Inland fish farming alternatives for Ghana: technical and economic aspects TABLE OF CONTENTS

Hyperlinks to non-FAO Internet sites do not imply any official endorsement of or responsibility for the opinions, ideas, data or products presented at these locations, or guarantee the validity of the information provided. The sole purpose of links to non-FAO sites is to indicate further information available on related topics.

This electronic document has been scanned using optical character recognition (OCR) software. FAO declines all responsibility for any discrepancies that may exist between the present document and its original printed version.

TABLE OF CONTENTS

1. INTRODUCTION

2. SPECIES

3. CULTURE SYSTEMS
3.1 Cages and raceways
3.2 Pens and ponds

4. FERTILIZERS AND FEED

5. THE PONDS
5.1 Pond size
5.2 Water supply
5.3 Pond construction
5.4 Pond operations

6. WATER

7. SUPPLY OF FINGERLINGS
7.1 Commercial production of fingerlings of O. niloticus

8. ECONOMIC EVALUATION OF ELEVEN SPECIES/CULTURE SYSTEM COMBINATIONS
8.1 Common assumptions for the economic analysis
8.2 O. niloticus and Chicken manure
8.3 O. niloticus, Clarias gariepinis and chicken manure
8.4 Tilapia and rice bran
8.5 O. niloticus (monoculture) with pigs
8.6 O. niloticus (male) and pigs
8.7 O. niloticus and Clarias in rice fields

9. REFERENCES

1. Introduction

The purpose of this report is to identify those fish farming alternatives which are best suited to Ghanaian conditions. The perspective is that of the next ten years.

In order to identify these alternatives, the authors review: species, culture systems and their economic consequences, water needs and costs, supply of fingerlings and finally specify the physical and economic aspects of the culture systems.

The report does not discuss the markets for the fish produced in the fish farm; that is done in Field Technical Report 3 Neither does it discuss the economic availability of inputs; that is done in Field Working Paper 4. However, the analysis of this report rests on the conclusions of those two sections.

2. Species

Tilapias account for at least 95% of farmed fish in Ghana. Total annual production has been estimated to be between 360 and 450 tons for the five year period 1984–1988¹. Harvest of other species in fish ponds is due to: (i) accidental introduction of these species into the ponds, or, (ii) stocking by the farmer. These species include: Clarias gariepinus, Heterobranchus spp, and Heterotis niloticus. However, even when these species are stocked, they are few in number compared to the tilapias. Thus, from the point of view of the farmer they are secondary. Heterotis do reproduce in the pond, but the rate of reproduction is low. Catfish fingerlings have to be caught in the wild.

A number of considerations reduce the number of species which are likely to be the subject of farming in the next five years. The main factors are (1) the market and (ii) the availability of fingerlings. The latter is the most drastic constraint. In fact it effectively limits fish farming in the early 1990's to tilapias and occasional use of catfish and Heterotis. These species are commonly consumed and will find a market throughout the country. However, even when the Clarias hatchery limititation has been removed there are few other immediate candidates for culture. Carps would have to be introduced, and would have to be accepted by consumers. For the first few years market price would have to be considerably below that now obtained for tilapias, which would make culture economically uninteresting, as the facilities and inputs used could just as well be used to grow the higher priced tilapias. The climatic conditions, of course, do not lend themselves to the culture of salmonids.

¹ FAO Fisheries Circular 815

3. Culture systems

A fish culture system consists of a combination of inputs: water, physical installations, feed, labour and fish. Each such combination yields a certain output in the form of fish (or other aquatic animals). The appropriate combinations are those for which the value of the inputs used during a specified period is less than the value the producer obtains for the fish produced. In what follows, the analysis will be organized around the four basic types of installations: ponds, pens, cages and race-ways.

3.1 Cages and raceways

Experience has shown that in cage and raceway culture the cost of feed is a large proportion (50% or more) of the cost per kg of fish produced. Also, in Ghana the cost of feed will be the determining factor.

Although, at least in theory, plankton-feeding tilapia (O. niloticus) would survive and grow in cages placed in waters with large masses of plankton, the growth per metre cube of water in a cage would be low, both because of the low growth rate (less than 0.5 g day) but primarily because of the low stocking density (0.1–0.2 kg of fish per cubic metre) which would be possible. To increase growth and stocking densities, in order to reduce the fixed costs (of cages) per kg grown, artificial feed in pelleted form, and containing a minimum of 20 % of proteins, would be necessary. The likelihood of obtaining such pellets at a competitive price is small. Given an ex-farm price for tilapias of 350 to 400 C/kg, a conversion ratio of 1.6 and a 50 % feed cost in the overall cost of production, the per kg cost to the fish farmer would be about (0.5 × 400/1.6 =) C. 125. Given the cost of ingredients and transport, the authors consider it very unlikely that feed could be obtained at the stated price.

Thus, the mission considers it extremely unlikely that cage culture of tilapia can compete with either capture fisheries or with pond culture.

