Farmers’ Goals and Efficiency in the Production of
Sugar Cane:
The Philippine Case
M. Dina Padilla-Fernandez and Peter Nuthall
Research Report 07/2001 August 2001
Farm and Horticultural Management Group Lincoln University
ISSN 1174-8796
Farm and Horticultural Management Group
The Farm and Horticultural Management Group comprises staff of the Applied Management and Computing Division at Lincoln University whose research and teaching interests are in applied and theoretical management and systems analysis in primary production.
The group teaches subjects leading to agricultural/horticultural commerce and science degrees, though the courses offered also contribute to other degrees.
The group is strongly involved in postgraduate teaching leading to honours, masters and PhD degrees.
Research interests are in systems modelling, analysis and simulation, decision theory, agnbusiness and industry analysis, business strategies, employment relations and labour management, financial
management, information and decision systems, rural development and also risk perceptions and management.
Research Reports
Every paper appearing in this series has undergone editorial review within the group. The editorial panel is selected by an editor who is appointed by the Chair of the Applied Management and Computing Division Research Committee.
The views expressed in this paper are not necessarily the same as those held by members of the editorial panel, nor of the Group, Division or University. The accuracy of the information presented in this paper is the sole responsibility of the authors.
Copyright
Copyright remains with the authors. Unless otherwise stated permission to copy for research or teaching purposes is granted on the condition that the authors and the series are given due acknowledgement.
Reproduction in any form for purposes other than research or teaching is forbidden unless prior written permission has been obtained from the authors.
Correspondence
This paper represents work to date and may not necessarily form the basis for the authors' final conclusions relating to this topic. It is likely, however, that the paper will appear in some form in a journal or in conference proceedings in the future. The authors would be pleased to receive correspondence in connection with any of the issues raised in this paper. Please contact the authors either by email or by writing to the address below.
Any correspondence concerning the series should be sent to:
The Editor
Farm and Horticultural Management Group Applied Management and Computing Division PO Box 84
Lincoln University Canterbury NEW ZEALAND
Email: [email protected]
Farmers' Goals and Efficiency in the P r o d u c t i : ~ n ~ f ~ uWMRSITV The Philippine Case ' LIBRARY
% 7 AUG 2001
Abstract
This research evaluated the link between Philippine Sugar cane farmersJ goals, values and attitudes (and some selected efficiency-related variables) with production efficiency.
The analysis was based on both information from informal interviews and formal primary data
collection. The Data Envelopment Analysis technique was used to determine the relative efficiencies of individual farmers and to identify the major factors that influence the efficiency of production. Pure technical, scale, overall technical, allocative and economic efficiency measures were derived for the sample of sugar cane farmers from the Central Negros area, The Philippines. Under the specification of variable returns to scale (VRS), the mean pure technical, scale, overall technical, allocative and economic efficiency indices were 0.7580, 0.9884, 0.7298, 0.7941 and 0.6025, respectively.
The farmers' characteristics and their associations with goals and attitudes were determined. The result shows that 'per cent of land owned' is correlated with farmers' decision-making and thus their production efficiency.
The study was unique in that it incorporated the farmers' values and attitudes towards farming and production efficiency. The Bootstrap regression method was used to determine the factors affecting the variations in farmers' efficiency. Factors positively associated with production efficiency include farm experience, exposure to extension and off-farm work; for goals and attitudes - the intrinsic independence goal, the instrumental aspects of farming, leisure orientation, optimistic attitude, and risk consciousness were all associated with efficiency.
The key policy options that must be considered for addressing inefficiencies include education and extension advice, developing the importance of the instrumental aspects of farming, developing group (block) farming as well as farmers' and millers' cooperatives, improved access to credit and improved technology (with emphasis on soil and fertiliser management and the use of improved varieties).
Key words: Data Envelopment Analysis, technical efficiency, scale efficiency, allocative efficiency, economic efficiency, frontier efficiency analysis, farmer's goals, sugar cane farmers, Philippines.
1 irstroducticsn
A very important aspect of farm management is the setting of goals and objectives. Goals are commonly referred to as objectives or aspirations for which a person has decided to strive. (0hlmer et al., 1998).
Basically, a person's choice of goals is influenced by hislher values and beliefs (Gasson, 1973). Values refer to a person's view of the goodness of objects, results, and situations.
They also express one's needs and motives; goals and objectives express the means to follow those values (0hlmer et al., 1998). Beliefs describe what people think is reality. A belief involves mental conviction, acceptance, confidence, or faith that a proposition is true.
Beliefs influence values. Values also influence beliefs (Gasson, 1973).
In farming, economic goals such as profit or output maximisation may be the farmers' primary goal, however non-economic goals are also important. This is usually the case on family farms when the unit of production is both family and business enterprise-based. This
* M.D. Padilla-Fernandez is at the Sugar Regulatory Administration-Department of Agriculture, Quezon City, The Philippines and Peter Nuthall is at the Applied Management and Computing Division, Lincoln University, Canterbury, New Zealand.
dual economic farming activity within the household causes a unique form of decision making as choosing a particular farm goal may preclude the choice of a family goal or
vice
versa.Fundamental decisions have to be made, say, in the allocation of family members' time between competing activities- production and 'reproduction' in the farm business, off-farm work and leisure.
Goals give purpose and direction to decisions and performance, therefore they must be defined and serve as a measure of farm success or failure (Osburn and Schneeberger, 1983). These factors have implications for farmers' efficiency analysis (Ellis, 1988).
Efficiency is an important characteristic in judging the performance of farmers. Production efficiency is the ability of the farmers to produce an output at minimum cost (Kopp, 1981) and to combine outputs for maximum profit. It may be divided into technical and allocative
efficiency. Farmers are technically efficient if they produce the maximum obtainable level of output from given inputs and technology. Farmers are allocatively efficient if their production inputs and output combinations give maximum profit for given resources. That is, farmers are allocatively inefficient due to their sub-optimal combination of inputs (Farrell, 1957, in Torkamani and Hardaker, 1996 p.76) andlor due to their inefficient combination of outputs.
Efficiency studies can determine if farmers could possibly raise productivity. If the farmers are not producing efficiently, their production can be increased through improved
management practices by transferring the experiences of the efficient farmers to the less efficient ones. If the farmers are operating efficiently, their production can be increased through the adoption of new technology (Abate, 1995), when available.
However, some farmers may be reluctant to move out of the less productive technology. This kind of attitude' towards new practices may develop from the value that the farmers place upon farming. If they value farming in its current form as a way of life, they may accept relatively quickly those practices they perceive as conserving their own time and physical effort, and possibly resist practices they perceive as involving pressure to reorganise their farm business, and change their way of life.
