Bachmann, Lorenz B. R.: Review of the Agricultural Knowledge System in Fiji - Opportunities and Limitations of Participatory Methods and Platforms to promote Innovation Development -

60

Chapter 5. The Agricultural Knowledge and Information System in Fiji

The examination of the Fiji AKIS will proceed in several steps. A first overall perspective describes all major actors and institutions of the AKIS. Then, all major actors are reviewed one by one. The analysis starts with a general view of the Ministry of Agriculture, Fisheries and Forest (MAFF) followed by a close up of the Division of Economic Planning and Statistics (EP&S). Then the Research Division (RD) and the Extension Division (ED) are investigated. As last actor, the focus is directed on the farmers and final users of the knowledge system. To complete the picture, the interplay of all actors, their linkages and information flows are examined.

5.1 Overview: organisations, institutions and actors within the Fijian Agricultural Knowledge and Information System

More than one might expect of a small island country, Fiji has an elaborate agricultural knowledge system with a considerable number of different organisations and institutions. For this reason, it would be outside the scope of this thesis to describe all system actors. However, the most important players in the governmental, educational, and private sector, with relevance to agriculture, will be reviewed briefly. The major players are depicted in Figure 14, which illustrates the overall setting and the linkages between these various actors.

At a first glance, the most striking feature is the existence of two separate research and extension sub-systems. The larger research and extension sub-system lies within the Ministry of Agriculture. The second research and extension system is operated by the Fiji Sugar Corporation (FSC). The latter and smaller system will be looked at first.

Fiji Sugar Corporation (FSC)

The corporation is a private sector company and covers all aspects of sugar cane production, milling and marketing. Agricultural research and extension for sugar cane farmers are other aspects of FSC‘s work. However, in relation to the milling and marketing activities, they are minor ones.

The sugar cane research centre is located close to the main sugar mill in Lautoka in the western division. The centre is small with only 6 permanent professional staff members. The qualifications of existing staff are low (1 MSc, 5 BSc), which reduces the potentials of the centre to produce innovations (Eyzaguirre 1996, 65). One of the main activities is the breading of new sugar cane varieties. In the past, the centre produced a number of excellent varieties that have found worldwide distribution, but at present the output is rather limited.


61

Figure 14: Simplified model of the Fijian AKIS

Source: own design.


62

During the research period the major work was concentrated on soil fertility and fertiliser trials, physiology and plant protection. The centres biggest problem was the lack of qualified staff.<43>

The research centre‘s linkages with the FSC extension are weak. Main areas of co-operation were the introduction of new sugar cane varieties and fertiliser recommendations. The information flow was not very interactive, with limited feed back from extension. A central problem was the limited understanding of growers‘ problems, their farm management decisions and preferences. Growers‘ yields were far below the potential of the varieties and researchers lacked understanding of why their recommendations were not followed.

Some reasons for the linkage problems could be attributed to the set-up of the FSC extension service. The service had two main goals. First, to provide farmers with a service for inputs and help in the organisation of the harvest. Second, to provide advice in the application of new recommendations. A good analysis of the extension set-up is given in the Landell-Mills report (1991, 30): “ there is little substance, at present, to the organisational links between the Chief Extension Officer located at the Sugar Cane Research centre in Lautoka and the Extension Officer at Lautoka and the three Technical Field Extension Officers at the three other factories; and the links between these factory based Extension Officers and the Farm Advisers in the field are weak, too. The later are sub-ordinates to the FSC‘s Field Officers and form part of the field staff reporting to the field manager of each mill. This leaves the Chief Extension Officer in a position of a captain without troops. Extension work is not the primary concern of the Field Officers, most of whose time is taken up by administrative and organisational responsibilities in connection with the distribution of farm supplies and the organisation of the harvest. Subordinate to the Field Officers, the Farm Advisers are also partly occupied with these tasks. Moreover, the Farm Advisers have a markedly inferior status to the Field officers in terms of educational qualifications, pay and fringe benefits. In all, it cannot be said that the extension service at present rates sufficiently high in the hierarchy of FSC functions.“

This set-up explains why most organisational and production target oriented tasks are performed well. In this respect extension has enabled a considerable expansion of the production of sugar cane. This achievement was acknowledged by the Government extension service, where FSC extension enjoyed a good reputation. However, the actual extension function of farm advice and knowledge exchange with growers lagged behind.

Sugar cane farmers have some control over the industry through the Sugar Cane Growers‘ Council. However, agricultural issues, which trigger innovations on farms, play only a minor part in its dealings, whereas overall industry matters, administrative, payment and price issues constitute the major part of the Council‘s activities. The Council also has a representative in the Sugar Cane Research Committee. But as the Committee has not met for several years, its input towards research policies has been minimal (Landell-Mills 1991, p. 29).


63

The focus of this study lies on the aspect of diversification of agricultural production. As the mandate of FSC is exclusively focused on sugar, a deeper review of the sugar cane research and extension would not provide relevant information. Instead the further analysis will be focused on the Ministry of Agriculture, which co-ordinates all attempts to diversify agricultural production.

Ministry of Agriculture, Fisheries and Forests (MAFF)

The Ministry of Agriculture, Fisheries and Forests (MAFF)<44> has the mandate for all farmers in Fiji. However, research and extension regarding sugar cane farming is excluded from its mandate and remains the unique responsibility of the FSC. The main clientele of MAFF are all non-sugar cane farmers, who represent about two thirds of all Fijian farmers. This entails a broad range of agricultural activities, as farmers are more or less subsistence oriented, with a variable degree of commercialisation into different crop and livestock enterprises.

On national level, agricultural policies are determined by Fiji Government and approved by Cabinet. The Ministry is responsible to Cabinet and has both executive and consultative functions. In this respect, MAFF is providing advice on policies to cabinet and implementing its policies. Directly contributing to this consultative function is, in particular, the top management level of the Minister, his permanent secretaries and all heads of departments including the directors of research and extension.

The Department of Economic Planning and Statistics (EP&S) has a specialised role in this function. It monitors the impact of agricultural policies, appraises and assists donor and national projects. Moreover, it advises on farm management issues and updates the Fiji Farm Management Handbook. This reference contains the current recommendations for all crop and livestock activities<45>.

The Ministry‘s research and extension set-up is composed of six divisions or sections. The Research Division (RD) and Extension Division (ED) are two large separate units that deal with crop production only. These divisions are the focus of this study. The department of Animal Health and Production (AH&P) accommodates veterinary services, adaptive livestock research and extension under one single roof. An assistance to both extension divisions, is the Training and Communications Section (TC&S) that compiles extension materials for both field work and mass media publication and conducts training workshops for Ministry staff and farmers. The Fisheries and Forests departments also both have own research and extension sections.

Educational Institutions

The University of the South Pacific (USP) and the Fiji College of Agriculture (FCA) offer training and education in agriculture. Most MAFF staff are former students of the three year course for Tropical Agriculture at FCA. USP‘s school of agriculture is located at Alafua Campus in Western Samoa. It offers a 3-year Bachelor of Agriculture (BAg) course and the opportunity of a two-year postgraduate Master of Agriculture (MAg) course. As the Alafua courses are outside the country, fewer Fijian students make use of


64

the opportunity. Since 1994, USP has offered a distance education programme in agriculture.

Both institutions keep close ties with the Ministry; however, FCA is linked tighter as it is located next to the main research station in Koronivia. Some of their student courses are taught by research or extension staff from MAFF. FCA also operates a large farm for agricultural experimentation, providing teaching opportunities and food production for the boarding students. FCA maintains very little direct research or extension contacts with farmers. The curriculum focuses on the basic scientific principles of tropic farming.

USP's School of Agriculture in Western Samoa works in close co-operation with FCA. The Institute for Research and Extension and Training in Agriculture (IRETA), located at Alafua campus, co-ordinates regional research activities and maintains an information and extension service to the Ministry staff in all member countries of USP<46> via satellite. IRETA also assists the national extension services mainly with publications and opportunities for training workshops, but does not provide active extension to farmers.

Regional and International Institutions

Another institution with regional mandate is the South Pacific Commission (SPC). With its Headquarters in Noumea (New Caledonia), SPC represents all Pacific countries including the rim countries Australia, New Zealand, Japan, several countries in North and South America and the former colonial powers Britain and France. SPC operates an agricultural centre in Suva. The centre plays an important role in assisting countries in the organisation of national plant protection and quarantine services. Moreover, SPC provides a modern agricultural information network with satellite links to all member countries. The centre also hosts the regions second largest agricultural library<47>. SPC and USP also maintain closer links with the agricultural Universities in Hawaii, New Zealand and Australia. International institutions with agricultural mandate are not present in Fiji. Therefore, linkages to these international centres are weaker.

Donor Assistance

Donor funded projects play an important role in the Fiji AKIS. Typically, two to five projects at a time are linked to the Ministry. Development assistance is provided to various domains of agriculture, but a certain focus lies on quarantine, control of pests and diseases and the save transfer of genetic materials between Pacific Island countries. Though many projects have a regional mandate, which requires the experts to share their working time between several countries, Fiji, as the regional hub, benefits considerably from outside technical assistance. Usually, the experts that hold MSc or PhD qualifications generally outnumber local staff with equivalent qualifications. Consequently, donor projects have a strong advisory role within the AKIS.


65

Private sector

Trading and industry still only play a minor role in Fijian agriculture. Mechanisation and the usage levels of other inputs outside the sugar sector are low. As capital is scarce for most farmers, business for traders is limited and consequently private sector technology development for farming purposes is underdeveloped. Some exporters have production contracts with farmers, and besides providing technology, have an impact on marketing and quality standards.

Few supporting organisations exist for the majority of subsistence farmers in Fiji. Most common are co-operatives that deal with marketing of commodities like cocoa, ginger, beef, dairy products or pineapples. Some co-operatives also organise input distribution. However, most farmers are not formally organised. One exception is the cane belt, where a number of farmers‘ organisations operate (e.g. National Farmers Union, Fiji Cane Growers Association).

Media

Radio broadcasting is an important media for agricultural extension. Fiji Radio transmits programmes weekly with agricultural topics. Print media is also important. Unfortunately, no specific agricultural journals for farmers exist. The major daily newspapers publish weekly articles on agricultural or horticultural topics. Several monthly magazines such as ’The Review‘, ’Island Business Pacific‘ or ’Pacific Islands monthly‘ occasionally publish articles on business ventures in agriculture. However, these magazines, unlike the newspapers, circulate little within the farming community.

5.2 Ministry of Agriculture, central player in the Fijian Agricultural Knowledge and Information System

5.2.1 Goals and organisational structure

The Ministry of agriculture is the central executive body to the implementation of the Fijian Government‘s policies and strategies for the agricultural sector. During the researchers fieldwork, the following policy goals were set out in the planning document, ’Opportunities for Growth‘ tabled in Parliament in February 1993, and emphasised:

  1. a move away from import substitution towards promotion of exports,
  2. the importance of markets in planning and management,
  3. the need to improve productivity and cost effectiveness,
  4. the importance of private sector investment,
  5. reorientation from production targets towards financial performance and price competitiveness,
  6. intensification and quality control,
  7. provision of essential services by MAFF (extension, research, marketing assistance, regulatory, infrastructure development),
  8. privatisation of selected operations, and
  9. soil conservation and sustainable land use (Woodward 1994, appendix 3, 1).


66

During the beginning of the research period, the Ministry was in the process of digesting this tough ’liberalisation package‘. In the research period, the Ministry showed difficulties to break with the past tradition of input subsidies, production targets<48> and protected domestic markets. The required changes towards quality control, competitiveness, and in particular service orientation, hardly seemed to gain ground.

MAFF, the Ministry has three main divisions: agriculture, fisheries and forests. In the past, the rank and importance of the various divisions shifted, which was reflected in the name and organisation of the Ministry<49>. The Ministry has a typical hierarchical structure with the Minister and his Permanent Secretary heading the organisation. On the next level there are Deputy Secretaries for Operations and Services and the Conservator of Forests. Also on this level, but lower in grade, is the Director of Administration and Finance. Figure 15 depicts the organisational chart<50> of the Ministry.

Under operations figure the Directors of Animal Health and Production, Director of Extension, Director of Fisheries, and in lower grade, the Principal of the Fiji College of Agriculture. Under services come the Directors of Research, Director of Drainage and Irrigation together with the Chief Economist. The Principal Agricultural Officer of the Training and Communication section is also attached to this level. Under the Conservator of Forests, figures the Deputy Conservators for Timber and Resource Development.

The division of labour between the various departments is very articulated, but several assignments are not very practical. A key issue is the separation of livestock and crop research and extension. Both livestock research and extension are grouped under the small Division of Animal Health and Production. Crop research and extension are split into two large, separate divisions each dealing only with crop specific research/extension. Additionally, the departments of Fisheries and Forests have their own research and extension sections. This organisational set-up is likely to increase the chance for duplication of efforts as it increases the needs for information exchange between the different departments. During the research phase, considerable efforts were made by the Ministry to reorganise and streamline the structure and procedural routines. However, while several procedural routines (in particular accounting) were revised, proposals for organisational changes have come to nothing.

The efficiency of the Ministry was effected by frequent changes in top management positions. In the period 1993-96 the position of the Minister was occupied by four different people, the positions of the Permanent Secretary and the Director of Fisheries by three different people, and the positions of Deputy Secretary Services, Conservator Forests, Deputy Conservator Timber, Director of Extension, Director of Drainage/


67

Irrigation and the Chief Economist by two different people. This high rate of change, in case of the position of the Minister, even annually, makes medium- to long-term planning very difficult.

Figure 15: MAFF organisational chart

Source: MAFF, 1996.

A further comprehensive review of the overall Ministry is outside the scope of this thesis. In line with the theoretical concept of the knowledge systems perspective, the research will now be focused on the agricultural departments within the Ministry. Firstly, the role of the department for economic planning and statistics will be examined. Then, the divisions of research and extension will be reviewed in detail.

5.2.2 Economic Planning and Statistics Division

The division for economic planning and statistics plays a central role in Fiji‘s MAFF. The role of the division comprised the of following responsibilities:

The long list of tasks illustrates the wide range of activities of the division. A key role lies in the advisory function for the Minister and Cabinet. Responsible for the formulation and evaluation of development projects and policy proposals for the primary sector, the division had a strong influence on capital investments and priorities for development. Through this advisory function, the division had a closer and more influential relation on Government and agricultural policy decision making than other divisions within the Ministry. The division is located in the Ministry headquarters in Suva.

The tasks within the division were split into four sections:

The commodity development and marketing section was responsible for the analysis of marketing and commodity information, marketing studies and marketing strategy formulation. In the research period, the focus of activities was on traditional markets such as dairy, rice and ginger. With cocoa and copra, the section undertook measures for consolidation and rehabilitation. Core activities consisted of the monitoring of international price developments and the potential impact of GATT for local producers.

The projects section has the leading role for planning, appraising, monitoring and evaluation for both existing and new projects within the Ministry. This is a very important role as the section analyses, the successes and failures of past projects and suggests consequences for future intervention. In this respect the section worked together with other divisions of MAFF. However, fulfilling this leading role, the section had several difficulties. The diversification of agricultural enterprises did not proceed substantially<51>.


69

Still, considerable emphasis was placed on traditional commodities, although the long-term international prospects were not very promising (e.g. cocoa, copra). A systematic search for alternatives and potential new commodities ideas did not take place. In the research period, only limited progress was made. While the number of new project proposals was still at zero in 1993, it grew to three in 1994 and seven in 1995. In 1998 two new donor funded projects could be attracted<52>.

The statistics section maintains an agricultural database with a wide range of information. Recording of local market prices constitutes a major part of the task. This is one of the strengths of the section. Organising the agricultural censuses in ten year intervals is another responsibility of the section. Apart from recording prices, the section largely relies on other Ministry divisions for data collection.

The major tasks of the farm management section are to prepare farm plans and farm feasibility studies, provide farm and project appraisal services and assist farmers with loan applications. A key function is the updating of the Fiji Farm Management Handbook (FFMH): the manual compiles the latest research findings on all crop and livestock enterprises. For this reason, the manual plays a central role for all divisions. An important problem with the manual was the specific economic language that made it difficult to understand for non-economists. All staff that the researcher contacted during the fieldwork, had very limited background in economics. During the research period, the section was lagging behind in the updating of the FFMH. The 1994 update appeared only late in 1995 and even then, uncompleted. The gross margin section on crops was missing completely, and for livestock, the 1990 figures were simply reprinted. Based on the PRAs, a number of recommendations<53> were made for the FFMH, including more detailed enterprise descriptions, consistent fertiliser and pesticide information, additional production figures in more customary area sizes (acre and sqchain instead of ha) and staff training requirements, for staff in basic economics.

Table 9: EP&S staff resources

Grade

Vacancies

Positions

Chief Economist

 

1

PAO/ PEO

 

1

SAO/ SEO

2

5

AO/ RO

2

6

GT

 

 

ATO

 

 

AA/ SAA/ STA/ TA

2

13

Other

 

2

Total

6

28

Source: own compilation from MAFF 1994, 72; MAFFA 1996, 85.

The size of the overall division with 28 permanent staff members was small to medium (Table 9). A particular critical point was the high number of vacancies. Already since the mid eighties, time and again a number of senior positions were vacant. The position of the Chief Economist could only be filled in 1995. In the same year, the positions of 2 Senior Economic Planning Officers and 2 Senior Agricultural Officers were vacant. Candidates with good economic qualifications were difficult to find. The curriculum at FCA includes only basic elements of agricultural economics. The division even assisted FCA in teaching,


70

which resulted in a kind of ’short circuit‘: As most staff within the division are college leavers from FCA, the level of economic qualification at the division can not be raised significantly with new college leavers.

A good and cost effective way to buy-in additional expertise was the recruitment of a senior economist via the British Overseas Development Institute (ODI). The division should try to expand this co-operation and also explore assistance from other countries that offer similar programmes (e.g. CIMIT, DED). The ODI support certainly strengthened the economic qualification within the division. The problems with understaffing and staff qualification were certainly the major reasons that constrained the overall performance of the division.

The financial situation of the division could be considered as rather stable. Since the mid eighties, with the exception of the year following the military coup, the allocated budget grew steadily. Table 10 depicts the development of the budget.

