar firms in Kenya. The decomposition will enable policy makers to trace lagging productivity to particular factors.
Analysis of productivity of sugar firms in Kenya is very important because of the threat posed to the industry by cheap imports and heightened competition from sugar produced in the COMESA countries. In order to compete effectively against international sugar producers, the productivity of the industry must be improved.
The research proposal is organized into three chapters. Chapter one deals with introduction and gives an overview of the industry, including background information, location and structure of the sugar industry, the role of the industry, performance of the industry, challenges, reforms, research problem, objectives and significance of the study. Chapter two covers the theoretical foundation of productivity measurement, empirical evidence and an overview of the literature. Chapter three describes the theoretical model, specification of the empirical model to be estimated, stages of estimation, description of variables and data sources.
This chapter has two sections: the theoretical and empirical literature review. The theoretical literature review gives an outline of the theoretical foundation on which the subject matter of the study is based while the empirical literature review is based on the studies that have been carried out on the same or related subjects. The chapter also has an overview of literature reviewed.
Productivity growth can be defined in terms of the improvement and technical change with which inputs are converted into outputs in the production process, (Shih-Hsun et al., 2003). Emphasis on productivity growth is enhanced by the recognition that increased productivity will lead to higher production and a sector’s contribution to economic growth.
The productivity of a firm or an industry is a measure of the relationship between its production of goods and services and the factors of production used. Productivity is therefore used as a tool to measure the performance of an economic entity. Increased productivity will be an indicator that the firm or industry is utilizing its scarce factors of production efficiently.
According to Chambers (1988), productivity can be used to measure the rate of technical change in production and can be divided into two sub-concepts, that is, Partial Factor Productivity (PFP) and Total Factor Productivity (TFP). Partial Factor productivity is the ratio of output to a single or specific input. It measures the contribution of one particular input to output while ignoring the contribution of other factors of production. The Total Factor Productivity (TFP) is the ratio of output to the aggregate measure of the inputs of all the factors of production. This is the ideal measure of productivity as it incorporates the contribution of all factor inputs.
The term productive efficiency is commonly used to describe the level of performance of a production unit in terms of its utilization of input resources in generating outputs while minimizing wastage of resources in the production process.
Economic efficiency is a term that refers to the optimal production and consumption of goods and services. According to Farrel (1957), the economic efficiency of a production unit is composed of two different efficiency measures; technical efficiency and allocative efficiency. Efficiency is concerned with the relation between scarce input resources (e.g. labour, capital, machinery etc) and either immediate or final outcomes. The physical relationship between input and output is called technical efficiency.
Allocative efficiency is the ability of a firm to use inputs in optimal proportions, given their respective prices. If a production process equates the marginal rate of technical substitution between each pair of inputs with the input price ratio, then it exhibits allocative efficiency. According to Yin (1999), the type of efficiency measured depends on data availability and appropriate behavioural assumptions.
Technical efficiency is the ability of a production unit to avoid waste by producing maximum possible output from available input or by using very little input as output production allows. Technical efficiency occurs when firms are obtaining the maximum output given certain inputs of production (Wolgin, 1973). It involves the transformation of the production function through introduction of new inputs and techniques of production. A technically efficient firm will, therefore be on the boundary of its production possibilities surface. The ratio of the observed output to potential output, given the available technology determines the technical efficiency of an individual firm.
There are many different methods of measuring productivity change. Some of the methods include Growth Accounting, Stochastic Frontier Analysis (SFA) and Data Envelopment Analysis (DEA). The choice of any particular method depends on the availability of data and the purpose for productivity measurement.
Growth accounting is the most commonly used method for measuring productivity change. It is based on the assumption that output is produced using labour (L) and capital (K), and the relative contributions to output growth of labour and capital are and respectively. Productivity change can be obtained as the residual of subtracting *L + *K from the change in output. The growth accounting is attractive because of its simplicity though it requires several restrictive assumptions to hold. Among them is that product markets must be perfect so that the factor shares ( and ) reflect their respective marginal products. Agents are assumed to be maximizing and production equilibrium is reached under an optimal allocation of resources. However the main drawback of this method is that parameters are average values and if features of the firms are heterogeneous and the analysis attempts at highlighting those (e.g. heterogeneous technological change), growth accounting will be an inappropriate tool.