Therefore, differences in personal, family and farm business goals may be considered in determining the factors as to why one farmer is more economically efficient than the other one. It can be argued that in the decision-making process of a farmer, a combination of economic, sociological and psychological considerations are all factors. Thus, research on farm efficiency may be more substantive if it can be seen from these perspectives rather than from the simple assumption of a single profit directed objective.
2 Historical Background of the Philippine Sugar Industry
The values and beliefs of the sugar cane farmers,* and how these attributes have interacted with the situations in the industry are also important factors that may have influenced the farming systems and therefore production efficiency. Therefore, any attempt to improve efficiency in the industry must allow for all these factors which are reviewed below.
1 'Attitudes are orientations toward or away from some objects, concepts, or situations, and a readiness to respond in a predetermined manner to these or related objects, concepts, or situations. The formation of attitudes may be attributed to such factors as culture, the home and family within a culture, and the individual's social interactions with formal and informal groups. Attitudes are, in part, determined by the culture in which the individual is reared' (Andres, 1981 p.104). Many different beliefs and values may underpin an attitude. This may be the cause of farmers' reluctance to move out of the less productive technology.
or the industry
-
accepted title- planters or hacenderos who are generally the owners of the land [Aguilar, 19841. Today,hacenderos can be lessees who pay rent in cash or equivalent to 18 to 20 per cent of the sugar produced.
During the Spanish regime, the colonial government encouraged the people to produce sugar by offering incentives e.g., easy credit, milling equipment and access to untitled lands.
At the farm level, the hacenderos provided, in varying degrees, credit, animals, equipment, houses, medical fees, clothing and money for social obligations to labourers to ensure and control them to stay on the farms. Thus the paternalistic management styles in the sugar industry developed.
The rise of sugar as an export crop attracted many foreigners and merchants into sugar cane farming. Foreigners and merchants became the new breed of planters, some of the pioneers in sugar farming became sharecroppers and others who lost their land to the new breed of planters became the hacienda (farms) labour forces. This had led to the development of the different management styles. As Billig (1994 p.664) noted, 'the proliferation of these people with different socio-economic and cultural backgrounds had generated the 'hustlers,
profiteers and entrepreneurs' in the sugar industry'.
During the American rule, the centrifugal mills were installed and the farms were grouped into mill districts. Consequently, a production quota system was imposed for each mill district. This arrangement organised the industry in a highly non-competitive way. Since sugar was transported entirely by rail, a planter could not choose to send the cane to a different mill (Quirino, 1974 in Billig, 1993). This impacted the entrepreneurial skills of the farmers as they had no incentive to increase andlor improve their production.
Meanwhile, the modernisation process changed the agrarian structure in the industry as consolidation of more wealth and power within the industry occurred. These factors
contributed to the planters' banding together into a 'sugar bloc' that gave them influence and protection, including politically.
During the Post-war years the insurgency in the countryside grew and this started the
'landlord absentee' type of management that generally caused production inefficiencies. The change in residence by the planters minimised the supervision on the farms and also
aggravated the weak relations among the sharecroppers and farm labourers which lessened the quality of the performance of the farms. This absentee landlordism doubled during the 1984 international sugar crisis as leftist insurgents grew increasingly bold in the countryside.
Over the 1950s, 60s and 70s, the dependency of the Philippines on the Americans deepened as the country was granted a high U.S. sugar quota until 1974. During these years, sugar producers enjoyed an improved lifestyle. However, had they invested money in their farms and mills, instead of relying on the sure U.S. market and protection from the government, they may not have felt the effect of the sugar crises from 1978 till the mid-1980s, and they may not have opposed the GATT policy.
In 1974, Martial Law was imposed and placed the sugar industry under government monopoly. The planters had their political powers reduced although a few became more influential because of their political connections. Later, sugar was changed to a free-trade enterprise and this should have been the incentive to increase production efficiency.
However, as the industry became more exposed to the domestic and international competition, planters became more vulnerable. Some left the business.
Many problems beset the industry and these affected the performance of the farmers as well as the people in the different sectors of the industry. The Comprehensive Agrarian Reform Program (CARP) is one policy that lessens the incentive to invest in the sugar industry and to pursue productivity gains. The CARP lowers the collateral value of agricultural land and this further reduces the flow of credit to agriculture. The fragmentation of land into small farm sizes countewails efficiency in farm production.
Currently a re-distribution of sugar land is occurring due to the full implementation of the CARP. The changed agrarian structure and the implication of the paternalism resulted in a large gap between big and small sugar cane farmers' performance. The big planters blamed the low production and productivity on the inefficiency of small planters. Considering big planters own and control around 71 per cent of the country's total sugar land means much of the cause lies with them. As big planters have access to information, modern technology and extension services, they should produce efficiently. Small farmers, on the other hand, with sub-marginal land areas, might be expected to produce sub-optimally. But as sugar cane is all they are familiar with, to them sugar is profitable as it entails far less risk compared to other crops. Moreover, it is a crop that does not require intensive care.
There are still some large sugar haciendas in the Philippines, but the trend now is toward smaller holdings. Apart from the CARP, one of the reasons is the 'natural land reform,' which is how landowners refer to the process of partitioning through inheritance, and because of this, different types of farmers and farm management styles exist. Billig (1 993) described the sugar cane planters of today in this manner:
'There are still many sugar planters in the Philippines who are knowledgeable and dedicated agriculturalists. These planters spend time on their farms, know the workers personally, and are involved in day-to-day management. Such haciendas tend to be the most productive and humane ones. But there are also many planters, perhaps the majority, who know little about agriculture. These are typically the ones who spend little time on the farm, delegate all responsibility to encargados, have no personal relationship with the workers, and concern themselves only with the expenditure and income of the farm. Some of these are educated professionals who inherited haciendas but prefer to devote their attention to their careers.
Others simply do not like farming and prefer golfing, socialising and travelling
-
all supported by profits from the hacienda. They are planters because they inherited land rather than because of any interest, competence or labour of their own' (p.131).3 Background of the Problem
Arguments against increasing the number of farms of reduced size units through the Comprehensive Agrarian Reform Program
CARP)^
in the Philippines are based o n the premise of economies of farm size and the prevention of investments in the commercial crop sector thus disrupting production of cane. It is argued that large farms perform betterbecause there is the opportunity for the optimal utilisation of resources. The use of modern machines like tractors and harvesters is considered to be more appropriate and economically efficient given large farms. It is also argued that land reform may lead to the beneficiaries using a large percentage of the earnings for consumption rather than investments. Such a reduction in investment ultimately leads to a decline in production with a further negative effect on revenue (Putzel and Cunnington, 1989).
Perhaps the most legitimate argument of the landed class relates to the management skills and attitudes of the land reform beneficiaries. As Putzel and Cunnington (1 989 p.25) described, 'tenantlfarm workers in any case do not have the skill and wherewithal required for cash crop production.' They are intrinsically lazy and should not, therefore, be entrusted to own land of their own (Hayami et al., 1990).