Table 10: Allocated budget of EP&S in ’000 FJD

Budget items

1984

1988

1989

1990

1991

1993

1994

1995

1996

1998

Established staff

442.2

275.0

313.0

393.5

363.8

511.3

516.3

516.3

529.2

399.2

Unestablished staff

17.2

15.1

20.8

20.8

24.6

27.4

27.6

26.8

28.7

22.0

Travel & Communication

16.1

7.6

11.7

16.8

15.8

18.0

18.0

18.0

18.0

18.0

Maintenance & operations

6.1

6.2

4.3

19.4

18.2

22.0

22.0

22.0

22.0

22.0

Purchase goods & services

1.5

6.9

10.2

14.4

14.7

20.0

20.0

20.0

20.0

20.0

Special Expenditure

0.5

8.1

215.8

567.8

728.4

25.0

25.0

25.0

25.0

175.0

Total

483.6

318.9

575.8

1032.6

1165.5

623.7

628.9

628.1

642.9

656.5

Source: own compilation from MPI 1985,73; MPI 1991,102; MAFFA 1999, 60.

In the period 1994-95, the budget stagnated, increasing again in 1997 and 1998 due to the new funds released for the commodity development framework (CDF).

In this respect Fiji, is rather fortunate compared to other development countries that faced rather tough declines in Government spending for agriculture in recent years. The present overall resource base can be considered as acceptable. The division was equipped with a good number of vehicles and good computer facilities.

5.3 Research Division

5.3.1 Mission and goals

The Research Division (RD) is one of the major players within MAFF. In the recent process of organisational review, the division‘s mandate was reformulated and a new mission statement was defined:


71

“The Research Division is committed to the provision of high quality and relevant applied agricultural research and support services, and the dissemination of research findings" (Kumar and Hine 1993, quoted in Dowling and Kumar 1995, 7).

The mission statement was further detailed by a set of objectives:

  1. develop elite lines of germplasm and make these available to Fiji‘s farmers;
  2. maintain and improve existing genetic resources;
  3. devise packages of crop husbandry and land management practices in response to farmer needs;
  4. protect crops in an economically feasible, socially acceptable and environmentally friendly manner, with emphasis on preservation and maintenance of the existing bio-diversity through natural resource management;
  5. provide a relevant and effective agricultural analytical service;
  6. provide timely information to the Government; and
  7. maintain an effective research capacity (Dowling and Kumar 1995,7).

The mission statement for the RD focused on good quality applied research services and their dissemination. This is certainly a very sensible overall goal for the research department of a small island nation like Fiji. Scarce resources are not wasted on basic or fundamental research. Another valuable aspect was the consideration of the dissemination of research findings. This broadened the mandate of research and took the necessity of disseminating results into account.

Taking a closer look at the Research Division‘s goals, a more diverse structure of tasks was apparent. Besides research, a number of services were of importance: maintenance and development of genetic resources, analytical services (chemistry laboratory) and the information provision for Government. Overall, the focus of the new tasks had a strong service orientation. This characterised the new working philosophy. The intention of the review team was clear: trying to convert the inward-oriented Government department towards a more service- or client-oriented organisation. However, some of the old philosophy was still visible in the tasks. In task four the division was ’to protect crops‘. Obviously, crop protection can only be done by farmers themselves and not by Ministry staff. But in the past, RD staff occasionally did even that: spraying crops for farmers<54>. A move forward, however, was visible in task three, where new farming practices should be ’in response to farmers needs‘. But whether this goal can be achieved by simple ’packages‘ remains questionable.

5.3.2 Organisational structure and tasks

The RD is headed by the Director of Research. The division was grouped into six main sections: Agronomy, Horticulture, Plant Protection, Fiji Agricultural Chemical Laboratory


72

(FACL), Land Use, and Administration. The organisational set-up and the location of research stations are depicted in the following figure.

Figure 16: Research Division, organisational chart and station locations

Source: own design.

Agronomy, was the largest section, and like all other sections, headed by a Principal Agricultural Officer (PAO). The section was responsible for the introduction, maintenance and improvement of germplasm in the country. This task required activities such as testing of new varieties for adaptability to local conditions, disease resistance screening and operating facilities for maintaining the distribution of seed and planting material. For selected elite germplasm, packages of environmentally sound cultural practices were developed. This work involved several aspects:

The horticulture section was located on the four larger research stations. The research programmes were commodity-oriented and concentrated on fruits, vegetables and cereals. The tasks were comparable to those of the agronomy section. Besides farmers, the horticultural industry was also an important client of the section.


73

The two sections‘ research programmes were largely commodity-oriented and followed recommendations of periodical ISNAR reviews (1982, 1985, and 1991). The technical programmes for 1993 had the following content:

The plant protection section was located at the two main research stations. Its main purpose was to reduce losses due to pests and diseases. Typical activities of the section included:

The section specialised in a number of fields. Most prominent was the work in biological control and IPM in co-operation with GTZ and biological control of taro beetle in co-operation with the EU. Of particular importance was the work on fruit flies that was financed by FAO. The existence of a number of fruit fly species in Fiji made it impossible for Fiji to export fruits to Australia and New Zealand<55>.

The Land use section was located at the main research centre in Koronivia. The section was more service-oriented rather than being involved in adaptive research. The main tasks could be summarised as:


74

Special fields of work included a new soils mapping of Fiji, and the build-up of expertise in geographic information systems (GIS). In co-operation with IBSRAM the section was conducting research on soil erosion.

The Fiji Agricultural Chemistry Laboratory (FACL) was located in Koronivia and provided laboratory services for several Fiji Government departments. Chemical and physical analysis was offered for soils, plants, animal feeds, irrigation waters, and food and forensic samples. FACL was assisted by Australian and New Zealand Aid, with a focus on training and quality assurance.

The Administration section was responsible for the financial monitoring of all programmes. It was also in charge of the management of all research stations. The section also hosts the specialist library and information services for all staff.

5.3.3 Available resources and staff qualifications

Among the small island nations in the South Pacific, Fiji has the largest agricultural research department<56>. In 1995 the RD had 118 established and 154 unestablished staff (Table 11). In the past decade, the department managed to increase the total number of staff positions. The number of established staff increased from 109 in 1988 to 118 in 1995.

Table 11: Development of Research Division staff and remuneration levels

 

1988

1995

Shift in

Salary

Grade

Positions

Vacant

Positions

vacant

positions (%)

levels (FJD)

Director

1

0

1

0

0

33.461

PAO

4

0

4

1

0

26.255

SRO/RO

17

7

23

4

35.3

14.111-19.111

SAO

3

1

3

1

0

19.111

AO

3

1

3

0

0

14.911

GT

4

2

2

0

-50.0

12.073

ATO

10

5

10

4

0

11.212

AA/ SAA/ STA/ TA

50

12

54

10

8.0

5.153-8.660

Admin. staff

17

0

18

1

5.9

5.100-11.212

Total established

109

28

118

21

8.3

 

Unestablished

130

 

154

 

18.5

 

Grand total

239

 

272

 

13.8

 

Source: own compilation from MPI 1988,76; Dowling and Kumar 1995, 24 and A1a.

The trend for unestablished staff was similar. The largest increase occurred in middle management (SRO/RO) with additional 6 positions; while the overall largest group of


75

field staff (AA/SAA/STA/TA) grew less by only 4 positions. This size suggested that the department had a substantial potential to conduct research programmes. However, in this case, size was not a good direct indicator for institutional output. The division faced several problems. A major problem was vacancies. As a consequence of the military coup in 1987, many qualified staff left the country. This problem was still felt during the research period. The high number of 28 vacancies (1988) could only be reduced to 21 in 1995. A considerable number of officers were filling posts in an acting capacity basis only, as they did not have the necessary formal qualifications for the post.

Table 12 highlights staff qualifications in relation to positions. In brackets are the numbers of officers undertaking additional studies. An obvious important fact was the very low level of formal qualifications in the department. Among the 118 established staff, only 2 held a masters qualification, and only one officer held a doctorate qualification. In total, 20 officers held a bachelor or postgraduate qualification. The large majority of staff held the FCA diploma title.

Table 12: Position and highest qualification of Research Division staff

Highest qualification

DR

PAO

SRO/ RO

SAO/ AO

GT

ATO

SAA/AA STA/TA

Clerical Staff

Total

Doctorate

 

1

 

 

 

 

 

 

1

Masters

1

 

1

 

 

 

 

 

2

Post graduate

 

 

3

 

 

 

1(1)

 

4(1)

Bachelor

 

1(1)

12(6)

1(1)

2

 

 

 

16(8)

Diploma

 

2

7(1)

2(1)

 

11(2)

33(7)

 

55(11)

Certificate

 

 

1

 

 

 

1

 

2

Not stated

 

 

1

2

 

1(1)

9

17

30(1)

Source: adapted from Dowling and Kumar 1995, 31.

This situation was far from desirable. The very nature of a research post requires the clear ability to work independently, develop new ideas, put existing facts and information into question, and to creatively analyse and combine different factors. This ability is generally only acquired properly in a master‘s course, and only to a certain extent in a bachelor course. Activities and outputs of the RD, that will be discussed later, need to be seen in this context of low staff qualification.

The budget development of the RD is shown in Table 13. The total budget amounted to 3.9 million in 1998. Compared to 1.75 million in 1984 this corresponded to an average annual increase of 5.9 %. This can be considered as an increase above the necessary compensation for inflation. However, the real development was more dynamic. While the budget grew at a higher rate until 1993, it was on the decline in the following two years and grew stronger again from 1997.

The most expensive item on the budget was staff salaries. Taking the average for 1994 to 1998, these accounted for close to two thirds of the budget. The next biggest budget item was capital construction (12.5 %). This item included new funds that were released in 1997 under the commodity development framework programme. After this came value-added tax. This tax was newly introduced in 1993.


76

Table 13: Research Division‘s budget development

 

1984

1988

1989

1990

1993

1994

1995

1998

Ø 1994-1998 in %

Budget items

in '000 FJD

Established staff

721

566

811

757

1162

1174

1203

1353

35.9

Unestablished staff

844

488

588

613

784

791

811

971

24.8

Travel & Communication

22

18

33

47

49

49

49

41

1.3

Maintenance & operations

48

33

49

80

72

72

72

60

2.0

Purchase goods & services

50

30

36

78

75

75

75

52

1.9

Operating grants and transfers

20

15

15

30

 

0

3

0

0.0

Capital purchase

46

128

 

 

 

 

 

0

0.0

Capital construction

 

95

333

30

275

120

150

1034

12.5

Special expenditure

 

 

 

 

 

 

300

0

2.9

Capital grant and transfer

 

 

127

782

949

657

200

0

8.2

Value added tax

 

 

 

 

426

317

337

428

10.4

Total

1751

1373

1992

2417

3792

3256

3200

3939

100.0

Aid in kind

 

600

2200

600

1090

1846

893

 

 

Source: own compilation from MPI 1985,57; MPI 1989,77, MPI 1990,85; MPI 1991,87; MAFF 1994,59; MAFF 1995,100; MAFFA 1996,103, MAFFA 1999, 73.

Foreign aid in form of capital grants and transfer played a major role (8.2 %). But these contributions showed a very high variability over the years. In some years (1992), they contributed more than 40 % to the budget. Equally important, if not more important, was aid in kind (constructions, vehicles, laboratory equipment, etc.). Aid in kind ranged from 0.6 million to 2.2 million per year. Considerable constraints for the work of the division were the low budgets for travel, communication, maintenance, and operations (1.3 and 2.0 %). These budgets did not permit many activities, and reduced the mobility of staff. This was certainly one reason why researchers ’had to focus‘ on on-station trials rather than seek close co-operation with farmers in remoter locations. Here, donor funded assistance played a vital role. Without external assistance, the division would have faced more difficulties to function.

The available physical infrastructure of the RD can be considered as comparatively good. In 1995 the division disposed of 47 vehicles, 27 tractors and a number of other farm machinery and implements. Furthermore, the division maintained 124 staff quarters (MAFFA 1996, 104). The available annual reports only occasionally specified the actual conditions of the vehicles and other machinery. However, own observations made during the fieldwork suggested that the conditions were reasonable. Calculating the number of established staff (118 in 1995), about 2.5 officers shared one vehicle. This was certainly a very good ratio for a research division.


77

Summarising the strengths of the division, they lay in the good infrastructure, and, with some limitation, the financial resources of the division. The main weaknesses were, the high numbers of vacancies and the generally low level of formal education of the staff.

5.3.4 Research priority setting

The current chapter examines how these priorities are determined, and which actors play an important role in this process. For this purpose both research and extension staff were requested to rate the influence of selected actors in the formulation of research priorities. The results are listed in Table 14.

Table 14: Actor influence in the formulation of research priorities

Actors

Strength of influence as seen by

RD1)
(n = 7)

ED2) (n=21)

Fiji Government policies

4.1

4.1

Extension

3.8

3.1

Donors and projects

3.7

3.5

Private sector

3.6

2.3

Farmers

3.4

2.4

Research division as a whole

3.4

4.0

Regional/intern. Research centres

2.7

3.2

Personal preferences of researchers

2.5

3.4

1) Research Division, 2) Extension Division; Ranking scale: 1= none, 2= weak, 3= medium, 4= strong, 5= exclusive.

Interesting enough, was the result that both researchers and extensionists rated the Fijian Government as the most important player. Researchers, however, ranked the role of extension, donors and the private sector as close to strong (3.6 to 3.8). Surprisingly, they rated themselves and farmers only as medium players (3.4). The influence of regional or international research centres and the personal preferences of researchers were considered only to be weak to medium (2.5-2.7).

Extensionists saw the influence of the Research Division quite differently. They rated researchers as the second most important player, with a strong influence of 4.0 (whereas researchers attributed this role to extensionists with 3.8), followed by donors and the researchers personal preferences, with a medium to strong influence (3.4-3.5). Regional or international research centres were considered as medium players (3.2), whilst farmers and the private sector were considered to play only a weak role (2.3-2.4).

While interpreting these ratings, it appeared necessary to take a look at how these decisions had taken place in practice. Dowling and Kumar (1995) reviewed the developments in the RD over the past years. They draw the attention to the fact that ISNAR, as an International research centre, had a considerable influence on the formulation of research plans in Fiji. The authors reveal that the “current (1995) research program has many similarities to that proposed in 1985, principally in terms of its scope and the commodities being investigated“ (Dowling and Kumar 1995, 8). Besides recommendations for priorities, ISNAR also made proposals for the overall set-up of the RD as earlier as 1981. An extract of the recommendations made by the ISNAR 1981 review, are summarised here:

ISNAR was again invited to review the RD in 1985 and 1991. The later report stated that little attempts had been made to implement the earlier recommendations. The authors drew particular attention to the fact that, “there was still no formal mechanism in place for identifying the value or worth of research being undertaken, or for recognising if the work follows any pre-set, clearly identifiable priorities“(Dowling and Kumar 1995, 8).

Returning to the interpretation of the influence rating of RD and ED staff, the lack of an official procedure apparently led to reduced clarity and transparency for all parties involved. Decisions on priorities seem to take place as need arises, e.g. as running programmes come to an end or new funds become available. Thus, changes take place in a more long-term perspective. This may explain why both staff groups considered the Government as the strongest player. They themselves felt rather weak, as they could not influence priorities in the short term and, therefore, any changes taking place appear to be effects of long-term policy changes. The civil service hierarchy, where all decisions come from the top (i. e. the Government), strengthens this perception.

Summarising the results, it becomes evident that the real influence of actors involved in the formulation of research priorities is difficult to determine. While staff from both research and extension consider the Government as the major play, this appears at least doubtful, as Dowling and Kumar 1995 have shown that the Government was unable to implement recommendations made by ISNAR for almost two decades. This, in the contrary, indicated a weak Government, a weak MAFF top management and the overall malfunction of the organisation. An important point raised was the role of farmers in the process of determining priorities. Both staff groups and the 1981 ISNAR report criticise that farmers‘ needs and problems were given much too little weight.


79

These findings clearly stress that there is a need to establish a transparent official procedure to determine research priorities for short-, medium- and long-term. A proposal of what these procedures could look, will be made in chapter 5.5.4.

5.3.5 Main activities and working approaches

The main activities and working approaches are described in two subsequent ways. In the present chapter, some more general and descriptive aspects of the working approach are provided. The topic is conferred separately for the RD in this chapter and for the ED in chapter 5.4.3. Furthermore, different working approaches and co-operational aspects will be discussed with the help of a number of case studies of selected projects implemented by MAFF in chapter 6.3.

At first, it should be noted that ’the approach‘ of the Research Division did not exist. A large number of different research techniques and methods were applied. Depending on the individual task and context, a selection of methods were applied. A number of regulations on these issues existed, but these were of a more administrative nature and were generally rarely consulted.

This made it difficult to discuss the approaches applied. In an attempt to provide an overall characterisation of the RD, the working approach could be called ’classical‘ or ’traditional‘. Traditional or classical refer to the fact that the division was strongly influenced by what could be called the general ’agricultural wisdom‘ of the sixties to the eighties. The best proof for this can be seen in the strong focus on experimental on-station work. Nevertheless, newer ideas made their way into the system. These were promoted by externally funded technical assistance projects. In this respect, the division could also be characterised as, ’in the search process for the new approaches‘. This search process was very open and ongoing. ’Open‘ characterises the situation in Fiji very well. People listen very well with considerable respect for new ideas promoted by outside assistance. There was little resistance to new approaches, and a good climate for experimentation. This climate was also beneficial for the research activities of this study. While this openness facilitated the introduction of new methods, it still does not say anything about the final acceptance of these methods and their institutionalisation. This aspect will be looked at in more depth in chapter 6.2.5.

As the first aspect of working approaches, the main sources of information used by researchers were reviewed. Table 15 displays the results.

Journals and books ranked as the most common sources of information. This was followed by training workshops, private sector, experts and the Fiji farm management manual on one level. Farmers, other colleagues and subject matter specialists followed in last position.


80

At first sight, these findings appear rather typical of researchers, who use books and journals as their main sources of information. Based on observations during the field phase, the researcher would question these findings. During the seven month that the research team worked in the main agricultural library in Koronivia, not a single research officer carried out a literature search<57>. The observation that researchers were not ’up to date‘ with international developments in agriculture was also mentioned by several experts with whom the researcher had contact<58>.

