METHODS FOR ASSESSING ECONOMIC, ENVIRONMENTAL AND SOCIAL IMPACTS OF AQUACULTURE TECHNOLOGIES: ADOPTION OF INTEGRATED AGRICULTURE-AQUACULTURE IN MALAWI John Antle
Dept. of Agricultural and Resource Economics
Oregon State University, Corvallis OR 97331 USA
firstname.lastname@example.org Roberto Valdivia
Dept. of Agricultural Economics and Economics
Montana State University, Bozeman MT 59716 USA February 1, 2011 Abstract
There is a growing demand for assessment of economic, environmental and social impacts of new food-related technologies, including the impacts of new methods for aquaculture management. This paper presents a new “minimum-data Tradeoff Analysis” (TOA-MD) model that can be applied to assess economic, environmental and social impacts in a wide array of agricultural systems that incorporate aquaculture, crops, and livestock (Antle 2011; Antle and Valdivia 2010). This model is widely applicable to assess impacts because it utilizes a generic model structure that can be parameterized with data available from a variety of sources, including farm surveys, experimental data, simulated data from bio-physical simulation models, and expert judgment. A key feature of this model is that it takes into account the fact that farmers systematically selected themselves into adopting and non-adopting groups. Analysis shows that this selection must be taken into account to obtain accurate estimates of impact.
To illustrate the use of the TOA-MD model, we use it to implement an impact assessment of integrated agriculture-aquaculture (IAA) systems in southern Malawi developed by the World Fish Center, using a WorldFish farm survey data collected in 2004, together with data from other public sources. We use the TOA-MD model to demonstrate how it is possible to use available data to move a conventional economic impact assessment “along the impact assessment pathway” to estimate adoption rates in the relevant populations, and to quantify impacts on distributional outcomes such as poverty, environmental impacts such as soil and water quality, and social and health-related outcomes such as nutrition or gender impacts. The analysis predicts an adoption rate of about 44%. In two districts, there is a substantial increase in protein consumption associated with the adoption of IAA and substantial reductions in poverty, whereas in others the effects are smaller.
1. Introduction One of the great challenges in impact assessment is to “move assessment along the impact pathway” to quantify distributional, environmental impacts, and social impacts of agricultural technologies being developed and disseminated. As noted in a recent report sponsored by the Standing Panel on Impact Assessment of the CGIAR (Walker et al. 2008), a major impediment to meeting the growing demand for broader impact assessments is their cost in time and other resources, particularly when donors expect impact assessments to be carried out as part of a technology-related project. As Walker et al. (2008) observe, “In terms of both budgetary support and human capital, a disaggregated multi-dimensional impact study can be quite demanding and costly….The supply of these studies is more likely to be constrained by lack of funding than the other types…. (p. 7). Nevertheless, Walker et al. (2008) conclude, “The desirability of moving … along the impact pathway is unquestioned. As donors want to see ever more comprehensive impact assessments, so ways have to be found to accommodate their wishes… even when resources for carrying out these … studies are not forthcoming.” (p. 14).
A recent development in impact assessment methodology is the use of a “parsimonious” approach that moves the focus from site-specific, processes based models and data, to the use of simulation models parameterized with population data (Antle 2011). This approach has been implemented in the form of a generic “minimum-data Tradeoff Analysis” (TOA-MD) model that can be applied to assess impacts in a wide array of agricultural systems that incorporate crops, livestock and aquaculture (Antle and Valdivia 2010). The TOA-MD model is a unique simulation tool that uses a statistical description of a heterogeneous farm population to simulate the proportion of farms that utilizes a baseline system (in this case, farms not using integrated agriculture-aquaculture, or IAA) and the proportion of farms that would adopt an alternative system (in this case, farms using IAA) within defined strata of the population. We apply this model using those data from the World Fish impact assessment of IAA in Malawi made public by Dey et al (2010), together with data from other public sources such as the national agricultural census. We use the TOA-MD model to demonstrate how it is possible to move a conventional economic impact assessment “along the impact assessment pathway” to estimate adoption rates in the relevant populations, and to quantify impacts on distributional outcomes such as poverty, environmental impacts such as soil and water quality, and social and health-related outcomes such as nutrition or gender impacts.