Eduardo Locsin, who implemented land reform in his own hacienda (plantation), also believed the major cause of failure of his experiment was the mentality and management styles of the people (McBeth, Far Eastern Economic Review, 25 January 1990). Ledesma and Montinola (1 988 p.38) expressed this sentiment succinctly: '... there is one hindrance to
The CARP was aimed at establishing owner cultivatorship of economically sized farms that would ultimately improve productivity and provide equity among the farmers, tenants and farm workers.
an effective land reform programme, it is the persistence of the dependency mentality among the potential beneficiaries
...
they do not have the initiative to look for their own solutions to their problems and difficulties.'It can be inferred that the fixed social arrangements developed in the hacienda by the
landlords centuries ago (when labour had the bargaining power and the sugar industry had a favourable world market) had caused the shadow of paternalism and dependency in the lives of the farming people. These formed the farmers' framework of values and these also
negatively influence their farming performance.
Locsin gave a more positive comment: 'It all comes down to a fear of the unknown and an ingrained lack of self-confidence.' He added: 'the worker was reduced to an automaton. He can only think in terms of 24 hours. He doesn't budget because he can always ask for an advance.' He added: 'they don't know how to own anything because the hacienda always did it For them. They would even come to my wife and ask her how they should name their babies' (McBeth, Far Eastern Economic Review, 25 January 1990 p. 26).
This also confirms the findings of Ledesma and Montinola (1 988). The first year of the land reform programme did not show positive changes and its beneficiaries said they preferred having a landlord.
In 1997, a Presidential Task Force was created to determine the cause of low sugar
production. The Committee noted the large number of small farmers (84 per cent) and their performance. Small farmers have a lower yield (46.45 tonnes cane per hectare as against 61.37 for large farmers) and in terms of total cane production, only 25 per cent comes from them. The Committee reported that the fragmentation of farms had resulted in inefficient and
many uneconomic farm sizes and further noted that some aspects of economies of scale in sugar cane farming are difficult to overcome. For larger farms the Committee also reported that CARP had decreased farm investments due to uncertainties in land ownership and valuation (Report of the Presidential Task Force on the Sugar Industry, Sugar Regulatory Administration, 1997).
The performance of the small vis-a-vis large farms and the trend towards increasing the number of the small farms due to the full implementation of the CARP have important implications in the production of sugar. The current economic reform of moving towards a free world market economy (e.g., GATT and AFTA) would make sugar cane farmers
uncompetitive if exposed to world market prices. The Philippine sugar production cost of US- 28.60 cents a kilogram is unprofitable, as the world sugar price is 29.04 US-cents a kilogram.
The 19.8 to 22 US-cents per kilogram production costs of Australia, Brazil and Thailand (Sugar Letter, Sugar Y Azucar, 1996) means it would be cheaper to import all supplies, but the balance of payments would be affected.
The Philippines has 18 sugar-producing provinces, 36 (out of 41) operational mills and 16 refineries. Around 556,000 workers are employed directly on the farms and around 25,000 in the processing plants. About 5 million people are dependent on sugar for their economic existence. There are around 348,000 hectares of sugar cane land and approximately 37,000 sugar cane farmers. (Report of the Presidential Task Force on the Sugar Industry, Sugar Regulatory Administration, 1997).
In view of the vital contribution and role of the industry to the Philippine economy, the production of sugar must be given proper support by the government if it is to be made economic and for the farmers to improve their efficiency markedly. The problem is to decide whether these farmers can operate at a level of economic efficiency that will ensure their future survival. Agricultural policy makers must consider whether support and restructuring will achieve these ends.
Knowledge of the productive efficiency of sugar cane farms will indicate whether agricultural production under the present conditions can be increased without the use of high investment capital. This study investigates whether farmers are efficient in their resource utilisation, whether their decision behaviour is rational in an economic sense, and the significance of non-economic goals as well as their associations.
A knowledge of farmers' goals, values and altitudes, say in the maintenance of traditions, or valuing leisure more than work, are necessary in understanding the efficiency variations. In this study, the farmers' goals, values and attitudes are determined and along with the socio- economic variables, they are related to farmers' efficiency levels to identify relationships that have implications on farmers' productive efficiency. This will allow a more effective sugar industry policy to be formulated.
4 Conceptual Framework
Osburn and Schneeberger (1983 p.7) believed that 'it is in this managerial gap' which distinguishes why some agricultural businesses have grown and prospered, while other similar ones have failed and gone out of the business.'
Physical resources are not productive unless they are organised and co-ordinated effectively.
The management may be provided by a single individual- the farmer, or by a hierarchy of decision-makers in a corporate farm. 'How that decision-maker will react in a given situation, basically how helshe thinks, can be viewed as a psychological question in an economics context' (Howard, 1997 p.39).
One instrument that has been used to distinguish a number of psychological characteristics of farmer's is their value system4 (or orientations). Gasson (1 973) developed these
orientations from Cambridgeshire and Suffolk farmers by examining the goals and values that the sample farmers held. An instrumental orientation implies that farming is viewed as a means of obtaining income and security with pleasant working conditions. Farmers with a predominantly social orientation are farming for the sake of interpersonal relationships in work. Expressive values suggest that farming is a means of self-expression or personal fulfilment while an intrinsic orientation value means that farming is valued as an activity in its own right.
Many characteristics may be associated with value orientations. A case could be made for including human capital investment (e.g., education, farming experience, association with a particular farm, attendance in seminarsltraining) and socio-economic characteristics (e.g., age, family size, off-farm work, type and size of farm, income level and indebtedness and so on) [Gasson, 19731.
Pemberton and Craddock (1 979 p.23) found in the Carman region of Manitoba in Canada that 'high income farmers are more oriented to economic and monetary goals and have higher levels of aspiration than the lower income farmers', who seem to be more oriented to economic survival. Scales (1 990) also found that top New Zealand farmers emphasised maximum profit as important for success, while average income farmers emphasised getting
4 Goals and values can predict human behaviour by determining where they stand relative to one another. They are organised in systems or value orientation. Variation in the rank order of common value components, all of which may be present, cause value systems (or orientations) to differ between individuals and between sub-groups of society. That is, people desire to achieve all valued ends but in situations where these are mutually exclusive, it is the relative ordering of values which
determines how they decide to act or perform (Gasson, 1973). A value system, therefore, is a method that people use to solve their problems, and cope with their environment (Andres, 1981).
an average standard of living. Not all farmers aspire to be top producers (Fairweather and Keating, 1990).
Perkin and Rehman (1 994) correlated the 'monetary', 'lifestyle' and 'independence' goal components with the socio-economic status of British farmers and the main findings were: (1) age and education were related to 'lifestyle', (older people are more likely to want to remain on the farm and less likely to want time away to do other things. The converse is true for those respondents who have received formal higher education at a college of agriculture or university); and (2) long-term debt was related to 'independence', particularly the purchase of land, that is, the 'ownership of property.' The lifestyle component exerts an influence over the level of intermediate debt.
