Information on current and past fish consumption patterns, and how they are likely to change as production/supply, prices and incomes change, is required to assess the impact that technological change, infrastructure development and economic policies will have on food security and the distribution of fish. (Dey, 2000a). Against this background, it is necessary to examine the trends and current consumption patterns and their sensitivity to changes in production, prices and income. Such information must be species-specific or at least specific for different types of fish, as fish are not a homogeneous commodity (Westlund, 1995; Smith et al., 1998; Dey, 2000a). However, in most cases such information by species is not available. For this chapter the FAO database on food balance sheets was used to discuss the trends in per caput annual fish consumption and in per caput annual fish food protein intake of consumers in the selected Asian countries. Trends and changes in fish consumption by source (e.g. freshwater, pelagic, demersal, etc.) and their contribution to total fish consumption and the importance of fish food protein to total animal protein intake are also discussed. The remaining section of this chapter discusses current consumption patterns and prices of different types of fish by income class and by location (rural/urban) based on household surveys conducted by ICLARM in recent years. In examining the demand for fish in general and freshwater fish species in particular in these countries, elasticities from previous studies (Dey, 2000a; Chern, 1997; Chern and Wang, 1994; Huang and Bouis, 1996; Estrada and Bantilan, 1991; and Bhatta, 2000) were compiled and presented.
5.1 Trends in fish consumption
5.1.1 Fish food protein
Table 9 shows the trends in consumption of fish and fishery products and the contribution of fish to total animal protein. For over four decades, on average, consumers from the Philippines have had the highest daily per caput supply/intake of fish food protein in particular and animal protein in general. Daily per caput supply of fish food protein during 1997 was 11.1 g and over the last 27 years (1961-1997) it averaged at about 11.33 g. India, where almost 50% of the population are “vegetarian” and are non-fish eaters, has the lowest average daily per caput fish food protein intake (1.5 g in 1997), followed by Bangladesh and Viet Nam.
There is however a decreasing trend in daily per caput fish protein intake among Filipino consumers over the last ten years. After a record high of 13.7 g in 1990, per caput fish protein intake decreased drastically with an average annual rate of -3.36%. During the same period, fish protein intake in Thailand increased drastically from 5.9 g to 10.2 g in 1997 at an average annual rate of 5.21%. With the observed pattern, it is likely that Thailand will surpass Philippines in fish protein intake in the coming years. A drastic increase in fish protein intake is also observed among consumers in China and Viet Nam during the period 1980-97. It was estimated that average annual growth rates in these countries in the same period were 9.91% and 3.42%, respectively. While India and Indonesia have a linear growth rate from 1961-1997, a negative growth rate was observed in Bangladesh after the East Pakistan period (1973) and it was only during 1990-1997 that it began to exhibit a positive average annual growth rate of 4.85%.
5.1.2 Animal protein
The importance of fish in the diet can be estimated by the extent to which it accounts for the animal protein intake (Kent, 1997). As mentioned earlier, average per caput daily animal protein supply in the period 1961-1997 is highest in the Philippines (20.61 g) followed by Thailand (16.37 g) and lowest in Bangladesh (5.28 g). However, the share of animal protein other than fish (e.g. meat, chicken, beef, etc.) to total animal protein intake in China was increasing during this period with an average annual growth rate of 8%. In 1997, China's daily per caput animal protein consumption was 26.2 g compared to 25.8 g in the Philippines. Thailand also exhibits the same pattern during the same period with an average annual growth rate of 3.10%. Also during the same period, animal protein supply in Bangladesh and Viet Nam started to increase from a gradually decreasing trend before 1980. These two countries have average annual growth rates of 1.63% and 3.41%, respectively. Indonesia and the Philippines have almost minimal annual growth rates, averaging 1.52% and 1.28%, respectively.
5.1.3 Share of fish food protein to total animal protein
Fish is indeed an important source of protein in these countries, as it contributes about 15-53% of the total animal protein intake. As far as the trends in share of fish protein to total animal protein are concerned, most of the countries exhibit similar patterns except for China, Indonesia and the Philippines.
Indonesia, whose share of fish protein to total animal protein is highest among the countries considered (53% in 1997 and averaging at about 59% in the period 1961 to 1997), has an almost decreasing trend throughout the period. All countries except China have an increasing share of fish protein during the period 1961 to the early 1970s, then decreasing up to 1990. While the share of fish to total animal protein increased after this (1990-1997) in most of the countries, it continues to decrease in the cases of Indonesia and the Philippines. India, which has consistently the lowest fish protein supply throughout the period 1961-1997 (and the lowest fish consumption) also consistently has the lowest share of fish protein to animal protein.
