WPS4498 Policy ReseaRch WoRking PaPeR 4498 Quantifying Institutional Impacts and Development Synergies in Water Resource Programs: A Methodology with Application to the Kala Oya Basin, Sri Lanka R. Maria Saleth Ariel Dinar The World Bank Development Research Group Sustainable Rural and Urban Development Team January 2008 Policy ReseaRch WoRking PaPeR 4498 Abstract The success of development programs, including water Kala Oya Basin in Sri Lanka in order to evaluate the resource projects, depends on two key factors: the role impacts of three water-related programs and the roles of underlying institutions and the impact synergies from of 11 institutions in the context of food security. The other closely related programs. Existing methodologies results provide considerable insights on the relative role have limitations in accounting for these critical factors. of institutions and the flow of development synergies This paper fills this gap by developing a methodology, both within and across different impact pathways. The which quantifies both the roles that institutions play in methodology can also be used to locate slack in impact impact generation and the extent of impact synergies chains and identify policy options to enhance the impact that flows from closely related programs within a flows. unified framework. The methodology is applied to the This paper--a product of the Sustainable Rural and Urban Development Team, Development Research Group--is part of a larger effort in the department to mainstream research in water resource economics. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at r.saleth@cgiar.org and adinar@ worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team QUANTIFYING INSTITUTIONAL IMPACTS AND DEVELOPMENT SYNERGIES IN WATER RESOURCE PROGRAMS: A Methodology with Application to the Kala Oya Basin, Sri Lanka* R. Maria Saleth and Ariel Dinar** Keywords: Development Programs, Econometric Analysis, Food Security, Impact Pathways, Impact Synergy, Institutional Analysis, Institution-Impact Matrix, Kala Oya Basin, Millennium Development Goals, Perception Data, Sri Lanka, Stakeholder Evaluation. This paper is a product of the study "The Institutional Matrix of the Millennium Development Goals: An Empirical Study of Food Security Goals in Kala Oya Basin, Sri Lanka" funded by the World Bank Research Committee and the International Water Management Institute. Earlier versions of this paper were presented at the International Seminar on Water Management "Technology, Economics and the Environment", Fundacion Ramon Areces in Madrid, Spain, January 19-20, 2007 and in a seminar at the International Food Policy Research Institute, Washington DC, November 29, 2007. We thank the participants in the two seminars for their comments. We would like to thank Susanne Neubert, Bandi Kamaiah, Sarath Abayawardana, Seenithamby Manoharan, Ranjtih Ariyaratne, and Shyamalie de Silva for their support in data collection and analysis, and Carlos Espina and Mahfuz Rahman for their help with the MAPLE and MATHEMATICA software application used for derivative analysis. We also thank the 67 respondents for their valuable time and insightful information without which this study would not have been possible. This version of the paper benefitted greatly from comments provided by Daniel Bromley, K. William Easter, Elinor Ostrom, and Vernon Ruttan. The authors are respectively Principal Researcher, International Water Management Institute, Colombo and Lead Economist, World Bank, Washington, DC. ii CONTENTS SUMMARY...............................................................................................................................................................III 1. MOTIVATION AND CONTEXT..........................................................................................................................1 2. IMPACT SYNERGIES AND INSTITUTIONAL ROLES: AN ILLUSTRATION ..........................................3 3. THE METHODOLOGICAL FRAMEWORK.....................................................................................................5 3.1. CONCEPTUAL SETTING .......................................................................................................................................6 3.2. BUILDING BLOCKS OF THE METHODOLOGY........................................................................................................6 3.3. INSTITUTION-IMPACT MATRIX............................................................................................................................9 4. THE EMPIRICAL CONTEXT: THE KALA OYA BASIN, SRI LANKA......................................................12 5. EMPIRICAL SPECIFICATION OF THE MODEL .........................................................................................13 6. DATA SOURCES..................................................................................................................................................18 7. RESULTS AND INTERPRETATIONS..............................................................................................................20 8. EVIDENCE FOR DEVELOPMENT SYNERGIES AND INSTITUTIONAL EFFECTS .............................28 9. CONCLUSIONS AND POLICY IMPLICATIONS...........................................................................................34 ENDNOTES...............................................................................................................................................................38 REFERENCES ..........................................................................................................................................................41 ANNEX A: TECHNICAL NOTES..........................................................................................................................57 ANNEX B: SURVEY INSTRUMENT.....................................................................................................................58 ANNEX C: DERIVING THE REDUCED FORM EQUATION...........................................................................65 iii SUMMARY With increasing investments in development programs in general and water-related programs in particular, there are obvious concerns regarding their actual impacts on development objectives. Two persistent aspects hamper their proper evaluation both in economic literature and in development policy: (a) the specific roles that institutions play in the process of impact generation and transmission, and (b) the impact synergies that a development program derives from past, ongoing, and planned programs. Exclusion of these aspects is very important, particularly in the evaluation of meta development goals such as food security, where the realization of the final objective is linked with the progress of several intermediate but related targets of a hierarchy of programs spanning across sectors and time. This paper develops and applies a methodology that explicitly captures the effects of both institutional roles and development synergies within a unified framework and quantitative context. The development of the framework is illustrated by (a) referring to a set water and agriculture-related development programs (system rehabilitation, bulk water delivery, and crop diversification), (b) tracing their impact pathways and interaction points, (c) locating relevant institutions in these points and pathways, and (d) linking them all with the final goal of food security--one of the key targets of the Millennium Development Goals. This framework is, then, translated into a system model with 21 sequentially linked equations, comprised of a set of development, institutional, and impact variables. For the practical application of the framework, the Kala Oya Basin in Sri Lanka is taken as the empirical context and the perception-based information