4.1 Changing consumption patterns
Economic reform in China succeeded in the sense that both the nation's GDP and citizens' real income have substantially increased. Higher income has spurred greater demand for aquatic products. Both, price and per caput consumption of aquatic products have risen over the last two decades.
Table 12 presents per caput aquatic consumption over time for urban and rural areas. Figures shown in this table do not include consumption away from home. A food market emerged in China, for the most part, after the early 1980s. We presume that the consumption levels in the early 1980s presented in Table 12, are more or less consistent with actual consumption levels during that period. Results from our recent consumption survey in both rural and urban areas (consumption away from home was about 15-20% in 1998) indicate that per caput consumption of aquatic products might have reached more than 6 kg in 1997, more than a 50% increase from 1981 (2.49 kg) for the nation as a whole.
Urban residents consume considerably more aquatic products than do consumers in rural areas. In 1978, urban residents consumed an average of 7.26 kg of aquatic products, 5.7 times more than rural consumers. The ratio of urban consumption of aquatic products to rural consumption declined significantly over time due to the rapid development of food markets in rural regions (Huang and Rozelle, 1999). By the 1990s, this ratio declined from 2.5 to 3.6 (Table 12).
Several factors minimized the gap that existed in consumption patterns between rural and urban areas: urban residents' income remained about 2-2.5 times higher than rural residents' during the entire period of the reform; relatively cheap prices for aquatic products are another factor often cited in the literature. However, Huang and Rozelle (1999) showed that a wider choice of available foods, such as more aquatic products in urban markets, is the other important factor underlying the lower consumption of aquatic products in rural areas. This implies that, as food markets continue to develop in rural areas, consumption is expected to rise, even if income and prices are held constant.
The large difference in rural and urban consumption patterns for aquatic products also implies that there will be structural shifts for aquatic products, and that other patterns of food demand will emerge, as people move from rural to urban areas. Several of our studies show that shifting consumption patterns, induced by urbanization, can explain the observed changes in food consumption, including changes in the consumption of aquatic products, that have been taking place over time (Huang et al., 1991; Huang and David, 1993; Huang and Bouis, 1996).
We can see the effect income has on consumption, when we compare the consumption of aquatic products across income groups. Table 13 shows that consumption of aquatic products increased with higher income for urban consumers. Consumers in the top 10% income group consumed nearly twice as many aquatic products than consumers in the lowest 10% income group. Contribution of aquatic products to nutrient intake is still very low at the current level of consumption. However, it is expected to rise, as income grows. In 1995, aquatic products contributed only 2-3% to total protein intake in China (Table 14).
Few studies have estimated demand elasticities for aquacultural products in China. Three recent studies by Huang and Bouis (1996), and Chern and Wang (1994) and Chern (1997) give empirical estimations of demand, including demand for products from aquaculture. Applying the LA/AIDS model and using household food-consumption data provided by the State Statistical Bureau, Huang and Bouis found that expenditure elasticities of demand for aquatic products are 1.45 for the nation as a whole, 1.39 for rural areas, and 1.48 for urban areas. The authors' own price elasticity estimated in the study is 1.48. Applying the quadratic expenditure system (QES), Chern (1994) estimated expenditure elasticities ranging from 1.86 to 2.85 for aquatic product demand; our own price elasticities range from -1.78 to -2.37. A more recent study by Chern further indicates that expenditure elasticity of demand for aquatic products is high, but varies significantly with the models used. In the LES model, he estimated an expenditure elasticity of 2.26 for aquatic product demand, while expenditure elasticity is only 0.37 in the LA/AIDS model.
It should be noted that the above studies only estimate expenditure elasticities. As saving rates in China have been rising over time due to growing income, we expect income elasticities to be much lower than those presented here. Based on income and expenditure relationships, our best guess for income elasticity of demand for aquatic products is about 0.9 to 1.1 for the current level of fish consumption. These modified figures are indeed closely comparable to those found in the rest of Asian countries. They are also more consistent with the consumption figures of aquatic products by income groups presented in Table 13.
We assume an income elasticity of 1.0 for aquatic product demand. However, considering the rapid growth of income in China over the past 2 decades, figures presented in Table 12 could easily be under-reporting levels of consumption of aquatic products. Therefore, it is important to collect more consistent time-series data on consumption of aquatic products, not only for estimations of demand, but also for the crucial development of a reliable database for aquatic product demand in the future.
4.2 Structural changes
Since trade in the fishery sector is still small (see the next section), and the market is very competitive, production figures for various species closely resemble consumption figures. Species production data for the period 1990-1999 (see Appendix 7) indicate that there was a boom in the mid 1990s when growth rates peaked, which was shared by most of the aquatic species surveyed. Comparison across the board reveals that common species, such as black carp, grass carp, silver carp, common carp and crucian carp, still dominate the market and have been maintaining a decent growth rate (10%-30% on average) over the years. High-value small species, such as freshwater crab and soft-shelled turtle, however, enjoyed the highest growth rate (10%-80%, Table 15). This trend is also shown in the change of per caput production (Table 16). Common species still dominate per caput production of aquatic products, but shares of high-value species (e.g. crab and turtle) have increased dramatically. These data further support the income-effect hypothesis we discussed earlier. High-value species enjoy higher increases of consumption as a result of higher per caput income. The same phenomenon can be observed in marine aquaculture. High-value species, such as abalone, sea cucumber and marine pearl, have been showing impressive, yet fluctuating, growth rates. Their per caput production and consumption have also grown significantly.
4.3 Trade balance
China is a net exporter of aquatic products. The annual balance for aquatic products reached more than US$ 1 billion in the 1990s and has been growing over time (Table 17). In 1997, total exports of aquatic products reached 0.72 million t, but accounted for only 2% of total production (if corrected for over-estimation in official production statistics, exports might have reached 4% of total production). Unlike the many restrictions China has imposed on agricultural products, the fishery industry remains largely unregulated, facing very low tariffs and no quotas on imports. Comparative advantages and market characteristics are decisive factors defining the status quo.
Frozen fish and fish slices are major trade products accounting for more than half of both imports and exports (Appendix 7). Freezing of fishery products protects them from rotting during long distance shipment.
Species that have seen their shares in imports increase are hairtail, shelled prawn and salmon. Appendix 7 shows details of quantity and value of major species in China's fishery imports and exports in 1997.