5.2 Trends in fish consumption/supply
As for fish food protein and animal protein intake, the Philippines used to have the highest annual per caput fish supply (Table 10). Average annual per caput supply during the period 1961-1998 was 31.69 kg, followed by Thailand (21.33 kg) and Viet Nam (11.21 kg). But over the last decade, average per caput annual fish consumption decreased from 37.6 kg in 1991 to 29.6 kg in 1998. Over the same period, per caput fish consumption in Thailand increased from 23.8 kg to 31.1 kg. In 1998, Thailand had the highest per caput annual fish consumption among these countries, while India had the lowest average per caput annual fish consumption (4.6 kg) followed by Bangladesh (10.4 kg), Indonesia (13.8 kg) and Viet Nam (17.1 kg).
In Bangladesh, fluctuations of annual per caput fish consumption are mainly due to fluctuations in per caput freshwater fish consumption. This indicates that freshwater fish are the main factor driving the increase and decrease in total per caput fish consumption of Bangladeshi people, and it accounts for an average of 83% of total per caput fish consumption during the period 1973-1998 and ranges from 80% to 90%. Over a longer time span, looking at the East Pakistan period (1961-1972), it was observed that per caput fish and freshwater fish consumption were increasing during this period. Then they dropped drastically to 8.3 kg and 7.1 kg, respectively in 1975 from 11 kg and 7.1 kg, respectively, in 1974. They continued to decrease steadily down to 7.4 kg and 6 kg in 1990. But with the increase in freshwater fish consumption during the period 1991-1998 (with an average annual growth rate of 4.9%), per caput fish consumption increased as well from 7.4 kg to 10.4 kg with an average annual growth rate of 4.7%. It is worth noting that, while per caput annual consumption of other fish types remained constant throughout the period, per caput annual consumption of crustacean fish increases from 0.2 kg in 1986 to 1.0 kg in 1998 with an average annual growth rate of 14.42%.
In China, the study of Hishamunda of the FAO Fisheries department, cited by the State of World Fisheries and Aquaculture (FAO, 1999b) showed that the high correlation between economic growth and fish consumption in China is an effect of the responsiveness of freshwater aquaculture to the stimulus of the market. In particular, the increase in annual per caput supply of carp and other fish species from aquaculture has meant an increase in annual consumption of aquatic products from 2.67 kg in 1952 to 7.29 in 1992, even though the population grew from 575 million to 1 172 million (Wang, 1996; Williams and Bimbao, 1998). Huang and Qiao (2000) on the other hand, reported that the successive economic reforms in the country in favour of market development, and the technological revolution that started in 1972 have substantially improved productivity and production of aqua products. Per caput fish consumption also increased subsequently. Interestingly, current FAO data shows that mollusc are also becoming an important contributor to per caput fish consumption. Per caput annual mollusc supply increased from 0.5 kg in 1980 to 6.5 kg in 1998 with an average annual growth rate of 15% compared to 4.19% during the period 1961-1998. Though consumption shares of other fish types (demersal, pelagic, other marine, crustaceans and cephalopods) are almost negligible and are decreasing, it was observed that per caput supply of these fish types were also increasing during the same period. Other marine fish in fact exhibit an annual average growth rate of 6.43% compared to 0.92% during the period 1961-1980.
India's per caput fish consumption and composition pattern is somewhat similar to that of Bangladesh. An increase in freshwater fish consumption is the only factor that influences the increase of per caput annual fish consumption from 1961 to 1998. In that period, per caput freshwater fish consumption increased with an average annual growth rate of 2.66%, resulting in an increase in per caput fish consumption from 9.5 kg in 1961 to 17.9 kg in 1998, with an average annual growth rate of 1.95%. As other fish types exhibit almost constant growth rates during the same period, shares of these fish types in total per caput annual fish consumption decreased while the share of freshwater fish increased. The increase in per caput fish consumption in India may be attributed to the increase in aquaculture production due to the intensification of the use of large ponds and reservoirs (also in China) through stocking and feeding of carp polyculture, which has supplied the domestic market especially over the past decade (Prein and Ahmed, 2000).