collected from a sample of 67 experts is used as the data source. The estimation of the model coefficients provides considerable insights on the nature of both the roles that different institutions play at various points of the impact pathways as well as the synergies that development programs derive from each other. The sensitivity analysis performed with the reduced form equation suggests that, in terms of the marginal effects on food security, market institution has the highest effect, followed by others such as price regulation and trade policy. Unlike these institutions with a positive effect, there are others with a negative contribution such as land tenure and rural development policy. From the perspective of practical policy, this paper has two main contributions. First, it demonstrates the importance of accounting for the institutional impacts and development synergies when planning and implementing any new development program in a given region. Second, it provides a diagnostic tool both for locating the weak spots and slack links in various impact pathways of a development program by identifying the institutions and impact chains that are to be strengthened to improve the impact flows of development programs. 1 1. MOTIVATION AND CONTEXT The motivation for this paper comes from two major gaps persisting in the theory and practice related to the critical subjects of development impact and institutional analysis. First, there is a lack of proper treatment of the synergies inherent among development programs, projects or policies with common or closely related goals.1 These synergies occur not just among ongoing programs but even flow from those completed in the past and planned for the future. The evaluation of the impacts of development programs cannot be complete without accounting for these synergies. These synergies are particularly important for composite or meta development goals (e.g., Millennium Development Goals,2 combating climatic change, and governance reforms). This is because the realization of these goals is critically linked with the realization of several intermediate but related goals of a hierarchy of development programs. Second, institutions, defined as a system of legal, policy, and organizational components (Bromley, 1989; Ostrom, 1990), play a central role both in the facilitation and transmission of development impact. Although the general roles of individual institutions are being evaluated in various contexts and details (Saleth and Dinar, 2004 and 2008), there is an insufficient attention on the individual and joint roles several institutions in the specific context of impact generation and transmission. These two gaps are obviously serious in view of the bias and gaps they could cause in the planning, implementation, and assessment of development programs. Ironically, the issue of the lack of or insufficient treatment of impact synergies and the institutional roles in development impact is not entirely new as are its consequences to impact assessment, institutional analysis, and development planning. But, the problem persist essentially due to the absence of an empirically applicable methodological framework that can bring together the multiple impact pathways of two or more development programs within a common analytical framework and single evaluation context. These pathways are very important as they capture the various routes through which the impacts of a program are transmitted on to the final goal and these routes can be characterized by a chain of sequentially and functionally related development, institutional, and impact 2 variables.3 Existing impact assessment approaches are of no or little help in this context due to their inherent analytical limitations.4 Since they do not elaborate the impact process to capture the entire set of impact pathways, they miss the opportunity to locate and evaluate the impact role of institutions in the specific contexts of different pathways. With their ex-post orientation and reliance on objective data, the existing approaches also become unsuitable particularly in the context of multiple and time-lagged projects with continuing, lagged, and uncertain flow of impacts. Here, ex-ante approach and subjective information are unavoidable. This paper aims to develop and empirically illustrate a methodology that can directly capture both the development synergies and the institutional roles within a unified framework and quantitative context. The methodology is based on an analytical framework that traces the major impact pathways between the development programs and the development goal and characterizes these pathways in terms of the sequential and functional linkages among the development, institutional, and impact variables involved. Since these linkages can be translated into a system of structurally linked equations, each capturing different impact pathways, the framework can be mathematically translated for empirical application and quantitative evaluation. This paper demonstrates the application of this methodology in the empirical context of the Kala Oya Basin in Sri Lanka by taking (a) the food security related to the first MDG as the development goal, (b) a set of three water-related programs--namely, system rehabilitation, bulk water delivery, and crop diversification--as the candidate development programs,5 and (c) the ex-ante information from a sample of 67 stakeholders--consisting largely of government officials and national experts--as the data source. From here, the paper is structured as follows. Section 2 provides a graphical illustration of the welfare implications of both the development synergies and institutional role and indicates the policy value of their ex-ante evaluation. Section 3 sets the analytical framework with the conceptual foundation and building blocks of the proposed methodology. It also presents an institution-impact matrix developed in a generic context that can serve as a platform to operationalize the conceptual model into an empirically applicable form. Section 4 describes the study 3 area and highlights its major development challenges. Section 5 applies the institution-impact matrix to the development and institutional context of the study region. It provides both a graphical presentation of the impact pathways in terms of a flow diagram and a mathematical representation of these pathways as a set of functional relationships evident among development, institutional, and impact variables. Section 6 describes the data source and also provides empirical precedence and theoretical justification for using stakeholder-based ex-ante information. Section 7 presents and analyzes the results of the econometric models of institution-impact interaction. It provides statistical evidence for the relative role and significance of the development, institutional, and impact variables in difference equations representing various impact layers and also demonstrates the structural linkages among these equations or layers of impact pathways. Section 8 provides numerical evidence for the size and flow of development synergies and institutional impacts based on a sensitivity analysis of the reduced form equation in terms of the local and system-wide impacts of a marginal change in different variables. The final section concludes with the analytical and empirical insights of the paper, its limitations, and the scope for its future extension and refinement. 