The economic aspects of race-way culture are similar. The race-way alternative is also unlikely because of the lack of suitable sites, other than possibly in the major irrigation canals.

3.2 Pens and ponds

Most fresh-water bodies in Ghana are subject to large (several metres) seasonal fluctuations in water levels. This makes them unsuitable as sites for pen-culture. For the reasons given above concerning the need for feed in cage culture, there is also a need to feed in pen culture. Without feed, the stocking densities will not be high enough for the production to cover the fixed costs of the installations.

In addition, the pen-culturist would have to compete with the local fishermen, who are able to capture the tilapia in all the fresh water bodies that are potential sites for pen culture of tilapia.

Thus, the mission believes that for the next five to ten years the pond is going to remain the base for the development of any significant commercial fish farming in Ghana.

4. Fertilizers and feed

There are two main arguments in favour of pursuing a fish farm development strategy which makes extensive use of animal manure, rather than agro-industrial by-products and agricultural waste.

The first is that livestock-feed is in short supply in Ghana. This would be so even if all the agricultural waste and by-products now produced were used as feed for livestock. An integral part of the present agricultural development policy is to make by-products available to livestock as feed. This means that during the next few years the real cost of agro-industrial by-products will increase. A similar policy is not being pursued in regard to animal manures.

The second argument in favour of manures is that of cost of transport. Transport is relatively expensive in Ghana. The use of a 5 ton truck, for a day, costs the equivalent of about (35.000/500 =) 70 man-days of unskilled farm labour. Thus, the less transport needed to transport either inputs or outputs, the better off the farm will be. As about 10 times as much animal by-products (measured in weight) as animal manure is needed to obtain a given amount of tilapia (or Clarias) in any one pond, it is preferable to base the culture on the use, and possible transport, of animal manure rather than on the use of agro-industrial by-products.

5. The ponds

There are several aspects to a well-functioning pond: size, construction method, water supply, drainage, and maintenance. In the following sections, the authors discuss the main aspects of importance to the future fish culture extension effort, but do not repeat the information which is commonly available in fish culture manuals.

5.1 Pond size

Several factors will combine to determine the common size of future fish ponds. The more important would seem to be: land-area available; access to water, manure and agro-industrial by-products; the preferred construction method; access to capital.

More than half of holdings are below 1.6 hectares in size. A holding could then consist of more than one farm (contiguous piece of land). The fact that only 12 % of Ghana's surface is cultivated should not be interpreted to mean that there is considerable space available for fish farming.

Although availability of land for the individual farmer means that he will not have a fish farm of several hectares, the land limitation is not likely to be the effective one. The effective limitation is going to be the supply of inputs: water, fertilizer or agro-industrial by-products.

Those who have access to, or can finance the payment of, a bull-dozer are likely to build several, and possibly quite large ponds at a time. Large in this context means a pond between 1000 and 4 000 square metres in size. The smaller farmers, those with a total holding of less than 1.6 hectares, might start more carefully by building one pond of a few hundred square metres, using manual labour.

Harvesting of fish by seining is facilitated if ponds are rectangular and not wider than 25 to 30 metres. For smaller ponds a width of 10 to 15 metres is recommended.

The pond bottom should slope gently, from a minimum depth of 0.5 metres to a maximum of 1.5 metres at full water supply. At that point the difference between the crest of the dikes and the water surface should be about 0.6 metres.

5.2 Water supply

Water inlets. The typical pond will be fed by water taken from a nearby stream, or river. The supply canal will be constructed so that it is possible to regulate, including completely block, the inflow of water to the pond. The common situation will be that of an earthen canal. This implies maintenance, but this is justified due to the high cost of cement. The water is taken from the supply canal into the pond through a plastic tube, or through a piece of bamboo.

Drainage There are three alternatives for drainage: (i) a monk, (ii) turn-down drain pipes, or, (iii) siphoning in combination with an overflow (in case the inflow can not be fully controlled).

The mission estimates the construction cost for a monk to be of the order of C. 12.000 to C. 15.000. A siphon can be made for little to no cost using discarded exhaust pipes and a short rubber hose. A turn-down pipe need not cost much more.

Thus, the smaller the pond, the more expensive - relatively -becomes the monk. In a pond of between 1 and 2 ares (100 and 200 m. square) the monk would be as expensive as the digging of the pond itself. The authors recommend that monks not be installed in ponds smaller than 10 ares or 1.000 metres square, but recommend that turn-down drain pipes be used whenever possible.

5.3 Pond construction

Ponds can be constructed either by hand or by utilizing a bulldozer. Experience shows that a pond of 100 m square needs between 16 and 25 mandays depending upon the soil conditions. Assuming average soil-conditions and taking into consideration that in a larger pond there is proportionally less work to do on pond-walls, a one acre pond would need about (20 × 4.000/100=) 800 mandays. At a salary rate of C. 500 per man and day (4 to 5 hours-work; morning to shortly after midday) the expense comes to C. 400.000. for the pond, not including the monk.