Indonesia's annual per caput fish consumption started to increase steadily from 9.4 kg in 1974 to 17.9 kg in 1998, with an average annual growth rate of 2.67%. During the same period, consumption of pelagic fish increased, with an average growth rate of 3.37% and surpassed consumption of freshwater fish, which decreased gradually until 1980 with a negative average annual rate of -2.39%. It was only during 1981 that freshwater fish consumption started to increase and exhibit a positive annual average rate of 2.41% up to 1998. It is worth noting that per caput consumption of demersal fish had been increasing gradually in the period 1961 to 1998, while per caput consumption of other marine fish was steadily decreasing.
As in Indonesia, the total annual per caput fish consumption in the Philippines was influenced mainly by pelagic fish, which contributed an average of 41% of the total annual per caput fish consumption during the period 1961-1998. Annual per caput fish consumption was observed to be increasing at an average annual growth rate of 3.53% during the period 1961 to 1975, with an increase in annual per caput consumption of pelagic fish. With the decline and increase in pelagic fish consumption, total fish consumption decreased and increased as well. Total annual per caput fish consumption registered a record high in 1991 (37.6 kg), then it declined drastically with an annual average growth rate of -3.45%. Consumption of all fish types decreased during this period. Per caput annual consumption of freshwater fish, which contributed an average of only 29% during the period 1961-1998, was observed to be increasing during the period 1961-1975, with an average annual growth rate of 3.44%. During the following years however, the average annual growth rate decreased at an average annual growth rate of 3.69%.
In Thailand, the average annual growth rate of per caput fish consumption of 11.25% during the period 1961-1971 can be attributed mainly to the increasing per caput consumption of other marine fish. As annual per caput marine fish consumption declined in the following years, per caput fish consumption decreased and fluctuated at around 20 kg even though per caput pelagic fish consumption was increasing. Annual per caput consumption of freshwater fish was almost constant at around 3 kg until 1988 when it started to increase with an average annual growth rate of 9.36%. During the same period, with the continuing increased in per caput pelagic fish consumption, per caput total fish consumption also increased with an average annual growth rate of 6%. Per caput consumption of demersal fish also increased during the same period with an average growth rate of 11.85%. Compared to other countries, annual per caput total fish consumption in Thailand is well distributed, with pelagic fish accounting for 29% on average of the total per caput fish consumption, followed by freshwater fish (18%), pelagic other marine fish (16%) and crustacean (11%).
In Viet Nam, annual per caput consumption of other marine fish, which contributed an average of 59% of the total per caput fish consumption during the period 1961-1998, is the main factor that influenced total annual per caput fish consumption. As in Thailand, annual per caput fish consumption of freshwater fish is almost constant at around 3 kg, accounting for 27% of the total per caput fish consumption (the average annual growth rate during that period was 0.07%). Freshwater fish consumption started to increase from 2.9 kg in 1992 to 5.7 kg in 1998, with an average annual growth rate of 12%. During the same period, annual per caput consumption of marine fish and total annual per caput fish consumption increased drastically with an average annual growth rate of 5.3% each, compared to a declining average annual rate of -2.97% and -1.43%, respectively, during the period 1961-1991. It is worth noting that annual per caput consumption of crustaceans also increased during the period 1985-1998 with an average annual growth rate of 9.36%. Per caput annual consumption and share of other fish types remained constant and insignificant.
5.3 Prices of different fish species
As stated earlier, demand for fish is influenced mainly by price and consumers' incomes. Price is an essential factor in the consumers' choice of the species that is within their reach. To understand better the demand for fish by species or source (capture or culture, marine or freshwater), one needs to examine the prices of different fish species in each country.
In general, prices of freshwater fish species are lower as compared to fish species from other environments/sources (Table 11). Among the different freshwater fish species in Bangladesh, silver barb seem to be the cheapest, followed by silver carp and assorted small fish. Among the Indian major carps, rohu is the most expensive, while mrigal is the cheapest. Although tilapia is one of the cheapest fish in the Philippines, its price is very much higher compared to other countries. In fact, the market price of tilapia in the Philippines is higher than the price on the international market, making it difficult for the Philippines to compete in the export of tilapia (Dey and Eknath, 1997).