The adoption and impacts of World Fish Center research on integrated agriculture-aquaculture (IAA) systems in Malawi has been studied and publicized by Dey et al. (2006, 2010), Russell et al. (2008), Government of Malawi, (2005), NSO-GoM (2010), FAO (2008) and related sources. The impact assessment that was carried out by World Fish Center analyzed factors influencing adoption, and using an assumed adoption rate, evaluated aggregate economic impacts and estimated a rate of return on investment. While the study collected some data on outcomes such as nutrition, this was done for a stratified random sample of a small number of farms not using and using aquaculture. Thus, while it was possible to conclude that if farms adopted IAA they would be better off in terms of income and nutrition, it was not possible to estimate an overall adoption rate or make statements about the overall impacts of IAA in the relevant population of farms that potentially could adopt the IAA technology. Moreover, the assessment also did not provide estimates of environmental impacts (CGIAR Science Council 2007). In this report we show how the TOA-MD approach can utilize data that was collected in the original impact assessment surveys, along with other publicly available data, to carry out a disaggregated, multi-dimensional impact assessment. The TOA-MD approach also provides the basis for carrying out sensitivity analysis to parameters that cannot be estimated with the available data, thus providing guidance about the types of data that should be collected in future impact assessments.
2. Impact Assessment using TOA-MD The TOA-MD model is a unique simulation tool that uses a statistical description of a heterogeneous farm population to simulate the proportion of farms that utilizes a baseline system (in this case, farms with small ponds and low integration) and the proportion of farms that would adopt an alternative system (in this case, farms with larger ponds, higher integration, and vegetable production) within defined strata of the population. Based on the predicted adoption rate of the alternative system, the TOA-MD model simulates associated economic, environmental and social impacts on adopters, non-adopters and the entire population. The version of the model used for the analysis presented here is a new version developed for impact assessment (Antle 2011). It is based on the earlier version developed for ecosystem service analysis (Antle and Valdivia 2006, 2010).
One unique feature of the TOA-MD model is its capability to exploit statistical relationships between technology adoption and the environmental, economic and social outcomes associated with adoption. Economic research shows that taking these inter-relationships between adoption and outcomes is critical to obtain accurate estimates of impact. This fact also has important implications for data collection that we discuss in the conclusions of this study.
Another unique feature of the TOA-MD model is its parsimonious, generic structure, which means that it can be used to simulate virtually any farm system. One virtue of this model design is that, unlike many large, complex simulation models, it is easy to relate results to particular features of the system. TOA-MD is also well-suited to address the uncertainty in impact assessments, by using sensitivity analysis to explore how results change with different assumptions. The TOA-MD model is programmed in Excel, and is easy to learn and use.
The model utilizes the following types of data:
population means and variances of production, output price and cost of production, by crop, aquaculture and livestock activity
population means and variances of environmental and social outcomes associated with each system
correlations between system returns and environmental and social outcomes
population means and variances of farm household characteristics (farm size, pond size, household size, off-farm income).
Population and Strata The population represented is farms in southern Malawi that could adopt aquaculture, or that have aquaculture operating at a low level of integration that could be improved. The strata are 5 districts where survey data were collected : Zomba East and West (pop. 670500), Mulanje (pop. 428322), Mwanza (pop. 138000), Thyolo (pop. 458000), and Mangochi (pop. 610000), see Figure 1.
Brooks (1992) estimated the potential areas for aquaculture in Malawi based on some physiographic factors (land formations, altitude, temperature, precipitation, run-off and soils). Brooks estimated that the areas under or that have potential for aquaculture in Malawi was about 11,650 km2 and about 7,200km2 corresponded to the southern regions of Malawi. More recently, the project “Determination of High-Potential Aquaculture Development Areas and Impact in Africa and Asia (funded by the Federal ministry of Economic Cooperation and Development of Germany and in coordination with the WorldFish and other partner institutions) developed a decision-support package that can be used to identify areas where aquaculture is feasible. One of the studies of this project was carried out in southern Malawi where they identified areas with existing and potential aquaculture adoption. They used biophysical (e.g. water availability, land conditions) and socio-economic (market, knowledge and inputs, labor and finance) criteria to evaluate the suitability of the area. The overall area suitable for Southern Malawi estimated was about 35,400km2 (Kam et al., 2008; Kam and Teoh, 2008).
Agriculture in Southern Malawi is characterized by small farms (average 0.89 ha)1 growing mixed crop systems of maize, beans and some vegetables. Poverty rates are very high, approximately 70% based on the Malawi poverty line of about $0.41/person/day. Some farms have non-agricultural income. Various NGOs have been involved for several years in encouraging adoption of more highly integrated agriculture-aquaculture (IAA) systems. Adopters of IAA tend to be larger farms growing irrigated vegetables and have higher incomes and lower poverty.