If the construction is made with the use of a bulldozer, the following is the basis of calculation. The bull-dozer, type D6 Caterpillar, equipped with a ripper, will be able to move 100 cubic metres of earth per hour (provided distance is not more than 80 metres) (Voss, personal com.). Given that it will not be necessary to excavate down to 1 metre over the whole pond area, in order to obtain a pond-depth of 1 metre, the calculation builds on an average excavation of two thirds of a metre. The volume to be moved is then about (4.000 × 2/3=) 2.667 cubic metres. At a charge of C. 13.000 per hour of actual operation of the bull-dozer (all expenses included) the cost for the pond excavation comes to (2.667/100 × 13.000 = 346.710, or say) C. 350.000. To this should be added expenditure for labour occupied in compacting and planting grass on pond-walls. This will cost about Cedis (8×280/20=112×500 =) 56.000.

In conclusion, it seems that there is little money to save by using either of the two construction methods. However, in areas where labour is available for hire, it will, of course, be desirable from a social point of view to build the ponds by hand. Furthermore, certainly this will be the case when smaller ponds (below 1000 square metres) are to be built. The cost of transporting the bull-dozer will be too high.

The complete cost of the pond should include that for a monk and for water inlets and water canals, depending on which is used. The monk will use one bag of cement, plus gravel and sand at a cost of C. 2000 per bag of 50 kgs and ten mandays of labour, five of which should be for a mason. Total cost for the monk is thus about C. 12.500. To this should be added the eventual cost of a water inlet. This will depend on the length and construction method. The maximum cost for the inlet is placed at that of the monk.

Thus, the total costs for the one acre pond will be of the order of C. 0.425 million. In order to cover miscellaneous and unforeseen expenditures we add 10 %. In the calculations that follow the standard cost for a one acre pond has been rounded to C. 0.5 million (as of April 1990).

5.4 Pond operations

Pond preparation, filling and draining should follow standard procedures, summarized as follows:

• pond bottom should be levelled with a gentle slope allowing complete draining of the pond at harvest;
• inlets and outlets should be adequately screened to prevent unwanted fishes from entering the pond, as well as preventing cultured fish from escaping;
• to allow silt to deposit, ponds should be filled one or two days prior to stocking;
• once the pond is filled, manure or compost should be applied, the quantities to be determined by pond size and the species to be stocked;
• stocking should be done according to the prescribed stocking rates, handling fish as little as possible and letting them adapt gradually to the water temperature of the pond, at given stocking rates;
• monthly sampling is recommended to check the growth rate of fish;
• ponds should be drained early in the morning to avoid fish being exposed to high temperatures:
• once drained, unwanted vegetation should be removed (with the roots) and excess silt taken out (can be used as fertilizer) and grasses growing on the dikes should be cut.

Normal precautions (sticks in the pond bottom, barbed wire) may be used to dissuade poachers. However, as will be seen later, any substantial fish farming operation (more than 20 ares of water surface area) will be able to afford a watchman fulltime.

6. Water

Ponds are filled by gravity. Given the high evaporation rates of 5 to 6 mm daily in Ghana (pers. com. Voss), there is the need for constant replenishement of some (0.005 × 4.000 × 100 =) 20.000 litres per day (or 20 cubic metres of water) to replace losses to the atmosphere from a 4.000 square metre pond. Losses due to seepage should be minimal. Thus a total replenishment of some 20 cubic metres per day will be needed. During grow-out, water should not be allowed to escape through the monk.

A one acre pond with an average depth of just under one metre is equivalent to 3 acre-feet of water. This quantity of water is the quantity needed to irrigate one crop of paddy from planting to harvest. Given that the same amount of water is enough to replenish losses to evaporation during six months, the total water needed to raise on crop of fish during six months will be equivalent to that needed for two crops of paddy.

At the moment, farmers growing paddy on land irrigated by the Ghanaian Irrigation Development Authority (IDA) pay an equivalent of C. 22.500 per year and acre. The amount is not linked to the quantity of water used, yet. However, it seems likely that in the future the charges for water will be revised and farmers be made to pay in relation to both quantity used and to the actual cost of delivering the water. The mission has used an amount of C. 45.000 per acre and 6 months period, for those species/culture method combinations for which the typical, or most advantageous location, is believed to be the irrigation scheme.

7. Supply of fingerlings

As stated previously the mission considers tilapias and Clarias to be the fish which may be subject to culture during the next 5 to 10 years in fresh waters in Ghana. Any reasonably sized fish farm will depend on a regular supply of fingerlings.

While it was possible at the time of the mission's visit to obtain fingerlings of tilapia, produced in ponds, the same was not true for Clarias gariepinis. No Clarias producing hatchery existed in the country. The technology for reproducing Clarias, however, is being commercially used in the Central African Republic and in Cote d'Ivoire. For the economic analysis of culture alternatives, presented later in this appendix, it is essential to have a realistic estimate of the cost of hatchery produced fingerlings of C. gariepinis. A hatchery and its estimated cost are sketched out in Field Working Paper 3. A price of between C. 8 and C. 12 per fingerling seems reasonable.