2. IMPACT SYNERGIES AND INSTITUTIONAL ROLES: AN ILLUSTRATION When selecting development programs, policy makers usually make an ex-ante assessment of their effects both on overall welfare and on its distribution across groups in the society. But, such assessments often ignore the issue of how this welfare and its distribution would change when the roles of institutional impacts and synergies from related programs are taken into account. The practical importance and policy value of considering these changes in such an ex-ante assessment is graphically demonstrated using Figure 1, which is an adaptation of a framework suggested by Just et al. (2004). Figure 1 depicts a simple economy with two individuals (or groups), i.e., I (rich) and J (poor), who, with a given bundle of resources, can produce/consume two goods, i.e., food (F) and recreation (R). Given current technologies and institutions, the production possibility frontier for the economy is OP. Assume that the economy is in a status quo at (i, j)0with a corresponding welfare levels for the 4 two-person society. Based on the Edgeworth Box analysis, J's welfare is: JF(0)+JR(0) and I's welfare is [P-JF(0)]+[O-JR(0)]. Now, suppose the government wants to take the economy towards the frontier OP and improve, thereby, both the total welfare and its distribution. For this, it considers two programs, which are expected, a priori, to achieve such economic and social objectives, i.e., a `dashed' (dashed line) program (D) and a `solid' (solid line) program (S). As can be seen from Figure 1, the `dashed' program moves the economy from (i, j)0 to (i, j)D and the `solid' program moves the economy to (i, j)S . Both policies are Pareto optimal in the sense that they satisfy the condition of utility maximization for both individuals/groups. But, the `dashed' program is less efficient as it falls short of the production possibilities frontier (OP) and ends with an inner frontier, O'P'0.5 (see Hair et al. 1995). Since none of them were used together as independent variables in any equation, the multicollinearity can be taken not as a serious problem. 22 The result is somewhat surprising because the bulk water provision, as being piloted in the study region, is only to farmer groups and not to individual farmers as needed for promoting independent crop decisions. 23 Note that the strong positive effect of CROPATEN on LABPRODY cannot be interpreted simply as food crops are more conducive for labor productivity. It needs to be explained in the light of the inverse relation between WAGERATE and RURALEMP seen in Equation [13], which means that high wage limits labor use and with a given farm productivity, lesser labor use causes labor productivity to be higher than otherwise. It is this effect occurring in the face of a labor scarcity and paddy domination that is behind the positive association seen between CROPATEN and LABPRODY. 24 The inverse relation between RURALEMP and LIVSTOCK suggests the tradeoff in labor time allocation between wage employment and livestock rearing, which is particularly so among the groups which need both sources of income. 25 It is important to note that even though bulk water delivery policy and water institution are part of institutions, they remain only endogenous because the former is specified as a function of system rehabilitation and the later is specified as a function of bulk water delivery policy. 41 REFERENCES Baker, Judy L., 2000, Evaluating the Impact of Development Projects on Poverty: A Handbook for Practitioners, Directions in Development Series, World Bank, Washington DC. Bandara, K. R. Neil, undated, "Assessment of Kala Oya River Basin Project", PowerPoint Presentation, Kala Oya Basin Secretariat, Mahaweli Authority of Sri Lanka, Colombo. Barrett, Scott, and Kathryn Graddy, 2000, "Freedom, Growth, and the Environment", Environment and Development Economics, 5: 433-456. Bourguignon, F., and L. A. Pereira da Silva, 2003, The Impact of Economic Policies on Poverty and Income Distribution: Evaluation Techniques and Tools, Washington, DC and New York: World Bank and Oxford University Press (Available at: http://www.worldbank.org/psia). Bromley, D.W., 1985, "Resources and Economic Development", Journal of Economic Issues, 19(September): 779-96. Bromley, D.W., 1989, Economic Interests and Institutions: The Conceptual Foundations of Public Policy, New York: Basil Blackwell. Centre for Global Development, 2006, When Will We Ever Learn? Improving Lives through Impact Evaluation, Report of the Evaluation Gap Working Group, May. Coudouel, Aline, Anis A. Dani, and Stefano Paternostro, 2006, "Lessons from the Implementation of Poverty and Social Impact Analysis of Reforms", in Aline Coudouel, Anis A. Dani, and Stefano Paternostro, eds., Poverty and Social Impact Analysis of Reforms: Lessons and Examples from Implementation, World Bank, Washington DC, pp. 1-28. Commons, J.R., 1934, Institutional Economics, New York: Macmillan De Silva, S., Ranjith Ariyaratne, and Sarath Abayawardana, 2006, Kala Oya Basin, Sri Lanka: A Resource and Development Profile, IWMI Working Paper, Colombo, Sri Lanka. Department of Census and Statistics (DCS) and UN World Food Program (WFP), 2005. Vulnerability of GN divisions to food insecurity, Anuradhapura District 2004. Colombo, Sri Lanka Department for International Development (DfID) and World Bank, 2005, Tools for Institutional, Political and Social Analysis in PSIA: A Sourcebook, Washington, DC: World Bank and DfID (http://web.worldbank.org/wbsite/external/ topics/extsocialdevelopment/exttoppsisou). Dinar, Ariel and R. Maria Saleth, 2005, "Can Water Institution Be Cured? A Water Institutions Health Index", Water Science and Technology: Water Supply, l5(6):17-40, 2005. Douglas, Mary, 1986, How Institutions Think?, New York: Syracuse University Press. 42 Gray, Cheryl W., and Daniel Kaufmann, 1998, "Corruption and Development", Finance & Development, 35(1): 7-10. Gujarati, D., 1995, Basic Econometrics, 3rd Edition, New York, McGraw-Hill. Hair, J. F., R. E. Anderson, R. L. Tatham, and W. C. Black, 1995 Multivariate Data Analysis. New Jersey: Prentice Hall. Hausman, J. A., 1978, "Specification Tests in Econometrics", Econometrica, 46: 1251-1271. Just, Richard E., Darrell L. Hueth, and Andrew Schmitz, 2004, The Welfare Economics of Public Policy: A Practical Approach to Project and Policy Evaluation, Cheltenham, UK: Edward Elgar. Kahneman, Daniel, and Amos Tversky, 1984, "Choices, Values, and Frames", American Psychologist, 39(4): 341-350. Kaufmann, Daniel, Aart Kraay, and Massimo Mastruzzi, 2006, Governance Matters V: Governance Indicators for 1996-2005, Research Department, World Bank, Washington DC (http://ssrn.com/abstract=929549) Kennedy, Peter, 1987, A Guide to Econometrics (Second Edition), Cambridge, MA: The MIT Press. Knack, Stephen, and Philip Keefer, 1986, "Institutions and Economic Performance: Cross-Country Tests Using Alternative Institutional Measures", Economics and Politics, 7(3): 207-227. Neubert, Susanne, 2000, Social Impact Analysis of Poverty Alleviation Programmes and Projects, Illford, Essex: Frank Cass. Neubert, Susanne, 2006, "Impact Analysis in Development Cooperation and Approaches Adopted in Research and Practice", German Development Institute, Bonn. North, Douglass C., 1990, Institutions, Institutional Change, and Economic Performance, Cambridge, MA: Cambridge University Press. Ostrom, Elinor, 1990, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge, UK: Cambridge University Press. Ostrom, Vincent, 1980, "Artisanship and Artifact", Public Administration Review, 40(July-Aug): 309-317. Saleth, R. M., and Ariel Dinar, 2004, The Institutional Economics of Water: A Cross- Country Analysis of Institutions and Performance, Cheltenham, UK: Edward Elgar. Saleth, R.M., and Ariel Dinar, 2008, "Linkages within Institutional Structure: An Empirical Analysis of Water Institutions", Journal of Institutional Economics (Forthcoming). Saleth, R. M., Ariel Dinar, Susanne Neubert, Bandi Kamaiah, Seenithamby Manoharan, Sarath Abayawardena, and Ranjith Ariyaratne, 2008, "Institutions, Impact Synergies, and Food Security Goal: A Methodology with 43 Results From Kala Oya Basin, Sri Lanka", Research Report No: 124, International Water Management Institute, Colombo, Sri Lanka. Tool, M. R., 1977, "A Social Value Theory in Neo-institutional Economics", Journal of Economic Issues, 11(December): 823-849. 