Table 15: Main sources of information used by the Research Division

Information sources

Frequency
(n=8)

Journals

3,0

Technical books from the library

2,8

Training workshops

2,5

Private sector, companies, traders

2,5

Fiji farm management manual

2,5

Project experts

2,5

Progressive farmers

2,3

Other colleagues

2,3

Subject Matter Specialists

2,3

Ranking scale: 1= rare, 2= occasional, 3= common.

The fact that other parties such as experts, farmers and colleagues also functioned as sources of general information, indicated that researchers exchanged views with their environment quite regularly.

The main activities of RD staff were collected with the help of an activity profile. In the follow-up questionnaire the respondents were asked to rate the time spent on various activities for an average month with the help of 40 points (1 point equivalent to1 half working day). Figure 17 illustrates the findings expressed as a percentage of total time allocation.

Officers spent considerable time with the design and planning of research activities (10-15 %). The executive field officers spent about one quarter of their time with activities on-station, while the middle management staff spent only about 15 % of their time. Surprising, was the fact that the executive officers spent more than 20 % of their time with on-farm activities. This result could be seen as a first sign of the impact of the training inputs provided through the various donor funded assistance projects. Researchers started to change their role and enlarge their methodological toolbox. Nevertheless, on-station trials still played the dominant role.

Analysis of trials and activities was another important activity field. Here, the senior staff spent about 18 %, while the executive officers spent about 12 % of their time. These figures reflect that senior staff needed more time for discussing and assisting the analysis of the various ongoing research projects. While the executive staff were mainly responsible for the field implementation and initial steps of analysis, the senior staff assisted in particular with more difficult steps of analysis or help in the interpretation of


81

findings. Rather little time was spent on report writing (11-14 %). This was certainly a weak point in the system. A look at a number of reports revealed that these were often written in a very basic format and could by classified as simple trial descriptions. In particular interpretation and conclusions were weak.

Figure 17: Time allocation of Research Division staff

Source: own data, follow-up questionnaire.

Field days and extension accounted for a very little part (5-6 %) of the work of researchers. But as this was not part of the official mandate, this was also well understandable. However, from the AKIS perspective, enforced co-operation with extension would be desirable. That this was taking place to a considerable degree, could be concluded from the generous time allocated to on-farm activities. These were generally carried out in close co-operation with extension staff.

In-service training was the key instrument to improve the knowledge base and abilities of staff. The questionnaire results indicated that senior officers make far more use (11 %) of training opportunities than lower grade staff (4 %). From informal discussions with research officers another aspect emerged. Making use of training was very person specific. While some officers tried to attend all possible training opportunities, others were positively uninterested and might not participate for years on a single course.

Other activities accounted for close to 10 % of total time. This figure does not require further discussion. The above figure illustrated in broad categories how researchers spent their working time. It did not yet say much about the methodology, or in other words, what and how things were done.

Research work was organised in several sections or small working groups. Mainly commodity-oriented, these groups were headed by an SRO or RO officer. One to four executive officers constituted a group. Each group worked on a number of different research topics. The number of research topics dealt with by individual officers was very


82

variable. Most officers worked on 3 or more research topics per year. Only one officer mentioned 2 or less topics. The average number of research topics per researcher was four.

It can take a long time for a research topic to lead to a recommendation. Many research questions never lead to a useful final recommendation. This was also the case here. While each officer worked on three or four topics per year, the average number of recommendations produced was only 1.6 per year. Again, there was a considerable variation. Four officers only produced one recommendation once every second or third year, while two officers claimed to be very productive with 2 or more recommendations per year.

As the final step, a recommendation had to be approved by an official RD meeting. The other steps necessary to reach this stage are shown in Table 16.

Table 16: Steps before recommendations release

Methodological steps

Number of researchers

Rarely done

Occasionally done

Always done

On-station trials

0

0

7

On-farm verification

 

1

6

Test for environmental safety

1

3

3

Analysis of market potential

1

2

4

Economic analysis

0

2

5

Documentation in research notes, technical bulletins

0

1

6

Publication in Fiji Agricultural Journal

1

3

3

Preparation of farmer leaflets, extension materials

0

4

3

Meeting for official approval

0

1

5

The first prerequisites were on-station trails. This was mentioned by all 7 researchers. If these trials proved successful, then, on-farm verification trials were conducted. This was always done. Only one researcher mentioned it being done only occasionally. This indicated a positive trend. Informal discussion with researchers at the beginning of the research phase, and the linkage workshop held in 1994 still, considered on-farm trials as an exception.

Tests for environmental safety of new recommendations were becoming more frequent. Three officers mentioned them as done occasionally and another three as always done<59>. An assessment of the market potential and an economic analysis was mentioned by 4 and 5 officers respectively, as a step that was always performed, while 3 and 2 respectively, considered it as a rare or occasional routine. Although this ranking by RD staff indicated that economic and market aspects were considered, a few comments appear necessary.


83

Dealing with economic aspects was a rather new field for the RD. Only a single socio-economist was assigned to the work field. One person is far from adequate and this was also seen so by RD top management (cf. Dowling and Kumar 1995, 42). This is an area, to which the RD will have to attribute considerable more attention in the future. More farm economists will be necessary to enable a sound check of the economic viability of new technologies.

The documentation of research findings was another weak area of the RD. Six researchers stated that research findings were always documented as research notes or technical bulletins. These documents, however, are in-house documents of the RD and little read by extensionists or other interested outside parties. Proper publications in the Fiji Journal of Agriculture were much less frequent: one officer rated it, as rarely done, four as occasionally done, 3 officers and as always done<60>. Similar, was the rating for the preparation of farmer leaflets or other extension materials: 0 rarely, 4 occasionally, and 3 always done. For the practical success, extension and farmer oriented documents are much more important than scientific publications. To ensure better adoption rates for recommendations, this was an unsatisfactory result. In the AKIS terminology this could be considered as a crucial underused interface.

5.3.6 Summarising conclusions

The new mission statement of the Research Division focused the work on applied agricultural research and the dissemination of research findings: This statement can be seen as a modern and suitable mandate. This represents certainly a step in the right direction for the division.

For the assessment of the divisions resources and staff development, a look back at the hypotheses formulated at the beginning is necessary:

H 3

Lack of funding or strong fluctuations in funding are a major reasons for low output and adoption of innovations.

H 4

A critical mass of well-educated staff is a prerequisite for effective development of innovation and their dissemination to farmers.

The conditions formulated in hypothesis three (lack of or fluctuating funding) were certainly not given at the RD. The budget had a positive development, constant funding was available and the existing facilities were adequate to conduct sound research. Therefore, a lack of resources could not be seen as the limiting factor for performance. Considering that the output of the AKIS as outlined in chapter 4 was very limited for the last two decades, this means that despite good funding, only limited output was achieved. Considering that the conditions of the hypothesis were not met, the hypothesis cannot be validated here definitely. The case will be examined a second time, in respect of the situation for the Extension Division (chapter 5.4.4).


84

Hypothesis four, however, could be seen as confirmed. The RD lacked the ’critical mass‘ of well-educated staff. Sufficiently qualified staff were rare, thus affecting the quality of work. Therefore, low staff qualifications could be seen as a major reason for the low output of the AKIS as a whole.

The consequences of low staff qualifications can be illustrated by a look at research trial designs<61>. Many field trials conducted by the RD were simple copies of earlier work. Officers did not dare, or have the abilities, to conduct new ’explorative‘ research. With such a ’business as usual‘ approach, where trials are always done in the same traditional way, no progress can be achieved in research. Work of the RD could be characterised by too much routine, too little creativity and research in the real sense of researching.

Based on this problem description, it is clear that the RD has to focus on increasing the qualification of its staff. The division should facilitate all opportunities for further education, promote overseas leave for study purposes, and increase training opportunities through project linked activities. In the medium to long term, benefits would also be possible through better curricula at the local college of agriculture (FCA). This might raise the entry qualification of staff at the Ministry. In the medium to long term, the division should attempt to make sure that at least half of all researchers attain a Bachelors, or better, a Masters degree. With low qualifications, not much more than ’routine research‘ can be expected.

In the follow-up questionnaire, researchers indicated that there was no sound system in place for the determination of research priorities. Determination of priorities was intransparent and research priorities were largely perceived as imposed by the Government. ISNAR played a major role in suggesting priorities for research, and stressed the need for more farmer orientation in their first review as early as 1981. A systematic analysis of farmers‘ problems has still not taken place. To better focus the direction of priorities, this would be urgently required. A model of how that could be undertaken, will be presented in chapter 5.5.4.

The review of researchers main sources of information indicated, that they lacked up to date information. Researchers would need to review international developments more closely and make better use of the existing library services. Preuss (1994, 222) came to similar conclusions for small research systems in Africa. He called for the need to increase the ’borrowing capacity‘ of research systems. Small counties have to rely much more on existing information elsewhere rather than trying to do everything themselves. He further pointed out the need to allocate staff and funding, for increasing the borrowing capacity. This recommendation is certainly also very relevant for research in Fiji.

The review of research activities and methodologies used, indicated that the division started to adopt some of the new methodologies promoted through donor projects. At the beginning of the research phase in 1993 the RD was still almost exclusively involved with on-station research. However, the training workshops held by technical assistance projects, seem to have shown some effects in the meanwhile. The follow-up questionnaire findings indicated that on-farm research played a growing role in the division. Direct co-


85

operation with farmers and participatory technology development, were gaining ground in the division. This finding was very encouraging and should be further consolidated in the future.

5.4 Extension Division

5.4.1 Mission and goals

The extension division (ED) is, size-wise, one of the major players within MAFF. However, compared to research, extension had a lower reputation. This lower profile was also visible in the lack of a clear mission statement and outline of organisational goals. In an interview with the Director of extension<62>, he explained that there was no official paper that highlights the division‘s policies or goals. He himself, unhappy with the reply added:

“We have no clearly defined methodology. Our work comes close to the T&V system, but it is not quite that. Strictly speaking, we do not follow any approach by the word. We try to find our own Fijian way."

However, screening the annual reports form 1984 to 1995, something close to a short mission description could be found in the 1994 report:

“The ED has been the main arm of the Ministry responsible for promoting and facilitating the production of different types of food for various needs and the actual production of primary commodities for export, through the efficient use of the country‘s agricultural resources. The objectives of the extension task have been to guide the growers on proper utilisation of agro inputs in order to reduce production cost, and to diffuse integrated technology (traditional and improved) that is appropriate to the financial situation of Fijian farmers" (MAFF 1995, 90).

The two statements give a first notion of the work of the division. Goals were not clearly set. The division rather stood in the second line and waited for orders. Whatever directives came from Government or MAFF top management, the division was the ’main arm‘ to implement these. This was a very submissive understanding of its function. Technologies ’are to be diffused‘ and farmers ’are to be guided‘. This approach was very far away from modern concepts such as participatory technology development (PTD) or participatory extension work. The vague and unclear definitions of goals represent a very poor framework for a successful extension work.

5.4.2 Organisational structure, available resources and staff qualifications

The ED is structured according to geographical areas. The director of extension is based in headquarters that are located in the capital of Suva. Four geographical divisions (central, western, eastern and northern) have their respective offices in the main provincial cities. Besides these main offices, a number of smaller extension offices are


86

located across the country. Some of the offices are under one roof with research stations (Sigatoka, Taveuni coconut centre, Seaqaqa). Each division is headed by a PAO. The practical fieldwork is carried out by the lower grade staff ATO and AA/SAA. In the middle of the hierarchy are SAOs and AOs that have more co-ordinating and supervising functions and usually join field staff only for major field events. The staff development in the ED is depicted in Table 17. Unfortunately, the available statistics were not consistent over the years. For several years, only total staff number figures were available.

Table 17: Development of staff in the Extension Division

Grade

1984

1989

1990

1992

1994

1995

1998

Director

1

 

 

 

 

1

 

PAO

4

 

 

 

 

5

 

SRO/RO

 

 

 

 

 

 

 

SAO

1

 

 

 

 

14

 

AO

11

 

 

 

 

11

 

ATO

28

 

 

 

 

37

 

AA/ SAA/ STA/ TA

112

 

 

 

 

86

 

Admin. staff

49

 

 

 

 

 

 

Total established

206

147

115

167

155

154

206

Unestablished

 

122

91

182

145

 

172

Grand total

 

269

206

349

300

 

378

Vacant positions

17

 

 

 

 

11

 

Source: own compilation from MPI 1985, 65; MPI 1990, 92; MPI 1991, 95; MAFF 1993, 45; MAFF 1995, 90; MAFFA 1996, 90. MAFFA 1999, 66.

Unlike the Research Division, that had a slow, but gradual increase in staff numbers, development in the Extension Division was characterised by high fluctuations. Established staff positions were above 200 in the early eighties and dropped sharply to 115 in 1990.

Since then, staff numbers have climbed, with several ups and downs, to 206 in 1998. The number of unestablished staff, followed in large terms the same trend, and ranged from 91 to 182 positions. Several positions were occupied with staff filling posts in an acting capacity basis only, as no officers that met the necessary requirements for promotion (formal qualification and years in service) were available.

These high fluctuations in staff numbers compared to the RD are an indicator that the ED had less influence in top management of MAFF. Not all divisions were equally affected by budgetary cuts, and some divisions obviously managed to negotiate better conditions.

In the practical work, such heavy fluctuations in staff are difficult to digest and certainly have negative effects on the quality of work. High numbers of new staff joining and losses of qualified older staff, affect the knowledge base and disrupt learning processes over time. Work becomes more short sighted and less guided by a long-term perspective. Consequently, the accumulation of experience and knowledge is reduced. These


87

unfavourable conditions were one main reason for lower staff morale in the division. This could be felt at any time while working with extension staff.

No official statistics were available regarding the level of formal qualification at ED. Instead the qualification and position of ED staff that participated in the follow-up questionnaire are depicted in Table 18.

Table 18: Qualification and position of Extension Division staff

Highest qualification

Director

PAO

SAO

AO/ RO

GT

ATO

SAA/ AA STA/ TA

Total

Doctorate

 

 

 

 

 

 

 

0

Masters

 

 

 

 

 

 

 

0

Bachelor

 

 

(1)

2

 

(2)

(1)

2(6)

Diploma DTA

1

 

1

2

 

5

7

15

Certificate

CTA

 

 

1

 

 

 

2

3

ODILT

 

 

 

 

 

 

1

1

Total

1

 

2

4

 

5

10

21

Figures in brackets indicate multiple qualifications.

The sample can be considered as representative of the overall divisions. With the exception of the position of PAOs, all position levels are presented. The figures indicate that the level of qualification in the ED was considerably lower than in the RD. Within the division there was not even a single officer that held a Master‘s degree. The majority of staff, including the director, held the DTA obtained at FCA. Only six officers held a Bachelor degree.

It is self evident that the overall low level of formal qualification represented a serious constraint for the development of the division and limits the actual and potential quality of the output of the division. The average qualification of a DTA has to be considered as more of a technical qualification only. The college course teaches technical information and skills, but it does not enable a critical scientific approach or train the ability for independent and self-determined working or learning. Thus, staff qualifications had to be seen as a major limiting factor for the development of the division.

The budget development of the ED is shown in Table 19. Unfortunately, the budget statistics were not consistent over the years and need to be interpreted with care. Budget development was very volatile, with a total budget of 4.78 million in 1983, a peak of 6.68 million in 1989 and a decline to 4.43 million in 1995. This corresponded to an average annual growth of 5.7 % in the 1980s that was followed by a sharp annual decline of 9.5 % in the 1990s. This decline was also apparent, when comparing both budgets of the RD and the ED. While the budget of the ED still exceeded the RD budget by 3.1 times in 1984, it declined to only 1.4 in 1995. This shows how drastically, the ED has lost importance compared to the RD.

At the end of the 1980s the largest item on the budget was capital expenses (4.48 million). This stood for foreign aid contributions that were shown as part of the Ministry budget.


88

Table 19: Extension Division‘s budget development

 

1983

1984

1989

1990

1992

1993

1994

1995

Budget items

'000 FJD

Established staff

1522

1439

 

 

 

 

 

 

Unestablished staff

473

523

 

 

 

 

 

 

Travel & Communication

111

111

 

 

 

 

 

 

Maintenance & operations

185

196

 

 

 

 

 

 

Purchase goods & services

37

37

 

 

 

 

 

 

Operational expenses

 

 

2205

2837

2878

3591

3354

3898

Operating grants and transfers

358

558

 

 

 

 

 

 

Capital expenses

 

 

4482

3783

2864

1027

788

177

Capital purchase

10

 

 

 

 

 

 

 

Capital construction

736

 

 

 

 

 

 

 

Special expenditure

 

68

 

 

 

 

 

 

Capital grant and transfer

1352

2458

 

 

 

 

 

 

Value added tax

 

 

 

 

 

421

105

360

Total

4784

5388

6686

6620

5742

5039

4247

4435

Source: own compilation from MPI 1985,57; MPI 1989,77, MPI 1990,85; MPI 1991,87;MAFF 1993,45; MAFF 1994,59; MAFF 1995,100; MAFFA 1996,103.

These had declined to less than 0.2 million in 1995. However, this was mainly a consequence of different accounting procedures. Foreign aid contributions were no longer shown as part of the Ministry budget. Development aid to agriculture had declined in that period too, but by far not as drastically as this budget line suggests. Rather little can be said about the budget breakdown itself, as in recent years, only operational and capital expenses were differentiated. Operational expenses grew steadily during the 1990s. This indicated that the functioning of the division was still safeguarded. From discussions with staff, the researcher concluded that the budgets were tight at all times, but they still provided sufficient funds for a certain number of field activities.

Adequate funds are certainly an important aspect for successful extension work. However, this is only one aspect among many others. Even with funds, it is still more important to promote the right extension messages with the right methodologies. In a discussion among the research team Mr. Naerecoko made the point on the issue:

“We (MAFF) are always short of funds. Often there are no funds to run the vehicles or implement activities. But when we had the JICA (Japanese aid) rice project, we had plenty of funds and we could carry out all activities we liked. But we were not any more successful. All the support that went in to the rice production schemes is considered a failure now."


89

The available physical infrastructure of the ED could be considered as comparatively good. In 1995 the division disposed of 86 vehicles, 17 tractors and other farming machinery and implements. Furthermore, the division maintained 144 staff quarters (MAFFA 1996, 91). The available annual reports did not specify the actual conditions of the vehicles and other machinery. However, own observations made during the fieldwork, suggested that the conditions are generally reasonable. Calculating the number of established staff (154 in 1995), less than 2 officers (1.8)<63> shared one vehicle. This is certainly a very good ratio that many NARES in African countries could only dream of.