In the beginning of 1990 a small trade in tilapia fingerlings existed in the Kumasi area. The fish ponds operated by the Department of Fisheries at the Vea Irrigation Scheme also produced some. The reported capacity is about 70.000 per year for four “fingerling” ponds.

Farmers who choose to culture tilapia (specifically O. niloticus) in monoculture will have an automatic supply of fingerlings. As a rule it can be expected that a culture that relies on manure (either from chicken or pigs) at the specified rate (see Section 8) and which therefore produces fish at the rate of 1.600 to 4.000 kgs/are/year, will produce enough fingerlings of the 5 to 10 gram size to permit restocking of the pond. However, this implies that the farmer has access to at least two ponds, or to hapas in a reservoir, in order to store fingerlings until the production pond has been desilted, filled with water and manured.

Farmers who choose to culture O. niloticus and C. gariepinis jointly, or who choose to culture only male tilapias, will depend on a commercial, regular supply of uniform-sized fingerlings of both species.

7.1 Commercial production of fingerlings of O. niloticus

In order to produce fingerlings of O. niloticus a hatchery in the true sence of the word is not needed. This is so as the brood stock will reproduce - under the temperature conditions prevailing throughout most of Ghana - continuously without any artificial stimulation, as long as stocking ratios and feeding are adequate. Thus, in the following the authors will refer to a “fingerling production facility”.

7.1.1 Components of a fingerling production facility

The main components of the fingerling production facility are: three ponds with a year-round supply of uncontaminated water; proven brood stock, labour and an experienced fish farmer.

7.1.2 Size of fingerling production facility

A commercial fingerling production facility should not be so small that it can not recover its fixed costs - in addition to all direct costs for each fingerling produced. However, neither should it be so large that it can not utilize its capacity fully. The fixed costs of the production facility will consist essentially of capital costs and staff. Thus the smallest economic unit will be assessed in relation to these. It is more difficult to predict what the maximum will be: it should be that at which accessible fish farmers buy all the fingerlings produced. It is thus essential to make certain that the minimum facility in fact is not larger than the maximum that the market is likely to absorb.

The minimum staffing for a fingerling production facility would have to be: (i) an experienced fish farmer or aquaculturist; (ii) two labourers and (iii) one watchman. This staffing will ensure competent technology and a day/night presence on the facility. While the experienced fish farmer need not work full-time at the facility, the others would have do. The salary for this component is of the order of 0.5 million a year ². Thus, in order to recover this cost for a 100.000 fingerlings per year facility, the production would have to be sold at a minimum of C.5/each (not counting other costs). The “reasonableness” of this price in part depends on the weight of the fry/fingerling; at 5 grams it is the equivalent of C. 2.000 per kg of fish, which is about 10 times the present landed price of tilapia.

As a first approximation, the fingerling production facility will be calculated for a yearly production of 200.000 fingerlings. Such a size will be sufficient to supply about 5 ha of ponds during one year, assuming an average stocking rate of 2 fingerlings per square metre. Five hectares may not seem much. However, at an average fish pond surface area of 1.000 square metre per farmer, the fingerling production facility would be enough to supply 50 farmers.

7.1.3 Size of fingerlings

Given that the farmers who engage in polyculture with Clarias will want fry of 10 grams, the facility will have to have ponds to enable the production of these sizes. Also, those farmers who want to grow all male tilapia will be supplied with hand-sexed tilapia of about 20 grams in weight. Also farmers doing mono-culture of tilapia will be pleased to have larger fingerlings, at a uniform per kg price. The facility will be designed to produce 40 % fingerlings of 20 grams, 40 % of 10 grams and the remaining 20% at 5 grams.

7.1.4 Timing of fingerling production and capacity

From the manager's point of view the ideal sales pattern of course would be an even stream of fingerlings. It is unlikely, however, that farmers will demand the fingerlings evenly over the year. They will want to time their harvests to coincide with those periods of the year when fish prices are highest: lean marine season - March, April; and peak flooding of rivers - July/August. Thus it would be unrealistic to design the capacity of the facility to the “average” demand. Instead, the facility will be dimensioned as if it were to supply all the fingerlings in half a year; that is, the capacity will be double the calculated need at an even demand.

² C. 25.000/month for aquaculturist during 6 months; C. 10.000/month to the other three staff for a year each.

7.1.5 Minimum facilities

Land. The production of larvae (0.5 g) can be calculated at the rate of 20/m2/month (Kestemont, Micha and Falter, 1989; p.22). The average need for the facility is (200.000/6=) 33.333/larvae/month. Thus the brood-ponds should measure aprox (33.000/20=) 1.666 square metres. Three ponds, each of 600 square metres (15 × 40), would be sufficient; six ponds of 10 × 30 m would be better.