One way of avoiding the averaging problem in growth accounting is to use Stochastic Frontier Analysis (SFA). The approach is attractive because it constructs a frontier of efficient observations, which envelops the relatively inefficient observations. An important advantage of the method is that it is able to handle outliers and that hypotheses can be tested in the usual econometric way. The approach permits assessment of maximal output subject to input levels and therefore appears to be an output oriented measure. SFA is a base or non-orienting measure which means that the assessment of efficiency is not conditional on holding all inputs or all outputs constant. However, there are important drawbacks of this approach as well. The production function is assumed valid for all observations and technological change is the same for all observations. It also imposes an explicit functional form of the production function and distribution assumption on the data and therefore prone to mis-specification. A major criticism of the method is that it cannot adequately handle multiple inputs.
In contrast to the above two methods, Data Envelopment Analysis (DEA) does not require any assumption about the functional form of a production function or economic agent’s behavior. Furthermore there is no need to assume any specific distributional form of the error term and there is no need to assume perfect capital markets or optimal allocation of resources. Depending on the objectives of a study, DEA can either be input or output oriented. The input based measure defines the frontier by seeking the maximum possible proportional reduction in input use while output is held constant. The output based method seeks the maximum proportional increase in output production while the input level is fixed.
The most comprehensive measure of aggregate or sectoral productivity is Total Factor Productivity (TFP). TFP can be estimated using different techniques for instance Malmquist and Tornquist indices. Malmquist index has gained popularity since Fare et al (1994) applied the linear programming approach to calculate the distance functions that make up the Malmquist index. Since Data Envelopment Analysis can be directly applied to calculate the index, the Malmquist index has the advantage of computational ease. The method does not require information on cost or revenue shares to aggregate inputs or outputs. It is less data demanding and allows decomposition into changes in efficiency and technology. The method does not attract any of the restrictions of SFA assumptions. It is however susceptible to the effects of data noise and can suffer from the problem of unusual shadow prices when degrees of freedom are limited (Coelli and Rao, 2005).
Empirical Literature Review
This sub section presents empirical studies that have been done by researchers who have applied Data Envelopment Analysis in the manufacturing sector, the sugar industry and other related sectors.
Wadud (2007) assessed the productivity growth and efficiency change in the Malaysian Manufacturing industry. The study estimated productivity growth in the industry for the period 1983-1999. Malmquist Productivity Indices (MPIs) were computed using non parametric Data Envelopment Analysis (DEA) which showed that productivity growth was generated by efficiency change and growth in technology. The results of the study indicated that a majority of the industries operated with low levels of technical efficiency with little or no improvement overtime.
Tarimo (1998) did a study on sugar cane production, processing and marketing in Tanzania. It reviewed the agronomic, production and marketing aspects of sugarcane in Tanzania by focusing on the main factors associated with variation in production during the previous ten years. The results of the study showed that sugar cane production in Tanzania had shown reduced growth during the mid-1980s but showed slight improvement in early 1990s due to favourable economic situation in the country following trade liberalization. The study concluded that sustained sugar production in the country would depend on improved production technology, marketing and storage infrastructure at factories and regional centres. There was need to strengthen research in sugar cane industry to develop high yielding and disease resistant varieties.
Raheman et al, (2009), conducted a study on the Efficiency Dynamics of the sugar industry in Pakistan. The study analyzed the performance of 20 sugar firms using panel data and estimated the Malmquist total factor productivity growth indices using non parametric approach. Sales revenue was used as an output variable while cost of goods sold; operating expenses, total assets and shareholder’s equity were input variables. The results of the study showed that the overall sugar industry improved technological progress by 0.8% while managerial efficiency change put a negative effect on the productivity by the same percentage. As a result, the overall total factor productivity during the study period remained almost static with a decline of 0.1% indicating a divergent trend between TFP and its components. The research suggested that the sugar industry was facing serious productivity growth problems where no increase was recorded in total factor productivity during the period.