44 R e (i, j) S c r O e I a JR(S) (i, j) D t O' i JR(D) o n JR(0) (i, j) 0 0 P' P J JF(0) JF(D) JF(S) Food Figure 1: Ex-ante Evaluation of Alternative Programs and Social Welfare 45 EVALUATION INTERFACE (The Institution-Impact Matrix) BASIN INSTITUTIONAL MATRIX Water Development Institutions Program-1 Land Institutions Development Food Security Program-2 Agricultural Institutions Goal Development Other Policies & Program-3 Institutions Figure 2: Conceptual Frame for Institution-impact Interface 46 Development Program 3 (Watershed Development) Development Program 2 (Introduction of New Crop Varieties) Development Program 1 (Water Development/Dam Construction) Impact Institutional Configurations Development Goals Pathways (Defined by combinations of variables capturing the land, (In terms of variables capturing poverty, water, agricultural, and resource/environment institutions) food and resource conservation goals) Water Land Agricultural Res/Env Income/ Food Price/ Efficient Use Institutions Institutions Institutions Institutions Jobs Output of Resources Water rights/ Land tenure/ Input system/ Water/soil quality Irrigation User organs Tenancy extension codes ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) Regional Basin institutions Land Farm wage Pollution Growth ($$$) markets policy regulations ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) Sectoral water Farmland Rural-urban Wastewater Urbanization Allocation use rules markets regulations ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) Pricing & cost Property Wastewater use Sanitation Water Supply recovery ownership practice Policy ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) Ecological Project choice Policy on Policies on fragile Forest laws Effects & scale commons areas & policies ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) ($$$) Note: $$$ = Values of actual or perception-based information on one or more variables designed to capture both the status and effectiveness of each of the institutional aspects as well as the level of impacts on each of the food security aspects. Custom Institution Figure 3: Institution-impact Matrix: A Simplified Presentation 47 Figure 4: The Kala Oya Basin, Sri Lanka 48 Subsidy Market Policy Cultivation Agricultural Institution Costs Income Water Institution Land Productivity Land Farm Wage Tenure Legislations Income Bulk Water Water Rural Distribution Productivity Employment Farm Input Labor Institution Wage Rates Income Land/Soil Labor Health Productivity Crop Food Food Diversification Crop Pattern Availability Security Food Production Trade Samurdhi Food System Policy Policy Prices Rehabilitation Custom Non-farm Institution Activities Price Regulation Rural Dev.Policy Fodder/Feed Livestock/ Supply Poultry Development Programs Institutions Development Impacts Impact Transmission Pathways Development Goal Impact Dimensions (Variables) Institutional Impacts Impact Transmission Feeders Figure 5: Institution-Impact Interactions with 3 Development Programs 49 EQN 1 EQN 2 EQN 3 EQN 4 EQN 5 EQN 6 EQN 7 EQN 8 EQN 9 EQN 10 EQN 11 EQN 12 EQN 13 EQN 14 EQN 15 EQN 16 EQN 17 EQN. 18 EQN 19 EQN. 20 EQN. 21 Figure 6: Structural Linkages within the Model 50 Table 1: Variables in the Institution-impact Model Categories of No Names of Variables Acronym Variables Used Development Goal 1 Food Security FOODSECT Development Programs 1 System Rehabilitation SYSREHAB 2 Bulk Water Distribution BULKWATD 3 Crop Diversification CROPDIVR Impact Variables 1 Crop Pattern CROPATEN 2 Land Productivity LANPRODY 3 Water Productivity WATPRODY 4 Labor Productivity LABPRODY 5 Rural Employment RURALEMP 6 Wage Rates WAGERATE 7 Cultivation Costs CULTCOST 8 Agricultural Income AGLINCOM 9 Land Quality/soil Health LANHELTH 10 Food Production FOODPROD 11 Non-farm Enterprises NFAMENTS 12 Fodder & Feed Supply FEDSUPLY 13 Livestock/Poultry LIVSTOCK 14 Farm Income FAMINCOM 15 Labor Income LABINCOM 16 Food Availability FOODAVAL 17 Food Price FOODPRIC Institutional Variables 1 Land Tenure LANTENUR 2 Water Institutions WATINSTN 3 Customary Institutions CUSINSTN 4 Farm Input Institutions FAMINSTN 5 Market Institutions MKTINSTN 6 Price Regulations PRICREGL 7 Wage/Labor Legislations WAGELAWS 8 Rural Development Policy RDVPOLCY 9 Trade Policy TRDPOLCY 10 Farm Subsidy Policy SUBPOLCY 11 Samurdhi Policy SAMPOLCY 51 Table 2: Descriptive Statistics for Model Variables No Endogenous Mean Standard Deviation Minimum Maximum Variables 1 BULKWATD 6.32 1.75 1.00 9.00 2 CROPDIVR 6.04 1.79 2.00 10.00 3 CROPATEN 5.60 1.00 2.79 7.57 4 WATINSTN 5.03 1.88 1.00 9.00 5 WATPRODY 7.29 1.42 4.00 10.00 6 LANHELTH 7.62 1.33 3.50 10.00 7 LANPRODY 6.84 1.40 2.63 10.00 8 FEDSUPLY 5.32 1.43 1.00 8.00 9 LIVSTOCK 3.64 1.62 0.90 7.90 10 NFAMENTS 7.07 1.29 2.25 9.50 11 LABPRODY 4.94 2.21 1.00 9.00 12 WAGERATE 6.10 1.27 2.50 8.50 13 RURALEMP 5.31 2.08 1.00 10.00 14 CULTCOST 5.66 1.68 1.00 8.00 15 AGLINCOM 6.90 1.49 3.00 10.00 16 FAMINCOM 5.50 1.09 3.00 9.00 17 LABINCOM 4.64 1.31 2.00 8.00 18 FOODPROD 5.22 1.23 2.33 7.67 19 FOODAVAL 5.24 1.36 2.50 8.50 20 FOODPRIC 4.37 1.31 1.50 7.50 21 FOODSECT 5.07 1.59 0.75 8.00 No Endogenous Mean Standard Deviation Minimum Maximum Variables 1 SYSREHAB 6.75 1.19 1.67 8.83 2 LANTENUR 6.20 1.15 3.56 8.33 3 CUSINSTN 4.71 1.28 1.40 7.60 4 FAMINSTN 5.52 1.68 1.00 9.00 5 MKTINSTN 5.10 1.35 1.67 9.33 6 PRICREGL 4.62 1.57 1.00 8.75 7 WAGELAWS 3.51 1.74 1.00 8.50 8 RDVPOLCY 5.07 1.85 1.50 9.00 9 TRDPOLCY 6.57 1.41 3.00 9.00 10 SUBPOLCY 6.82 1.38 3.00 10.00 11 SAMPOLCY 5.12 1.97 1.00 10.00 52 Table 3: OLS Results for the Single Equation Model Dependent Independent Estimated T-Ratio Level of Elasticity R2 Variable Variables Coefficient Significance at Means FOODSECT BULKWATD 0.317 1.485 0.146 0.395 0.447 SYSREHAB -0.053 -0.152 0.880 -0.070 CROPDIVR -0.038 -0.251 0.803 -0.045 CROPATEN -0.267 -0.652 0.519 -0.295 WATINSTN 0.255 1.351 0.185 0.253 WATPRODY 0.013 0.056 0.955 0.018 LANHELTH -0.212 -0.864 0.393 -0.318 LANPRODY -0.135 -0.459 0.649 -0.182 FEDSUPLY 0.124 0.554 0.583 0.130 LIVSTOCK -0.111 -0.633 0.531 -0.080 NFAMENTS 0.057 0.258 0.798 0.080 LABPRODY 0.253 1.606 0.117 0.247 WAGERATE 0.506 1.752 0.088 0.610 RURALEMP 0.077 0.475 0.637 0.081 CULTCOST 0.125 0.715 0.479 0.140 AGLINCOM -0.785 -2.707 0.010 -1.068 FAMINCOM 1.128 2.910 0.006 1.225 LABINCOM 0.452 1.907 0.065 0.414 FOODPROD -0.118 -0.342 0.735 -0.122 FOODAVAL 0.125 0.487 0.629 0.129 FOODPRIC -0.215 -0.898 0.375 -0.185 LANTENUR -0.139 -0.613 0.544 -0.170 CUSINSTN -0.102 -0.402 0.690 -0.095 FAMINSTN -0.129 -0.618 0.540 -0.141 MKTINSTN -0.253 -1.184 0.244 -0.255 PRICREGL -0.003 -0.020 0.984 -0.003 WAGELAWS -0.053 -0.256 0.800 -0.037 RDVPOLCY 0.107 0.634 0.530 0.108 TRDPOLCY 0.184 0.729 0.471 0.239 SUBPOLCY -0.134 -0.720 0.476 -0.180 SAMPOLCY 0.173 0.876 0.387 0.175 53 Table 4: System Model of Institution-Impact Interaction: 3-SLS Results Eqn. Dependent Independent Estimated Asymptotic Level of Elasticity R2 No Variables Variables Coefficient T-Ratio Significance at Means [1] BULKWATD SYSREHAB 0.886 10.240 0.000 0.947 0.333 LANTENUR 0.056 0.591 0.555 0.055 [2] CROPDIVR BULKWATD 0.594 5.993 0.000 0.621 -0.366 FAMINSTN 0.385 3.572 0.000 0.352 [3] CROPATEN CROPDIVR 0.438 7.381 0.000 0.473 -0.354 LANTENUR 0.141 1.897 0.058 0.156 CUSINSTN 0.446 7.107 0.000 0.375 [4] WATINSTN BULKWATD 0.851 6.548 0.000 1.068 0.148 LANTENUR -0.014 -0.097 0.923 -0.017 CUSINSTN -0.058 -0.465 0.642 -0.054 [5] WATPRODY CROPATEN 1.099 8.446 0.000 0.845 -0.104 WATINSTN 0.167 1.174 0.240 0.115 FAMINSTN 0.047 0.553 0.581 0.036 [6] LANHELTH CROPATEN 1.000 4.747 0.000 0.736 -0.089 WATPRODY 0.358 2.450 0.014 0.343 LANTENUR -0.102 -0.800 