It takes about a month for the larvae to grow to a weight of 5 grams (Kestemont, Micha and Falter, 1989; page 24). The larvae can be stocked at a density of 50 to 100 per square metre during this grow-out period. Assuming the lower stocking density, a total of (33.333/50=) 666 metre square of pond would be needed to handle the larvae production of one month.

To obtain fingerlings of 10 to 15 grams a further months would be needed; to reach 20 to 25 grams one more month. In order to grow all fingerlings to 20 to 25 grams size, a maximum of (3 × 666=) 2.000 square metres of pond surface area would be needed in three or six ponds. In the latter alternative, the ponds could measure approximately 13 × 26 metres. If fingerlings are sold at a weight lower than the 20 to 25 grams, less surface area would be needed for grow-out.

Thus, the facility as a whole would have nine ponds, with a combined water suface area of 3.828 square metres, just under one acre. Total la d area occupied would be less than 2 acres.

Water. An average of 5 1/minute will be needed to replace evaporatior losses. But more water is needed to fill ponds with reasonable speed.

A 600 m2 pond will need a supply of about 150 1/minute in order to fill up within the space of 24 hours. Thus, it would seem that the water supply should not be much less than 100 1/minute. Of cours, it is important that the ponds be arranged in such a manner that water drained from one pond can be used to fill another.

7.1.6 Economic aspects

(i) Investment

The cost of construction of a one acre (4.000 m2) pond has been estimated by the mission to be 0.5 million cedis. Given the relatively small size of the ponds proposed for this fingerling production facility, monks will not be constructed. However, the work needed to prepare the dikes for the 9 (or 12) ponds will be at least double that of a one-acre pond. The pond construction costs for the facility are placed at 0.65 million cedis.

In addition, the unit will need seines, scales, hand-tools and a shed, with a small office, a store and a place for the watchman. The total investment for the fingerling production facility is placed at one million cedis.

Also, the unit must purchase broodstock. One female can produce at the rate of 200 to 300 larvae per month (Kestemong, Micha and Falter, 1989; p 22). Thus, to have the capacity to produce 33.000 fingerlings per month, the facility would need (33.000/500 =) 66 females (average weight 200 grams). At the ratio of one male (average weight 350 grams) to 4 females, a further 14 males would be needed. Thus, to be on the safe side some 100 broodstock should be purchased. The price of live broodstock, at 2.5 times that for live fish or C. 500 per kg, would mean an investment of about ((20×0.35 + 80 × 0.2) × 500=) C. 11.500. To this add transport and the total costs for broodstock will be of the order of, at the most, C. 50.000.

(ii) Yearly costs

The main yearly costs will be labour, capital costs (interest charges and depreciation) and feed. It is assumed that the fingerling production facility is located at a perennial source of water and that it reaches the ponds by gravity ³.

Labour. As indicated earlier, with a staff of four - of which the aquaculturist is part-time - the yearly bill will come to about 0.5 million cedis.

Capital costs. The working capital is relatively small: an average of Cedis 50.000 for the broodstock, add the same amount for odd items in store (buckets, scales, chicken manure). The fixed investment is larger, about C. 1 million.

While the charges for the operating capital is that charged by banks on short term loans (26% per year) the charges for the fixed investment is that offered investors on their fixed accounts (16 %).

Feed The feed costs in effect will be in the form of the costs of chicken manure, purchased for the unit. The cost is placed at C. 100 per kg including transport (see Field Working Paper 5 for a discussion of cost of transport). For a one acre unit the total use for a year is just under one metric ton. Thus the total cost for manure is C. 100.000, a minor part of the total.

Cost per fingerling The fingerling production unit will make a handsome profit at a sales price of C. 5 per 20 gram O. niloticus. This calculation is based on the production, throughout the year, of 20 gram fingerlings, the most demanding alternative in regard to space requirements.

Return to management, land and water. The return to management, land and water will be some 200.000 per year for the two acres of land used.

³ If situated in an irrigation scheme, the costs for the facility would be increased by about C 90.000/year.

8. Economic evaluation of eleven species/culture system combinations

The following species/culture system combinations are analyzed below:

• O. niloticus using chicken manure.
• Polyculture of O. niloticus and Clarias gariepinis using chicken manure.
• O. niloticus using rice bran.
• O. niloticus using pig slurry.
• O. niloticus males, using pig slurry.
• O. niloticus and C. gariepinis in rice fields.
• O. niloticus and press-cake of fibres from fruits of oil palm.
• O. niloticus using cow manure.
• O. niloticus with chemical fertilizers.
• O. niloticus males, with chemical fertilizers.
• O. niloticus using compost.

At the outset to the mission, the authors judge these to be likely to be the most promising commercial alternatives.

Before going into the details of the analysis, the basis of the analysis will be reviewed. However, it should be noted that the above “single feed” and “single fertilizer” strategies are naturally not the only ones possible and, in fact, they may not always be the best. The basic reason for this tyoe of analysis is that it is adequate to permit the authors to obtain an idea of the relative economic merits of the systems. Their use should not be interpreted as a recommendation that these “single” strategies should be pursued.