Therefore, farmers' human capital investment and socio-economic characteristics may be hypothesised to be related with farmers' value and attitude orientations, and thus, may also influence farmers' management or decision-making skills.
Some researchers e.g., Salamon (1985) and Salamon and Davis-Brown (1 986), Ploeg (1 985), Fairweather (1 987), Olsson (1 988) and Pomeroy (1 987) used these value
orientations to determine the psychological characteristics of the farmers that are most likely to survive an economic downturn (Table 1). The asterisks beside the management styles listed in Table 1 are those thought to be most likely to survive an economic downturn.
Olsson and Pomeroy both see the entrepreneur as best suited to adapt to changes in primary production, and Salamon and Davis-Brown believe that cautious production best suits an economic downturn. However, Ploeg notes that each style can be economically successful while Fairweather leaves open the issue of which strategy is best for survival and notes that there is no consensus on which management style best equips a farmer to survive an economic downturn (Fairweather, 1987). Thus, the literature is not particularly helpful.
Table 1. Different management styles.
Management Styles Source Country
Entrepreneur Yeoman * Salamon, 1985 US
Salamon & Davis-Brown, 1986 Extensifier * Intensifier * Ploeg, 1985 Italy Financial Manager Individualist Worker
Productivity lncreaier Lifestyler Fairweather, 1987 New Zealand
Entrepreneur* Cautious Strategist Olsson, 1988 Sweden
Accumulator * Sufficer Pomeroy, 1987 New Zealand
* Farmers considered most likely to survive an economic downturn.
Source: Fairweather and Keating (1 990)
However, little is known about the efficiency of the farmers under each management style.
Some studies equate efficiency with survival and describe who is best equipped to suwive in the long run.5 Therefore this study tried to determine the efficiency of the farmers b y
measuring their production efficiency and relating this to their management style.
Production efficiency may also be associated with human capital, socio-economic characteristics, farm environment and the adoption of technology by the farmers. For example, technical inefficiency may be explained by factors such as the use of an obsolete production technique, or the inappropriate operation of a modern one. This may b e due, for instance, to a lack of technical information or the poor organisation of production tasks.
A farm, which survives and is able to make acceptable profits in a competitive world, is likely to be considered efficient in some sense. Profits are one measure of this. Another measure is the ability and willingness of the farm to make new or expand investment [Sheperd et al. (1983) Microeconomic Efficiency and Macroeconomic Performance, in Silberston (Ed.)].
Therefore, efficiency is expected to be related to variables such as the education and technical skills of the farmer, and possibly age (Hallam and Machado, 1996). A number of variables might at least explain part of the differences in efficiency between and among farms.
Battese et al. (1996) considered the age of the primary decision-maker, the maximum years of formal schooling for members of the household, and the ratio of adult males to the
household size, as explanatory variables to the inefficiencies of production of wheat farmers in the four districts of Pakistan. They found that in one district, age and schooling of farmers are significantly related to the efficient production of wheat.
Factors like farm size,. credit availability and extension contacts were also introduced to explain the causes of farm inefficiency (e.g., Kalirajan and Flinn, 1983, Lingard, Castillo and Jayasuriya, 1983). Meanwhile, Parikh and Shah (1 994) added the value of farm assets and the degree of land fragmentation to determine the variations of technical efficiency in the North-West Frontier Province of Pakistan. The analysis suggested that younger farmers with easier access to credit, more education and larger assets are most likely to operate
efficiently. They further suggested that increased education and availability of credit along with land consolidation would lead to improvements in efficiency.
The relation between efficiency and farm size has received the most attention in the literature (e.g., Britton and Hill, 1975; Pasour, 1981 ; Abate, 1995; Piesse, 1996; Adesina and Djato, 1996; and Tadesse and Krishnamoorthy, 1996). Yet, there is no consensus among the available studies on the age-old debate of efficiency differences in the small vs large-scale farm (Tadesse and Krishnamoorthy, 1996).
The conceptual paradigm proposed is presented in Figure 1. It is assumed that the
independent variables on the left of the diagram exert a certain influence directly or indirectly on the criterion variable. Each arrow in the figure represents a presumed path of influence.
Farmer's goals, valuesand attitudes
Farm and farmer's characteristics Human Capital
1. Years in school
2. Years of farming experience 3. Exposure to extension
Socio-economic Characteristics 1. Age
2. Household size 3. Off-farm work 4. Tenurial status 5. Access to credit 6. Farm size
Farm Environment 1. Location
2. Topography 3. Soil types
Adoption of Technology l .Improved variety
2.Application of fertiliser
technical, scale, overall
technical, allocative and
,
economic efficiency 1 levelsFigure 1. A schematic paradigm showing the associations of farmer's goals, values, attitudes, some selected-efficiency variables and farmer's production efficiency levels.
5 Analytical Framework
Data envelopment analysis (DEA), introduced by Charnes, Cooper and Rhodes (1 978) and further formalised by Banker, Charnes and Cooper (1 984) is a non-parametric approach that estimates 'production' technologies and thus measures efficiencies in production using the observed inputs and outputs of the sampled farms.
In this study, the individual technical and scale efficiency levels were derived and analysed using ou'rput-based DEA frontier analysis, while in the allocative efficiency measurement, the input-based DEA frontier was used. DEA models were estimated using the warwick6
Windows DEA program. The DEA linear programming models are specified below. Each linear programming problem is solved separately for each respondent in the database.
Under output maximisation and variable returns to scale (VRS) the software solves models A1 .l and A1.2 below.
Model A l . 1
Min Z = t Q l - Q 2
Where Xij and yrj are the ith input and the output level at DMUj, Jo is the DMU being assessed, the U, are weights associated with the outputs, the vi associated with the inputs, and are the unknowns to be solved for.
Let Z* be the optimal value of Z in the above model. The minimum and maximum limit of the
Cl
range are obtained by solving in turn the following two models:Model A1.2
M i n l M a * . Q , - Q ,
Notation is as in model A I . l
'Warwick Windows DEA User's Guide, 1996.
Under input minimisation and variable returns to scale the software solves models A2.1 and A2.2 below.
Model A2.1
Max Z -
zryrj0
+ Q , - Q 2Where Xij and y, are the ith input and the output level at DMUj, Jo is the DMU being assessed.