0.424 -0.083 [7] LANPRODY CROPATEN 0.520 2.323 0.020 0.425 0.285 LANHELTH 0.584 4.039 0.000 0.650 FAMINSTN -0.093 -1.515 0.130 -0.075 [8] FEDSUPLY CROPATEN 0.821 7.167 0.000 0.864 -0.004 CUSINSTN 0.152 1.160 0.246 0.134 [9] LIVSTOCK FEDSUPLY 0.613 3.295 0.001 0.901 -0.686 TRDPOLCY 0.028 0.192 0.847 0.052 [10] NFAMENTS CROPATEN 1.164 20.910 0.000 0.922 -0.471 RDVPOLCY 0.107 2.001 0.045 0.076 [11] LABPRODY LANPRODY -0.425 -1.230 0.219 -0.588 0.223 CROPATEN 1.413 3.354 0.001 1.602 [12] WAGERATE LABPRODY 0.138 1.158 0.247 0.111 -0.122 NFAMENTS 0.650 8.605 0.000 0.753 WAGELAWS 0.222 2.969 0.003 0.128 [13] RURALEMP LANPRODY 0.671 1.282 0.200 0.865 -0.120 WAGERATE -0.450 -1.474 0.141 -0.517 NFAMENTS 0.752 1.847 0.065 1.001 LIVSTOCK -0.515 -2.931 0.003 -0.351 [14] CULTCOST CROPATEN -0.066 -0.134 0.893 -0.065 -0.337 WAGERATE 1.048 1.905 0.057 1.130 FAMINSTN -0.017 -0.105 0.916 -0.017 SUBPOLCY -0.045 -0.312 0.755 -0.054 54 Table 4 (Continued) Eqn Dependent Independent Estimated Asymptotic Level of Elasticity R2 No Variables Variables Coefficient T-Ratio Significance at Means [15] AGLINCOM LANPRODY 0.609 3.963 0.000 0.605 0.030 CULTCOST 0.394 2.909 0.004 0.323 MKTINSTN 0.097 0.950 0.342 0.072 [16] FAMINCOM AGLINCOM 0.136 1.131 0.258 0.170 0.116 NFAMENTS 0.437 3.773 0.000 0.561 LIVSTOCK 0.412 6.153 0.000 0.271 [17] LABINCOM RURALEMP -0.437 -2.545 0.011 -0.501 -0.649 NFAMENTS 0.550 2.560 0.010 0.839 LIVSTOCK 0.778 5.153 0.000 0.608 SAMPOLCY 0.061 0.758 0.449 0.067 [18] FOODPROD CROPATEN 0.823 5.501 0.000 0.883 0.400 LANPRODY 0.312 1.949 0.051 0.409 WATPRODY -0.206 -2.115 0.034 -0.287 [19] FOODAVAL FOODPROD 0.451 2.816 0.005 0.449 0.149 TRDPOLCY 0.319 3.629 0.000 0.400 MKTINSTN 0.160 1.426 0.154 0.156 [20] FOODPRIC FOODPROD 0.474 4.329 0.000 0.566 0.096 PRICREGL 0.131 1.812 0.070 0.139 MKTINSTN 0.257 2.504 0.012 0.300 [21] FOODSECT FOODAVAL 0.520 2.280 0.023 0.538 -0.670 FOODPRIC -0.965 -2.932 0.003 -0.833 FAMINCOM 0.767 1.603 0.109 0.833 LABINCOM 0.507 1.622 0.105 0.465 Sample Size 67 Endogenous Variables 21 Exogenous Variables 11 System R2 0.878 Chi-Square (with 61 degrees of freedom, P=0.000) 140.9 Notes: (a) This model is estimated with no constant term in all equations. (b) Bold coefficients are significant at 10 percent or better. (c) Elasticity at means are the weighted coefficients with the weights being the ratio of the means of the concerned dependent and independent variables, This standardization enables a comparison of the relative importance of the independent variables both within and across equations. (d) Unlike OLS, where R2 has the range of 0-1, the R2 in the case of 3-SLS can range from - to 1. The relevant statistics to be considered in the case of 3-SLS estimation are the System R2, which captures the explanatory power of the model as a whole and Chi-Square, which constitutes a test of the overall significance of the model. 55 Table 5: Size and Flow of Development Impacts and Synergies Endogenous Equation Development Programs Total Variables Numbers System Bulk Crop Effects Rehabilitation Water DiversificationReceived Delivery BULKWATD y1 0.886 - - 0.886 CROPDIVR y2 0.526 0.594 - 1.120 CROPATEN y3 0.231 0.261 0.438 0.930 WATINSTN y4 0.754 0.851 - 1.605 WATPRODY y5 0.379 0.427 0.481 1.287 LANHELTH y6 0.366 0.346 0.172 0.884 LANPRODY y7 0.334 1.289 0.583 2.206 FEDSUPLY y8 0.189 0.213 0.395 0.797 LIVSTOCK y9 0.116 0.130 0.220 0.466 NFAMENTS y10 0.268 0.302 0.509 1.079 LABPRODY y11 0.374 0.035 0.522 0.931 WAGERATE y12 0.280 0.310 0.484 1.074 RURALEMP y13 0.229 0.261 0.426 0.916 CULTCOST y14 0.278 0.417 0.655 1.350 AGLINCOM y15 0.313 0.106 0.166 0.585 FAMINCOM y16 0.262 0.081 0.137 0.480 LABINCOM y17 0.275 0.782 0.249 1.306 FOODPROD y18 0.239 0.296 0.390 0.925 FOODAVAL y19 0.108 0.133 0.176 0.417 FOODPRIC y20 0.113 0.140 0.185 0.438 FOODSECT y21 0.011 0.226 0.395 0.632 Total Effects Generated 6.531 7.200 6.583 20.314 56 Table 6: Size and Flow of Institutional Impacts Endogenous Equation Institutional Variables Total Variables Number Effect LANTENUR CUSINSTN FAMINSTN MKTINSTN PRICREGL WAGELAWS RDVPOLCY TRDPOLCY SUBPOLCY SAMPOLCY Received BULKWATD y1 0.056 - - - - - - - - - 0.056 CROPDIVR y2 0.033 0.385 - - - - - - - - 0.418 CROPATEN y3 0.156 0.446 0.169 - - - - - - - 0.771 WATINSTN y4 0.034 -0.058 - - - - - - - - -0.024 WATPRODY y5 0.177 0.481 0.232 - - - - - - - 0.890 LANHELTH y6 0.117 0.618 0.252 - - - - - - - 0.987 LANPRODY y7 0.149 0.593 0.142 - - - - - - - 0.884 FEDSUPLY y8 0.128 0.518 0.139 - - - - - - - 0.785 LIVSTOCK y9 0.078 0.318 0.085 - - - 0.028 - - 0.509 NFAMENTS y10 0.181 0.519 0.196 - - - 0.107 - - - 1.003 LABPRODY y11 0.145 0.648 0.129 - - - - - - - 0.922 WAGERATE y12 0.119 0.493 0.111 - - 0.222 0.015 - - - 0.960 RURALEMP y13 0.140 0.391 0.148 - - -0.114 0.073 -0.013 - - 0.879 CULTCOST y14 0.114 0.487 0.088 - - 0.233 0.016 - -0.045 - 0.427 AGLINCOM y15 0.136 0.553 0.121 0.097 - 0.092 0.006 - -0.018 - 0.987 FAMINCOM y16 0.146 0.498 0.147 0.040 - 0.038 0.049 0.004 -0.007 - 0.915 LABINCOM y17 0.174 0.451 0.186 - - -0.089 0.116 -0.022 - 0.061 0.877 FOODPROD y18 0.152 0.395 0.182 - - - - - - - 0.729 FOODAVAL y19 0.069 0.178 0.082 0.319 - - - 0.160 - - 0.808 FOODPRIC y20 0.072 0.187 0.086 0.131 0.257 - - - - - 0.733 FOODSECT y21 -0.003 0.055 0.004 0.332 0.130 0.106 -0.086 0.146 -0.004 0.059 0.739 Total Effects Generated 2.373 8.156 2.499 0.919 0.387 0.250 0.296 0.329 -0.074 0.120 15.255 57 ANNEX A: TECHNICAL NOTES Institutional Ecology Principle: This principle extends the `ecosystem' concept to institutional systems to analytically show (a) the linkages and synergies among institutions across domains (law, policy, and organization), spheres (land, water, agricultural, rural, and environmental), and scales (basin, region, and national) and (b) the nested and embedded character of institutions within the social, economic, political, and resource systems. Institutional Decomposition and Analysis Framework: This framework unbundles institutions into a set of interrelated rules, characterizes them using quantitative and qualitative variables, and formalizes the relations and linkages among these rules (Saleth and Dinar, 2004). The approach is similar in spirit to the Institutional Analysis and Development framework developed by Ostrom (1990) for application to local level institutions for common pool resources management. Ex-ante Approach: This approach tries to evaluate the futuristic changes and expectation aspects related to institutions based on the convergence in stakeholders' perception. Such consensual perception can summarize objective evaluation, learned judgments, aspirations, and expectations of participating stakeholders. Unlike the post mortem approach underlying the ex-post evaluation and analysis, the ex-ante approach is very useful for designing anticipatory and coping strategies that would allow enough lead time for policy/program adjustments and modifications. Adaptive Instrumental Evaluation: Unlike other evaluation approaches in economics relying on normative and absolute concepts such as `efficiency' based on the assumption of individual rationality and perfect information, the adaptive instrumental evaluation is based on a positive and relative approach (Tool, 1977; Kahneman and Tversky, 1984; Bromley, 1985). It allows the evaluation of events/aspects with respect to relevant reference points (e.g., best practices, desirable conditions, and stated objectives) rather than with respect to ideals or absolute conditions. It also allows the reference points to be flexible and changeable within the evaluation process itself (Saleth and Dinar, 2004). This approach is very pertinent for evaluating aspects such as institutions and their performance involving considerable level of qualitative and subjective considerations. 