8.1 Common assumptions for the economic analysis

The analysis of the various combinations of species and culture-systems should be standardized so that, when completed, the economically most rewarding combinations can be identified. This means that the assumptions should be applied to all alternatives under consideration. The manner in which price levels, depreciation, interest rates and inflation have been handled in the analysis is discussed below.

8.1.1 Price levels.

Price levels of both aquaculture produce and factors of production are important. Both are determined by a variety of factors. Amongst these are (i) the rate of inflation; (ii) the exchange rate of the Cedi against the U.S. dollar and (iii) the future growth of the aquaculture activity.

The rate of inflation has been high in Ghana. During the period 1983 – 1987 it was consistently closer to 30 % per year than to 20% (British Overseas Trade Board, 1989; p 4). The predictions for 1988 were a decline to 22 %.

No doubt inflation will continue. As, however, there is no strong reason to believe inflation will cause costs for inputs to increase more or less than prices for cultured species, no attempt is made to project what prices will be like in 1991 or later.

The exchange rate has stabilized but it is predicted that the Cedi will continue to weaken against the US dollar. As recurrent inputs are not imported and tilapia are sold locally, the effect on the economics of the fish farm will be no more than marginal. In the unlikely case that an export-oriented marine shrimp-, or fresh water prawn-, farming were to develop, a weakening Cedi would favour export prospects.

8.1.2 Depreciation

Pond farming of tilapia and/or clarias is considered below. In this technology costs of the pond represent more than 90% of the investment. Commercial farmers will also invest in seines. It has been considered that drainage can be done with siphons and turn-down drain pipes.

The question is to what extent the pond should be depreciated. There is no real reason to depreciate the ponds in order to protect the owner against fictitious profit making. The ponds, with normal maintenance, should last for the foreseeable future. However, there is another argument for depreciation. Given that economic development will occur, with attendant tendencies for specialization, only a few of existing or future fish farmers are likely to specialize in farming alone. Therefore, by the time the ponds are abandoned, the costs that have gone into their construction should have been recovered, preferably. For this reason the analysis includes a 5 % yearly depreciation of the costs of pond construction. At this rate, the investment will be recovered in 20 years - if the activity just meets costs. As seen later, cultures based on the use of manures in fact recover the investment much more rapidly.

8.1.3 Interest on working capital and investments in ponds and equipment.

From the point of view of the owner of the fish farm, there are two types of investment: the capital sunk in the construction of the fish pond and the funds deployed in order to pay for the recurrent inputs - including labour - required to produce fish from the pond. Both these “investments” should yield a return.

Farmers have been able to borrow from commercial banks the sums needed for initial expenditures on seeds and fertilizers. For these “crop loans” farmers are used to paying the going rate of interest, generally 26% per year. It seems likely that they might, or at least will have the possibility of, obtain this type of financing to cover expenditures on seed, feed and labour for fish culture. For this reason, the analysis includes expenditures on loans to cover these costs at the annual interest rate of 26 %.

The return on costs of constructing the fish pond can be addressed in various ways. In the Ghanaian context there are at least two situations: (i) that of the multi-faceted, fully commercial farmer who employs labour to run his various activities, and (ii) the small commercial farmer who does most of his work with family members and, in particular, uses them to construct the fish ponds. The first category generally must find the funds to finance the construction. That individual, therefore, has at least an hypothetical possibility of placing the same funds in the bank at the going rate.

Farmers in the second category are more likely to use family labour to construct ponds. They are then, in fact, converting a resource into capital. While the first category is likely to include, amongst the cost of the fish farm, the imputed (foregone interest) income, the latter is less likely to do so.

In the economic evaluations that follow, this cost has been included. It is based on the annual return of 16% which fixed deposits earned in Ghanaian banks at the time of this study.

8.1.4 Inflation

The impact of inflation on the price levels has been discussed above. That, however, is not the only impact the inflation has on the analysis of costs and returns.

The analysis presented below consists of a yearly cost and income statement for a fish farm once it has been established and is operating well. That analysis is static. It does not take into account all the effects of inflation. It does consider the effect it has on the value of fixed assets by the requirement of a 16 % return on the investment, but it does not account for the fact that prices of inputs and outputs change over time. The analysis below attempts to cater to that through the inclusion of a “inflation correction factor” (ICF) that is equal to the rate of inflation times receipts less expenditures. As expenditures occur before receipts, normally the ICF is positive.

8.2 O. niloticus and chicken manure

The typical situation for this fish farming alternative is the Ashanti region, around Kumasi. The farmer who establishes this fish farm also should have at least 600 egg-producing hens, or broilers, the droppings from all of which are used to fertilize the one acre fish pond.

The operations and inputs/outputs are described in Tables 1 and 2. As can be deduced from Table 2, it is postulated that the farmer can not use or sell the chicken droppings; that is, they have no value.