Let Z* be the optimal value of Z in the above model. The minimum and maximum limit of the Cl range are obtained by solving in turn the following two models:
Model A2.2
Min1Ma.x Q , - Q ,
Notation is as in model A2.1
The efficiency calculated from the VRS model is pure technical efficiency (PTE). Thus, PTE
= 1 indicates that the farmer is in the frontier and is pure technically efficient and PTE < 1 indicates that the farmer is inefficient. Banker, Charnes and Cooper (1 984) extended the original model [the Charnes, Cooper and Rhodes- CCR model or constant returns to scale (CRS) model
'1
to disentangle the effect of scale efficiency and showed that the CCR overall technical efficiency measure can be regarded as the product of the technical and scale efficiency measures (Banker and Thrall, 1992). Thus, scale efficiency is the ratio of the overall technical to the pure technical efficiency score.Economic efficiency can be calculated as the product of the overall technical (CRS) and allocative efficiency (VRS).
'This can be calculated in the DEA software by selecting the CRS model.
6
The
B00lZ~Erap Multiple Regression ModelsThe Bootstrap method was used to overcome the dependency problem in the DEA efficiency scores (they are relative to each other - and thus dependent). The technique applied was with sampling replacement so that there was a probability that a certain unit would be sampled again (in this study the probability was 111 27). The efficiency scores for each bootstrap sample of size n was recalculated i.e., the DEA model was re-run 100 times.
A bootstrap multiple regression technique was used to obtain the parameters of the models designed to account for the variations in the productive efficiency of the farmers. The models were specified on the basis of important considerations such as suspected collinearity
among the independent variables, particularly the goal factors. Therefore, the total average score (i.e., the item scores or the goals measured using the Likert scale), calculated from each goal and behaviour component, were introduced first along with the other explanatory variables (Model l a). (See below for the variable definitions)
Model l a
8
= a.+
blEDUC+
b 2 EXPE+
b 3 EXTN+
b 4 AGE+
b5HHSZE+
b 6OFFWORK
+
b7Dow+
b8Dcrdt+
b gDbac+
b IoFLAT+
b 1SROL+
b12ROL
+
b13CLAY+
b14SCbAY+
b15SANDY+
bI6N+
bI7P+
bI8K+
blgD~yv+ b2oGOALI
+
b21GOAL2+
b22GOAL3+
b23GOAL4+
b24GOAL5+
b31ATTII
+
b32ATT12+
b33ATT13+
634 ATTI4+
b35 ATT15+
b36 ATTI6+
U*In Model I b, factor scores from a factor analysis (see ssection 9) were used instead of the item scores in explaining variations in farmers' efficiency. Thus,
Model I b 0 = a.
+
blEDUC+
b 2 EXPE+
b3 EXTN+
b 4 AGE+
b5HHSZE+
b60FFWORK
+
b7Dow+
b8Dcrdt+
bgDbac+
b loFLAT+
bllSROL+
b q2ROL+
b 13CLAY
+
b i4SCLAY+
b 15SANDY+
b 16N+
b 17P+
b I8K+
b tSDHY"+
b20GOAL1
+
b21GOAL2fs+
bZ2GOAL3fs+
b23GOAL4fs+
b24GOAL5fs+
b 3 1 A T T l l f ~
+
b32ATT12fs +b33ATT13fs+
b34 ATTI4fs+
b35 ATT15fs+
b36 ATTI6fs
+
U*In Model 2a, the total average score (i.e., the item scores or goals measured by ranking) calculated from each goal and behaviour component is included along with the other explanatory variables.
Model 2a 0 = a.
+
blEDUC+
b 2 EXPE+
b3 EXTN+
b 4 AGE+
b5HHSZE+
b60FFWORK
+
b7Dow+
b8Dcrdt t- b9Dbac+
blOFLAT+
bllSROL+
b12ROL+
b13CLAY
+
b14SCLAY+
b15SANDY+
b 16N+
b17P+
bI8K+
blgDHyv+
bZ5RGOAL1
+
b26RGOAL2+
b27RGOAL3+
b28RGoAi-4+
b2gRGOAL5+
b 30RGOAL6+
b 3IATTl 1+
b32ATT12+
b 33ATT13+
b 34 ATTI4+
b35 ATTI5+
636 ATTIG
+
U*While in Model 2bJ factor scores (see section 9) were used instead of the item scores (in goal ranking) in the estimation. Thus,
Model 2b 0 = a.
+
blEDUC+
b 2 EXPE+
b 3 EXTN+
b 4 AGE+
b5HHSZE+
b60FFWORK
+
b7Dow+
bsDcrdt+
bgDbac+
blOFLAT+
bllSROL t b12ROL+
b13CLAY
+
b14SCLAY+
b 15SANDY+
b 16N+
b17P+
b18K+
b 1 g D ~ y ~+
b 2 5 R G O A L l f ~
+
b26RGOAL2fs+
b27RGOAL3fs+
bZ8RGOAL4fs+
b2gRGOAL5fs
+
b30RGOAL6f~+
b3,ATTllfs+
b32ATT12fs+
b33ATT13fs+
b34 ATT14fs
+
bg5 ATT15fs+
b36 ATTI6fs+
Uwhere 8 = efficiency scores- technical, scale, overall technical, allocative and economic ao = constant ( 0 intercept)
b i = regression coefficients EDUC = years in school
EXPE = years of farming experience
EXTN = number of exposures to extension (within 2 years) AGE = respondent's age (years)
HHSZE = household size
OFFWORK= number of hours in off-farm work (per year)
D,, = 1 if the respondent is a land owner, otherwise zero
Dcrdt = 1 if the respondent has access to credit, otherwise zero Dbac = 1 if the farm is near Bacolod City, otherwise zero FLAT = fraction of an area with flat topography
SROL = fraction of an area with slightly rolling topography ROL = fraction of an area with rolling topography
CLAY = fraction of an area with clay loam soil SCLAY= fraction of an area with sandy clay loam soil SANDY = fraction of an area with sandy loam soil
N = total amount of Nitrogen (kgs) applied per hectare P = total amount of Phosphorus (kgs) applied per hectare K = total amount of Potassium (kgs) applied per hectare DHYV = 1 if the respondent planted new varieties, otherwise zero GOAL1 = farm and social status
GOAL2 = instrumental orientation GOAL3 = independence orientation GOAL4 = family orientation
GOAL5 = leisure orientation RGOALI = farm status
RGOAL2 = business/development orientation RGOAL3 = social and intrinsic orientation RGOAL4 = social status
RGOAL5 = independence
RGOAL6 = country living orientation
ATTII = aggressive/openness in farming ATT12 = easy care farmer
ATT13 = optimistic
ATT14 = risk conscious & stressed attitude ATT15 = farm extension believer
ATTI6 = family and socially oriented
U = error term
All the explanatory variables that showed associations with efficiency in all models were combined and tested interchangeably taking into account the possibility of substitution among the components derived from Gasson (1 973) as they were measured twice, though differently. Moreover, substitution may also occur between the items and the factor scores.
By combining and substituting these components, it may be possible to better explain the variations in farmer's efficiency level.
The level of significance in hypothesis testing was set at 5 per cent. However, since this is an initial study of farmers' goals and efficiency, the level of significance in the goal and behaviour variables was set at 10 per cent.
