(Luận văn) income and income diversification of farm households in chau thanh a district , a case study of tan phu thanh village

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLANDS ng hi ep VIETNAM- NETHERLANDS PROJECT FOR M.A IN DEVELOPMENT ECONOMICS w n lo ad ju y th yi pl al n ua INCOME AND INCOME DIVERSIFICATION OF FARM va HOUSEHOLDS IN CHAU THANH A DISTRICT: n fu ll A CASE STUDY OF TAN PHU THANH VILLAGE oi m at nh z z ht vb Thesis submitted in partial fulfillment of the requirement for the degree of k jm Master of Arts in Development Economics an Lu Supervisor: MSc NGUYEN HUU DUNG om LE TAN NGHIEM l.c gm By n va ey t re t Ho Chi Minh City, December 2003 CERTIFICATION ng hi ep "I certify that the substance of this thesis has not already been submitted for any degree w and it is not being currently submitted for any other degree n lo ad I certify that, to the best of my knowledge, and help received in preparing this thesis, y th ju and all sources used, have been acknowledged in this thesis" yi pl Le Tan Nghiem n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re ACKNOWLEDGEMENT ng hi ep I would like to express my special thanks to my supervisor, MSc Nguyen Huu Dung for his patient attention, guidance, and precious advices in my completion of the thesis w n lo ad My sincere regards to Dr Mai Van Nam and Dr Le Khuong Ninh, the lecturers of ju y th School of Economics and Business Administration, Can Tho University for their enthu- yi siastic support in my study pl al n ua I would like to thank Ms Tran Thi Ngoc Huan, the researcher of Cuu Long Delta Rice n va Research Institute for her permission of using the data fu ll I wish to express my appreciation to all my teachers in the Vietnam-Netherlands Project oi m for Master Program in Development Economics for their knowledge and support My nh at thanks also go to the project staffs and all of my friends for their helpful advices z z vb Finally, I especially thank my family I would have never made this thesis without their jm ht loving and support k om l.c gm an Lu n va ey t re 11 ABSTRACT ng Diversification is a basic strategy that farm households in the Mekong River Delta rely hi ep on This study highlights this issue by using the sample of 217 farm households in Tan Phu Thanh Village, Chau Thanh A District, Can Tho Province The theory of farm w n household model and the literature on livelihood diversification are reviewed to set out lo ad factors affected on income diversification of farm households A log-linear model and a y th logit model are then employed to seek the determinants of farm households' income and ju yi income diversification in this area pl al n ua The study finds that diversification significantly increased the farm households' total n va income Participating in non-farm activities brings higher income to farm households ll fu Among significant influencing factors, labour ratio has a strongest positive effect on m farm households' income Finally, although land size is one of the most important inputs oi nh of farm production, its positive effect on the total income of farm households is small- at z est z ht vb Identifying the determinants of farm households' income diversification, the study em- jm k ployed the logit model of farm households' participation in non-farm activities There- gm l.c sults show that labour ratio is the most decisive factor inducing farm households to om work in non-farm activities It has the strongest positive effect on households' income an Lu diversification Farm households' decision to engage in non-farm activities is negatively affected by landowning Low agricultural income per capita pushes farm households to ey 111 t re ties n faced difficulties in agricultural production are more likely to take up non-farm activi- va stabilize their income by joining in non-farm work Finally, farm households recently TABLE OF CONTENTS ng hi CERTIFICATION i ep ACKNOWLEDGEMENT ii w n ABSTRACT iii