Table 2 shows that the returns to management, land and water are very handsome indeed, and this in spite of the high interest charges on invested capital. The investor should have recovered his investment in ponds after three harvests (within 18 months). In fact, even were the farmer to pay as much as C. 100 per kg of chicken droppings, the returns would be reduced only marginally; the investment would be recuperated after four, maximum five harvests (30 months).

Modifications. There would not seem to be any a priori reasons why a scale-up of the above activity would significantly modify the relationships described above. Also smaller farmers - say operating a pond surface area of only a tenth of that assumed, and having only 60 hens - could recover their investment in the time indicated.

TABLE 1: O. niloticus and Chicken Manure

TABLE: 2 Typical cost/income for established cultures at early 1990 prices: O. niloticus and Chicken Manure.

Period of culture: 6 months
Size of culture unit (m2 of pond surface area): 4.000 sq.m.
Region or area: Ashanti, outside Kumasi

⁴ Opportunity cost

8.3 O. niloticus, Clarias gariepinis and chicken manure

The typical situation of this fish farmer is as follows. Located in Ashanti and having a multifaceted farming operation including the raising of at least 600 egg-producing hens, or broilers. All the droppings from the hens are used to fertilize the one-acre fish pond.

The operations and associated inputs/outputs are described in Tables 3 and 4. Also in this case the basic analysis supposes that the fish farmer has no alternative use for the chicken droppings. However, given the substantial returns that this type of culture ought to give, the farmer in fact could afford to pay rather substantially for the manure. C. 150 per kg would not harm the benefits unduly.

The model, however, depends on a supply of fingerlings of Clarias gariepinis. This, in the spring of 1990, is totally unrealistic. Although it would be possible to obtain fingerlings of clarias through fishing, it would time-consuming indeed to obtain 4.000 of about 10 g each. The uniformity is important to avoid cannibalism as well as predation on the 20g fingerlings of O. niloticus to be introduced simultaneously. A functioning clarias hatchery as a precondition for this type of culture. Appendix gives a brief analysis of the likely costs for hatchery produced fingerlings of clarias. The conclusion is that the cost will be somewhere between C. 8 and C. 12 for each fingerling.

Table 4 shows that the returns (net profit) to management, land and water are better than with only O. niloticus and the investment should be recovered within two harvests (12 months). A charge of C. 150 per kg of chicken manure would delay the recovery of the investment with one harvest, no more.

Modifications can be done to this model without significant effects on the relationship between income and costs. This is so basically because the fixed type of cost - mainly salaries - are such a relatively small proportion of the total costs (less than 10 %). Other charges are all directly related to the scale of the operation.

TABLE 3: FISH AND CHICKEN MANURE

TABLE 4 FISH AND CHICKEN MANURE

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit (m2 of pond surface area) : 4000
Region or area: Kumasi

8.4 Tilapia and rice bran.

The typical situation for the culture of O. niloticus feeding on rice bran is the irrigation scheme, here modeled on the situation at Vea in Upper East Region. The farmer would convert a rice field to a fish pond ⁵. The basic situation includes one where the farmer has no alternate use for the rice bran, and thus he can consider it a free good.

The operations and inputs/outputs are described in Tables 5 and 6. The inputs required should be available. The Vea fish ponds, operated by the Ministry of Agriculture, can be used to supply a few hundred thousand fingerlings per year.

The economics of the operation are much less promising than that described for the variation using chicken droppings. The income basically matches the projected costs (including return on invested capital).

Modifications. Variations to scale, for reasons discussed earlier, will not change the basic conclusions. They would change, however, once the interest rate and rate of inflation are reduced.

⁵. The farmer would have to obtain the approval of IDA. This issue needs to be settled between the Deparmtent of Fisheries and IDA before any effort is made to extend this technology to rice farmers.

TABLE 5 TILAPIA MONOCULTURE WITH RICE BRAN

TABLE 6: TILAPIA MONOCULTURE WITH RICE BRAN

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit (m2 of pond surface area): 4000
Region or area: Tono irrigation scheme, Upper East Region

8.5 O. niloticus (monoculture) with pigs

The typical situation also for this culture system alternative is the established pig farmer in the centre of Ghana, around Kumasi or Koforidua. The farmer has some 40 pigs which he can put into pig-sties on the pond-bank, or from which he can transport the manure to the ponds. The pig-farming is presumed to be an economically rewarding activity on its own.

The operations and inputs/outputs are described in Tables 7 and 8. As is indicated in Table 8, the calculation of profitability assumes that the farmer is not selling, and does not have the possibility of selling the pig-manure.

The return to management, land and water, are substantial. Given the rather large imputed interest cost on the funds invested in the ponds, the cash flow position is even better; a surplus of about C. 540.000. Thus, the farmer should be able to recover any funds invested in the pond within a cycle of culture.