The relatively high standard of the infrastructure could be considered as a strength of the division. The high fluctuations in both budget and staff numbers, combined with the low level of formal qualifications, have to be seen as major weaknesses of the division.

5.4.3 Priority setting, activities and working approach

To determine extension priorities, the same procedure as for research priorities was applied. Both research and extension staff were requested to rate the influence of selected actors on the formulation of extension priorities. The results are depicted in Table 20.

Table 20: Actor influence in the formulation of extension priorities

Actors

Strength of influence as seen by

ED1) (n=21)

RD2) (n = 7)

Fiji Government policies

4.3

4.3

Extension

3.7

3.6

Donors and projects

3.4

3.5

Research division as a whole

2.8

2.8

Farmers

2.7

3.4

Private sector

2.7

3.7

Personal preferences of extensionists

2.5

2.2

Regional/intern. Research centres

2.5

2.5

1) Extension Division, 2) Research Division, Ranking scale: 1= none, 2= weak, 3= medium, 4= strong, 5= exclusive.

Both the RD and the ED rated the Fiji Government as the most important player. They equally agreed on extension in second position (3.6-3.7), with medium to strong influence. Donors and projects were seen closely in third position, as medium to strong actors (3.4-3.5). The RD as an actor, reached a medium role by both groups (2.8). Controversial, was the view on farmers and the private sector, that were given a rather strong role by research (3.4-3.7), while ED staff rated their influence as only weak to medium (2.7). The influence of personal preferences of extensionists was rated weak to medium by both groups (2.2-2.5). The influence of regional or international research centres was rated on the same level (2.5).


90

The rating showed that the Government was perceived as the main player. The earlier rating of the Research Division on research priorities, yielded the same result. Using the Engel typology presented in the theory chapter 2.1, the Fiji AKIS could be classified as a ’policy driven‘ configuration. Recalling Engel‘s basic characterisation of policy led configuration, several aspects could also be observed in Fiji: the Government is the main source of finance for research and extension. Co-ordination is achieved via the existing hierarchical structures and policies. Industrial actors (and farmers) generally play a secondary role. The configuration is characterised by a rigid definition of tasks and limited flexibility.

These findings stress that there is a need to establish a transparent official procedure to determine extension priorities. Such an official procedure could increase the understanding of all actors and raise their motivation to implement and achieve these goals. Priorities that were determined in a more transparent process rather than being hierarchically imposed, would have better chances of becoming successful. A proposal, of what such a procedure could look like, will be made later in chapter 5.5.4.

Activities and working approaches of the division are described in two subsequent ways. In this chapter, some more general and descriptive aspects of the working approach are discussed. The presented findings are based on the follow-up questionnaire, supplemented by experiences and observations gained during the fieldwork. Furthermore, different working approaches and co-operational aspects are conferred with the help of a number of case studies of selected projects implemented by MAFF in chapter 6.3.

At first, the main sources of information used by extension staff will be reviewed. Table 21 depicts the findings.

Table 21: Main sources of information used by the Extension Division

Information sources

Frequency
(n=21)

Own knowledge, experience

3.0

On-farm trials

2.4

Fiji farm management manual

2.4

Extension leaflets, handouts

2.3

Progressive farmers

2.2

Training workshops

2.2

Subject matter specialists

2.1

Technical books, library searches

2.0

Other colleagues

1.9

Personal contact with researchers

1.8

Daily papers, journals

1.8

Project experts

1.7

Private sector, companies, traders

1.3

Ranking scale: 1= rare, 2= occasional, 3= common.

The most important source of information for extension staff was own experience and knowledge. The second most important source were the Fiji farm management handbook and on-farm trials. Extension leaflets and handouts also played a major role. Occasional sources of information were, progressive farmers, subject matter specialists and training workshops. Technical books, personal contacts with researchers and extension colleagues ranked a little lower. Journals, papers, project experts and private sector actors were seen as rare sources of information.

The fact that extension officers relied more or less on themselves, has to be seen in the context of the fieldwork. Frequently out in the field, it was more difficult for extenionists to access other sources of information. On-farm trials


91

with farmers naturally were their best sources for new information. As main sources of written information, extension staff used only the official Ministry publications the Fiji farm management handbook (FFMH) and farmer leaflets. It was shown earlier that the FFMH, edited by EP&S, would require an update as it contained many outdated information. As a consequence, extension staff are not well up to date with the latest developments, in particular with new developments outside Fiji. This is further aggravated by the fact that very few scientific books or journals were read<64>. As most projects were attached to research, extension staff also had less exposure to foreign experts. This further limited access to new information.

Without any doubt, good extension work should be focused on farmers and their specific conditions and problems. This requires a good definition of the client group. As will be pointed out later in chapter 5.5.1 on farmers, the ED lacks a proper definition of its target group. There were a number of classification criteria that were used in the day to day discussions by extensionists (part-time/full-time, subsistence/commercial farmers, small/ medium/large farms), but none of these terms or categories were properly defined or used as a general rule. Consequently, each extension sub-office had its own list of farmers that were considered as their target group. This situation encountered during field work was also confirmed by the follow-up questionnaire. The number of farmers per extension area are depicted in Table 22. These numbers varied extremely from as low as 120 to 4,000. As a consequence, the ratio of farmers per extension officer varied equally greatly from 28 to 2,000. The average figures, with 187 as the rating by the group of middle management and 660 as the rating of field staff, appeared more realistic. Nevertheless, the ratings of both groups should match better and the discrepancy can only be explained by the very different estimates of the number of farmers per area. Using the total number of non-sugarcane farmers (65,000) according to the last Census, and the total number of extension officers in lower grades (124), the ration comes to 524. This ratio lies in between the two former estimates. This counter-check indicates that after all, the different estimates lay in the same magnitude.

Table 22: Ratio of farmers to extension officers

 

Middle management (n=6)

Field staff (n=15)

Mean

Range

Mean

Range

Number of farmers per extension area

1880

700-4000

962

120-3000

Ratio farmers per extension officer

187

28-270

660

89-2000

To cross-check the above estimates, the extension officers were also asked to indicate in as far as they felt that the current number of extension officers were adequate to deliver a good service to farmers. Half of the officers (10) rated the ratio as adequate while the other half (10) observed a slight shortage. None of the officers considered the options severe shortage or too many extension officers. These findings suggested that the number of extension workers in relation to their client-farmers was medium to good. As calculated above, about 500 farmers per extensionist was certainly too large a number to


92

provide good individual advise to all farmers, but under practical conditions, only a small percentage of all available farmers actually seek advise or participate in extension activities. Thus, the number of extensionists appears sufficient to cover the demand of farmers.

To determine which group of farmers benefited most, the extension officers were asked to rate out of 10 points which group of farmers was most active in requesting advise. The largest group was medium scale farmers with 42 %, followed by subsistence farmers with 30 % and large producers with 28 %. These results showed a clear bias in favour of large farms. Although large farms probably did not represent more than 10 % <65>of all farms, they obtained close to 1/3 of all advice.

Such a focus on larger farms is certainly no problem per se. But the approach has some risks. It is not certain that larger farms have the best productivity and can be considered as model farms for others. However, all too often this was done by extenionists. During the fieldwork for the RRA/PRAs, extension officers had a preference to introduce the research team with their ’biggest and most advanced‘ farms first. But copying those practices of big farms is usually out of the economic reach for medium or small-scale producers. Thus, it may be concluded here that extension had a tendency to promote the wrong model farms.

An insight into the work of the division is further provided by the activity profile of the extension officers. In the questionnaire the officers were asked to rate the time spent on various activities for an average month with the help of 40 points (1 point equivalent to half a working day). Figure 18 illustrates the findings expressed as a percentage of total time allocation. Field staff spent about 50 % of their time actually in the field, while this constituted only about 20 % for middle management staff. Regardless of the position, individual advice as method ranked much higher than group extension work and accounted for about 2/3 of all time spent in the field. This focus on individual advice certainly reduces the overall number of farmers that can be reached. It may also indicate that participatory extension tools, that are predominantly group-oriented methods, still only play a minor role in extension practice. As could be expected, middle management staff allocated about twice as much of their time to planning activities (25 %) than the field officers with about 13 %. Their co-ordinating and supervising roles explained this difference. Reporting duties required about 11 to 13 % of the time, with little differences between the management levels. Middle management spent about 18 % of their time in office meetings, while the field staff spent less than half of that figure. Important to note was the time for in-service training. Middle management spent about 11 %, while the time spent by field staff only came to slightly more than half of that level. This trend was already observed for the RD as well. Spending that much time on training was certainly a very positive aspect. This was an indication that the low level of qualifications was acknowledged by the Ministry and that attempts were made to reduce this weakness.


93

Figure 18: Time allocation of Extension Division staff

Source: own data, follow-up questionnaire.

Other duties took up about 10 % of their time. This figure does not require further discussion. The above figure illustrated in broad categories how extension staff spent their working time. It did not yet say much about the methodology, or in other words, what and how things are done. A few aspects of these issues are discussed in the following paragraphs.

Table 23: Main reasons for farm visits

Main purpose of farm visit

%

Discuss, present new or older recommendations

22.3

Follow-up activity

27.1

Data collection

11.9

Visit demonstrations or on-farm trials

17.1

Distribute inputs or materials

6.7

Announce meetings, training or other activities

7.6

Other

7.6

To gain a better idea about the main reasons for farm visits, the extension officers were asked to rate their reasons with the help of 20 points. The findings are depicted in Table 23 and expressed in percent. The first striking result was that the discussions about new or older recommendations only occurred in 22% of farm visits. In other words this meant that four out of five farm visits had other reasons. An explanation for this low rating could be seen in the absence or seeming absence of innovations. In several meetings, extension officers blamed researchers for not providing them with suitable new messages for extension. While this aspect was certainly valid to a certain extent, better two-way communication, would also require that extensionists search more actively for the existing literature to extract useful messages for farmers. This is still an activity that would require more promotion. Extensionists seemed to wait for researchers to become informed, rather than to inquire actively themselves about new information.


94

The most frequent reason for farm visits was the follow-up of some earlier activity with 27 %. The next major reason, with 17 % were field demonstrations or on-farm trials visits. As such trials require constant observation and management this time allocation appeared realistic. Another main objective for farm visits was data collection with close to 12 %. Considered as the ’long arm of the Ministry‘, extension carried out or assisted in many surveys. Common reasons were national statistics or surveys to co-ordinate disaster relief measures after cyclones. The distribution of inputs or material was a minor reason for farm visits with 8 %. This was certainly a positive development, as such services should be provided by the market and not extension. In the past, in particular during the ’heyday‘ of the rice development project, extension assisted with the distribution of fertilisers and pesticides, provided tractor ploughing services and rice harvesters. Since the introduction of the new liberal policies in 1993, many of these services were closed down.

The announcement of meetings, or other training activities for farmers together with other reasons accounted for about 15 % of farm visits. This highlighted that these organisational duties can be very time consuming. This was felt in particular when transport means were unavailable, or farming communities were in distant locations. PRA activities and group extension work also required frequent preparatory visits to inform farmers about the activities to come.

In order to gain some insight into the technical questions extension have to solve during their work, the officers were asked to rate the difficulty of farmers‘ problems with the help of a 10 point system. The findings showed that the majority of farmers‘ questions (62.5 %) were rather easy and could be answered straight away. 19.5 % of the questions were trickier, but they could be solved with the help of other colleagues. 14.5 % of problems required the consultation of research officers to identify solutions. Finally, in 4.5 % of the cases, no solutions to the raised problems could be identified. This self-evaluation of extension officers seemed to indicate that they could solve matters in 95 % of cases. It would be interesting to know if farmers ranked the advice received as equally successful<66>.

The last question investigated the way recommendations were discussed with farmers. It was of interest to see, if and how extension officers modified or adopted messages to suit them to specific farm conditions. The result was interesting, as more than three quarters of all officers (77.5 %) stated that they made such farm specific adaptations frequently. Only about one fifth (22.5 %) added such options after farmers had raised difficulties with the application of recommendations. Not a single officer stated discussing recommendations without specific adaptation. Satya Bahn Singh (senior extension officer based in Nausori) explained how he adapted a recommendation to specific farmer requirements:

“If we, for example, talk about chemical fertiliser, I first find out if the farmer has ever used it before. If, he has no experience, then I suggest starting with the smallest packet available at the MH (annotation: supermarket chain) and advise the farmer to apply it on a very small piece of land and to compare it with his

95

usual practice. Once the farmer gains experience, I suggest increasing the area and the amounts to be applied.“

The example showed that extension officers could deal with recommendations quite flexibly. The experience of the officer certainly played an important role. The above example quoted an officer with 20 years experience on the job. A younger colleague might not have had the experience to deal with recommendations equally flexibly. And one should not forget, modifying a recommendation too much or in the wrong way may result in undesired effects. Only experience can help to identify the range of acceptable modification. According to the follow-up questionnaire, extension staff seemed to be rather experienced in this respect<67>.

5.4.4 Summarising conclusions

The extension division is a weak actor within the Ministry. First of all, this was visible in the lack of goals and a clear mission statement. The division must attempt to gain more profile. A clear set of goals and a mission statement should be elaborated on urgently.

The review of staff and budgets showed that the division was to a certain extent ’the play ball‘ of the Ministry. Both staff numbers and budgets showed very strong fluctuations over the years. A look back at the two hypotheses already applied to RD in the last chapter shall be made here:

H 3

Lack of funding or strong fluctuations in funding are a major reasons for low output and adoption of innovations.

H 4

A critical mass of well-educated staff is a prerequisite for effective development of innovation and their dissemination to farmers.

Unlike the RD which had a constant growth of budget, the ED suffered from strong fluctuations in budget. In this respect, the conditions formulated in hypothesis three were given. Thus, funding problems could be considered as a reason for the low output of the division. Looking at the divisions from the wider systems perspective, it is clear that the system output would depend on the adequate or balanced funding of both the RD and the ED. This has not been the case. As only the RD was well funded, the ED suffered from considerable fluctuations. This gives a very plausible explanation for the low output of the AKIS, thus validating the hypothesis.

However, other factors also play an important role for the output of the AKIS. Staffing and staff qualifications are important factors that need to be considered. These factors are considered in hypothesis four.

The qualifications of staff at the ED has to be seen as the other major constraint factor. To enable good extension work, qualifications of staff should not be far behind those of researchers. This, however, was not the case. Qualifications at the ED were much lower


96

than at the RD, and already qualifications at RD were considered as inadequate. The fact that, besides qualifications, the numbers of established staff fluctuated heavily (115-206) enforced the problem. Therefore it is well justified to speak of a critical shortage of well-educated staff in the ED. Consequently, the hypothesis could be considered also as valid.

Evident conclusions of the above analysis are that budgets in the Ministry require better co-ordination and orientation towards medium to long term objectives. Staff qualifications require improvement. For the ED this means that in particular middle management staff should be required to have at least a Bachelors degree or better a Masters qualification. The division should facilitate all opportunities for further education, promote overseas leave for study purposes and increase training opportunities through project linked activities. Medium to long term benefits would also improve through better curricula at the local college of agriculture (FCA). This might raise the entry qualification of staff at the Ministry. Job rotation of research and extension officers could also help to build linkages and enable better learning opportunities.

Another serious constraint for the ED is the lack of a sound system for the determination of extension priorities. A systematic analysis of farmers‘ problems and a profound screening for potential solutions and appropriate extension messages has not taken place. A model showing how that task could be undertaken, will be presented later in chapter 5.5.4.

The review of ED staffs‘ main information sources revealed that they lack up to date information. Extension staff relied too much on existing Ministry material. Officers would need to review international developments more closely and make better use of the existing library services. Extension staff need to acquire information more actively. This would also make their interaction with farmers more effective, as they could draw on a larger stock of potential solutions for discussions with farmers.

The review of extension activities and methodologies used, indicated that the division started to adopt some of the new methodologies promoted through donor projects. At the beginning of the research phase in 1993, the ED had rather poorly developed client-orientation and the methodology was mainly inspired by a top down ’transfer of technology‘ mentality<68>. However, the many training workshops held by technical assistance projects seem to have already borne some fruits. The follow-up questionnaire findings indicated that extension officers have learnt some lessons and know what a more client-orientated and participatory extension could look like. This finding was very encouraging and should be further consolidated.

5.5 Farmers, the knowledge users

The first section of this chapter describes some socio-economic characteristics of farms with the aim to develop a farm classification system. Particular emphasis is given to farmers‘ needs and problems. In the last section of the chapter, a proposal is made to how farmers‘ problems could be integrated into the priority setting of the Ministry.


97

5.5.1 Towards user classification; some socio-economic characteristics of farms

According to the last agricultural census 95,000 farms operate in Fiji. The mandate of research and extension at MAFF are restricted to all non-sugarcane farmers. This group is comprised of approximately 65,000 farms (MPI 1992, 71). Compared with the mandate of many African NARES, this is certainly a rather small number, but the number is still much too high to take into account all farmers‘ problems individually. Thus, a problem of distinguishing between individual farmer‘s problems and problems that concern a larger number of farmers arises. It can be seen as a problem of prioritisation. Which of the many farmers‘ problems should be researched or addressed by extension. Commonly, NARES focus their activities on problems that concern a larger number of farmers. The practical problem is how to determine the common problems of these larger groups of farmers. Farm classification systems are often used to solve this dilemma. These systems are based on the working hypothesis that all farmers that fulfil the same criteria (e.g. farm size, agro-ecological zone, etc.) have similar problems and conditions for farming. These farmers in one group (or class) are then considered as likely to require the same solution (recommendation) for a given problem. Evidently, such systems are a strong reduction of reality, and the fewer classes there are, the stronger the reduction factor is. Keeping this limitation in mind, they still represent a means to structure problems and come to more manageable solutions. This was also the main reason why the reseacher started an analysis of Fiji farmers‘ problems stressing farm classification systems. The search revealed that no such standardised farm classification system existed. Most professionals in either research or extension used their own set of criteria to define farming classes. Therefore, the research team decided to start with screening potential criteria for useful farming systems categories. In an initial brainstorming exercise with a number of resource persons, the research team collected potential criteria. In the first round, the following criteria were discussed:

Farm size, cultivated land, agro-ecological zone, main commodity grown, type of settlement, total farm income, full-time or part-time farming, educational status and ethnic affiliation.