lo ad TABLE OF CONTENTS iv ju y th LIST OF TABLES vii yi pl LIST OF FIGURES ix ua al n CHAPTER INTRODUCTION I va n 1.1 Statement ofthe problem fu ll 1.2 Objectives ofthe study oi m 1.3 Research questions nh at 1.4 Research methodology z z 1.5 Scope ofthe study vb jm ht 1.6 Organization ofthe study k CHAPTER LITERATURE REVIEW gm l.c 2.1 Relevant concepts om 2.1.1 Farm household an Lu 2.1.2 Sources of farm households' income 2.1.3 Income diversification 10 IV ey 2.3 Reviews of relevant studies 23 t re 2.2.2 Income diversification (Ellis, 2000) 16 n 2.2.1 Theory of farm household behaviours 11 va 2.2 Theoretical backgrounds of farm households 11 2.3.1 A Case study of rural Southern Mali (Abdulai and CroleRees, 2001) 23 2.3.2 A Case study ofNorthem Ethiopia (Woldehanna and Oskam, 2001) 24 2.3.3 A Case study ofEthiopia (Block and Webb, 2001) 25 ng hi CHAPTER ANALYTICAL FRAMEWORK AND MODEL SPECIFICATION 29 ep 3.1 Sources ofthe data 29 w 3.2 Sources of income, income diversification and income inequality 30 n lo ad 2.1 Classification of farm households' income 30 ju y th 2.2 Measurement of income diversification 31 yi 3.2.3 Measurement of income inequality 32 pl 3.3 Determinants of farm households' income 33 ua al 3.4 Determinants of farm households' participation in non-farm activities n va 4.1 Supplementary tests of the model 40 n fu ll 4.2 Marginal effects 41 oi m CHAPTER GENERAL INFORMATION ON CHAU THANH A DISTRICT nh at AND TAN PHU THANH VILLAGE 43 z z 4.1 Chau Thanh A District 43 vb jm ht 4.1.1 Geography 43 k 4.1.2 Overview of socioeconomic context 44 gm 4.2 Tan Phu Thanh Village 49 om l.c CHAPTER INCOME AND INCOME DIVERSIFICATION OF FARM an Lu HOUSEHOLDS IN TAN PHU THANH VILLAGE 51 5.1 Socioeconomic characteristics 51 n ey t re 5.3 Income diversification- Herfindahl index 54 va 5.2 Sources of farm households' income 52 t 5.3.1 Number ofhouseholds' income generating activities, households' characteristics and total income 55 v 5.3.2 Herfindahl index across quintiles of total income 56 5.4 fucome inequality of farm households 51 5.5 Factors influencing farm households' income 58 ng hi 5.5.1 Land size, labour ratio and households's income 58 ep 5.5.2 Access to credits, risks in farm production and households' income 59 w 5.5.3 Herfindahl index and households' income 60 n lo ad 5.5.4 Main features of pure farm households versus non-pure farm households 61 ju y th 5.6 Factors influencing farm households' participation in non-farm activities 62 5.6.1 Labour ratio and non-farm activities 63 yi pl 5.6.2 Land size and non-farm activities 64 ua al 5.6.3 Household size and non-farm activities 64 n va 5.6.4 Farm income per capita and non-farm activities 65 n fu 5.7 Farm households' income and its significant determinants 67 ll m oi 5.8 Households' participation in non-farm activities and its significant determinants 70 nh at CHAPTER CONCLUSIONS AND RECOMMENDATIONS 75 z z 6.1 Conclusions 75 vb jm ht 6.2 Policy recommendations 78 k REFERENCES 80 gm om l.c APPENDIX 84 an Lu n va ey t re Vl LIST OF TABLES ng hi ep Table 2.1: Typical Variables used in Empirical Studies Explaining Household Income w Diversification 27 n lo ad Table 3.1: Variables of the Farm Household Income Model 36 y th ju Table 3.2: Variables of Farm Households' Participation in Non-farm Activities .40 yi pl Table 4.1: Structure of Gross Domestic Product (GDP), 1995-2002 44 ua al n Table 4.3: Trend of Land per Agriculturist in Chau Thanh A, 2000-2002 .48 va n Table 5.1: Socioeconomic Characteristics of the Sample 52 ll fu oi m Table 5.2: Variety of Sources ofHouseholds' Income 53 nh at Table 5.3: Numbers of Income-generating Activities And Households Characteristics 55 z z Table 5.4: Herfindahl Index across Quintiles ofFarm Households' Total Income 56 ht vb k jm Table 5.5: Distribution of Households' Income across Diciles 57 gm Table 5.6: Land Size and Labour Ratio across Total Income Quintiles 58 l.c Table 5.7: Access to Credits and Risks