Amri (2013) analyzed the impact of competition on total factor productivity, efficiency and technical change in Tunisian manufacturing firms. The main aim of the study was to measure the impact of competition on productivity growth and its components (technical change and efficiency change). Panel data from the firms covering the period 1997 – 2012 was estimated using non parametric approach to generate TFP and its components. The results showed that, the main determinant of TFP was technical efficiency. Technical change which was a second source of TFP growth had a negative role on productivity growth up to the year 2000. It was found that competition had a strong positive effect on TFP growth, efficiency improvement and technological progress.
Oliveira, Bornia, Silveira, Oliveira and Drummond (2013) conducted a study to evaluate the production efficiency of sugarcane farming in Brazil using DEA. A total of 17 DMUs comprising (2010 – 2011) harvest were considered. The inputs used in the study were; land, raw material, costs of the harvest, loading and transport of sugar cane. The output was taken to be revenue from sugar cane sales. Results of the study revealed that there was a relationship between crop productivity and profit. Six DMUs which were found to be efficient among the 17 DMUs were considered as a benchmark for other DMUs.
Kumar (2014) investigated the efficiency measurement of Indian Sugar Manufacturing firms using the DEA approach. DEA was used to calculate the technical and scale efficiency measures of the public and private sugar manufacturing firms of the Indian sugar industry (2006 to 2010). The study used a sample of 43 firms which controlled a major portion of the market share. Total sales of the firm during the year and total profit after tax of the firm during the financial year were taken as the output variables while total cost of goods sold, total operating expenses and total assets held by the firm during the year were taken as inputs. Empirical results using a five year panel data showed that Indian sugar firms achieved an average technical efficiency of 86-90 per cent.
Nazmul (2015) assessed the production efficiency of sugar factories of Bangladesh using DEA. In measuring efficiency, the amount of sugar produced was used as the dependent variable (output) while MT of sugar cane crushed and crushing days were used as the input variables. The study results showed that 99.6 per cent of variation in the output variable was explained by the explanatory input variables. Using DEA under a CRS technology assumption led to an average production efficiency score of 0.97 in the sugar firms. This was an indication that on the average the firms could increase their output by 3 per cent with the existing level of inputs. This is according to investigation by Odhiambo, Nyangito and Nzuma (2004), as indication of sources and determinants of Agricultural Growth and Productivity in Kenya for the period 1965 – 2001. In furtherance, sources of growth were identified through growth accounting approach while the determinants were assessed using econometric techniques. The results of the study showed that most of the agricultural growth in Kenya was due to factor inputs – land, labour and capital. Ten per cent of the growth was not influenced by either factor inputs or total factor productivity for the entire period. Labour accounted for 48 per cent of the total growth and was the most important source of growth. Land was also explicated as a very important determinant of agricultural growth and productivity. The study further revealed that Kenya’s trade policy, climate and Government Expenditure were important determinants of agricultural total factor productivity growth.
Mulwa et al, (2009), estimated the productivity growth in small holder sugarcane farming in Kenya using Malmquist TFP decomposition. The output considered in the estimation was the sugarcane in MT per hectare. Five inputs were postulated to influence cane production and namely; seed-cane (MT per hectare), fertilizer (kilograms per acre); hired labour (man-days per acre); and land preparation (machine hours per acre). The calculations of the study included 95 farmers from three sugar zones, Mumias, Chemelil and West Kenya. The results of the study indicated that Mumias had a problem of continued use of old technologies which was causing declining technical change. As such, Chemelil had both technical change and efficiency problems. Farmers were not utilizing the available technologies fully hence declining efficiency level in addition to a downward trend in the adoption of newer sugar cane production technologies. The study indicated that one factor that was hindering the change in efficiencies and technical progress was the continued land sub division which brought with it diverse management styles.
Irungu, Wambugu, and Simon (2008), investigated the technical efficiency of sugar factories in Kenya, using the Stochastic Frontier Approach. The study measured technical efficiency overtime and explored sources of technical inefficiency of sugar mills in Kenya. A time varying translog stochastic production frontier was simultaneously estimated with inefficiency effects model based on panel data. The mean technical efficiency level of the five sugar factories was found to be 79.83%. This suggested that factories on average were 20.17% off the efficient frontier. The results also suggested that decreasing returns to scale (0.23) prevailed in the sugar processing industry. In addition, technical change was driving the sugar factories off the frontier at an annual rate of 1.25%. The results further suggested that capital-labour ratio, market share, sucrose content in sugar cane delivered and factory age had significant impact on technical inefficiency of the sugar factories.