The discussions revealed that the criteria were to a certain extent interrelated. But little was known about the real interrelations in the Fijian context. Most common practice was a grouping of farmers according to the main commodity marketed. However, it was felt that a single commodity based classification of farms did not reflect the diversity of farms sufficiently well. The key points for the selection of classification criteria were the practicality in terms of easy observeability in the field, and the question of the criteria enabling a relevant distinction in terms of farmer specific requirements for a given technology. After long discussions, the research team finally agreed to test the following three criteria during the first RRA:

  1. a) agro-ecological zone (lowland, upland)
  2. b) total resource wealth (low, medium, high)
  3. c) type of settlement (village, individual settlement)


98

The first criteria distinguished between the more densely populated lowland areas with very flat topography, and the very hilly upland areas with less population and more forest cover. The second criteria divided the farms according to total income into three groups with low, medium and high income. The last criteria described the major type of land ownership and indirectly indicated the ethnic group. The ethnic Fijian farmers generally live in villages and own their land (mataqali land), while other ethnic groups (Indians, Chinese, and fewer Fijians) live on individual farms in settlements and lease their land.

To assess the suitability of these three classification criteria a number of farm characteristics were explored. The selection of variables included the three main factors in economic theory: land, labour and capital. Investigations tried to reveal to which extent these factors represented major constraints in the different farming systems. Table 24 lists the variables investigated during the three RRA/PRAs.

Table 24: RRA/PRA exercises and specific variables investigated

 

RRA 1

PRA 2

PRA 3

presented here

Labour availability

x

xx

 

3

Farm income

x

 

 

3

Crop profitability

 

xxx

xx

 

Farm management

x

xx

xx

3

Attitude towards Innovation

x

xx

xx

3

Cropping

x

xxx

xx

3

Marketing

x

xx

xxx

3

Livestock

x

 

 

 

Problems

x

xx

xxx

3

Depths of investigation: x = low; xx = medium; xxx = high.

The first RRA had a very explorative nature and covered the largest width of variables. The following PRAs were then smaller in width, deeper in subject and more geographically focused. An extract of the findings is given in the following paragraphs. The complete findings are documented in separate publications.<69>

Before starting with the actual description of farms, it is essential to recall a few of the relevant weaknesses of the approach used. While analysing the farmer sub-system, theoretically, it would be most meaningful to know farmer‘s views as undistorted as possible. However, such a neutral perspective is difficult to realise. Various PRA tools helped farmers to express their own views. However, it needs to be added that some tools, in particular informal interviewing is always guided strongly by the interests of the person questioning<70>. Thus the farmers‘ perspective presented here could in fact be considered as


99

a blend of perspectives of farmers and those of researchers and extensionists interested in working with farmers. This bias should be kept in mind for the assessment of the farmer sub-system.

5.5.1.1 Labour availability on farms

As part of their daily routine farmers executed a number of different activities. Table 25 gives an overview of these activities for a lowland settlement. Annex 1 depicts the situation in an upland settlement in the central division.

Farmers‘ time allocation on these various activities showed a large variation. Farmers in the lowland settlements seemed to be working slightly longer hours than their colleagues in the upland areas (around 55 hours weekly compared to 47 hours weekly). These times included marketing and collection of wild food sources. Farmers in average group worked somewhat less than farmers in advanced groups. Women spent less time on agricultural activities (in between ¼ to ¾ of the time of their husbands). Farmers worked six to seven days a week. They spent one or two days weekly on marketing (2-18 hours). The most common market days were Friday and Saturday. Farmers have quite different marketing strategies. While some sit at the market and wait until all their produce is sold, others market more regularly via middlemen or exporters.

Important differences existed in the extent to which farmers spent time on education (0 to 12 hours a week). It was found that advanced farmers did not necessarily study more than average farmers.

Other important activities are communal and religious activities. Farmers allocated between 5-15 hours a week to both of these. There were no prolonged periods of labour shortages during the year. The only exceptions were rice farmers who faced difficulty during the rice harvest periods. Some farmers occasionally hired casual labourers. Recruiting labour was not considered a major problem. In the prioritisation process of problems, farmers scarcely mentioned labour as a problem, and the few cases that did ranked it with low priority.

Conclusions on labour availability. A very large labour variability between individual farms could be observed. This large variability made it impossible to draw clear trends for labour availability on farms in relation to the three classification criteria. Family size appeared as the main factor that determined labour availability. Therefore, the three classification criteria did not appear as suitable categories to distinguish farms according to labour, or, predict farmers‘ choices for new technologies in relation to labour.


100

Table 25: Daily labour activity profile for farmers in lowland settlements Serua

Activity

Group

Mon.

Tue.

Wed.

Thur.

Fri.

Sat.

Sun.

Av. per day

Total per week

Total per year

Farming

Average 1

6

7

4

9

4

1

2

4.7

33

1650

 

Average 2

6

7

5

5

5

2

2

4.6

32

1600

 

Advanced

7

7

7

7

10

1

4

6.4

45

2250

 

Women

4

3.5

3.3

2.7

3.7

1.5

0

2.7

18.7

935

Household

Average 1

1

1

0

0

1

0

0

0.4

3

150

 

Average 2

2

2

2

2

1

1

0

1.4

10

500

 

Advanced

1

1

1

1

0

1

1

0.9

6

300

 

Women

4

3

3.3

3

4.7

3

3.5

3.5

24.5

1225

Meals

Average 1

3

3

3

3

3

3

3

3

21

1050

 

Average 2

3

3

3

3

3

3

3

3

21

1050

 

Advanced

2

2

2

2

1

2

-1

1.9

13

650

 

Women

2.5

2.5

2.5

2.2

1.7

3

3.2

2.5

17.6

880

Relaxing

Average 1

2

2

2

1

0

3

5

2.1

15

750

 

Average 2

0

2

2

2

2

1

6

2.1

15

750

 

Advanced

3

2

2

2

2

3

-1

2.4

17

850

 

Women

1.5

2.5

2

1.5

1.7

0

5

2

14.2

710

Marketing

Average 1

0

0

4

0

5

4

0

1.9

13

650

 

Average 2

2

0

0

2

2

3

0

1.3

9

450

 

Advanced

0

1

1

1

0

6

0

1.3

9

450

 

Women

0.7

0.5

0.3

2.5

0.5

2.5

0

1

7

350

Shopping

Average 1

0

0

2

0

0

3

0

0.7

5

250

 

Average 2

0

0

0

0

0

0

0

0

0

0

 

Advanced

0

0

0

0

0

1

0

0.1

1

50

 

Women

0

0

1

0.7

1.3

0.5

0

0.5

3.5

175

Religious

Average 1

1

1

1

1

1

1

3

1.3

9

450

 

Average 2

1

1

1

1

1

1

6

1.7

12

600

 

Advanced

1

2

1

1

1

1

-4

1.1

8

400

 

Women

1.7

3

1.3

1

1.3

0.7

3

1.7

12

600

Communal

Average 1

2

0

0

0

0

2

2

0.9

6

300

 

Average 2

1

2

2

0

0

2

0

1

7

350

 

Advanced

0

1

0

0

1

2

1

0.6

4

200

 

Women

0.5

1.3

1.5

1

0.7

1

0.5

0.9

6.5

325

Education

Average 1

1

1

1

1

1

0

1

0.9

6

300

 

Average 2

1

0

1

1

1

2

0

0.9

6

300

 

Advanced

2

2

2

2

2

1

-1

1.6

11

550

 

Women

0.5

0.3

0.7

0.5

1.3

0.7

1

0.7

5

250

Sleeping

Average 1

6

7

7

7

7

7

8

7

49

2450

 

Average 2

7

7

7

7

7

7

7

7

49

2450

 

Advanced

8

6

8

8

7

6

-2

7

49

2450

 

Women

8

7

7.4

7.5

6.5

9

7.3

7.5

52.7

2635

Wild food

Average 1

2

2

0

2

2

0

0

1.1

8

400

 

Average 2

1

0

1

1

2

2

0

1

7

350

 

Advanced

0

0

0

0

0

-2

3

0.4

3

150

 

Women

0.5

0.5

0.7

1.3

0.5

1

0

0.6

4.5

225


101

5.5.1.2 Farm income and expenditure

Farm income was a very difficult question to determine. The proxy-indicators and wealth ranking procedures used, led to indecisive results. Income calculated on the basis of interviews in the first RRA did not always match well with the pre-selected three wealth groups. While farmers were very reluctant to disclose information on income, information on expenditures were disclosed rather openly. Calculated income ranged from 900 $ to 10,000 $ per farm. Income categories could be grouped as 1,000-1,999 $ (low), 2,000-3,999 $ (medium) and above 4,000 $ (high)<71>. However, this grouping represented ideal types. Several farmers in each group had to be considered as exceptions. Some farmers (in particular in the lowland areas) had substantial off-farm income. These part-time farmers with high total income often made low investments in agriculture compared to farmers that gained all their income from farming. This problem made it difficult to draw clear relations between income and farm investments. No significant differences in income between the various agro-ecological zones were observed. Equally uncertain were income differences between villages and settlements.

Expenditure for farm inputs was generally low. Middle and low group farmers spent up to 200 $ per year while expenditure on big farms reached 600 $ or more. Only a single farmer actually used an agricultural loan. A second farmer was still in the loan application process. Credit was generally limited to settlement farmers that were able to offer securities.

Conclusions on Income groups. The large variation in income between farms did not allow statements on the relation between income and the classification criteria, ’type of settlement‘, or, ’agro-ecological zone‘. Investment in agricultural inputs showed a certain relation to farm income. High income group farmers invested more in farming. The practical difficulties with the wealth ranking procedures<72>, however, led to the decision to distinguish only two income groups: average and advanced farmers. Low and medium income groups showed rather little difference in spending on agricultural inputs. For this reason they were combined to one group of ’average‘ farmers.

5.5.1.3 Cropping activities

The majority of farms cultivated in between 0.5 to 3 ha of land. Lowland farmers were better mechanised and used animal traction (rarely tractors), and thus, were able to


102

cultivate larger areas (2-3 ha), whilst upland farmers hardly used ploughs and cultivated in the range of 1-2 ha. Advanced farmers cultivated about 50 % more land than farmers in the average group.

The main crops grown are illustrated in Figure 19. Taro was grown in all areas as an important staple food and cash crop that achieves high prices. Cassava was the main staple crop. Sweet potato was mainly cultivated as a subsistence crop but in the lowlands it also played a role as a cash crop. Rice was almost only found in the lowland areas. Many farmers grew rice for home consumption because the prices are too low to market the crop. This was also the main reason why advanced farmers did not cultivate rice. Ginger is an important upland cash crop. Kava is the traditional upland cash crop and national drink. The production of vegetables had expanded considerably in the last decade and constituted an important cash crop in the lowlands. Amongst the advanced farmers some cultivated more than 1 ha of vegetables.

Figure 19: Cropping pattern by area and farm type

Source: own data, PRA 2.


103

Farmers practised several crop rotations for different blocks depending on the soil and topography. Common rotations were:

To determine the cultivation intensity and amount of fellow land was very difficult. A number of different PRA tools were used but some of the results remained contradictory. A central problem was the fact that farmers had no accurate idea of their farms and individual plots. A plot size estimation game showed that most farmers overestimated the sizes of their plots.

Field size estimates ranged from an under-estimation of 70 % to an over-estimation of up-to 400 %. Most common were over-estimates of 30 to 50 %. Big plots were estimated somewhat more accurately than small plots. Consequently, farmers‘ information on the size of their farms needed to be reviewed very carefully. The same problems applied to the mapping exercises, where farmers drew sketches of their farms. Plots in the uplands had very irregular shape, and in addition with the slope, were difficult to estimate. Distinguishing fallow and natural bush was another difficulty that added to the problem. However, keeping the difficulty in mind the results suggested that farmers kept between 20 to 50 % of their land as fallow to restore soil fertility. The typical fallow vegetation was grass or light bush in the upland villages, where longer periods of fallowing were recorded. To gain some insight into soil fertility, yield measurements were taken. Farmers showed problems in understanding the meaning of ’yield‘. It was very common for them to confuse the size of the tubers with yield. They expected to harvest a certain number of tubers of a certain size, but they did not consider how much land they required for growing. However, most farmers, and in particular the women had quite a good idea of how much they harvested in their respective sales units (e.g. bundles, dozens, bags, cartons etc.). For the rapid yield estimate, three to five plants of taro or cassava were dug out at random and then the average planting densities were measured. With the help of the tuber weights and the planting distances, yields were calculated.

The yields measured were higher than expected. Cassava yields ranged from 9<73> to 59 t/ha. Taro yields ranged from 25 to 28 t/ha. These yield estimates exceed considerably the average yield figures given in the Fiji farm management handbook<74> (cassava 20-23 t/ha and taro 9-15 t/ha). This indicated that farmers managed to obtain good yields with their current farming practice. Some samples even exceeded yields achieved on research stations<75>. Comparing the yield level for cassava with the world average (9 t/ha), yields


104

achieved in Fiji may be classified among the best in the world (cf. Rehm and Espig, 1984, 47).

Overall, yields in the upland areas seemed higher than in the lowland areas. However, the variation between measurements observed, did not permit to identify a clear trend regarding yield and type of settlement or yield and income groups.

Conclusions on cropping activities. Cropping activities showed a large variation between farms. Differentiating between average and advanced farmers appeared justified. The latter usually cultivated larger areas. Also, cropping patterns between up- and lowland farmers differed substantially. This affected the type of crops grown, fallow and soil fertility management, and the degree of mechanisation. A clear influence of the type of settlement on cropping activities, was less evident.

5.5.1.4 Conclusions on farm classification

The results of the RRA/PRAs allowed only a limited answer on the validity of the classification criteria. The main reason, had to be seen in the large variability between farms. However, a few tentative conclusions shall be presented here.

The agro-ecological zones upland and lowland, showed a rather high number of differences between the zones that appeared as certain. For this reason this criteria appeared as suitable. The wealth criteria, with the three wealth groups, proved very difficult to work with in practice. Farmers were reluctant to disclose income information, and investment in agriculture apparently did not depend on income only. Separating only two groups of average and advanced farmers proved easier and seemed to accommodate differences between farmers sufficiently well. Instead of ’wealth‘, ’progressiveness‘ would be a more appropriate title for this criterion. Regarding the last criteria type of settlement, the observed differences were less evident. Differences related to a certain degree to ethnic affiliation, but these also overlapped with wealth group effects. Therefore, the criteria did not appear as useful independent classification criteria.

Consequently, the two criteria agro ecological zone (upland/lowland) and progressiveness (average/advanced) were considered as the most suitable classification criteria. Two agro-ecological zones multiplied by two progressiveness groups, create four farming system classes. Considering the diversity encountered at the farmers‘ level, it may be asked if this diversity can be represented sufficiently well in only four classes. The answer remains difficult. On the one hand, more classes enable to address more factors separately, but on the other hand, more factors make it more difficult to distinguish these in the field, and more factors also reduce the overall lucidity and manageability of such a system. Considering the small size of the Ministry in Fiji and the limited resources available, the later aspect gained in importance.

Recalling the purpose of a classification system, structuring the many different farmers‘ problems, and, facilitating the return of recommendations, it is evident that in particular, the diffusion of recommendation is a process that has to attempt to acknowledge the existing farm diversity. If the process of innovation development works correctly, the end results should offer something that fits at least roughly to the needs of farmers in the different classes. The remaining fine-tuning to the individual situation of a given farmer


105

could be seen as a process of the farmers‘ own experimentation that could, where necessary, be further facilitated by extension officers<76>.

For this reason, the research team decided that the two criteria would be a reasonable starting point for a classification system. It was considered necessary that the usefulness of the criteria should be further investigated by additional research<77>. The last sub-chapter will take up the question of farm classification again, and combine it with the findings on farmers‘ problems.

5.5.2 Farmers‘ ways to acquire skills and information and attitude towards innovation

The common way for farmers to learn agricultural skills during their youth, was through the teachings of the father and elders of the family. All farmers interviewed, acquired their agricultural skills this way. This generation linked knowledge transfer is not free of problems. The case of the two upland settlements, Mataikadawa and Vunisoco, revealed the problematic nature. In the remote location Mataikadava, family ties were strong, and children remained in the location. Elders had a very profound agricultural knowledge including diverse agroforestry and soil conservation practices, and passed this on to the youth. Vunisoco, however, is a more recent settlement closer to the capital Suva, where many farmers originated from the outer islands group of Lau. Coming from the small atoll islands, these farmers had no experience with the mountainous rainforest vegetation on the main islands, thus, their traditional knowledge was of little use in the new environment. They had to learn again how to cope with the new environment.

Among sources of new information, most farmers ranked the agricultural extension service as the best source. Only one group in Serua mentioned lack of agricultural advice as a problem. This surprisingly very good result may be explained in the first instance by farmers‘ politeness. Another reason may be that the RRA/PRAs were conducted in areas that have good extension coverage. But extension coverage should not be played down. The islands are well administrated, and there are certainly no ’white spots‘ left, where extension activities have not taken place. Though the extension service was well known by all farmers, knowledge exchange between both groups was rather one sided. Farmers stated to have very few possibilities to take influence on the technology generation process, or, to articulate their problems.

The second best sources of information for farmers on agricultural innovations, were other farmers. Ranked third were the agricultural programmes in the radio. Other sources of information, like newspapers, magazines or the market place, were not mentioned frequently. In addition, the weekly marketing day was mentioned as an important source


106

of new information. This gave farmers an opportunity to discuss novelties with other farmers, traders and consumers.

Farmers‘ learning from other farmers was consequently an important issue in the innovation process. The key activity seemed to be observation. Farmers observe their colleagues and copy practices they consider interesting. This copying process may be more or less accurate. A farmer in the first RRA was an interesting example: The farmer (high income group) built a piggery for 30 sows based on his own observations during a farm visit in New Zealand. He did not take any plans. He reconstructed the stable based on his observations and discussions with the farmer in New Zealand. He also copied the materials (concrete, corrugated iron sheet roof, etc.) and the feeding and cleaning management system. At the time of construction (10 years ago) he could not get specific advice by MAFF extension, as they were unfamiliar with the size of the enterprise. During the use of the stable, the farmer realised that several aspects of the stable were not ideal for the conditions in Fiji (in particular the expensive construction). However, he could not afford any further modifications. This example shows the importance of observation in the behaviour of farmers. It also shows some of the risks. Farmers had difficulty to assess the long-term viability of investments. This may also be an explanation why so few farmers invested larger amounts in their farms. They have difficulty to assess the associated risk.