7 The Research
Locale
and the Selection of the Study AreaThe investigation was conducted in Negros, a small island in the Philippines (Appendix 1).
The province has two pronounced seasons, the wet and the dry. The dry season is from late December to May for the northern part, and from November to May for the southern portion.
The rainy season starts in June, reaches its peak in September and ends in October for the northern part. For the southern portion, the wet season begins in June, attains its peak in August and tapers off towards November (Aguilar, 1984).
The soil is considered to have come from two distinct origins: coraline and volcanic. The northern part of the province, largely influenced by the proximity of the seacoast, is of coraline origin. The southern part, especially the interior, strategically influenced by the presence of Manla-on volcano, is of volcanic origin. In terms of slope, the northern and western parts of the province are generally considered to be largely level plains and gently rolling slopes while the remaining portion is practically a land of sierras (mountains) of varying elevation (Aguilar, 1984).
As shown in Appendix l, the island is divided into two: Negros Occidental and Oriental.
Negros Occidental has a total area of about 792,610 hectares. Of its total land area, 64 per cent is devoted to agriculture. Sugar comprises 55 per cent of the land use, thus accounting for it's largely mono-crop character (Guide to Negros Occidental, 1997). Negros Occidental occupies around 48 per cent of the total area planted to sugar cane (Sugar Regulatory Administration (SRA) Annual Report, 1997 and Extension Services Annual Report, 1997). It consists mainly of moderately sloping to rolling lands with slopes ranging from 0-1 8 per cent comprising about 70.9 per cent of the 563,100 hectares of the provincial land area.
Currently, Negros Occidental has 11 sugar milling districts which are divided into three areas
-
North, Central, and South Negros (Table 2). In CY 1996-97, the Central Negros area obtained the highest average yield per hectare, 11 1.91 fifty-kilogram bags (Lkg); while the South Negros area obtained the lowest, 83.05 Lkglha.Table 2 Production Statistics for Negros Occidental Sugar cane Milling Districts, CY 1996-97.
Total Area Total Sugar Yield Sugar Milling District (hectare) (50 (L) kg bags) (LKg/ha.)*
Central:
Hawaiian Phils. / Aidsisa 1 1,202 1,500,060 133.91 Bac. Murcia & Talisay Silay 23,270 3,057,445 131.39
La Carlota 16,065 1,578,830 98.28
Ma-ao 9,928 834,511 84.06
North:
San Carlos 9,835 1,063,858 108.17
Lopez 10,931 1,031,921 94.40
Victorias 30,097 2,770,767 92.06
SagayIDanao 1 5,027 1,226,717 81.28
South:
Biscom 27,271 2,516,531 92.28
Sonedco 11,209 971,727 86.75
Dacongcogon 8,935 626,557 70.1 2
Average 97.52
Source: Sugar Regulatory Administration- La Granja Agricultural Research and Extension Centre, Extension Office Annual Report, 1997.
* This measurement is generally used in the sugar industry.
Since the aim of this study is to explain the economic, social and psychological factors influencing the variation in the production of sugar cane, it is necessary therefore, to
minimise the differences in productivity due to environmental (ecological) factors particularly soil topography and types. It is important to ensure that all the holdings selected are similar, or that any variation is at a minimum. In view of this, Central Negros was selected. The variation in climate including temperature, sunshine, rainfall and humidity can be assumed to be small compared to the North and South Negros areas.
Within each area, the mill districts' sugar yields per hectare vary. Looking at the Central Negros area, Hawaiian-Phils./Aidsisa and Bac. MurTTalISilay produced more than Ma-ao and La Carlota. Ma-ao's production per hectare is below the province's average production level while La Carlota's production is only slightly above the level. Therefore in order to have different levels of production efficiency, these four sugar mill districts were considered as they exhibited different levels of productivity. Another consideration was the accessibility of these four sugar mill districts given the time and budgetary constraints.
8 C~llection of Data
A stratified random sampling procedure was applied. The sugar cane planter as a sampling unit was limited to the head of a farming household who is an owner-operator andlor lesee- operator, except for a farm manager.
The size of the sample was determined using the simplified formula for n in sampling for proportions given by Cochran (1 977). The calculation of the sample size was also based on the cost and time invested with the acceptable error being set at 25 per cent.
The list of the respondents was taken from the SRA Planters' Directory CY 1997-98. Some respondents were replaced and the replacements taken from the same strata and in the same location.
The majority of the interviews occurred in the house, while a small number (especially for large planters) were interviewed in their non-farm work place. The data collection process was completed within 93 days (23 July 1998 to 23 October 1998).
A structured questionnaire was used in the interview. The questionnaires were pre-tested on a sample of farmers in the study area. The questionnaires comprised farm management factors (cultural practices, cost of production, etc.) and farmers' goals and attitudes.
Gasson's (1 973) goals were used and two methods of goal elicitation were applied: the 7- point Likert scale and goal ranking. In the first method, the respondents were asked to state the extent to which they believed such goals affected their operations on a scale of 7 (very important) to 1 (not important). In the second method, the respondents were asked to rank the goals from 1 (most important) to 20 (least, or not important). Questions relating to the farmer's attitude towards farming, technologies, government policies, farming business and decision-making were taken from Edinburgh Farming Attitude Scale (Willock, 1997) and Fairweather and Keating (1 990) and were used after revision to suit sugar cane farmers socio-economic environmental conditions. A Likert 1 to 7 scale of importance was also used.
At the end of the survey, 44 planters were excluded. Some due to incomplete information, and some questionnaires were not returned. The remaining 127 respondents were used in the analysis. The collated data was encoded using FoxPro data base programming.
9 Treatment
sf
Data9.1 Farmers' Goals, Values and Attitudes. A principal component analysis was carried out on the 20 variables rated by the respondents. After each run, the components that explained the least proportion of variance were deleted. After the third run the remaining 13 sorts were again factored and the varimax rotation revealed that 68.1 per cent of the total item variance was explained by 5 factors with eigenvalues of 1.0 or larger (Table 3).
The first factor has articulated primarily by the five expressive values and accounts for a large proportion of the variance. This dimension (or factor) can be regarded as the farmer's social status or identity (GOALI). Two of the instrumental values defined the second dimension. These objectives are essentially financial in nature (increase the family living standard and increase maximum farm income) (GOAL2). Two of the intrinsic values defined the third dimension: doing the work you like and being able to arrange hours of work. This dimension can be called 'independence' (GOAL3). The mixed social and instrumental values (i.e., spend time with the family and save for children's education, respectively), which load on the fourth factor are attributable to a farmer's family-oriented values (GOAL4). The fifth factor comprised two of the intrinsic values and since the factor with the larger loading on it is that of more leisure time than country living, this dimension can be termed as 'leisure
orientation' (GOAL5).