Mulwa (2001) estimated the technical efficiency in sugar processing for the period 1980-2000 using Aigner, Lovell and Schmidt (ALS) 1977 model to investigate the impact of structural Adjustment Programmes (SAPs) on Mumias Sugar Company. Metric Tonnes of cane produced per year were the dependent variable while capital, labour, cane, chemicals power and fuel were the explanatory variables. The study used time series data. All the variables except cane were found to be experiencing diminishing marginal returns. Most of the variation in sugar output was explained by cane. A unit increase of cane was associated with 0.966 units increment in sugar in the SAP period and 0.964 in the pre SAPs respectively.
Mulwa et al (2007) investigated the impact of liberalization of the sugar industry in Kenya based on technical and scale efficiencies using DEA and SFA. The study mainly focused on Mumias sugar factory as a representative of the sugar industry. The results of the study indicated that liberalization had a negative impact on efficiency. Using DEA, the technical efficiency levels declined drastically from 100% in 1991 to lower levels of 85.4% in 1997 before picking up again while scale efficiencies largely remained unchanged. The SFA results showed a decline in technical efficiency between periods, 1984-1988 and 1992-1997 with a lowest of 88.5% being recorded in 1998.
Obange, Onyango and Siringi (2011), did a study on the determinants of sugar market performance under imperfect conditions using the industry competitiveness model in Kenya. The study investigated the effects of market factors on high prices which has influenced the performance of the sugar industry in Kenya. The results of the study indicated that price related factors significantly contributed to poor performance of the local sugar manufacturing firms under the then prevailing imperfect conditions in Kenya. The study recommended that diversification was crucial for the sugar industry if the sugar firms had to maximize revenues and become more competitive both at local and regional markets.
Waswa and Netondo (2014) did an investigation on integrating sustainability ethics in commercial sugar cane farming in Lake Victoria, Basin Kenya. The study focused on the trend in commercial sugar cane farming which was increasingly eating into areas traditionally known as Kenya’s food crop baskets. The paper recommended dynamic policy and legislation adjustments to control the expansion of sugar cane farming into areas known for their contribution to household and national food security. The sugar companies should contribute to sustainability concerns by integrating establishment of indigenous forest cover and food crop production in the nucleus estates.
The theoretical literature on productivity is diverse and describes different aspects of productivity. Productivity may be discussed in the production functions using inputs such as labour and capital. Productivity measures are intended to identify changes in the level of production that cannot be explained by changes in inputs or the characteristics of the original production process. Growth in productivity holding other factors constant implies improved performance. There are different methods of measuring productivity.
A review of empirical literature suggests that DEA is widely preferred technique in measuring comparative efficiency, productivity and performance across various firms and industries. The use of DEA in evaluating productivity and performance has been limited generally in Kenya and in particular to the sugar industry.
This study uses DEA to analyze productivity of the industry to identify causes of productivity changes through a Malmquist index decomposition which never been done. There are no reported productivity studies at the firm level using financial data for the sugar industry in Kenya. This study attempts to fill this gap by estimating the firm level total factor productivity and its components, for the sugar firms in the industry and to assess the variations in TFP growth between firms and over time.
This chapter describes the theoretical framework which explains the theoretical foundation, on which the study is based, stages of estimation, model specification, and description of variables, tools of analysis employed in the study and methods of data collection.
Data Envelopment Analysis (DEA) is a non-parametric technique which does not require specification of a particular form of a production function. DEA methodology was initiated by Charnes et al. (1978) who built on the frontier concept started by Farrell (1957). The methodology used in this study is based on the work of Fare et al. (1994) and Coelli et al. (1998). The DEA Malmquist Index is used to calculate the total factor productivity growth of sugar firms where each firm in the sugar industry is treated as a Decision Making Unit (DMU).
DEA differs from simple efficiency ratio in that it accommodates multiple inputs and outputs. It provides significant additional information about where efficiency improvements can be achieved and the magnitude of these potential improvements. DEA technique defines productivity measure of a production unit by its position relative to the frontier of the best performance established mathematically by the ratio of weighted sum of outputs to weighted sum of inputs. The estimated frontier of the best performance is also referred to as efficient frontier or envelopment surface.