Integrated Aquaculture-Agriculture (IAA) Farming Systems The IAA farming system is based on the utilization of organic wastes and by-products, such as crop residues, as feed inputs to the fish pond, and the recycling of pond mud and water containing nutrient wastes back to cropland. In Southern Malawi, maize bran is the most common pond input. (Dey et al. 2007).
The Malawi data differentiate the farms according to a) the adoption or non-adoption of IAA, and b) the level of integration of IAA. Based on this data we defined the systems for our analysis as:
System 1: Crop-based system of maize, beans and some other crops, with the addition of small ponds with a low level of integration with agriculture. The level of integration is defined by the number of bio-resource flows in the farm, where 2 or less bio-resources is considered to be the low integration case.
System 2: more highly integrated system with larger ponds and irrigated vegetables that utilize water from ponds.
Impact Indicators Mean farm income and per-capita income
The mean farm income in the study area is about $420/yr while the per-capita income is about $160/yr. Fish culture contributes in average between 8% and 10% to the annual farm income (Dey et al., 2007, 2010)
Poverty rates are high in the region, several sources indicate poverty rates ranging from 65% to 78% (NSO-GoM, 1998, IFAD, 2006). According to the survey data, using the official poverty line of 16,165 Malawi Kwacha/year and an exchange rate of 108 Kwacha per US dollar, the poverty rate of the farms in the survey is about 90 percent.
A survey performed by the GoM’s National Nutrition in December 2005 for the rural Malawian population, concluded that the national average number of meals per day for an adult was 2.0; 45% of adults had two meals the day before the survey, roughly one-third of adults had three meals and 19% had only one meal the day before the survey. The percentage of households reporting at least one member regularly reducing the amount of food they consumed at mealtimes was 82% and 49% of households reported that at least one member did not eat during a whole day in the last month due to lack of food (GoM and UNICEF, 2005). An important component of diet is protein consumption.
3. Data The TOA-MD model utilizes statistics (means, variances, correlations) estimated from the data. The model set-up with all of the data used for the analysis is available from the authors.
The average farm size in the population is about 1.9 ha, with an average household size of 5 people. The low integrated farms is characterized by having small ponds averaging 150m2 (0.015 ha), while high integrated farms own larger ponds with an average of 300m2. Non-agricultural income per farm varies across strata and ranges from $44 to about $100, with an average of $76/farm.
Maize is the main staple crop in this region, but farmers also grow a combination of other crops (e.g. beans, pigeon peas, cowpeas, etc.). For this analysis we use maize and beans to represent the crop activities in the farms with low integration. Farmers with high integration of IAA grow vegetables in addition to maize and beans. Data on yields as well as production costs and prices were obtained from GoM (2005), Chilongo (2005) and NoS (2010).
Using the survey data, the protein consumption in each household (kg/person/month) was calculated. The data showed that the average for non-IAA farms was about 1.32 kg/person/month, whereas the average for IAA farms was about 1.64.
4. Results Results of the analysis are presented in Figures 2, 3 and 4, and summarized in Table 1. Figure 2 presents curves showing the simulated adoption rate of IAA as a function of the opportunity cost of changing from System 1 to System 2. The rate that would occur if farmers are behaving economically rationally and maximizing expected returns to their farms, is the point where the curves cross the horizontal axis. This rate ranges from 38 to 49 percent.
Figure 3 presents the predicted poverty rates in relation to the adoption rate of IAA. The baseline poverty rates are at the zero adoption rate, and as noted above, average about 90 percent, and range from 74 to 99 percent. At the economically-efficient rates of adoption (the rates where the adoption curves cross the horizontal axis in Figure 2), the poverty rates decline by 8 to 14 percent, when averaged over the entire population of adopting and non-adopting farms. However, when only adopting farms are considered, the poverty rates decline for adopters by 19 to 35 percent (Table 1).
Figure 4. shows the impacts on protein consumption. The baseline (the rate at zero adoption) shows that protein consumption varies substantially across the regions. Adoption of IAA has relatively small impacts on those areas where consumption is relatively high, but has substantial impacts in Mulange and Mangochi, the two districts with the lowest protein consumption. In those areas, protein consumption among adopters increases from less than 1 kg/person/month to over 2 kg/person/month.