Historically, brackish water fish have higher farm-gate prices than those from freshwater aquaculture. In spite of the increasing trend of production, prices are also generally increasing in most of these countries - particularly in Bangladesh. The price of rohu (a dominant Indian carp species) increased by 3.71% (wholesale) and 1.88% (retail) during the period 1985/86-1995/96, although production increased significantly during the period (Alam, 2000). In the Philippines, from 1988 to 1998, consumer price indices in the urban area (the National Capital Region or NCR), outside of NCR and for the country as a whole, have increased and purchasing power has been reduced by 10.62%, 8.94% and 9.42%, respectively (Olalo, 2000). Farm gate prices of milkfish, prawn and tilapia have registered positive growth rates, while carp species have registered negative. In Thailand, prices of cultured species are decreasing in general as compared to captured species (Piumsobun, 2000). As more and more farmers adopt the GIFT (Genetically Improved Farm Tilapia), prices of tilapia will decrease in these countries (Dey, 2000b).
5.4 Fish consumption preferences
The DEGITA and carp consumers' field surveys (ICLARM, 1998 and 2000) showed that annual per caput fish consumption rates are 19.92 kg in Bangladesh, 31.08 kg in China, 15 kg in India, 15.81 kg in Indonesia, 46.0 kg in the Philippines, 28.80 kg in Thailand and 12.86 kg and 37.80 kg in northern and southern Viet Nam, respectively (Table 12). FAO data however showed that the average national annual per caput fish consumption rate during 1998 was highest in Thailand (31.10 kg), followed by the Philippines (29.0 kg), China (25.70 kg), Indonesia (17.9 kg), Viet Nam (17.1 kg), Bangladesh (10.40 kg), and India (4.6 kg) in 1998 (FAO, 2000b). Except for Indonesia and Thailand, figures from the field survey are considerably higher than what is reported in the national average. Lower figures for the country average may be mainly due to the fact that sampled areas are located in fish-producing areas, where per caput fish consumption is higher than in non-fish-producing areas.
The carp consumer field survey (ICLARM, 2000) showed that silver carp is in top position of the fish consumption basket in Bangladesh. Basically, Indian major carps such as rohu, catla, and mrigal are the preferred species in Bangladesh by all yardsticks. It may be that field demonstrations of different projects and NGOs over the past couple of years have convinced farmers that silver carp grows faster. The silver carp therefore, dominates the stocking in these areas. People consume the species as it is available and at an affordable price. Chinese consumers have indicated that grass carp, silver carp, crucian carp and common carp are the favoured species. In India, rohu and catla are the common and preferred species. In the Philippines, tilapia is the preferred freshwater species. Among marine/brackish species, milkfish is preferred. In general, marine fish species are preferred among consumers in the Philippines. Consumers in landlocked areas of Thailand on the other hand, are very fond of tilapia, silver barb, snakehead and catfish. In these areas, marine fish constitutes only about 8% of the total fish consumed. In northern Viet Nam, rohu, grass carp, silver carp, silver carp, tilapia and common carp are the preferred fish species. On the other hand, marine fish are preferred in southern Viet Nam. Among freshwater fish species, snakehead is preferred in southern Viet Nam.
To see the importance of freshwater fish species in the consumer's budget, Table 13 shows the proportion of each species in total fish expenditure. In Bangladesh, assorted small fish, rohu river shad and catla, which are all freshwater fish, dominate fish expenditure among Bangladeshi consumers regardless of income class. As expected, on average, lower income group consumers buy cheaper species such as silver carp, silver barb, tilapia and river shad, while higher income groups consume relatively expensive species like catla, exotic carp and live species. In China, crucian and grass carp are the two most important species bought by all income groups, but higher income groups spend more on other/marine fish (others) than lower income groups. This is also true in India, the Philippines and Thailand. Indian carps (rohu and catla) are the freshwater species preferred by consumers in India. In Thailand, snakehead, tilapia, silver barb and marine fish are the species on which most consumers spend their money. Marine fish accounts for 15% on average of total fish expenditure. Lower income groups spend more on tilapia and silver barb on average compared to higher income groups. In northern Viet Nam, rohu, tilapia and common carp are the preferred species among the upper and medium income groups, while rohu and silver carp are preferred by the lower income group. Unlike in Thailand, in northern Viet Nam the higher income group spends more on average on tilapia than the lower income groups. In southern Viet Nam, consumers spend more on snakehead. On average, marine fish accounts for 11% of total fish expenditure.