When farmers were asked if they experimented with new technologies they all first denied the questions. Probably, with research station trials in mind, they replied that they did no such experiments. After rephrasing ’experimenting‘ to ’trying out new practices, or, making changes to current practices‘, they began to understand better. However, they still had difficulties with the concept. Concluding from the discussions, it seemed that farmers do three things: observe, reason and act. They observe the process, try to understand what is happening, and based on their experience act accordingly. From their point of view there is no neat theory building and hypothesis formulation that are consequently confirmed or rejected through a research trial. They do not have the time and space for such experiments. A complete failure cannot be afforded. A farmer is interested in the success of his efforts. Therefore, whenever something is tried out, he observes the process very carefully and changes factors as soon as he thinks it may be necessary. Comparing new technology with his own experience, he can then judge if things are working better or not. This approach contrasts radically with the scientific approach whereby factors are kept constant and variation is screened out as far as possible. In contrast to scientific trials, farmers‘ management could be seen as a permanent ongoing experiment. He adapts permanently to the various outside factors that may be changing. This requires constant reasoning and decision taking<78>. This also explained to some extent the fact that the majority of farmers stated that they experimented rather little with new technology. Farmers have enough trouble to run their farm, and therefore, they are reluctant to take additional ’challenges‘ on board<79>. This attitude seemed particularly common in the


107

average farmer groups. The advanced farmer group appeared slightly more open towards new ideas. Most frequently, farmers mentioned trying out new varieties or experimenting with fertilisers. Asked about their criteria in technology evaluation, farmers again had difficulties to reply. On this abstract, general level farmers could not reply. They ’felt‘ if something worked or not, but it was very difficult for them to desegregate feeling into reasons or criteria. What turned out on this general level, was that farmers preferred easy, workable solutions. In terms of crops this meant easy sowing or planting, good natural growth and disease resistance. In terms of labour input not the total amount of time was most relevant, but rather the physical hardness of the work was reflected. The physical, less hard less complicated option seemed to be preferred. When asked more specifically regarding certain crops, farmers could express their evaluation criteria much better. This shall be further illustrated in the example of tomato cultivation. Different groups of farmers were asked to rank tomato variety characteristics according to their preferences. The results are depicted in Table 26.

Table 26: Farmers‘ ranking of tomato variety characteristics

Evaluation criteria
for tomato varieties

Farmer groups

Average

Advanced

Women

Market demand

1

1

1

Shelf - life

2

6

7

Ease of raising

3

 

3

Labour requirement

3

6

8

Taste

4

4

9

Ease of harvesting

4

4

2

Fruit size

4

4

11

Disease and pest resistance

5

2

4

Availability of seeds

6

5

5

Yield

6

3

10

Cost of seeds

7

4

5

Input cost

 

4

6

Several findings were of interest. Depending on the criterion selected, the groups had different ranking preferences. Most important to all groups was the market demand or price that could be achieved at harvest. This choice can be explained by the strong price fluctuations from as low as 20c to 2$ /kg. Prices are only good in the early and late parts of the season. During the main season, prices are forced down. For this reason farmers tried to sell early or late and looked for possibilities to spread the harvesting period as long as possible. This was also the reason for the high ranking of shelf- life by the average farmer group. A variety with good shelf life helps farmers to sell earlier or later.

Advanced farmers already had a different perception. For them disease and pest resistance ranked second and yield third. This could be explained by the fact that they earn their money via quantity and thus have a bigger interest in yield. Women, generally responsible for the harvest, considered ease of harvesting as second most important. Input costs were of more concern to advanced farmers (as they use them more often) than average farmers. Of particular interest was the fact that average farmers (rank 6) and women (rank 10) attributed very low priority to the specific yield of a variety. This finding showed that the common practice of researchers to screen variety trials according to yield as the single most important criterion, is not necessarily in line with farmers interests for new varieties. This PRA, was the first time


108

actual farmer preferences in respect of a single crop were assessed in Fiji. The participating researchers were quickly impressed by the wealth of information and the surprising results that were gained. Some even felt embarrassed for not having done similar research earlier. Ranking exercises for all crops grown in Fiji would represent an interesting possibility to better focus the search for innovations.

In Fijian villages, the traditional chiefs may have an important influence on the uptake of new technologies. Two villages in the Navua District illustrate the case. The first village Nangarawai, is a rather remote village that had received road access a few years ago. The village chief had a very open and friendly character. The social ties in the village were very good and the village had a large community hall and a joint van for transports. The houses looked prosperous with many flower gardens next to the ’bures‘<80>. The village had tap water, fish ponds, improved pastures, produced vegetables for the market, and had even installed recently flush toilets for all houses. The second village Kalokolevu is located closer to the main coastal highway with good road access. However, the chief was an introverted, difficult person. The communal activities in the village were poor and there was no joint community hall. Several young people mentioned that they wanted to leave the village. The central village lawn appeared neglected. The standard of housing was low, and little modern technology was visible. No modern farming practices were visible in the vicinity of the village. These two examples were probably somewhat extreme cases, but they illustrate the existing spectrum and influence of traditional leadership very well.

To gain an idea about the potentials for future innovation uptake, farmers were interviewed about their future plans for their farms. The majority of farmers were very satisfied with their present situation<81>. They had no specific plans for the future development of their farms. One explanation for this ’lack of plans‘ could be seen in the fact that farmers‘ basic needs were well satisfied and consequently there was little urgent need ’to change something that apparently worked well‘. Another explanation might be seen in the limited supply of innovations by the Ministry. Farmers had the impression that there were no attractive alternatives to what they already did. As a future plan, a few farmers mentioned the wish to improve their houses or the education of their children. The only exceptions of farmers with concrete plans, were two who had applied for agricultural loans. This was not surprising as detailed farm development plans are required as part of the loan application scheme for agricultural credits at the Fiji Development Bank (FDB).

Implications for MAFF

The question about farmers‘ information sources revealed the important role of the Ministry. Extension officers ranked number one. This is certainly a good indication that extension is reaching farmers. The obvious problem seems to be the lack of relevant innovation. The recommendations supplied are apparently not relevant enough for farmers. The rather limited overall degree of farmers‘ experimentation can be seen as an indication of this. Other farmers as sources for new information can not compensate this


109

lack of relevant innovation. The remoteness of the islands and the limited exposure to other farmers (only on market days) means that there is rather little to learn from other farmers. Every farmer in the village is doing more or less the same thing. There is little to see that is both better and works. The media plays only a minor role for farmers. Another reason may be that no farmer specific publication exists. All existing media provide only a very limited range of agricultural information. A quarterly journal written in easy language and well illustrated with interesting examples from the Pacific region, might be the right means to circulate more farming ideas.

The example of the tomato variety ranking highlighted that farmers have very specific interests in particular crop characteristics. Only if innovations that meet these characteristics are made available, will farmers be motivated to make changes to their farming systems. As shown above, farmers carry on as usual, as long as this is not happening. This should not be interpreted as that farmers are not being willing to make changes. In chapter 4.2 it could be shown that Fiji farmers responded very quickly to the new export opportunities for taro when the neighbouring country Western Samoa was struck by taro blight that wiped out their production. Farms make use of relevant opportunities. The Ministry must learn to address this need more effectively.

5.5.3 Basic needs satisfaction and farming problems

Determining farmers‘ needs or problems is a difficult problem in itself. The problem starts with the question: What are needs or problems? A farming family may need many different things for life. Only a few needs may be directly related to farming. For this reason it was decided to look at both aspects: farmers‘ basic need satisfaction, and their specific farming problems. Recording both aspects helped to better understand the overall situation of the farm family and to assess the nature and scale of problems.

Figure 20 provides an overview of farmers‘ basic needs satisfaction in three settlements. The situations in the two old settlements Serua and Mataikadava could be considered as very good while conditions in the more recent settlement Vunisoco were less favourable. All farmer groups indicated that they were satisfied with the current quality of life. Some considered their conditions even as excellent. Regarding food, almost all groups ranked their present situation as excellent. Compared to other developing countries Fiji is certainly in a very fortunate situation. Farmers do not face major problems to produce sufficient food for subsistence. There is certainly no problem of food scarcity or hunger<82>. This makes life to some extent ’easier‘ and reduces immediate pressures to sustain a living. Water also is generally available in abundance. Most villages and settlements have tap water. Except Vunisoco, where some households still used water from a surface creek, water was ranked from satisfactory to excellent (3-5). Somewhat lower (3-4) but still in the range of satisfactory to good was the situation for clothing and housing.


110

Figure 20: Basic needs satisfaction in selected locations

Source: own data PRA 2.

A little more controversial was the situation regarding health.

In Serua the women considered the situation only as fair whilst the men regarded it as excellent. In the other two settlements the groups ranked health from satisfactory to excellent (3-5).


111

Education also scored very high. Most groups considered it as excellent<83>. Only in Vunisoco, did the people complain about the distance to the next school. All farmers interviewed were literate and about one third had even attended secondary school.

Once again, the income ranking created difficulties. All groups were reluctant to rank their cash income and only two finally gave a scoring. However, farmers indicated verbally that they considered their cash income as not satisfactory and that they would like to earn more money.

The ranking showed that farmers‘ basic needs were largely fulfilled. This is certainly a positive finding. From the perspective of a change agent, however, it also has some undesired aspects. As most farmers are basically content, there is little need to change things - in particular not quickly. This lack of pressing problems probably creates a less favourable climate for developing innovations.

But moving from this general view to the real farming problems, the situation looked somewhat different. An overview of problems is depicted in Table 27.

In Serua farmers ranked the following problems as most important: lack of machinery (in particular for ploughing: bullocks, implements, tractors), high costs of inputs, and excessive rain combined with inadequate drainage. Less important problems were: lack of advise, availability and quality of fertilisers, pest and diseases and shortage of land.

At the end of the priority list were availability of casual labourers and land rent. Marketing seemed to be a major problem for average farmers, while the advanced farmers ranked it low. In the upland settlements, problems were quite different. In Vunisoco problems were more general, less linked to agriculture. Key problems were poor conditions of access roads, water access and low income. Main agricultural problems were lack of equipment, difficulty to clear forests for farming and marketing of produce. In Mataikadawa the main problems were: too few taro planting material due to the taro beetle, and wild pig and cattle damage due to poor fencing. Medium problems were: the cassava white fly, flooding and poor drainage, lack of tools and shortage of manpower. Minor problems were: financial constraints, weather and difficulty of land clearing.

Comparing the problems of the three settlements revealed a few issues. There were no common problems in all three settlements. However, a number of problems were common in at least two settlements (e.g. high input costs, fertiliser, lack of equipment) while some problems were purely location specific (e.g. access road, water -Vunisoco; rice threshing - Serua). Problem perceptions of the different groups corresponded only to a certain degree. In between a quarter, to half of the problems corresponded among the various groups of each settlement. The ranking of problems also indicated some parallels. Among the top four problems, one to two problems corresponded between the various groups of the same settlements. Only in a few cases, did top ranked problems of one group be found to rank much lower by other groups of the same settlement. Hence, it may be concluded that the different groups shared to a certain degree the view on common problems. However, the overall high variability on problem perception was still striking. The groups generated more different problems than common problems.


112

Table 27: Farmers‘ problem ranking in selected settlements

Vunisoco settlement

Women

Average 1

Average 2

Advanced

1. Little farm equipment

1. Low cash income

1. Poor water quality

1. Poor roads

2. No tap water

2. Lack of good farming knowledge

2. Poor roads

2. Education of children

3. Poor roads

2. Poor roads

3. Few farm tools

3. Water shortage

3. Hard to clear forest

3. Marketing of produce

4. Delivery to market

 

3. Taro beetle

3. Price of fertiliser

5. Availability of seeds

 

4. Lack of planting material

4. Lack of equipment, tools, bullocks

 

 

4. Hard to get to market

4. Shortage of labour

 

 

4. Hard to find jobs

 

 

 

4. High input costs

 

 

 

4. No electricity

 

 

 

Mataikadawa settlement

Women

Average 1

Average 2

Advanced

1. Too few taro suckers

1. Drainage

1. Financial constraints

1. Taro beetle damage

2. Poor fencing

2. Water supply

2. Shortage of man power

2. Hard to clear new land

3. Cassava white fly

3. Flooding

3. Poor fencing

3. Poor drainage

4. Wild pig damage (up to 15%)

4. Planting materials

4. Taro beetle

4. Damage to crops by stray animals

5. Occasional flooding

5. Farming tools

5. Crop damage by cattle and wild pigs

 

6. Too few pair of bullocks

6. Land clearing

6. Cassava white fly

 

7. Health (transport to Suva)

7. Changing weather

 

 

8. Education (transport)

8. Poor drainage

 

 

9. Price fluctuations

9. No shelter

 

Serua settlement

Women

Average 1

Advanced

1. Tractor, land preparation costs

1. Lack of machinery for land preparation

1. Price of inputs (fertiliser)

2. Extension advisory service

2. Difficult to hire tractor

2. Price of pesticides

3. High rice production costs

3. Cost of farm inputs

2. Poor drainage

4. Low price of rice

4. No extension service visits

3. Cost of tractor hire

5. Too wet

4. Marketing

4. Poor fertiliser quality

6. Expensive Agro - inputs

4. Availability & price of seeds and planting materials.

5. Lack of advise

7. Expensive & scarce labour

5. Poor germination of seeds

8. High rice threshing cost

 

6. Availability of labourers

 

 

6. Lack of rice harvesters

 

 

7. Marketing of produce

The cases where two average groups in the same location could be compared showed no better correspondence of problems. An explanation for this high variability may be seen in the very broad nature of the question used in the PRA exercise: Think about your farms. What are your main agricultural problems? This question was so open, that naturally many different matters could be raised. But even daily or seasonal peculiarities may induce farmers to articulate certain problems. Such a case occurred in the first RRA


113

with excessive rain and associated drainage problems. On the two rainy days of the survey, the farmers mentioned the problem more often and discussed it in more detail than during the dry days.

The situation changed when the problem ranking was done with a more narrow and concrete topic. In PRA 3 problem ranking was done for problems related to vegetable production only. The results are displayed in Table 28.

Table 28: Problems related to vegetable production by farmer group and location

Naduri

Women

Average

Advanced

1. Water source is far

1. Seeds are not available when needed

1. Marketing

2. Low soil fertility

2. Input prices are high

2. No tractor

3. Lack of knowledge

3. Marketing

3. High cost of agro input

4. Poor seeds quality

4. Middlemen dictate low prices

4. Virus in watermelon

 

5. Water source is far

5. High cost of papaw seedling

 

6. Diseases

 

Barara

Average

Advanced

1. Flooding

1. High land rent

2. Expensive fertiliser

2. Marketing

3. Marketing

3. Diseases

4. Lack of water

4. Mixed seeds in packets

5. Seed storage

5. High price of chemicals

6. Middlemen give low prices

6. Supermarket seeds are unreliable

Siminilaya

Average

Advanced

1. Marketing

1. High input cost

2. High chemical costs

2. No fixed market

3. High seed costs

3. High market fees

4. High fertiliser costs

4. Low prices on all vegetables

5. No irrigation system

5. Poor irrigation services

6. Poor seed germination

6. Land rent is high

 

7. Market competition with sugar cane farmers

Dubalevu

Women

Average

1. Low return of farm produce

1. Marketing during main season

2. High input costs

2. Expensive input costs

3. Damping off in tomatoes

3. Poor land preparation

4. High seed costs

4. Poor quality seeds

5. Irrigation

5. Water

The problem perceptions of the different groups matched better. In between half and three quarters of the problems corresponded among the various groups of each location. The ranking of problems also indicated more parallels. Among the top four problems, one to three problems corresponded between the different groups of the same location. Similar to the general ranking, the results of the womens‘ groups and the average and advanced


114

groups did not differ substantially in their problem assessment. These findings underpinned the trustworthiness of the ranking results, as the different groups did not contradict each other.

After this topical problem ranking, the three most important problems were analysed in more detail. As an example Table 29 summarises the causes and solutions identified in relation to the problem ’high agro input costs‘. Looking into the causes of the problem, revealed some of the ’underlying‘ problems: no fallow practised; low soil fertility; no organic fertilisers etc. These causes provided an opportunity to better understand the key problem and also highlighted potential options to solve the key problem. The following discussion about solutions provided first indications in which direction farmers and MAFF staff saw potential solutions. Some farmers‘ solutions to the problem, were, in fact expectations of the Government to make changes to policies: introduce subsidy, MAFF to supply seeds and chemicals (cheaper) and introduce price control. Other solutions indicated agricultural option: e.g. practice crop rotation, green manuring. The suggestions of researchers and extension equally included policy options e.g. liberalising the market; agronomic options e.g. the use of organic manure, apply integrated pest management (IPM), or social development options e.g. buy in bulk through co-operatives.

Table 29: Problems, their causes and solutions

Causes

Problem

Farmers‘ solutions

Ext./Res. solutions

No organic fertilisers

 

Practice crop rotation

Increase yield through small scale farming practices.

No fallow practised

 

Green manuring

Use organic manure.

Heavy duty on imported items

 

Introduce subsidy

Apply IPM

Few suppliers only

HIGH AGRO INPUT COSTS

MAFF to supply seeds and chemicals

Buy in bulk through farmers co-operative

Low soil fertility

 

Good cropping practices

Liberalise the market

Sold by private companies

 

Reduce quantity

Improve farmers cash-flow problems

No subsidies

 

Introduce price control.

 

Inputs are all imported

 

 

 

The high number of solutions to a single problem illustrated that there may be different strategies to solve a problem. It also showed that it might be necessary to address several points at the same time in order to solve the problem. Possible areas for intervention are not only agronomic, but may also be on policy level or involve socio economic aspects (e.g. founding of co-operatives). Agronomic solutions might require additional research e.g. green manure, socio economic problems might be areas where extension could provide direct assistance without the need for prior research e.g. co-operative.