This study uses the Malmquist total productivity (TFP) index to examine the productivity change in the sugar industry in Kenya. TFP change can be explained as the growth in output net of growth of inputs used. This study measures TFP based on the output –distance function according to Malmquist (1953). By using the output –oriented version of DEA, the study follows the approach of Fare et al. (1994) in calculating productivity change in the sugar industry in Kenya.
The Malmquist TFP index measures the change between two data points by calculating the ratio of distances at each data point relative to a common technology. Its major benefits are that price data are not required and the TFP indices maybe decomposed into two components; technical efficiency change (firms getting closer to the frontier) and technical change (shifts in the frontier itself).
Distance functions are used to define the Malmquist index. They allow one to describe a multi-input, multi-output production technology without the need to specify a behavioural objective (such as cost minimization or profit maximization). Distance functions can either be input oriented or output oriented. An input distance function describes the production technology by looking at proportional reduction in input usage; while output is held constant. An output distance function considers a maximal proportional increase in outputs, with inputs held constant. Under the assumption of CRS, the two measures will generate equal value while under VRS; the results will vary (Fare et al, 1994).
If period t technology is used as the reference technology, the Malmquist TFP change index between period s (base period) and period t can be written as
If the period s reference technology is used, then the Malmquist index will be defined.
The two indexes appear to be identical in the above illustration. To avoid arbitrariness in choosing the benchmark technology, Fare et al. (1992 and 1994) specify the Malmquist productivity Index as a geometric mean of the above two indices. According to Fare et al. (1994), the Malmquist output oriented TFP change index between period s (base period) and period t (the subsequent period) is calculated as follows.
In equation 3, represents the distance from period t observation to period s technology in which y represents output while x represents input. In the interpretation of Malmquist Index, when m is greater than 1 it implies that the TFP index has grown between periods s and t. If m is less than 1, it implies that TFP has declined. If m=1, then there is no change in TFP index.
According to Fare et al. (1994), the Malmquist productivity index can also be written in the following way;
When the Malmquist index is expressed in the above format, two important components are derived. The ratio measures the change in the output oriented measure of technical efficiency between period s and t. The ratio inside the bracket measures the technical change which is measured as a geometric mean in the shift in the production technology between the two periods. In the above model, the efficiency change (catching up effect) measures how much close a firm is to the frontier by capturing the extent of diffusion of technology or knowledge of technology use. The technical change (frontier effect) measures the shift or movement of frontier between two periods with regard to the rate of technology adoption or innovation.
3.5 First Stage Estimation
To investigate productivity change, we first derive the estimates of total factor productivity (TFP) growth. The Malmquist TFP index is estimated and decomposed into efficiency change and technical change. Equation 6 is the estimable equation in which the ratio outside the brackets measures the change in technical efficiency (EFFCH) between the periods’s and t. The geometric mean of the two ratios inside the square brackets captures the shift in technology (TECH) between the two periods evaluated at and . MPI is therefore the product of efficiency change and technical change
There are independent variables that may influence productivity in the industry but cannot be controlled by the firms. An OLS regression of TFP scores got in stage one is done on a vector of such variables. This is to explain the variation of the TFPCH scores derived from the first stage. A general form of the formula can be given as.
Where y is the TFPCH index and , is the vector of explanatory variables. The general relationship between TFPCH and the variables takes the following form;
Where is the market share, is the firm age, is the cane quality, is number of firms in the industry and is a dummy variable for the ownership structure where 1= privately owned sugar firms and 2= government owned sugar firms and μ is the error term.
3.7 Input and Output Variables
DEA can be applied to assess the performance of revenue producing firms. In stage 1 estimation, sales revenue and profit/loss after tax are used as the output variables while cost of goods sold, operating expenses and total assets as input variables.
The study uses panel data on financial performance of sugar firms in Kenya. Data for the study is obtained from secondary sources in the form of annual year books of statistics from the Kenya Sugar Directorate, Economic Surveys and other publications.