58 ANNEX B: SURVEY INSTRUMENT "THE INSTITUTIONAL MATRIX OF THE MILLENNIUM DEVELOPMENT GOALS: AN EMPIRICAL STUDY OF FOOD SECURITY GOALS IN KALA OYA BASIN, SRI LANKA" (Research Preparation Work funded by WB and IWMI) PART-A: INSTRTUCTIONS (1) The conceptual framework is generic, but captures as much as possible the relevant aspects of KOB basin in particular and Sri Lanka in general; (2) It is focused on the impact of the three development programs on food security, particularly from the perspective of small farmers, farm workers, and other rural poor; (3) `Impact pathways' are the routes through the economic impacts of development programs are transmitted to the development goals. These impact transmissions are carried out by the `impact variables'. In the present context, three development programs (i.e., crop diversification program; system rehabilitation, and bulk water allocation policy) and one development goal (i.e., food security) are considered. (4) Before asking questions, the conceptual framework is briefly explained to give adequate background for the respondents. First, the 3 development programs and their role in food security, then, their impact pathways defined by the impact variables, and, finally, the role of institutional factors in affecting these pathways are all explained to them. (5) The respondents are also informed that the questions to be asked are related to different components of the framework and answers are expected with respect to the conditions prevalent in KOB in particular and Sri Lanka in general. (6) More importantly, it is necessary to convince them that the evaluation is done in an ex-ante context and what they perceive or believe about various relationships in the conceptual framework are very important and valuable for the evaluation and analysis. Also, it is important to inform them that the development programs can both those that are implemented as well as those that are contemplated or potentially relevant for the KOB or Sri Lanka. (7) All questions are formulated as yes or no questions or questions requiring answers within the scale of 1-10, with `1' being low or weak and `10' being high or strong, depending on the context. For coding purpose, a `no' answer is treated as 0 and the `yes' answer is evaluated within the scale of 1-10. Thus, all answers are recorded within the scale of 0-10. 59 PART-B: BASIC DETAILS (1) Respondent's Details: (a) Name ........................................... (b) Qualification ........................................... (c) Discipline ........................................... (d) Professional Position ........................................... (e) Years of Experience ........................................... (f) Contact Details ........................................... ........................................... ........................................... ........................................... Email.................................... (2) Interview Details: (a) Interviewers Name .......................................... (b) Place and Date .......................................... PART­C: QUESTIONNAIRE 1. Food Security (FOODSECT) (a) How strong, in your opinion, is the food security status of small farmers?.........................................124 (b) How strong, in your opinion, is the food security status of farm workers? ........................................124 (c) How strong, in your opinion, is the food security status of the rural poor? ........................................124 (d) How strong, in your opinion, is the nutritional status of children and aged? ......................................124 2. Crop Diversification (CROPDIVR) (From low to high-value crops; e.g., paddy to vegetables, oilseeds, and fruits) (a) How bright are the economic and technical prospects for crop diversification? ..................................124 (b) How effective are the crop diversification efforts of the government? ..............................................124 (c) How important are customs in crop choice? ..................................................................................124 (d) How serious are customs in constraining crop diversification? .........................................................124 (e) How important is water delivery system for crop diversification? .....................................................124 (f) How serious is small farm size as a constraint for crop diversification? ............................................124 (g) How important is land and soil quality as a factor for crop diversification? ........................................124 60 3. System Rehabilitation (SYSREHAB) (a) How effective is the system rehabilitation program? ......................................................................124 (b) How far can rehabilitation improve land and soil health (by limiting salinity)? ...................................124 (c) How important is system rehabilitation as a contributing factor for land productivity? ........................124 (d) How far system rehabilitation is effective in facilitating bulk water allocation? ..................................124 4. Bulk Water Distribution (BULKWATD) (a) How far can bulk water distribution improve existing water allocation procedures? ............................124 (b) How far can bulk water distribution strengthen water user organizations? ........................................124 (c) How far can bulk water distribution contribute to crop diversification? .............................................124 (d) How far can bulk water distribution improve water use efficiency? ..................................................124 (e) How far can bulk water distribution contribute to land & soil health? ...............................................124 5. Crop Pattern (CROPATEN) (a) To what extent can crop diversification alter crop pattern? .............................................................124 (b) How far can diversification lead to the adoption of high-value crops? ...............................................124 (c) How far can the changes in crop pattern lead to water savings? .....................................................124 (d) Haw far can the changes in crop pattern improve land and soil health (via crop rotation)? .................124 (e) How far can the changes in crop pattern negatively affect foodgrain output? ....................................124 (f) How far can the changes crop pattern negatively affect fodder/feed supply? ....................................124 (g) How far can the changes in crop pattern raise cultivation costs? .....................................................124 (h) If crop pattern shifts towards high-value crops, how important is this shift for the development of rural non-farm activities? .......................................................................................................124 6. Land Productivity (LANPRODY) (Output per unit of land; it differs by crops) (a) How important is land productivity for farm employment? ..............................................................124 (b) How important is land productivity for farm income? .....................................................................124 (c) How important is land productivity for labor productivity? ..............................................................124 (d) Generally, higher land productivity leads to higher water productivity. How strong will be this relationship between land and water productivity? ..................................................................124 (e) Crop pattern changes, though reduce the area under food crops, can also improve the overall farm productivity. If so, how significant will be this effect? .................................................124 (f) System rehabilitation and bulk water delivery can improve water delivery and