Summarising, it may be concluded that the assessment of farmers‘ problems is a complex issue. Problems have to be seen together with their causes and potential solutions. A basic


115

farm classification system is necessary to structure the large group of farmers into meaningful classes with ’similar problems‘. At this stage it seems necessary to take a look back at the original working hypotheses made in the first chapter. Regarding farmers‘ problems and their interaction with MAFF the following hypotheses were formulated:

H 1

In small island countries farmers‘ problems can be addressed comprehensively and in a comparatively short time.

H 2

Analysis of farmers‘ problems will help to develop sound extension and research priorities.

The first hypothesis could not be confirmed. At the beginning of the study it was thought that in a small country like Fiji a comprehensive review of farmers‘ problems would be possible in a short time. But the large diversity encountered in the field soon made it clear that a comprehensive review would be much more time consuming. Thus, even in small island countries, socio economic and agro ecological factors may create a huge diversity that is not far from that in many much bigger countries. Therefore, the hypothesis has to be rejected.

The second hypothesis could be confirmed. For the areas investigated, the problem analysis and better understanding of farming systems made it easy to formulate a specific set of research and extension priorities. These were published in the PRA field reports<84>. With the help of the following model for integrating farmers‘ problems into the AKIS, it should be possible to revise research and extension priorities on the national scale.

5.5.4 A model for integration farmers‘ problems into the AKIS

In the chapter 2.2.1 a cyclic sequence of six main AKIS functions was presented: need/ problem identification, knowledge generation, knowledge operationalisation, knowledge dissemination, knowledge utilisation, and finally evaluation. In this chapter it will be shown how the first step of need/problem identification can be broken down into various operational steps and how these are connected with the other AKIS functions. An illustration of the model is depicted in Figure 21.

The starting point of the sequence of steps is located in the top right corner. The circle, ’need /problem identification‘, is extracted from the cycle at the bottom to symbolise the breakdown into operational steps. The steps are shown in boxes in the middle. The circles on the left of the boxes symbolise the key actors that play a role in each step. The size of the circles symbolises the importance of actors. The steps are connected with arrows to indicate the direction of the process. At the end of the steps, the arrows feed back into the cycle of the AKIS functions.

The main content of each step is as such: The first step represents the actual process of discussing problems with farmers. The PRA tools tested during the fieldwork appear very suitable for this purpose. This would ensure that farmers can express their views, while researchers and extensionists remain in the background and simply facilitate the process.


116

Figure 21: Model for feeding farmers‘ problems into the AKIS

Source: own design.


117

The process could start with a first look at basic needs, followed by the screening of main agricultural problems. To structure the discussions and to avoid responses that are too general, it would be recommended to suggest central problem areas for analysis (e.g. crops, livestock, marketing, etc.). Then, problems should be ranked and a small selection of major problems (3-5) should be analysed in detail with causes and potential solutions<85>. The selection of farmers and locations should be structured by the farm classification system elaborated earlier. This would include two agro-ecological zones and the separation of farmers in two groups according to progressiveness<86>. Coverage of 10 to 20 % of the locations in the main areas of intervention of the Ministry should provide a sufficient sample. The aim should be a representative selection of locations to ensure that the problems of the majority of farmers will be recorded. The final sample size would also depend on the Ministries‘ ability to deploy staff.

In the second step, the information would be compiled on a national level. The frequency with which problems were raised would provide a good overview about the real priority problems at the farmers‘ level. Compilation and analysis would be an important role for research and extension. Results would have to be compiled for all farming systems classes separately. This would result in priority problem lists for all farm classes. The tentative solutions collected at farmers‘ level, should then be compared with available solutions on the international scale. This enlarged set of problems and solutions, should, then provide the basis for determining research and extension priorities. In this process farmers‘ priorities should be compared with the existing priorities and policies. A joint Council of all divisions of MAFF would be necessary to discuss the findings and agree on a tentative list of research and extension priorities. At this stage, policy and market actors could also contribute to the discussions. This would help to accommodate all actors‘ interests as far possible.

The last step, would be the compilation of the national results, discussing tentative priorities with farmers again. No such exercise was as yet included in the field work trials. However, it seems that a PRA exercise with presentation of results and solutions and an open discussion would be the appropriate means. This would offer farmers an opportunity to comment on the suggestions made. At the same time, it could be the starting point for further co-operation. According to their interest, farmers could participate in the further technology development process for new research priorities, or farmers could start to experiment with the selected recommendations offered by extension. Depending on the branch extension or research, the respective divisions could then take control. In particular for extension, market actors could also play a growing role in the future. As the screening of farmers‘ problems showed, many problems do not necessarily require research. Options for development may have to do with better marketing, processing or exporting of farm produce. The better problem analysis could provide indications for private sector actors to place investments.


118

In the model these further developments are symbolised by the arrows that feed back into the cycle of AKIS functions. Participatory technology development feeds back into knowledge generation, while farmers start experimenting with selected solutions feeds back in between knowledge operationalisation and knowledge dissemination. Once the full cycle is completed for a certain innovation, the evaluation results could be fed back into an update of the need identification and start-up a new national problem compilation.

The implementation of the model in practice would provide indications in which, at intervals, an update of such a national problem census would be necessary. At this stage it could be estimated that this might be necessary in 3-5 year intervals.

Implementing the above model would represent a considerable challenge for MAFF. The model would represent a big step ahead in terms of farmer participation. In the actual situation, where no farmer structures or organisations exist, the full initiative to implement the approach would be required by MAFF. The methodology used for this study, participatory action research, might be a suitable way to accompany this introduction process. Implementing the model could thus become an important learning cycle for the Ministry. This would provide an excellent opportunity for Ministry staff to further raise their qualifications.

5.6 Linkages and information flow between the main actors

To reach the goal of developing and disseminating an innovation, the actors in a knowledge system have to perform a number of tasks (see system functions chapter 2.2.1). There are various ways to divide these tasks between the different actors. Some actors will be assigned to research, and some will have the job of extension, which are the key players in the system. Certain tasks will have to be carried out jointly by the two or more parties, with varying input from each group, depending on the nature of the task. But no matter which actor or group of actors performs the task, a good degree of co-operation will be necessary. Thus, to cooperate, the parties have to be linked in some way to one another. In this sense ’linkages‘ describe the organisational set-up or structure used to facilitate co-operation and the necessary communication or exchange of resources.

The existing linkages between the main divisions in the Ministry are examined in the first section. A closer look at the information flow between the main players will be given in the second section.

5.6.1 Existing linkage problems

A workshop was considered the most suitable means to gain a good insight into existing linkage problems in Fiji. The workshop was held in the Southern Cross Hotel in Suva in November 1994 and it was organised by the Soil and Crop Evaluation Project (SCEP), a donor funded project with the purpose to assist MAFF in the process of improving its crop research and extension capabilities.


119

A total of 35 participants of all Divisions of MAFF attended the one-day workshop<87>. Four working groups were set up to analyse the topic with the help of the following main question:

Are there linkage problems between the various groups responsible for developing and transferring technology?

As a further guideline, the groups were asked to examine linkage problems with particular reference to the following tasks<88>:

The results of each group were visualised on pinboards and discussed in the plenum. The summary of the four group works is presented according to the above guidelines below<89>:

Diagnosing farmers‘ problems

Extension was considered responsible for this task. If a problem was encountered, extension staff discussed the issue informally, firstly, with other colleagues, and secondly, with research staff. Then it is passed on to research. An important complaint of extension was that it took too long to get a response from research. Research defended their position with the difficulty to allocate research priorities to such requests and the general difficulty to identify quick solutions. Joint visits to farmers by both research and extension staffs were uncommon. Only one group highlighted the need that the farmers themselves should be involved in problem identification. There was a feeling of lack of co-ordination between research and extension, because there was no joint investigation of problems or evaluation of activities and no joint listing of priorities.

Design of a research program

Research programs were exclusively designed by researchers for on-station trials. Extension staff was not involved in this design process. A few mentioned that the diagnosis of problems should be done in conjunction with research and extension. However, some researchers considered that there was not a big need for extension to be involved. Another linkage problem raised was that research and extension meetings did


120

not achieve the aims of incorporating the views of the extension staff into research programmes. The majority felt that extension staff should have some input into the development of a research programme. Another key linkage problem discussed was that staff from both divisions were too busy doing their own work to talk to each other effectively.

Generate technologies

All groups agreed that there was an almost exclusive reliance on research to generate technologies. There was practically no involvement of either extension staff or farmers in this process. Researchers tended to focus on commercial crops with little emphasis on semi-commercial and subsistence farmers.

Consolidate technologies

While all groups agreed that there should be a good co-operation in relation to this task, it was identified that their was a lack of definition of which roles that each division should play. Extension staff was occasionally involved in the identification and layout of sites, but there was no real involvement. This led to a lack of ownership and the fact that little ended up happening in the field. All groups expressed the need that both extension and research should be actively involved in the consolidation of technologies.

Disseminate information and knowledge

Several groups highlighted, as an important problem, that there was no technology output and little new information to extend. Extension indicated a reluctance to extend incomplete packages and research, on the other hand, complained that there was no feedback from extension staff on the success of technology packages. A lack of training on this issue was seen as the main reason for this problem. Some groups reasoned that a technology package by itself was not sufficient. A system of extension attached to it would be lacking; e.g. field days, demonstrations. Extension further revealed that there was no easily understood technical information available. Others added that radio programs were not an effective means of extension and that the programs were rarely evaluated. It was agreed that the bi-directional flow of information between research, extension and the training and communication unit (TC&S) was inappropriate.

Approve and release technologies

One group remarked that the wrong recommendations were being used. Another group thought that all groups should be involved in the process of technology approval. It was agreed that at present only research was taking the major responsibility and that extension was not formally involved in the task. Extension officers were only expected to adopt and recommend the recommendations. It was felt by some, that this process should be a management decision. Once a technology was described in the Fiji farm management manual, then it was considered as approved. However, the actual process of how this currently happened, was not transparent to anyone.

Multiply improved genetic material and duplicate technology packages

Some comments were made that there was very little new information worth extending or multiplying, that it was furthermore expensive and the co-ordination on the tasks was poor between all divisions. Other groups questioned the division of tasks and attributed


121

the role to indicate needs for multiplication to extension, while research should be responsible for the actual implementation of multiplication.

Deliver technologies

The delivery and evaluation of recommendations was considered interrelated. The groups felt that there should be a structured system whereby the process of technology delivery could be reviewed. Currently, there was no such system in place to obtain feedback or to evaluate the uptake of technologies or their effectiveness in the field. Most groups expressed the view that extension, research and farmers should be involved in the transfer of technology. There was no way of knowing if technologies were being delivered or not.

Evaluate technologies

The groups stressed again the lack of a feed back mechanism and that extension, research and farmers should be involved in this task. There was a perceived need to have an agricultural economist to do some economic evaluation of the benefits of the technology package. It was rated that the current input of EP&S was not being used in the most appropriate way and that the section needed to be more consulted in the technology transfer process.

Analysis of workshop results on linkage problems

The comments of the groups revealed very broadly that the Fiji AKIS was in a desolate condition. The openness to report such negative facts may appear as surprising, in particular as all levels of management participated in the workshop. Apparently no problems (or only minor facts) were hidden or left unspoken. This showed a willingness to admit mistakes and engage in a sound process of reform. The participatory set-up of the workshop and the discussions in small groups probably created the positive atmosphere, which enabled such profound reflections<90>.

In an attempt to summarise the key problems raised, it may be said, that already at the starting point, farmers‘ real problems were not properly identified. Whatever problems were identified by extension, these faced difficulties in finding consideration by researchers. Consequently, research designs remained dominated by researchers and the topics focused on their research station biased perceptions of farming problems (commercial crops). These apparently did not match well with the real farming problems. Therefore, it was not ensured that the technology generation process was geared towards promising ends. The necessary consolidation process of new technologies was omitted. This in turn led to the fact that technologies were approved by research without a real validation of these technologies in the field. Dissemination of technologies was then hampered by several factors. Firstly, the overall output rate of innovations was very low. Secondly, the promoted technology packages were at least partly inappropriate. Thirdly, it was not clear to which group of farmers the technologies might be suitable. Finally, the extension approaches used seem unsuitable and poorly managed. Furthermore, it appeared that only information was disseminated by extension, as co-ordination problems between research and extension hampered the multiplication of necessary genetic materials.


122

Without proper monitoring of dissemination results, a meaningful evaluation of the overall process remained impossible.

The short summary illustrates the many linkage problems that cause the inefficiency of the system. The comments of the workshop furthermore indicated that there was, as yet, no culture of collaboration or joint system perspective. Each division‘s outputs were not considered as being part of an exchange relationship in service provision. The different divisions were not really aware of the need to work together. This was evident through the lack of role definitions and the fact that the different actors blamed each other for linkage deficits.

This attitude is devastating in a system context where all collaborating parties may benefit. This attitude may also explain why other outside actors were not noticed as potential partners for co-operation. Not noticing other actors also makes it impossible to gauge the potentials that other actors might offer in relation to improving one's own outputs. And there are many more actors that are relevant for developing agricultural innovations. In the Fiji context that could be particularly policy makers and donors, private sector and market players such as importers, exporters or processing industries and also educational institutions.

A cause for linkage problems could be seen in MAFF‘s management. This was specified as a lack of procedures, particularly enabling procedures and directives. From own experience with collaboration with MAFF, the researcher rated the lack of follow-up as the most important missing procedure. Innovation development requires many iterative steps or in other word cycles and loops to complete the necessary tasks. A key management problem is that tasks are started-up, but as soon as problems occur or other issues require urgent action, activities are paused or stopped. Works are not resumed, as follow-up mechanisms are not in place.

Comparing Fiji with other cases

ISNAR has investigated NARES in seven study countries. The following common linkage problems were identified: missing tasks, missing linkage mechanisms, duplication of effort, non operational linkage mechanism and ineffective linkage mechanisms (Eponou 1993,17)

Comparing these problems with the Fiji workshop results many of the problems raised fit under the above ISNAR categories:

Missing tasks. A clear definition of roles for all tasks is lacking. Farmers' problems are not fed into the system. Multiplication of genetic materials is not assured.

Missing linkage mechanisms. No formal linkage structure is in place. No officers (e.g. subject matter specialists) have a clear mandate to linkage roles.

Duplication of effort. This seems a minor problem, as both divisions follow their own agenda with very little overlap. Duplication of efforts is rather a problem of repetitions


123

over time; e.g. researchers repeat the same trials time and again, only because original results were never properly documented or staff have changed positions<91>.

Non operational linkage mechanisms Existing meetings between the divisions do not achieve intended results. Although research and extension are based jointly on some of the stations, too little real co-operation seems to take place.

Ineffective linkage mechanisms. Where existing linkages work, the results are not implemented. Recommendations are ignored or not followed-up.

The fact that the problems identified in the Fiji AKIS fit so easily in the ISNAR categories, indicates that the problems are well comparable to situations in other countries. This was also confirmed by an IRETA study that investigated the NARS in 5 Pacific Island countries (Kern 1994, 2-29). All countries were visited for the study and a total of 83 resource persons (mainly national Ministry staff), thereof 24 in Fiji, were interviewed. Many problems were similar throughout the five nations. The main problems identified were fully in line with the results of the Fiji workshop: weak structures of NARS and NAES, communication and information flow deficits, lack of sound monitoring of programmes, very weak linkages between actors, in particular between the key players research and extension and from them to farmers and policy makers. Some causes for the situation were seen in the prevalent public service rules that were considered as anti-productive. In addition, human relations problems, motivation deficits and apathy towards change were identified.

5.6.2 Information flow

During the first workshop on linkages, information flow problems were already mentioned as an important factor. A follow-up workshop was considered the most suitable means to gain a better insight into existing information flow deficits.

The aim of the workshop consisted of four tasks. Investigate information flow, create awareness among MAFF staff member on the potential of participatory methods and identify further training requirements in the above two topics. The workshop was organised jointly by the PRAP P11 teamleader Dr. Mechthild Kronen and the researcher. SCEP and IBSRAM attended with contributions. The workshop was held at the main research station in Koronivia in September 1995.

A total of 39 participants attended the workshop. The focus was on MAFF staff (27) and project experts (6). All MAFF management levels and the different divisions were well represented. In addition 3 representatives of FCA and USP were invited in order to include actors from the education sector. Furthermore one donor representative and two observers of regional organisations attended the workshop.


124

Five working groups were set up to elaborate different views on the topic. The composition of the working groups was pre-set. This consisted of trying to put together MAFF staff members according to similarities in working fields, and to evenly distribute project personnel as well as staff from external organisations. According to the majority of participants in a given group, these could be classified as top management, research, extension and two groups as cross-sections.

The workshop dealt with two main topics, for each of which one day was allocated. The first day examined the current information flow and MAFF‘s organisational interaction with other actors. The second day investigated in particular the information flow and information needs between farmers and MAFF with particular reference to data collection methods. First results of the researchers training course on participatory methods were presented and discussed together with other new local experiences with survey methods (RRA, household surveys and formal surveys). In this section mainly the results of the first day will be reviewed.<92>

Workshop results

Information flow was analysed in two ways. First, the groups worked on information flow within MAFF using such visual methods as flow charts and venn diagrams. Second, the groups were asked to visualise the Ministry‘s interaction with other actors using venn diagrams.

The flow chart diagram of group 4 is depicted in Figure 22. It gives a very good illustration of the current information flow at MAFF. Formal information flow is shown in full lines, while informal information flow is represented in dotted lines.

The chart shows clearly that information flow runs largely along the hierarchical structure in the Ministry. Information flows in a rather straight line from the lowest grade of Senior Agricultural Assistant (SAA) up step by step intermediate grades (Agricultural Technical Officer (ATO)-> Senior Agricultural Officer (SAO)-> Principal Agricultural Officer (PAO)) until it reaches Director level. Via Deputy Permanent Secretaries (DPSO and DPSS) information finally reaches the Permanent Secretary (PS). Information flows are bi-directional so that information may flow up or down the stream of hierarchy. The chart resembles a classical linear organisation. Looking closer at some details, several deficits and inconsistencies become apparent.

There is no direct information flow between the intermediate levels of research and extension; formal links are limited to the top of the hierarchy at director level only. This may be seen as a main reason why information flows are underdeveloped.