Lower income groups tend to spend more on food items (60-80% of total expenditure) compared to higher income groups (35-50%) (Table 14). However, the contribution of fish expenditure to total expenditure (food and non-food) is higher among lower income groups. This is also true for the contribution of fish expenditure to total food expenditure, except in China and Viet Nam. Although per caput fish consumption is higher among higher income groups, contribution of fish expenditure to total animal protein expenditure is higher among lower income groups. This shows that fish is an important source of protein among relatively poorer households in these countries. No wonder it is regarded as “poor man's protein” (Williams, 1996).
The consumers' survey also indicates that annual per caput fish consumption in rural areas is substantially higher than in urban areas (Table 15). Producer-consumers have the highest fish consumption, followed by rural-consumer and urban-consumers.
5.5 Consumer preferences for various freshwater fish
Price and income influence consumer demand and preferences. However, price is just one of many factors that determine preference. There are more fish-specific characteristics or traits that consumers consider when choosing fish. In the light of government efforts and the support of different funding and research institutions to come up with genetically improved species, it is essential to examine the preference patterns of consumers for different fish and fish traits.
Table 16 presents consumers' preferences for freshwater species based on the results of the carp consumer survey. The table shows that the species preferred by consumers in Bangladesh and India is rohu, followed by catla and mrigal. Common carp is preferred by consumers in Indonesia and southern Viet Nam, followed in the latter by snakehead and silver carp. Consumers in northern Viet Nam rank grass carp as the preferred species, followed by mud carp and common carp. Chinese consumers choose crucian carp first, followed by grass carp and common carp. In Thailand, the preferred freshwater fish is tilapia, followed by snakehead and catfish, while silver barb ranked fifth. The reasons for consumer preferences for these species are mainly good taste, reasonable price and easy availability (ICLARM, 2000). Good physical attributes are the reason for high preference of silver barb among Bangladeshi consumers.
The sample respondents were asked to rank the traits they prefer for individual species. The results of the exercise are presented in Table 17. Interestingly, the table shows that trait preference varies considerably among the three most preferred species within a country, except for rohu and catla in Bangladesh, where colour and higher dress-out percentage rank first and second, respectively for these two Indian major carp. Also, for the same species, preferred traits vary considerably among countries except for mrigal, for which higher dress-out percentage is the preferred trait among Bangladeshi and Indian consumers. Another exception is common carp, where its better flavour is preferred by Chinese and northern Vietnamese consumers, and also grass carp where its bigger size is preferred in both countries. Unlike Bangladeshi consumers, Indian consumers consider the body shape and flavour of rohu and bigger size and higher fat of catla as more important than their colour and higher dress-out percentage. As in India, body shape, size and flavour are among the traits that are preferred by Chinese consumers for their preferred species. Crucian carp is preferred because of its body shape and grass carp because of its bigger size.
Higher dress-out, bigger size, better flavour and body shape are the traits considered important by Thai consumers for silver barb. Vietnamese consumers from the south on the other hand, preferred the higher fat content of the same species. Higher fat is also the most important trait for common carp for the Vietnamese from the south. While the same consumers considered the colour of silver carp as most important, consumers from the north named its higher fat content.
Table 18 shows consumers' preferences for size, shape, colour and other parts of the fish. All carp species except silver barb are preferred in bigger size, up to 3 pcs/kg-1. Surprisingly, consumers in Bangladesh seemed to prefer bigger size (<1 to 1 pc per kg) compared to other countries. Southern Viet Nam and Thailand on the other hand prefer smaller fish (<2 to 3 pcs per kg).
Shape preference for carp species such as rohu, mrigal and grass carp varies across countries. Consumers in Thailand and India prefer long and thin shape for rohu and mrigal, but those in northern and southern Viet Nam prefer short and thick shape. For grass carp, consumers in China and India prefer long and thin shape, whereas those in northern and southern Viet Nam like short and thick. For common carp, silver carp, silver barb catla and crucian carp, consumers in all countries preferred short and thick and short and deep.
Another interesting finding is that colour preference for the same species also varies across countries. Rohu is preferred in silver (Thailand), light (northern Viet Nam) and bright (southern Viet Nam). Common carp is preferred mostly in yellow, with additional reddish colour for Chinese consumers. Mrigal is preferred in silver (Thailand), black-blue (northern Viet Nam) and bright (southern Viet Nam). Silver carp is mostly preferred in silver (China and northern Viet Nam) and bright colour (southern Viet Nam). For silver barb, silver (Thailand), light-blue (northern Viet Nam) and yellow fin colour (southern Viet Nam) are preferred. Grass carp is preferred in black/green/silver in China, whereas light and bright colours are preferred in northern and southern Viet Nam, respectively.