contribute, thereby, to overall farm land productivity. If so, how significant will be this effect? ..........................124 61 7. Water Productivity (WATPRODY) (Output per unit of applied water; it differs by crops) Generally, efficient water use contributes to land productivity, partly by minimizing the negative effects of water over use (e.g., waterlogging; Salinity) and partly by enhancing the efficiency and productivity of other farm inputs. If this is so, (a) How strong will be the impact of water use efficiency on land productivity? ......................................124 (b) How strong will be the impact of water use efficiency on the efficiency of other inputs? .....................124 8. Labor Productivity (LABPRODY) (Output per labor; it differs by crops) (a) Generally, higher labor productivity will lead to higher wage rate. If so, how strong (or weak) is the relationship between labor productivity and wage rates? .......................................................124 (b) Generally, efficient and productive workers do the same or more work. If so, how important is the role of productivity in determining the overall level of farm employment? ...............................124 9. Rural Employment (RURALEMP) (a) Generally, given the level of land productivity, more employment means less labor productivity. If so, how strong is this negative relationship? .............................................................................124 (b) Generally, for given wage rates, more employment means more income. But, with low or declining wage rates, more employment may not always lead to more income. How realistic is this fact? .........124 10. Wage Rates (WAGERATE) (a) How strong is the influence of higher wage rates on cultivation costs? .............................................124 (b) Are the wage rates high enough to provide incentive for improved labor productivity? If so, how strong will be this effect? ............................................................................................124 (c) Are the wage rates adequate enough to assure decent income for farm workers? If so, how strong will be this fact? ...............................................................................................124 11. Cultivation Costs (CULTCOST) (a) Given that higher cultivation costs reduce agricultural income, will the additional costs due to crop diversification reduce small farmers' income?. If so, how serious is this cost effect on farm income? ...124 (b) At the same time, the additional costs due to diversification can also be smaller in relation to the additional income from the same. If so, how important is this fact for crop choice? ..........................124 12. Agricultural Income (AGLINCOM) (a) While farm income is a necessary condition for food security, other non-income factors (e.g., food price and supply, its quality and composition, and family size) are also important. Given this, how important is the relative role of income in ensuring food security? ..................................................124 13. Land Quality and Soil Health (LANHELTH) (a) How important is land and soil health for land productivity, especially in the long-run? ......................124 (b) How important is the land and soil health for flexible crop choice? ..................................................124 62 14. Food Production (FOODPROD) (a) Normally, higher food production means more food supply in the market. But, export, procurement, and hoarding can reduce food availability. If so, how serious is this effect? .....................................124 (b) Similarly, higher food output means low food prices for consumers. But, the factors noted above may act against such price decline. If so, how serious is this effect? ..............................................124 15. Non-farm Enterprises (NFAMENTS) (e.g., small enterprises, petty trade, handicrafts, services) (a) Does labor scarcity affect farm wage rates? If so, how significant is this effect? ...............................124 (b) How important are non-farm activities for rural employment? ........................................................124 (c) Do non-farm activities create farm labor scarcity? If so how serious is this effect? ...........................124 16. Fodder and Feed Supply (FEDSUPLY) (e.g., rice straw, husks, and other farm by-products) (a) How important is the role of agriculture in supplying fodder and feeds? ...........................................124 (b) Does change in crop pattern (say from paddy to vegetables or oilseeds) will affect fodder supply? If so, how serious will be this negative effect? ..............................................................................124 (c) If the farm families with livestock rely on green fodder from public grazing lands and home gardens, crop pattern changes does not matter much. How realistic is this fact? ...........................................124 17. Livestock and Poultry (LIVSTOCK) (This does not relate to commercial enterprises, but only maintained by rural families) (a) How important are livestock & poultry for self-employment? ..........................................................124 (b) How important are livestock & poultry as an income source for small farmers? .................................124 (c) How important are livestock & poultry as an income source for farm workers and the poor? ..............124 (d) How important are livestock & poultry for the family consumption of milk & meat? ...........................124 (e) How important are livestock & poultry for the nutritional security of the children and aged? ...............124 18. Farm Income (FAMINCOM) (a) How food-secure are the small farmers? ......................................................................................124 (b) Is this security due to their cultivating food (paddy) crops? If so, how realistic is this fact? ................124 (c) Is food security role crops at the cost of crop diversification? If so, how realistic is this fact? .............124 19. Labor Income (LABINCOM) (a) How adequate are the wage income of rural workers to assure their food security? ...........................124 (b) How critical are the livestock and non-farm income sources for rural workers and the poor? ..............124 20. Food Availability (FOODAVAL) (a) How adequate is food availability to assure food security for rural workers and the poor? ..................124 63 21. Food Price (FOODPRIC) (b) How affordable are food prices to rural workers and the poor? ........................................................124 22. Land Tenure (LANTENUR) (Farm size; Tenure Security) (a) How important is farm size for adopting improved farm technologies and practices? .........................124 (b) How important is tenure security for adopting improved farm technologies and practices? .................124 (c) How important is land titles in securing farm credits? ....................................................................124 (d) How serious are small farms as constraints for efficient water delivery?............................................124 (e) Are smaller farms more efficient in water use? If so, how realistic is this fact? ..................................124 (f) Generally, small farms are unable to benefit from scale economies. If so, how serious is this fact in affecting their cultivation costs? .....................................................124 23. Water Institutions (WATINSTN) (Water release policy; allocation procedures) (a) How flexible is the water release policy for promoting diverse crops? ..............................................124 (b) How suitable are the existing water allocation practices for efficient water use? ................................124 24. Farm Input