Table 3. Factors resulting from items related to farmers' goals and behaviour (elicited using 7-point Likert scale).
Varimax Eiaen % of
Item Factors values variance
Factor 1 (GOAL 1
-
Farm and social status) 3.69622 28.4B Be recognised as an owner of the land. .78279
Be recognised as a top producer. .75689
m Be recognised as a leader in the adoption of modern .83919 technologies.
Be recognised as an adopter of modern technologies. ,82868
Be recognised as a sugar cane farmer. ,671 08
Factor 2 (GOAL 2- (Instrumental) 1.58077 12.2
s Increase standard of living. .83978
c Increase maximum farm income .83044
Factor 3 (GOAL 3- Independence) 1.30595 10.0
Doing the work you like. ,74083
c Being able to arrange hours of work. .86886
Factor 4 (GOAL 4- Family orientation) 1.21 747 9.4
c Spend time with the family. ,72092
m save for children's education. .84042
Factor 5 (GOAL 5- Leisure orientation) 1.07967 8.3
Live in a healthy, outdoor, farming life. 56348
Have more leisure time. .89403
A principal component analysis was also performed on the same 20 variables but using the ordered ranking. The results of the factor analysis are shown in Table 4.
Table 4. Factors resulting from items related to farmers' goals and behaviour (elicited through goal ranking).
Varimax Eigen % of
Item Factors Values variance
Factor 1 (RGOALI - Farm status ) 2.61755 20.1
Be recognised as a top producer. .79472 Be recognised as a leader in the adoption of .87577 modern technologies.
e Be recognised as an adopter of modern .66288 technologies.
Factor 2 (RGOAL2- Business/development 1.80898 13.9
orientation)
Q Increase standard of living. .64624
Increase maximum farm income. ,651 66
0 Expand the business. ,68433
Factor 3 (RGOAL3-Social & intrinsic) 1.48503 11.4
Q Have more leisure time. ,84778
e Leave business for next generation. -.66678
Factor 4 (RGOAL4- Social status) 1.20806 9.3
Be recognised as a sugar cane farmer. .73593
e Be recognised as an owner of the land. ,8261 6
Factor 5 (RGOAL5- Independence) 1.03293 7.9
e Doing the work you like. .70899
e Being able to arrange hours of work. .81212
Factor 6 (RGOAL6- Country living orientation) 1.01210 7.8 Live in a healthy, outdoor, farming life. -.79674
The analysis reveals that 70.4 per cent of the total item variance is explained by the six factors (five in the previous method). The farm and farmer's social identity was split into two:
The farm status (RGOALI) remained in factor 1 while the two expressive values were loaded in factor 4 and thus termed social status (RGOAL4). The original two instrumental values in factor 2 (in Table 3), now became three with the inclusion of another instrumental goal- to expand the business and since it has the largest loading in this component, this dimension was termed business/development orientation (RGOAL2).
The mixed social and intrinsic values which load on factor 3 are the farmer's leisure and family oriented values i.e., the objective of having more leisure time along with the long- term objective of maintaining the continuity of the farm business in the hands of the family
(RGOAL3). Note, however, that the goals of doing the work you like and being able to arrange hours of work (factor 3 in Table 4) loaded on to the same factor-factor 5 (RGOAL5).
The last factor (6) has only one goal this can be termed 'country living orientation' (RGOAL6).
The results of the factor analyses showed that there is some consistency between the Likert scale goal significance and the ranking method as shown in the principal components.
Therefore it is sufficient to support the tentative conclusion that for this set of principal components, they appear to be nearly congruent. This may infer that one of the methods can be disregarded, as the components derived from scaling will show some collinearity with the components derived from the ranking method. However, it would be interesting to know which of these components, if any, explain variations in the farmer's efficiency levels. Thus, all of these components were initially included in the efficiency explaining relationships presented later.
The last set of explanatory variables
-
the farmer's attitudes, were also analysed and the results of the factor analysis are shown in Table 5. The analysis reveals that 65.3 per cent of the total item variance is explained by six factors with eigenvalues of 1.0 or larger. The firstfactor expresses the concentration of the farmers toward technologies and obtaining farming business information
-
the Professional farmer (ATTII). In contrast, factor 2, whichexpresses easy care farmer behaviour (ATTI2), involves a 'lackadaisical, take what comes' attitude.
The third factor is articulated primarily by the faith in the goodness of farming. The variable with the largest loading on it is a belief in the good outlook for sugar cane farming, followed by a belief that farming is security for retirement. The third attitude, the need for a cautious farm planning is somewhat dichotomous. Overall, this dimension can be regarded as having an optimistic behaviour, rather than a cautious approach for successful farming (ATT13).
The fourth factor can be viewed as a risk related attitude involving both risk and the related stress (ATTI4). The fifth factor is expressed by farmer's attitudes towards farming
technologies. The variable with the larger loading is a belief in the extension workers and therefore this dimension can be termed the farm extension believer (ATT15). The sixth factor is articulated by the farmer's behaviours toward consultation in decision-making. This is basically family and social in nature, thus, its name (ATTIG).
Table 5. Factors resulting from items related to farmers' attitudes towards farming, new technologies, farm business and decision-making in the farm (elicited using 7-point Likert scale).
Varimax Eigen % of
Item Factor ~ a i u e s Variance
Factor 1 (ATTII- The Professional farmer) 3.35140 19.7
c New technologies improve the farm production. .64523
e It is important to read about farming .67336 technologies.
e It is important to make maximum farm profit. ,63959
0 It is important to pay attention to market prices. .66629
e It is important to i d n i t o r the farm production
level. .71833
Factor 2 (ATT12- Easy care farmer) 2.41312 14.2
e Farming is a lonely job. .68330
e Farming problems may be ignored until they go .85843 awav.
c Successful farming is often due to luck. ,69443
Factor 3 (ATTI3- Optimistic) 1.71054 10.1
a The ling-term butlook for farming is good. .81491
s Farming is likely to provide a secure retirement. .73883
c Successful farming is due to cautious planning. ,5081 5
Factor 4 (ATT14- Risk conscious & stressed attitude) 1.50358 8.8 Farming is too financially risky. ,84392
c ~ a t u r e of farming is stre-ssful: .75522
Factor 5 (ATT15- Farm extension believer) 1.091 88 6.4
0 Farming technologies can be sourced from .86340 extension workers.
c New technologies have reduced the cost of .80229 production.
Factor 6 (ATT16- Family and socially oriented) 1.04207 6.1
~amiiies could be consulted abbut farm financial .84228 decisions.
Sometimes farming neighbours should be .67834 consulted before taking major decisions.
It would be interesting to know if a simultaneous principal component analysis on all goal and behaviour variables would help explain efficiency. However, the separation of the variables will determine if the value orientations formulated by Gasson suits farmers in developing countries. Similar comments apply for the excerpts taken from the Edinburgh Farming Attitude Scale (Willock, 1997) and Fairweather and Keating (1 990) as revised to suit the farmers in the Philippines.