Modifications. Also for this alternative there is no a priori reason to expect that any change in scale (down to 100 m square) will substantially alter the rates of return. Given the favourable cash flow, the culture system is rather insensitive to inflation; that is, the returns will be about the same irrespective of the level of inflation. This is achieved as the imputed costs of interest on the capital invested in the pond is counterbalenced by the “Inflation compensation” obtained from the large positive capital flow, about equal to the size of the investment.

TABLE 7 TILAPIA MONOCULTURE WITH PIGS

TABLE 8: TILAPIA (MIXED SEX) WITH PIGS

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit (m2 of pond surface area): 4000
Region or area: Forest zone: Kumasi, Koforidua

8.6 O. niloticus (male) and pigs

The typical situation here is the same as that assumed for the culture of O. niloticus (monoculture) and pigs (Section 8).

The operations and inputs/outputs are described in Tables 9 and 10. Given the larger tilapia, the revenues are higher. However, as the farmer has to pay considerably more for the larger (20 gram) males than for the 10g mixed sex fingerlings, the net return for this culture alternative is about the same as for that of raising O. niloticus, of mixed sizes, in association with pigs.

Modifications. The scale of the operation can be varied without major effects on the results. Given the large sales revenue per culture period, when compared to the investment in ponds, the effects of inflation are not discouraging.

TABLE 9 O. niloticus (MALE) AND PIGS

TABLE 10: O. niloticus (MALE) AND PIGS

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit (m2 of pond surface area): 4000
Region of area: Forest zone: Kumasi, Koforidua

8.7 O. niloticus and Clarias in rice fields

The typical situation here is that of an irrigated rice plot in one of the IDA irrigation schemes, typically in the North. A one acre rice plot is modified into a rice-field suitable for rice-fish farming. This involves the excavation of a one metre wide dike along the perimeter of the pond. The cost of this could be of the order of (280 × 0.4/4 × 500=) 14.000 cedis. In addition, there is a loss of paddy production. This loss is the equivalent of about (4.000 × 0.4 × 0.07 =)112 kgs. In loss of revenue this is equivalent to (112/82 × 5.300 =)7.200 cedis. At the time of writing the typical situation is hypothetical (but technically feasible), as there are no clarias hatcheries to produce fingerlings for sale to farmers.

The minimum water depth is 15 cms in the rice field. It is assumed that the rice is transplanted from a nursery. This practice will drastically limit the need for weeding and the use of herbicides. The fish will eat insects, thus eliminating the need for using pesticides. These represent savings to the farmer.

The operations and associated inputs/outputs are described in tables 4.11 and 4.12. The revenue/cost situation is also depicted in figure xxx. The costs associated with this culture are basically those of acquiring fingerlings. There are little, if any, extra labour costs and no costs for feed or fertilizer. However, because no attempt is made to provide extra fish feed, the yields are naturally lower than in pond culture. The total combined production for a rice-field during four months is 360 kgs (or 360 x 3/0.4 =) 2700 kgs per hectare and year. However, as it is unlikely that more than two crops will be grown, the productivity is better placed at 1.800 kgs/hectare/year.

The culture scheme can be modified with regard to size without any major changes in the rate of returns.

TABLE 11: O. Niloticus AND Clarias IN RICE FIELDS

TABLE 12: O. Niloticus AND Clarias IN RICE FIELDS

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 120 days
Size of culture unit (m2 of rice field): 4000
Region or area: Irrigation schemes: Vea, Tono, Afife

TABLE 13: O. Niloticus AND PRESS CAKE FROM FIBRES OF OIL PALM FRUITS

TABLE 14: O. Niloticus AND PRESS CAKE FROM OIL PALM FRUITS

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit: 4000 m²
Region or area: forest zone

⁶ Cost of transport.

TABLE 15: O. Niloticus AND COW MANURE (WET)

TABLE 16: O. Niloticus AND COW MANURE

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit: 4000 m²
Region or area: forest zone and Northern Ghana

⁷ Inputed oportunity cost

TABLE 17: O. Niloticus AND CHEMICAL FERTILIZER (I)

TABLE 18: O. Niloticus AND CHEMICAL FERTILIZER (I)

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit: 4000 m²
Region or area: throughout Ghana

TABLE 19: O. Niloticus AND CHEMICAL FERTILIZER (II)

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit: 4000 m²
Region or area: throughout Ghana

TABLE 20: O. Niloticus AND COMPOST

TABLE 21: O. Niloticus AND COMPOST

Typical cost/income for one culture cycle for established fish farmers at early 1990 prices.

Period of culture: 6 months
Size of culture unit: 4000 m²
Region or area: throughout Ghana

⁸ The cost for compost is the time of the labour spent.

9. REFERENCES

British Overseas Trade Board
Country Profile, Ghana
The Dept. for Enterprise. GOUK; Sept. 1989.

Kestemont, P; J.C. Micha & U. Falter
“Les methodes de production d'alevins de Tilapia nilotica”
FAO, ADCP; 1989.