Institutions (FAMINSTN) (Credit, farm inputs, and extension institutions) (a) How effective and accessible is the farm credit system for small farmers? ........................................124 (b) How effective and accessible are the fertilizer and seeds supply systems for small farmers? ...............124 (c) How effective and accessible is the farm extension system for small farmers? ..................................124 (d) Are the farm input supply systems, including credit, too costly for small farmers? If so, how serious is this issue? ..................................................................................................124 (e) Are the farm input supply, including credit, focused on particular crops (e.g., paddy or coconut)? If so, how serious is this as a constraint for crop diversification? .....................................................124 25. Customary Institutions (CUSINSTN) (Local customs, conventions, traditions, and informal rules) (a) Normally, farmers' choice of food or traditional crops (e.g., paddy) is thought to be influenced by customary practices. If so, how limiting are local customs for crop diversification? ............................124 (b) How influential are local customs and conventions in water allocation and use decisions? ..................124 (c) Are there strong traditions in maintaining local commons as grazing areas for livestock? ...................124 26. Rural Development Policy (RDVPOLCY) (a) How effective are state policies in promoting rural non-farm activities? ...........................................124 (b) Are there special programs for developing specific non-farm enterprises (e.g., handicrafts; food processing units)? If so, how effective are they? ..........................................................................124 64 27. Market Institutions (MKTINSTN) (a) How effective are the agricultural markets in providing the right prices for farmers? .........................124 (b) How important is the role of traders and middlemen in the marketing of farm outputs? .....................124 (c) How effective are markets in stabilizing harvest and post-harvest price fluctuations? ........................124 (d) How effective is the procurement policy in supporting farm prices? .................................................124 28. Wage/Labor Legislations (WAGELAWS) (Legislations on wage rates and working conditions) (a) How effective are the minimum wage legislations in guiding rural wage rates? .................................124 (b) How strong are local customs and social pressures in influencing rural wage rates? ..........................124 (c) How effective are the special legal provisions (e.g., child labor; minimum working hour) in affecting rural labor supply and employment? ...........................................................................................124 29. Trade Policy (TRDPOLCY) (Farm import and export policies) (a) Do the trade policies on the import of milk and meat products limit livestock & poultry development? If so, how serious is this constraint? ............................................................................................124 (b) Do the trade policies on the import of food products add to domestic food availability? If so, how important is this policy for food and nutritional security? ................................................124 30. Price Regulations (PRICREGL) (a) How effective are price regulations in controlling the food prices for consumers? ..............................124 (b) Do price regulations distort agricultural markers? If so, how serious is this effect? ...........................124 31. Farm Subsidy Policy (SUBPOLCY) (Fertilizer and credit subsidies) (a) How effective are the subsidies for fertilizers and farm credits in reducing cultivations costs? .............124 (b) Do these subsidies have a favorable effect on farm income? If so, how significant are their effect? ....124 32. Samurdhi Policy (SAMPOLCY) (Special State program for Poverty alleviation) (a) How effective is the Samurdhi policy in supporting the income of the rural poor? ............................124 (b) How effective is the Samurdhi policy in improving the food availability to rural poor? .......................124 65 ANNEX C: DERIVING THE REDUCED FORM EQUATION Y1 = F1(X1, X 2) No Exogenous Variables Notation used 1 SYSREHAB X1 Y2 = F2(Y1, X 4) 2 LANTENUR X2 Y3 = F3(Y2, X 2, X3) 3 CUSINSTN X3 4 FAMINSTN X4 Y4 = F4(Y1, X 2, X3) 5 MKTINSTN X5 6 PRICREGL X6 Y5 = F5(Y3,Y4, X 4) 7 WAGELAWS X7 Y6 = F6(Y3,Y5, X 2) 8 RDVPOLCY X8 9 TRDPOLCY X9 Y7 = F7 (Y3,Y6, X 4) 10 SUBPOLCY X10 11 SUMPOLCY X11 Y8 = F8(Y3, X3) No Endogenous Notation used Y9 = F9(Y8, X9) Variables 1 BULKWATD Y1 Y10 = F10(Y3, X8) 2 CROPDIVR Y2 Y11 = F11(Y3,Y7 ) 3 CROPATEN Y3 4 WATINSTN Y4 Y12 = F12(Y10,Y11, X7 ) 5 WATPRODY Y5 6 LANHELTH Y6 Y13 = F13(Y7,Y9,Y10,Y12) 7 LANPRODY Y7 Y14 = F14(Y3,Y12, X 4, X10) 8 FEDSUPLY Y8 9 LIVSTOCK Y9 Y15 = F15(Y7,Y14, X5) 10 NFAMENTS Y10 11 LABPRODY Y11 Y16 = F16(Y9,Y10,Y15) 12 WAGERATE Y12 Y17 = F17 (Y9,Y10,Y13, X11) 13 RURALEMP Y13 14 CULTCOST Y14 Y18 = F18(Y3,Y5,Y7 ) 15 AGLINCOM Y15 Y19 = F19(Y18, X5, X9) 16 FAMINCOM Y16 17 LABINCOM Y17 Y20 = F20(Y18, X5, X6) 18 FOODPROD Y18 19 FOODAVAL Y19 Y21 = F21(Y16,Y17,Y19,Y20) 20 FOODPRIC Y20 21 FOODSECT Y21 Given these equations and their With the notations for all the exogenous sequential linkages depicted in Figure 6, and endogenous variables as assigned in the reduced form equation can be the above table, the 21 equations of the specified as a single but very long system model can be represented as equation shown below. follows: 66 Y21 = F21 F20 F18 F7 F6{ F5[ F4{F1[X1, X 2], X 2, X3}, F3{F2[F1(X1, X 2),X ], X 2, X3}, X 4 ], 4 F3[ F2[F1(X1, X 2),X ], X 2, X 3 ], X 2 }, F3{ F2[F1(X1, X 2 ),X ], X 2, X 3 },X , 4 4 4 F5 F4{F1[X1, X 2], X 2, X3}, F3{F2[F1(X1, X 2),X ], X 2, X3}, X 4 , 4 F3 F2[F1(X1, X 2), X 4],X ,X , 2 , X5, X6 3 F19 F18 F7 F6{ F5[ F4{F1[X1, X 2], X 2, X3}, F3{F2[F1(X1, X 2),X ], X 2, X3}, X 4 ], 4 F3[ F2[F1(X1, X 2 ),X ], X 2, X 3 ], X 2 }, F3{ F2[F1(X1, X 2 ),X ], X 2, X 3 },X , 4 4 4 F5 F4{F1[X1, X 2], X 2, X3}, F3{F2[F1(X1, X 2),X ], X 2, X3}, X 4 , 4 F3 F2[F1(X1, X 2), X 4],X ,X , 2 , X5, X9 3 F17 F13 F12 F11{ F7[ F6( F5{F4[F1(X1, X 2), X 2, X3],F3[F2(F1 < X1, X 2 >,X ), X 2, X3], X 4}, 4 F3{F2[F1(X1, X 2),X ], X 2, X 3}, X 2 ), F3( F2[F1(X1, X 2 ),X ], X 2, X 3 ),X ], 4 4 4 F3[ F2[F1(X1, X2), X4],X ,X 2 3 ] },F10{ F3{F2[F1(X1, X2),X4],X 2,X 3},X 8 },X 7 , F10 F3{F2[F1(X1, X2), X4],X 2,X 3},X 8 ,F9 F3{F2[F1(X1, X2), X4],X 2,X 3},X 9 , F7 F6( F5{F4[F1(X1, X 2), X 2, X3], F3[F2(F1 < X1, X 2 >,X ), X 2, X3], X 4}, 4 F3{F2[F1(X1, X 2),X ], X 2, X 3}, X 2 ), F3( F2[F1(X1, X 2),X ], X 2, X 3 ),X , 4 4 4 F10 F3{F2[F1(X1, X2), X4],X ,X },X 2 3 8 ,F9 F3{F2[F1(X1, X2), X4],X ,X },X 2 3 9 ,X 11 , F16 F15 F14 F12 F11{ F7[ F6( F5{F4[F1(X1, X2), X2, X3],F3[F2(F1 < X1, X2 >,X ),X ,X ], 4 2 3 X 4}, F3{F2[F1(X1, X 2),X ], X 2, X 3}, X 2 ), F3( F2[F1(X1, X 2),X ], X 2, X 3 ),X ], 4 4 4 F3[ F2[F1(X1, X2), X4],X ,X 2 3 ] },F10{ F3{F2[F1(X1, X2),X4],X 2,X 3},X 8 },X 7 , F3 F2[F1(X1, X 2), X 4],X ,X 2 3 ,X ,X 10 , 4 F7 F6( F5{F4[F1(X1, X 2), X 2, X3], F3[F2(F1 < X1, X 2 >,X ),X ,X ],X }, 4 2 3 4 F3{F2[F1(X1, X 2),X ], X 2, X 3}, X 2 ), F3{F2[F1(X1, X 2 ),X ], X 2, X 3},X ,X , 4 4 4 5 F10 F3{F2[F1(X1, X 2), X 4],X ,X },X 2 3 8 ,F9 F3{F2[F1(X1, X2), X4],X ,X },X 2 3 9 As can be seen, this single but long equation, which is representing the reduced form for the system model, has four main components, i.e., F20 (.), F19 (.), F17(.), and F16(.). Each of these components is based on the following general structure built using 12 different symbols for brackets: ..... ..... .......... .....{ ......[ .....( .....{....[.....(... < ..... > ...)....].....}.... )..... ]... }... .... ..... ..... .....