Pesticide use and fish harvests in Vietnamese rice agroecosystems.

While the Green Revolution dramatically increased rice yields across Southeast Asia, its implementation has had unanticipated costs in terms of agroecosystem stability and farmer health. Reassessments of the Green Revolution's impact have focused attention on the intensive use of agricultural chemicals, which were promoted as an essential accompaniment to high-yielding variety adoption (Pimentel and Pimentel 1990). Pesticides in particular have come under fire for their role in exacerbating pest problems, disrupting arthropod food webs (Way and Heong 1994; Schoenly, Cohen, and Heong 1996), and reducing farmer productivity due to illness (Rola and Pingali 1993; Antle and Pingali 1994; Huang et al. 2005).

Wild aquatic animals in flooded rice agroecosystems are another potential casualty of overzealous pesticide application. Fish, crustaceans, and other animals inhabit flooded rice paddies and are a common joint output of rice production in Southeast Asia. Many households harvest these fish as an additional source of food and income. They also provide natural pest control by feeding on insects and weeds.

Ecologists have documented the detrimental effects of pesticides on aquatic animals. Field experiments have shown that fish species commonly found in Southeast Asian rice fields exhibit high mortality from insecticide exposure and moderate-to-low mortality from herbicide and fungicide exposure (Cagauan 1990; Abdullah et al. 1997). These chemicals also indirectly harm fish populations by upsetting aquatic food webs (Bottrell and Weil 1995).

This article examines the relationship between fish harvesting and pesticide use in rice paddies from an economic perspective, accounting for farmers' choices in a multi-output system. The use of pesticides that are toxic to fish has both direct and indirect negative effects: farmers lose a potentially valuable food product for consumption or sale and an environmentally benign natural enemy of rice pests. Using survey data from rice farmers in the Mekong Delta region of Vietnam, we estimate the impact of spraying pesticides on harvesting fish, using instrumental variables to control for the potential endogeneity of these practices. Our approach allows us to examine whether farmers make their pesticide use and fish harvesting decisions simultaneously, taking into account pesticide toxicity to fish, or recursively, not accounting for potential damage to fish. In the latter case, they may be using pesticides inefficiently, failing to exploit services provided by stocks of fish and other aquatic organisms in the rice agroecosystem. We find that pesticide use has adverse effects on fish harvests but that these effects are not economically significant, suggesting that it is likely rational for farmers to ignore them.

Joint Rice-Fish Production in Vietnam

Semi-commercial rice production by smallholder peasant farmers is one of the main economic activities in the Mekong Delta region. Rice is the primary source of food and income for farmers, with many households both consuming and selling portions of their rice crop. Farmers typically grow two to three rice crops per year on small landholdings made up of multiple plots and vary their management practices by plot and by season. Increased rice yields have followed the adoption of the Green Revolution technology package throughout the region. Many farmers plant both short-duration modern varieties and long-duration traditional varieties, irrigate from nearby rivers and streams to flood their fields, and use some chemical fertilizers and pesticides. Modern varieties are typically sown by broadcasting rice seed in the fields, while traditional varieties are transplanted into fields as seedlings after germination. Seedlings must be transplanted within a short period of time, so women typically exchange labor, working in groups of six or more on each household's field until all are completed. Household labor is used for most other tasks, but some farms hire laborers or rent mechanized equipment or draft animals (Nguyen 2000).

A traditional activity among Vietnamese farmers is to produce rice jointly with aquatic animal foods--primarily fish, but also some shellfish, frogs, snakes, and even rodents. These fish and other animals are used for home food consumption, supplementing the staple food of rice. While all households maintain separate fish ponds, some also gather fish from flooded rice fields. Fish and other animals enter rice fields when farmers irrigate by channeling or pumping water from nearby streams. The flooded condition of the rice fields is both necessary for aquatic animal survival and advantageous for rice growth.

Wild fish species that enter paddies include snakehead (Channa striata), climbing perch (Anabas testudineus), and snakeskin gourami (Trichogaster pectoralis). (1) Most farmers rely on natural food sources present in the rice agroecosystem to sustain the fish population. Fish harvesting can occur during the course of other rice farming activities throughout the season but primarily takes place at the end of the season when the paddy is drained. Household members working on the farm may devote effort to gathering aquatic animals during the growing season if the value of the harvest to the household exceeds the opportunity cost of their labor.

Wild fish may provide an additional benefit while living in the rice agroecosystem, serving as a natural pest control by feeding on potential insect pests and weeds. Introducing fish into flooded fields is used as a deliberate integrated pest management (IPM) strategy in some rice production systems, where insectivorous fish may prey on the larvae of rice pests and defoliators (Lichtenberg 1987). The introduction of common carp, Nile tilapia, and silver barb into the Mekong Delta rice fields reduces the presence of aquatic weeds by 92-100% (Rothuis et al. 1999). This fish polyculture also significantly reduces the presence of rice caseworm, a rice plant defoliator (Vromant et al. 1998). The wild fish species mentioned above that are common to the Mekong Delta are primarily carnivorous or omnivorous, feeding on insects, weeds, and other small organisms living in rice paddies. Promoting fish growth in rice fields may therefore be a valuable part of an IPM strategy.

Other complementarities between rice and fish growth may also make joint production attractive to farmers. Fish may increase nutrient availability to rice by stirring up soil or providing excrement (Rothuis et al. 1998). A lower rice seeding density (100 kg/ha as opposed to 200 or 300 kg/ha) positively affects both rice and fish yields due to less mutual shading and increased food and oxygen availability (Rothuis et al. 1999). Increased fertilizer applications benefit rice and fish alike (Rothuis 1998).

While the potential for economies of scope in joint rice and fish production is well established conceptually, the extent to which they exist in practice has not been demonstrated definitively. Comparisons of rice yields between joint rice-fish production systems and rice monocropping show mixed results. Some studies show a significant increase in average rice yields in the rice-fish culture (Lightfoot et al. 1992, as cited in Vromant, Duong, and Ollivier 2002), while others find that overall rice yield is typically lower in rice-fish production than in rice monoculture when nonplanted areas such as trenches used in fish cultivation are taken into account (Berg 2002; Vromant, Duong, and Ollivier 2002). Higher water levels--necessary for fish habitat--may reduce weeds but lead to decreased rice yields.

Previous budget studies in Vietnam find that rice-fish culture can generate comparable or higher profits than rice monoculture. Berg (2002) compares input levels, costs, and net benefits between farmers producing rice alone and rice-fish jointly, using either IPM or conventional pest management in the Mekong Delta. He finds that joint rice-fish production, when combined with IPM practices, is more profitable for farmers than rice farming alone. Rothuis et al. (1998) find no significant difference in net returns between rice-fish production and rice monoculture. One benefit of producing rice and fish jointly noted by Rothuis et al. (1998) is farmers' decreased use of pesticides as compared to rice production only. Thus, rice-fish farming may be a more sustainable and diversified production system than rice alone even if rice yields or profits are not necessarily higher. However, these studies do not control for differences in soil quality, the opportunity cost of labor, and other factors that influence both rice productivity and fish harvests, nor do they examine whether farmers who harvest fish from rice fields deliberately decrease their pesticide use to protect fish stocks.