Table 2: Definition of Variables
|Sales Revenue||Output||Gross sales revenue (millions) in a year|
|Profit/Loss after tax||Output||Profit/loss (millions) in a year|
|Cost of goods sold||Input||Cost of goods sold in one year|
|Total assets||Input||Total value of firm assets in a year|
|Operating expenses||Input||Annual operating expenses|
|Market Share||Independent Variable||Ratio between a firm’s annual sales to industry annual sales.|
|Firm age||Independent Variable||Age of a factory as at the beginning of the study period (2004).|
|Cane quality||Independent variable||Factory performance compared to its rated capacity.|
|Size of the industry||Independent variable||Number of firms in the industry in a year|
|Ownership Structure||Independent Variable||Dummy variable 1= privately owned 2= government owned|
Data on sales revenue, profit/loss after tax, cane quality and total assets will be directly obtained from the KSB year books of sugar statistics. Data to calculate cost of goods sold, operating expenses and market share will be obtained from the KSB year books of sugar statistics. Information on ownership structure, firm age and size of the industry will be obtained from the KSB bulletins.
Amit, K.D. (2014), Efficiency Measurement of Indian Sugar Manufacturing Firms: A DEA Approach.
Amri, F. (2013), Impact of Competition on Total Factor Productivity, Efficiency and Technical change in manufacturing firms in Tunisia. A DEA Approach.
Baldwin, R. (1995), Productivity Growth, Plant Turn over and Restructuring in the Canadian Manufacturing Sector.
Banker, R.A., Emrouznejad, H.B., Lopes, A.L.M., M.R. de Almeida. (2012), Data Envelopment Analysis: Theory and Application: Proceedings of the 10th International Conference of DEA, August 2012, Natal, Brazil, 340pp, ISBN: 978 185449 437 5.
Banker, R.A., Emrouznejad, H. B., Alp, I.M., Ali, Cengiz. (2013), Data Envelopment Analysis and Performance measurement: Proceedings of the 11th International Conference of DEA, June 2013, Samsun, Turkey, ISBN: 978 185449 477 1.
Caves, D.W., Christensen, L.R., and Diewert, W.E. (1982). The Economic Theory of Index Numbers; Measurement of Input, Output and Productivity, Econometrica, 50:1393 – 1414.
Chambers, R.G. (1988). Applied Production Analysis: A Dual Approach, Cambridge University Press, New York.
Charnes, A., Cooper, W.W., and Rhodes, E. (1978). Measuring the Efficiency of Decision Making Units. European Journal of Operations Research, 2, 429-444.
Chisanga, B., Gathiaka, J., Nguruse, G., Onyancha, S., and Vilakazi, T. (2014), Competition in the Regional Sugar Sector: The Case of Kenya, South Africa, Tanzania and Zambia: Africa Competition Forum.
Coelli, T., Rao, D.S.P., and Battese, G.E. (1998). An introduction to Efficiency and Productivity Analysis. Kluwer Academic Publishers.
Coelli, T.J., Prasad, R., and Battese, G.E. (2005), An Introduction to Efficiency and Productivity Analysis, 2nd Edition, Springer, New York.
Cooper, N.W., Seiford, L.M., and Joe, Zhu. (2011), Data Envelopment Analysis (History, Models and Interpretation).
Diewert, W.E. (1992). The Measurement of Productivity, Bulletin of Economic Research, Blackwell Publishing, 44(3): 98 – 163.
Emrouznejad, A., Ibrahim, H.o., Abdel, L. Anouze. (2011), Performance Management and Measurement with Data Envelopment Analysis: Proceedings of the 8th International Conference of DEA, June 2010, American University of Beirut, Lebanon, 265pp, ISBN: 978 185449 4818.
Fare, R., Grosskopf, S., Norris, M., and Zhang, Z. (1994), Productivity Growth, Technical Progress and Efficiency Change in Industrial Countries, American Economic Review, 84:66-83.
Farrel, M.J. (1957). The Measurement of Productive Efficiency, Journal of the Royal Statistical Society, Series A, 120(3):253 – 281.
Gitau, M.W. (2009), Productivity of Commercial Banks in Kenya: Unpublished Masters Research Project Presented to the University of Nairobi, School of Economics.
Government of Kenya. (2010), Kenya Sugar Industry Strategic Plan (2010 – 2014): Enhancing Industry Competitiveness.