The contact to farmers is build via the above executive chain. It is important to note that only the lowest grade officers, the ones with the least professional experience, are responsible for communication with farmers. From the chart it appears as if the chain is attached only to EP&S, but it is clear that extension has the same structure and the larger number of officers. While EP&S staff mainly collect information from farmers and feed it into the system (predominantly up-ward flow: prices, market information), extension is


125

supposed to provide farmers with the latest recommendations (predominantly down stream information flow).

An important fact that was already noted in the linkage workshop is that research had no direct contact to farmers. This means that all information they receive is second hand (or if all hierarchy steps are counted at least 7th hand).

Figure 22: Model of information flow at MAFF

Source: Information flow workshop Koronivia, group 4 result.

The only information exchange seems to be via Subject Matter Specialists (SMS) on an informal basis to SAA level. This is certainly a key deficit, as field extension officers should have good direct contact to SMSs.

The Training and Communications Section (TC&S) is cut off from information. It has no direct information exchange with either research or extension. Only informal links seem to exist to SMSs and a single link to the PRAP information service (P9) project. Therefore, it may be wondered where they receive their information for further dissemination to farmers. The dissemination function, however, seems to be operational with information flows via several channels (broadcast, newspapers, journals, TV, pamphlets and training) to farmers.


126

Another weakness may be seen in the fact that all donor projects (SCEP, IBSRAM, SPC, PRAP) are only linked to the Director of research. In this configuration, knowledge and advice from projects benefits only one division, while the others are neglected. Further aggravating is the fact that information flows are indicated to be one way only. This implies that projects appear as ’teachers‘, rather than as partners with equal feed back and joint learning interests.

The libraries as important sources of information remain largely unused. There is no information flow to extension. This indicates that extension relies exclusively on research for new ideas. Research obviously use the library more, but as the analysis of the RD in chapter 5.3.5 showed, the researchers make to little use of existing literature.

The above chart permitted a very good look at the internal information flows at MAFF. In an attempt to widen the analysis of information flows, MAFF‘s interaction and information exchange with other actors was examined. All groups were asked to draw venn diagrams of their organisation and its interaction with other players.

The two most interesting charts are depicted in Figures 23 and 24.

Figure 23: MAFF and organisational interaction: management group view

Source: Information flow workshop Koronivia, group 2 result.


127

The chart of the management group shows a strong focus towards the cabinet (grey shaded circle). MAFF is seen as a closed body that is responsible towards the cabinet. In the center is MAFF top management with the Permanent Secretary and his Deputies (DPS).

Attached to this management circle are all divisions. This representation comes very close to the official hierarchy. Interestingly, crop research and extension are seen as overlapping, thus indicating good interaction. This view was not shared by the other groups. Outside MAFF are a large number of actors with which the Ministry is in contact. Closest are the relations to farmers and the private sector. These are split in a larger group of subsistence farmers and a smaller group of commercial farmers, exporters and traders. International markets and traders are seen behind the private- commercial exporter circle and linked with dotted lines only. This indicates that the Ministry has no good personal contacts to and understanding of outside markets and relies on the experience of local exporters that have practical trading experience. Further outside are all other Fiji Government Ministries that maintain links to MAFF. The most important one here is the Ministry of Finance, linked to Aid Co-ordinating Office and the Office of the Prime Minister (PSC). These Ministries often comment on the Ministry papers presented in Cabinet or tabled at Parliament.

The second venn diagram was prepared by one of the cross sectional groups. It shows a completely different view than that of management.

MAFF is not seen as a closed body and all divisions are positioned independently. Top management is not even included as a separate player. The extension division is seen as the largest MAFF entity in the center. Attached to extension with a small overlap is the farming community with farmers and farmers‘ associations. Interaction is enabled through SAA staff members. Research is smaller and has a little overlap with extension. Thus a certain degree of interaction is given. Donor projects are attached to research. The key feature in the chart is the yaqona bowl. All major actors are linked on an elliptical chain to the yaqona bowl. The yaqona bowl represents a symbol for communication and discussion (round table). As the traditional drink<93> and ceremony, yaqona is the starting point of all social interactions in the Fijian society. Next to the bowl are politics, which shows their strong dependency on traditional customs. Other MAFF sections are attached closely to the chain (TC&S, D&I), while EP&S, AH&P are a step outside, indicating that these are less involved in field work. Another important actor, the Fiji Development Bank (FDB) is in contact with extension. The bank provides loans to farmers.

Discussion of workshop results

For all workshop participants it was a striking recognition that there were so many different perceptions of information flow and interaction. Within this diversity, however, two main directions could be identified.


128

Figure 24: MAFF and organisational interaction: cross sectional group view

Source: Information flow workshop Koronivia, group 5 result.

First, there was the perception of management with the rather ordered and hierarchical understanding of information flow. This was visible at the example of the executive chain in the information flow diagram and again in the interaction diagram of management. The second direction was the view from within the hierarchy. Here, the various groups elaborated very different pictures, each mainly determined by the closer environment of each division. This could be seen very well in the yaqona bowl chart. But this influence was also visible in the management chart: management felt a direct responsibility to the cabinet through its frequent interaction, whereas the other groups did not have direct contact to the Cabinet and therefore did not include it in their representations.

This obvious mismatch of perceptions was felt as a concern by MAFF management. The Director of extension Mr. Samisoni made the point: "Five perceptions - and all are correct. Management expects all levels of the structure to have the same knowledge and understanding of policies. Apparently, this is not so. It seems that information flow is poor." The Chief Economist Mr. Taukei had a similar understanding and concluded: "The consistency of policy interpretation and understanding of operational norms from top


129

management level to the lower levels within MAFF furnishes some areas of concern. Overall there is a lack of total objectivity in policy implementation amongst the down-line functional divisions within MAFF. This may be suspected to affect the quality of service output and performance"(Prap P11 1995, 35).

Members of the other groups of these ’down-line functional divisions‘ saw the problem in the opposite direction. Instructions and policy goals were considered as not clear. There was not sufficient explanation as to why things were supposed to happen in a certain way. If attempts were made to forward difficult matters to management, feed back was not forthcoming. Some groups felt that too much information with too little relevance was circulating up and down the hierarchy<94>.

These problems are well known in organisational theory. Mintzberg (1992, 232) describes that bureaucratic organisations have a tendency to centralise decision making to such an extent, that the concerned staff have no or very little influence on the decision making process. This means, that the decision making managers do not possess the necessary information of the real situation at the base. Medium hierarchical levels, that might still posses this information, generally do not have sufficient influence to modify or adapt decisions made at the top. Their only function lies in the transfer of decisions from the top to the next lower hierarchy level in basically the same way only with a different minuteness of detail.

For the communication in the opposite direction, the case is reversed. Information has to be compressed from one hierarchy level to the next, so that the organisational top management receives a short and precise summary of proceedings at the base. Difficulties do occur whenever non-routine problems are encountered - as happens quite frequently in extension practice. These cases cannot be simply compressed and therefore have to be transmitted unfiltered to the next higher hierarchy level. Due to the limited physical capacities on higher levels, this leads to bottlenecks that may result in the fact that decisions are made very quickly, irrespective of the problem nature, or that wrong decision are made or that problems are not dealt with at all (cf. Mintzberg 1992, 248).

Comparing MAFF with the Mintzberg characterisations above, the parallels are obvious. It may therefore be concluded that MAFF suffers from the typical diseases of bureaucratic organisations everywhere in the world.

However, taking the different perceptions of actors in the system serious, it must be concluded that there is no such thing as a standardised system configuration that best fits the needs of all departments. Depending on the position of the observer within the system, a completely different picture exists and may be valid. This implies that from the point of view of optimising information flow and interaction, that a hierarchical configuration cannot provide adequate results. It reduces direct information flows and increases, at the same time, the chance of information distortions. In this line of reasoning the connection between information flow and organisational configuration is apparent. Indeed, the


130

subjects are difficult to differentiate. Considering linkages and information flow as interrelated, a strategy to improve linkages could also assist to improve information flow problems.

The main conclusions of the workshop were summarised in a plan of action to be implemented in the following 12-18 months after the workshop:

One group that worked on the policy guidelines for the action plan suggested the following philosophies to guide the process:


Fußnoten:

<43>

Personal communication 9.12.1994 with the head of the research station Mr. Jaishiree Gawander. In 1999 FSC realised the problem and set up a 6 million FJD programme for sugar cane research (Sunday Times 1999).

<44>

An organisational chart of the Ministry is provided in chapter 5.2.1.

<45>

cf. MAFF 1994.

<46>

Member countries of USP are: Cook Islands, Fiji, Kiribati, Nauru, Niue, Marshall Islands, PNG, Solomon Islands, Tokelau, Tonga, Tuvalu, Vanuatu and Western Samoa.

<47>

The largest agricultural library is maintained by the School of Agriculture at the Alafua Campus in Western Samoa.

<48>

In 1997, the Fiji Government partly returned to past wisdom, a commodity development framework (CDF) with input subsidies for exports commodities was approved (email from Dr. Kronen dated 19.1.1998).

<49>

Prior to 1991, the name of the Ministry was Ministry of Primary Industries (MPI); from 1992-95 and again since 1999 Ministry of Agriculture, Fisheries and Forests (MAFF), from 1995 to 1999 Ministry of Agriculture, Fisheries and Forests and ALTA (MAFFA). For simplicity, in this study the abbreviation MAFF is used.

<50>

The Fiji Government is on the Internet now. The latest Government structure can be viewed at the following address: http://fiji.gov.fj/cabinet/minstry.html.

<51>

Comparing the annual reports for the eleven year period 1984 to 1995, not a single new commodity was developed. On the contrary, the number of major commodities reviewed degreased from 25 to 16.

<52>

Calculated on the basis of the MAFF annual reports for 1994 and 1996 and 1998.

<53>

Detailed recommendations, see Bachmann et al. 1997 b, 41.

<54>

The Ministry has the role to inspect all traded goods in order to protect agriculture against the introduction of pests and diseases from outside the islands. This includes spraying farmers‘ fields if introductions of pests or diseases are observed.

<55>

As fruits were one area where major export chances were expected, post harvest treatment of fruits against fruit flies was an important area of research.

<56>

Eyzaquirre (1996, 63) points out that Fiji, due to its size in the region, resumes a regional responsibility to maintain and improve the genetic materials of the region.

<57>

This view is also supported by the official library statistics. In 1998 only 370 users were recorded of whom 97 % were students (MAFFA 1999, 84).

<58>

An interesting case was the Pineapple project at Seaqaqa research station. Local researchers had never heard anything about modern pineapple production using e.g. fertilisation according to plant leaf nutrient analysis, control of flowering periods with plant hormones. The project will be discussed later as one of four case studies.

<59>

Environmental awareness is promoted by a large number of NGOs (e.g. Greenpeace, SPACHEE) but also donors and international organisations (e.g. UNDP, EU; GTZ, etc.). It already plays a considerable role in public opinion. However, besides talks and discussions, still very few funds are allocated to this sector.

<60>

The fact that three officers rated it as always done, appears palliated. During the researchers stay in Fiji from 1992 to 1995, not a single issue of the Journal of Agriculture was published. Since then, attempts were made to revive the journal, but according to an email by Dr Kronen (21.3.1998), it still struggled heavily to receive articles.

<61>

In chapter 5.5.2 the problem will be highlighted by using the example of a tomato variety ranking carried out as a PRA field exercise.

<62>

Personal interview, 4.3.1993.

<63>

Calculating with all staff (established and unestablished) the ratio is still 3.5. This is still a good value. Given the fact that many areas are flat, it would also be possible to reach farmers with bicycles. But no bicycles or motorbikes are available. In particular for short distances, and at times when funds are shorter, this would be a good, simple alternative.

<64>

This finding was also confirmed by the two extensionists in the training course. Only one of them ever used the library once in his 20 years of service at the Ministry. During the seven months that the research team was located in the library, not a single extension officer requested a literature search.

<65>

Own estimate based on the National Agricultural Census 1991 (MPI 1992, 73).

<66>

This question was very difficult to assess. Generally, farmers did not criticise openly. RRA 1 included a question on the quality of extension services. All farmers rated the services (out of politeness?) as very good.

<67>

To allow a better judgement of extension approaches used, it would have been necessary to spend more time together with extension officers in the field and study their work in more detail. Such an approach was outside the scope of this thesis. Lühe 1996 used such an approach to study the extension system in Benin.

<68>

An example of this working approach will be provided in the first case study in chapter 6.3.1.

<69>

The complete results of RRA 1 and RRA/PRA 2 are published in Bachmann et al. 1997 a and Bachmann et al. 1997 b. RRA/PRA 3 is published in Bachmann et al. 1996.

<70>

Waters-Bayer (1994, 145) raises the question of ethics in rural research. She distinguishes extractive and enriching research. At the example of the Fulani dairy sector that is largely run by women, she shows that extractive research lead to market concentration, less flexibility and a loss of control for rural women. Attempts to enable women to learn from each other and develop innovations on their own (enriching research) were not undertaken. This small example shows that the perspective of research may have a considerable impact on both ethics and design of innovations.

<71>

In 1995 exchange rates were as follows: 1 Fiji $ = 1.15 DM or 0.71 US $. Calculated farm incomes were comparable with incomes in other professions. A housemaid in the capital Suva earned around 2500 $ a year, a factory worker 3-5,000 $ and a civil servant in middle management position 7-9,000 $.

<72>

In the first RRA the extension officers made the ranking of farmers into the three income groups. The wealth criteria used were: size of the house, quality of furniture, and visible agricultural investments. Extension officers had big problems to use 3 groups and felt that many decisions were very subjective. Therefore, in PRA two and tree only 2 groups were distinguished. Rather than using income and wealth as indicators, ownership of expensive agricultural inputs (e.g. tractors, no. of pairs of bullocks, etc) and the actual level of use of agricultural inputs were used as criteria to distinguish farmers. Furthermore, farmers were invited to group themselves according to these criteria. This approach proved both easier to apply in the field and the information received were more consistent.

<73>

A farmer in Vunisodo had a low yield of 9 t/ha due to very wide spacing of the crop. The tuber size was very good and therefore the farmer was satisfied with his ’yield‘. As he had access to more land than he could cultivate, he did not consider the lower yield per area as a problem. The production was satisfactory for him. Pointing to the fact that a higher yield per area might also mean less work (less area to cultivate), he picked up the idea and promised to try it out.

<74>

Production figures in the Fiji farm management handbook (cf. MAFF 1994) are based on field and research station findings.

<75>

Here the question may be asked: what can they still learn from research and extension? Regarding their traditional root crops, farmers certainly have a very high level of cultivation expertise.

<76>

Gündel (1998) describes the process of adapting a ’rough or semi-ready‘ recommendation to the specific needs of individual farmers. She points out that a recommendation must support the necessary flexibility, to enable farmer specific adaptation.

<77>

In this respect the experiences gained with the use of above criteria could be further analysed. Besides the use of explorative survey tools such as PRAs, formal surveys with a higher sample size would be suitable to validate existing findings.

<78>

Stolzenbach (1994, 157) examined farmers‘ experiments in Mali. He calls farmers‘ way of working: reflection-in-action. He sees an overlap of reflection and action as an ongoing-process.

<79>

But no case is without exceptions. The before mentioned pig farmer also experimented with crops. Advised by extension to plant new cocoa seedlings, he planted almost half of his land with this perennial crop. Only a few years later prices for cocoa collapsed and at the time of the survey it was even considered uneconomical to harvest the crop. Obviously, not all farmers can afford to experiment on a large scale on their farms. But this also protects them to a certain extent from making mistakes.

<80>

Fijian word for house.

<81>

In this respect see also following chapter 5.5.3 on basic need satisfaction.

<82>

On the contrary, Fiji already faces similar problems to industrialised countries with rising levels of people overweight, malnutrition, high fat and sugar consumption and related diseases e.g. diabetes, and heart attack (cf. Schulz 1992,17).

<83>

This finding is also confirmed by the high literacy rate of 91,6 % in Fiji (Collard, 1998).

<84>

cf. Bachmann et al. 1997 a and b.

<85>

Rauch (1996, 21) points out the importance of solutions. Development agents should add their views on potential ways to solve problems. Target groups should not be left alone in the discussion of problems and solutions. In this respect, a joint effort of all actors to look for solutions provides the best changes for success.

<86>

These criteria could be used to start up the process. If the criteria proved insufficient, additional criteria might be added on in due course.

<87>

The composition of workshop participants is shown in Table 5 in chapter 3.3 on p.41.

<88>

This list of tasks is extracted from the ISNAR research reports (Eponou 1993, 19) and was presented earlier in chapter 2.2.1 .

<89>

The workshop programme and detailed group results are documented as a SCEP project report (SCEP 1994).

<90>

The prospects to receive donor assistance to address existing linkage problems certainly also created a good motivation to analyse the situation properly. It might even be the case, that for this reason, some participants exaggerated existing deficits.

<91>

Personal communication by the biometrician of the PRAP Programme Dr. D. Morton 12.4.1994. The low rate of documentation of research findings is caused by several factors. Poor analysing skills, poor report writing skills and difficulties to interpret findings. As literature searches are not done systematically, many trials investigate research questions that are already well documented in other countries.

<92>

The entire results of the workshop are documented in a project report PRAP P11 (1995).

<93>

Yaqona is prepared out of the roots of a local plant. The drink is prepared in a large wooden bowl and its preparation ceremony is a prerequisite for all social events. The drink contains a narcotic substance that makes the consumer feel sleepy if larger amounts are consumed. Yaqona consumption is very frequent in daily life, even in Government offices.

<94>

According to the researcher‘s own observation, many of the internal documents circulated in the Ministry were heavily standardised papers (e.g. fixed report formats, tables, closed questions). Staff even joked about these formats and reports. A good example is the ’standard excuse‘ found in many documents: too much or too little rain. The weather is blamed for almost everything and this excuse is readily accepted by superiors without any further explanations.

<95>

This activity comes close to what Engel describes as creating information exchange protocols between different actors. In a new publication Engel and Salomon 1997 put together a comprehensive toolbox to facilitate network building. Several tools address in particular the subject of information exchange (cf. B3 Info-source-use exercise, B4 linkage matrix, B7 communication analysis). A recent brochure of GTZ (1997) also illustrates a set of tools that may be used for this purpose.



© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.

DiML DTD Version 2.0
Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML - Version erstellt am:
Thu Mar 29 16:23:51 2001