Like colour preference, consumers' preferences for different parts of fish vary significantly across countries. Table 18 shows the three most preferred body parts of the preferred species in each country. Consumers in Bangladesh mostly prefer the belly, tail and back portion of rohu and mrigal, while the head portion is most preferred for catla. Indian consumers on the other hand, prefer the back portion to the belly of rohu, while the egg is preferred to the tail of mrigal. In Thailand, the back portion is preferred for all species. Interestingly, consumers in northern and southern Viet Nam do not only differ in species preference, but also in preference for body parts of the same species. This suggests that the species and traits for genetic improvement should be different in these two areas.
5.6 Expenditure and demand elasticities for fish
Estimates on elasticity of demand for fish in Asia are very scanty due to scarcity of data. To date, the most promising work on the estimation of fish demand at an aggregate level in these countries is the work of Dey (2000a) on Bangladesh. Table 19 presents available information on demand elasticities for fish in these countries. Alam (2000) estimated expenditure and income elasticities for demand for fish in Bangladesh to be 0.79 and 0.65, respectively. These indicate that demand of fish increases with the increase of expenditure and income, respectively. But the increase in demand for fish is more sensitive to the increase in expenditure than in income. The estimates of Dey (2000a) on fish expenditure with respect to food expenditure vary from 2.67 for the poorest quartile group to 0.89 for the richest quartile. Dey's estimates for fish expenditure elasticity with respect to income vary from 1.52 for the poorest quartile to 0.62 for the richest quartile. These results show that expenditure and income elasticity for demand for fish fall with an increase in expenditure and/or income, suggesting that fish is a luxury commodity for the poor and a necessity for the rich. Compensated own-price elasticities for various types of fish (aggregate level) among quartile vary (absolute value) from -0.42 (poorest) to -0.59 (richest) for assorted small fish to -2.02 (richest) to -2.87 (poorest). This indicates that prices of carps are more elastic than any other fish types, while prices of assorted small fish are inelastic. It is worth noting that for carps, low-income groups are more sensitive to price changes than high-income groups, while the reverse is true for assorted small fish.
Alam (2000) estimated the own-price elasticities of demand for fish by species in Bangladesh. Provisional estimates are: -1.13 for rohu carp, -0.75 for catla carp, -0.91 each for mrigal and silver carp -1.07 for other exotic carp, -1.09 for silver barb, -0.91 for river shad. This is -1.10 for assorted species, -0.98 for live species and -0.93 for high valued species.
China has various estimates for elasticities of demand. Expenditure elasticities of demand for aquatic products in general are 1.45 for the country as a whole and 1.39 and 1.48 for rural and urban areas, respectively. Expenditure elasticity estimates derived from quadratic expenditure system range between 1.86 to 2.85. Ye (1996) estimates using a panel data (1978-91) showed that income elasticity of aquatic product is 1.93. In any case, one can conclude that fish consumption in China, just like in other countries is sensitive to income changes. Ye (1996) also concluded that the responsiveness of demand to income is stronger than price. The own-price elasticities of demand for fish were -1.48 (estimated from LA/AIDS model) and -1.78 to -2.37 (derived using quadratic expenditure system) which show that demand for fish in China is very elastic and sensitive to price changes.
In India, to the best of our knowledge there is no study that estimates elasticity of demand for a particular species of fish, or for fish as a separate group. The only available information is an aggregate figure comprising fish, eggs and meat. Different expenditure elasticity estimates range from a minimum of 0.63 to a maximum of 1.04 (Bahalla and Hazel, 1998). Estimates also show that the urban expenditure elasticities of demand for fish, meat and eggs are higher as compared to rural areas in India. Meenakshi and Ray (1999) estimated the meat-egg-fish elasticities to be -1.965 for rural areas and -0.913 for urban areas in India.
In the Philippines, own-price elasticities of demand are -0.65 for tilapia, -0.63 for milkfish, -1.50 for tuna, -0.41 for round scad and -1.52 for prawn (Olalo, 2000). Dey's (2000) estimates are -1.00 for tilapia, -1.2 for carps, -1.50 for crustaceans, -1.5 for other high value fish and -0.75 for other low valued species. Income elasticities of demand for fish in the Philippines tend to go up as income goes down, and this holds for both rural and urban areas.