Irungu, P., Wambugu, A., and Simon, G.N. (2008), Technical Efficiency of Sugar Factories in Kenya: A Stochastic Frontier Approach.
Kenya Anti – Corruption Commission. (2010), Review of the Policy, Legal and Regulatory Framework for the Sugar Sub Sector in Kenya. A Case Study of Governance Controversies affecting the Sub Sector.
Kenya National Assembly. (Third Session, March – 2015), The Crisis facing the Sugar Industry in Kenya. An Adopted Report of the Departmental Committee on Agriculture, Livestock and Cooperatives. Clerk’s Chambers, Parliament Buildings, Nairobi.
Kenya Sugar Board. (1998 – 2013), Year Books of sugar statistics. Kenya Sugar Board, Nairobi
Koopmans, T.C. (1951). An Analysis of Production as an Efficient Combination of Activities.
MAFAP (Monitoring African Food and Agricultural Policies). (2013), Analysis of Incentives and Disincentives for the Sugar Sector in Kenya.
Magati, P.O., and Muthoni, Ng’ang’a. (2012), Sectoral Productivity Awareness in Kenya. S
Mokhtarul Wadud., (2007), Productivity Growth and Efficiency Change in the Malaysian Manufacturing Industry: Recent Evidence from Disaggregated Data.
Mulwa et al. (2009), Productivity Growth in Smallholder Sugarcane Farming in Kenya: Malmquist TFP Decomposition.
Mulwa et al. (2007). Impact of Liberalization on efficiency and productivity of the sugar industry in Kenya. Nairobi.
Mulwa, R.M. (2001), Technical Efficiency in Sugar Processing. What difference have SAPs made in Mumias Sugar Company? Unpublished Masters Research Project presented to the University of Nairobi.
Mwanaongoro, S., and Imbambi, R.M. (2014), Assessment of Relationship between Plant and Equipment Maintenance Strategies and Factory Performance of the Kenya Sugar Firms.
Nazmul, I. (2015), Production Efficiency of sugar Factories in Bangladesh: An application of Data Envelopment Analysis.
Nishimizu, M., and Page, J.M. (1982). Total Factor Productivity Growth, Technological Progress and Technical Efficiency Change: Dimensions of Productivity Change in Yugoslavia, 1965 – 1978. Economic Journal 92: 920 – 936.
Nyongesa, F. (2005), Total Factor Productivity Growth in the Garment Industry in Nairobi: Unpublished Masters Research Project Presented to the University of Nairobi, School of Economics.
Nyoro, J.K. (2012). Agricultural and Rural Growth in Kenya. Tegemeo Institute, Egerton University.
Obange, N., Onyango, G.M., and Siringi, E.M.(2011), Determinants of Sugar Market Performance under Imperfect conditions: Empirical evidence from Kenya.
Odhiambo, W., Nyangito, H., and Nzuma, J. (2004), Determinants of agricultural Growth and Productivity in Kenya (1965-2011): KIPPRA Working Paper Series.
Oliveira, T.B.A., Bornia, A.C., Silveira, S.F.R., and Drumond A.M. (2013), Production Efficiency in Sugarcane Farming in Brazil. A Data Envelopment Analysis Approach.
Otieno, O., Kegode, P., and Ochola, S. (2003), The Challenges and way forward for the Sugar Sub Sector in Kenya.
Otiwa. J.O.(2013), Analysis of Consumer Demand for Sugar in Kenya (1981-2010): Unpublished Masters Project submitted to the School of Economics, University of Nairobi.
Raheman, Abdul and Talat,(2009), Efficiency Dynamics of the Sugar Industry in Pakistan.
Shi-Hsun, H., Ming, M. Yu and Ching- Cheng, Chang. (2003): Analysis of Total Factor Productivity Growth in China’s Agricultural Sector.
Tarimo. (1998). Sugar cane Production, Processing and Marketing in Tanzania.
Waswa, F., and Netondo, G.W. (2014), Integrating Sustainability Ethics in Commercial Sugar Cane Farming in the Lake Victoria Basin, Kenya.
Yin, R. (1999), Production Efficiency and Cost Competitiveness of Pulp Producers in the Pacific Rim, Forest Products Journal, 49: 43-49.
Source: Kenya Sugar Directorate Year Book of Statistics 2013