9.2 Farm and Farmer's Characterislies (Including the Technology Adoption).
The human capital investment includes: (1) the farmer's years of formal schooling (EDUC);
(2) the years in sugar cane farming (EXPER); and (3) the number of extension exposures (EXTN) for the past two years (e.g., number of visits of farmers to demonstration trials and research centres, group discussions, training on farm practices, and extension advice on various farm practices).
The socio-economic characteristics of the farmer included the age (AGE) of the farmer at the time of the survey, the household (HHSZE) variable which records all the people living in the
house, the off-farm work (OFFWORK) was the of hours spent on off-farm work per year. In addition dummy variables were incorporated, for the tenure status (D,, = l if the farmer was an owner operator, otherwise zero), for credit (Dcrdt = 1 if the farmer had access to credit, otherwise zero), and for location (D~ac = 1 , if the farm is near Bacolod City, otherwise zero).
The variables for topography and soil type were measured as fractions of the area with flat (FLAT), slightly rolling (SROL) and rolling (ROL) topography, and the fraction of the area with clay loam (CLAY), sandy clay loam (SCLAY) and sandy loam (SANDY) soil.
Some cultural practices were applied more or less the same by all respondents, particularly the frequency of cultivation, weeding and fertilisation. Therefore, these practices were not included. However, a dummy variety variable (DHYV = 1) was included if the farmer used a 1980s variety. Fertiliser was disaggregated into nitrogen (N), phosphorus (P) and potassium (K) variables to determine, as far as possible, which nutrients contributed to farm efficiency.
10 Results
10.1 Farm and Farmers' Characteristics (Including Technology Adoption). Around 44 per cent of the respondents graduated from college and this is reflected in the extent of the educational levels of the respondents, which is very high (12 years of schooling = second year in college) (Table 6).
Table 6. Selected farm and farmer's characteristics, including technology adoption.
l tem Mean Std Dev Minimum Maximum
Farmer's human capital
Years of education (EDUC) 12.54 3.3 3 2 1
Years of farming experience (EXP) 17.18 12.16 1 5 1
No. of exposures to extension (EXTN) in 2 years 9.44 21.85 0 200 Socio-economic
Age (AGE) (years) 51.42 11.01 25 78
Household size (HH) (people) 3.94 1.92 l 8
Hours. in off-farm worklyear (OFFWORK) 61 5.68 840.24 0 3120
Farm environment Hectares
Topography: Flat topography (FLAT) 18.31 31.09 0 156
Slightly rolling (SROL) 6.70 19.81 0 132
Rolling (ROL) 11.95 41 -88 0 31 0
Soil types: Clay loam (CLAY) 17.11 45.91 0 31 0
Sandy clay loam (SCLAY) 7.49 20.25 0 120
Sandy loam (SANDY) 12.37 24.85 0 109
Adoption of technology Hectares
New varieties 17.86 34.58 270
Old varieties 15.12 28.97 227
Mixed varieties 3.99 10.14 59
Kilograms per hectare
Nitrogen (N) 377.52 11 1.65 36 729
Phosphorus (P) 139.41 82.37 0 368
Potassium (K) 179.28 153.37 0 480
At maximum, the respondents obtained either two college degrees or attended post-graduate studies. Around 27 per cent of the respondents had no exposure to any extension service.
Although the maximum number of exposures to extension was high, the majority reported to have no more than 20 contacts (for two years) despite the average of 9.44. There were few younger sugar cane farmers, and, equally, few older ones. On average, the respondents were middle aged with a household size of around 4. Half of them have part-time jobs.
Seventy-one per cent of the respondents were landowners; 15 per cent were lessees, while the remaining 14 per cent were both landowners and lessees. The total cropped area was 4,694.29 hectares of which around 80 per cent was owned.
In terms of land topography and soil types, 49 per cent of the total area is flat, 19 per cent is slightly rolling while 32 per cent is rolling. The majority (46.3 per cent) of the total area is clay loam; around 20 per cent is sandy loam while 33.45 per cent is sandy clay loam. Only 48.37 per cent of the total area was planted to new varieties of sugar cane, 41 per cent t o the old varieties, while 10.63 per cent was in a mixed variety. Fertiliser application varied from as high as 729 kilograms per hectare to no application at all, except for N fertiliser.
10.2 Farmers' Goals, Values and Attitudes. Table 7 shows the farmers' responses to the Gasson value orientations using a 7-point Likert scale (7 as very important). Among the value orientations, the instrumental values obtained the highest total mean (31.25), followed by intrinsic (28.91), social (26.7) and expressive (24.84).
Table 7. Farmers' goals and behaviour [after Gasson (1973)l elicited using a '7-p~int bikert scale.
Variables Mean Std Dev Min Max n
Intrinsic
1. Independence
2. Live in a healthy, outdoor, farming life
3. Doing the work you like. 6.11 1.08 2 7 124
4. Being able to arrange hours of work. 5.96 1.16 2 7 124
5. Have more leisure time. 4.44 1.63 1 7 124
Expressive
6. Be recognised as a top producer. 4.85 1.73 1 7 124
7. Be recognised as a leader in the adoption of modern
technologies. 4.98 1.57 1 7 124
8. Be recognised as an adopter of modern technologies. 5.16 1.47 1 7 124 9. Be recognised as a sugar cane farmer. 4.87 1.55 1 7 124 10. Be recognised as an owner of the land. 4.98 1.82 l 7 124 Social
11. Involve family in decision making. 5.52 1.57 1 7 123 12. Leave business for next generation. 5.60 1.50 1 7 123
13. Employ more people. 4.52 1.53 1 7 124
14. Belonging to sugar cane farming community. 5.24 1.40 1 7 124
15. Spend time with the family. 5.82 1.37 1 7 124
Instrumental
16. lncrease standard of living.
17. lncrease maximum farm income.
18. Expand the business. 5.98 1.39 1 7 124
19. Keep debt as low as possible. 6.09 1.48 1 7 123
20. Save for children's education. 6.09 1.73 1 7 123
Note: The four value orientations were taken from Gasson (1 973) while some of the goals were revised to suit sugar cane farmers' socio-economic environmental condition.
The instrumental values that were placed as the most important by the respondents were 'to increase maximum farm income' and 'to increase family standard of living.' This was followed by the intrinsic values-'to live in a healthy, out door, farming life and being independent.' The least important goal was under the intrinsic (and not under expressive)- 'to have more leisure time.'
Among the social values, the goal was 'to spend time with the family' and 'to leave business for the next generation,' while among the expressive values, the most important goal was 'to be recognised as an adopter of modern technologies.'
Table 8 contains the results of the Gasson goal-ranking question.