Own price elasticity for freshwater cultured species in Thailand is -1.2, cross-price elasticity with respect to chicken price is 1.3 (Piumsonbun, 2000). These results show that freshwater fish and chicken are complementary commodities in Thailand. The income elasticity of demand for fish is 0.8. Magnitudes of price and cross-price elasticities of demand are different for different species. Both the own-price elasticity of demand for Nile tilapia and cross-price elasticity of demand for Nile tilapia with respect to striped catfish are -0.70. This means that striped catfish is a substitute for tilapia. Income elasticity of demand for Nile tilapia is however, very high (4.1). For silver barb the elasticities are -0.7 (own-price), 0.3 (cross-price with respect to striped catfish) and 2.2 (income elasticity). For walking catfish, the magnitudes are -0.9 (own-price), 1.1 (cross-price with respect to chicken price) and 2.5 (income elasticity). Price elasticity of demand for striped snakehead is 0.9, while the cross-price elasticity of the same with respect to beef, pork and chicken prices are 4.6, 2.4 and 1.1 respectively. Dey (2000a) estimated elasticities of demand to be -1.0 for tilapia, -1.10 for carps, -1.50 for crustaceans, -1.50 for other high valued species, and -0.50 for other low valued species. Income elasticity of demand for snakehead fish is 2.2.
5.7 Discussions on fish consumption and demand
FAO data show that fish has become an increasingly important source of protein over the last decade in most of these countries. The exceptions are Indonesia and the Philippines, where the supply of fish food protein is being replaced by other sources of animal protein, resulting in a decrease in the share of supply/consumption of fish food protein to total supply/consumption of animal protein. In the Philippines, even though the supply of fish food protein has been decreasing over the past decade, total supply in animal protein is still increasing. This implies that consumers in the Philippines do not depend on fish as a source of protein. In the case of Indonesia, both supply of fish food protein and total supply of animal protein is increasing, but the rate of increase in total supply of animal protein is higher than the increase in supply of fish food protein, resulting in the decrease of the share of supply of fish food protein to total supply of animal protein.
In Bangladesh, the per caput supply of fish food protein is the factor driving increases and decreases in the per caput supply of total animal protein. This indicates that fish has not been replaced by any other forms of animal protein (Prein and Ahmed, 2000; Kent, 1997) and that Bangladeshi people are very dependent on fish as a source of animal protein. In Viet Nam, it was not until the last decade that fish food protein became an important source of animal protein.
Results show that demand for fish increases as expenditure/income increases and that higher income groups tend to consume more fish than lower income groups. However, the share of fish (as protein) and share of fish to total food expenditure are higher among lower income groups, suggesting that lower income groups are the most dependent on fish. This result is also consistent with the generalization that although less developed countries are not the biggest consumers of fish, they are the most dependent on it (FAO, 1993; UNDP, 1993; Kent, 1997; FAO, 1999b).
With so many poor people highly dependent on fish, it is a matter of serious concern when their caput supply decreases over time. Though production is increasing, the continuing growth of population and the growth in disposable incomes means that production is not likely to keep up with demand (FAO, 1993a; Williams, 1996). So long as demand outruns supply, prices will go up and this affects the lower income groups most. Kent (1998) reported that this is already evident in nations such as India and the Philippines, where middle class people feel they can no longer afford to eat fish as part of their regular diets.
There is also cause for concern about the impacts of international trade on fish supplies for consumption in these countries. Fisheries trade can lead to declining food security, especially in those countries that eat more marine (pelagic and demersal) fish, such as Indonesia, the Philippines, Thailand and Viet Nam, since a large share of the exportable fishery products are coming from these sources. As international trade in fishery products grows rapidly, species once commonly eaten throughout the country are now exported.
In Bangladesh where freshwater fish accounts for an average of 83% of the total fish protein, international trade in fishery products does little harm to fish food supplies as reported by Kurien (1993). This is due to the fact that most of the freshwater fish species are not exportable and are only used for domestic consumption. Dey and Bimbao (1998) reported that shrimp accounted for about 91% of the total fishery exports in Bangladesh, which comprised about 69% of the total shrimp production in the country.
With its strongly increasing production, relatively lower price and relatively limited international market, freshwater fish is expected to become an increasingly important as a type of fish and as a source of animal protein, particularly for those in medium and lower income group of these countries.