5. Conclusions This paper demonstrates the use of the TOA-MD model to carry out an integrated impact assessment of technology adoption, using the case of integrated agriculture-aquaculture in Malawi. Using TOA-MD, it is possible to implement an integrated assessment of economic, environmental and social impacts at low cost relative to methods that rely on case-specific, complex bio-economic simulation models. Cost is reduced in two ways. First, by using a generic model that can be applied to virtually any system, the time and resources needed to design a new model for each case are largely eliminated. Second, by identifying in advance the indicators that need to be quantified, any data collection activities can be focused on the relevant information, thus eliminating the cost and respondent burden caused by the “kitchen sink” approach to survey design. Moreover, the TOA-MD approach shows that correlations between economic, environmental and social data are often needed to obtain accurate estimates of impact. By recognizing this need in advance, the cost of collecting data can be reduced, and the quality of impact assessment can be enhanced.
References Antle, J.M. 2011. “Parsimonious Technology Impact Assessment.” American Journal of Agricultural Economics, submitted (revised). Antle, J.M. and R.O. Valdivia. 2010. TOA-MD Version 4: Minimum-Data Tradeoff Analysis Model. www.tradeoffs.oregonstate.edu.
Antle, J.M., Valdivia, R. 2006. “Modelling the Supply of Ecosystem Services from Agriculture: A Minimum-Data Approach.” Australian Journal of Agricultural and Resource Economics 50: 1–15.
Brooks, A.C., 1992. Viability of commercial fish farming in Malawi - a short study. Central and Northern Regions Fish Farming Project, Mzuzu, Malawi
Chilongo, T. (2005) An Assessment of Smallholder Farmers' Access to Produce Markets in Malawi: The Case of Kasungu RDP. In: Tsutomo Takane (Ed) “Agricultural and Rural Development in Malawi: Macro and Micro Perspectives”.
Dey, M.M., P. Kambewa, M. Prein, D. Jamu, F. J. Paraguas, D. E. Pemsl and R. M. Briones. Impact of Development and Dissemination of Integrated Aquaculture-Agriculture (IAA) Technologies in Malawi, NAGA, WorldFish Center Quarterly Vol. 29 No. 1 & 2 Jan-Jun 2006
Dey, Madan M.; Paraguas, F.; Kambewa, P.; Pemsl, D. 2010. The Impact of integrated aquaculture-agriculture on small-scale farms in Southern Malawi. Agricultural Economics 41: 67-69
Dey, M.M., Kambewa, P., Prein, M., Jamu, D., Paraguas, F.J., Briones, R., Pemsl, D.E., 2007. Impact of the development and dissemination of integrated aquaculture-agriculture (IAA) technologies in Malawi. In: Waibel H., Zilberman, D. (Eds.), International Research on Natural Resources Management: Advances in Impact Assessment. CAB International, Oxfordshire, UK, pp. 118–14
FAO, 2008. Malawi Nutrition Profile: Nutrition and Consumer protection Division, FAO.
IFAD (2006) Enabling the rural poor to overcome poverty in Malawi. Rome, November 2006.
GoM(Government of Malawi), 2005. Integrated Household Survey 2004–2005. National Statistical Office, Zomba
Kam SP, Barth H, Pemsl DE, Kriesemer SK, Teoh SJ, and Bose ML. 2008. Recommendation Domains for Pond Aquaculture. WorldFish Center Studies and Reviews 1848. The WorldFish Center, Penang, Malaysia. 40 pp.
Kam, S.P. and S.J. Teoh (2008). Suitability Analysis & Query for Aquaculture (SAQUA). Tutorial Guide. The WorldFish Center.
National Statistics Office (NSO), Government of Malawi (2010) – Online statistics. http://www.nso.malawi.net/
Russell, A.; Grotz, P.; Kriesemer, S.; Pemsl, D. (2008). Recommendation Domains for pond Aquaculture: Country Case Study: Development and status of freshwater Aquaculture in Malawi. WorldFish Center Studies & Reviews No. 1869. The WorldFish Center, Penang, Malasya. 52p.
UNICEF. 2005. The State of World’s Children 2005. United Nations Children’s Fund. New York. USA.
Walker T., Maredia M., Kelley T., La Rovere R., Templeton D., Thiele G., and Douthwaite B. 2008. Strategic Guidance for Ex Post Impact Assessment of Agricultural Research. Report prepared for the Standing Panel on Impact Assessment, CGIAR Science Council. Science Council Secretariat: Rome, Italy.
Figure 1. Study area: Map of Malawi
Figure 2. Adoption Rate and Opportunity Cost of Adopting IAA in Southern Malawi – Predicted Adoption Rate is Point Where Curves Cross the Horizontal Axis
Figure 3. Poverty Rate and Adoption Rate of IAA, Southern Malawi.