in Production across Total Income Quintiles 59 om an Lu Table 5.8: Total Income across the quintiles ofHerfindahl Indices 60 n va Table 5.9: Some Different Features between Pure and Non-Pure Farm Households 61 t Vll ey Quintiles of Household's Labour Ratio 63 t re Table 5.10: Percentage of Households' Participation in Non-Agricultural Activities, by Table 5.11: Percentage of Farm Households' Participation in Non-Agricultural Activities, by Cohorts of Land Size 64 Table 5.12: Percentage of Households' Participation in Non-agricultural Activities, by ng hi Number of Household Members 65 ep Table 5.13: Percentage of Households' Participation in Non-agricultural Activities, by w n Quintiles of Farm Income Per Capita 66 lo ad ju y th Table 5.14: Determinants ofFarm Households' Income 68 yi Table 5.15: Logistic Regression Results for Determinants ofFarm Households' pl Participation in Non-farm Activities 711 ua al n Table 5.16: Estimated Marginal Effects at the Mean for Household's Participation in va n Non-farm Activities 72 ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re V111 LIST OF FIGURES ng hi ep Figure 2.1: Classification of Activities w n Figure 4.1: Structure of Employment in Chau Thanh A District, 2001-2002 45 lo ad Figure 4.2: Map ofChau Thanh A District .456 ju y th yi pl n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re IX The coefficient of the variable for labour ratio is positive and significantly different from zero, suggesting that households with higher ratios of working labours are more likely to earn income from non-farm activities This is consistent with Section 3.4.2 and Section ng hi 5.6.1 Among the significant factors, labour ratio has the strongest effect on the farm house- ep holds' participation in non-farm activities with the estimated marginal effect 0.30 (see Table w 5.16) This means that if the labour ratio increases percent at the mean, the probability of a n lo ad household participating in non-farm activities increases 0.30 This high probability of farm ju y th households to working in non-farm is resulted partly from the high dependent members ratios in households The mean labour ratio of 0.67 implies that every two labours had to sup- yi pl port one dependant in their farm households in this village n ua al n va Table 5.16: Estimated Marginal Effects at the Mean for Household's Participation in Non-farm Activities7 ll fu Means Coefficients Marginal effects oi m Variables Age 51.10377 Lab ratio 0.670594 at -0.0157555 Hhsize 5.169811 0.0710516 Credit 0.292453 -0 0803913 Land 0.644538 -0.9435935 Nlbinc 756.6982 -0.0000753 Risk 0.674528 1.120119 ** 0.0586415 Pcfinc 1143.551 -0.0006289 ** -0.0000329 nh -0.0008248 z 5.658177 *** 0.2962222 z -0.0042087 ** -0.0493999 k jm ht vb 0.0037198 gm -0.0000039 om l.c *denotes significant at the 0.10 level n va Notes: an Lu The average probability of participating in non-farm activities offarm households, P = 0.8396 ** denotes significant at the 0.05 level t re ey ***denotes significant at the 0.01 level t This Table is calculated by using the Section 3.4.4 72 The impact ofland size (area ofland cultivated) on households' decision on working off the farm is negative and conformable to Section 3.4.2 and Section 5.6.2 This is also consistent with the finding that larger landholding will generate higher income for farm ng hi households (see Section 5.8) The marginal effect ofland size on the decision to engage ep in non-farm of farm households is 0.05 This reflects the fact that farmers in the village w not have enough land to cultivate and the abundance in labours induces them to par- n lo ad ticipate in non-farm work y th ju Another significant factor causing farmers to seek income from non-farm work is the yi pl farm income per capita The negative coefficient of this factor shows that if the farm al n ua income per capita reduces, farm households will participate in non-farm activities Be- n va cause farm revenue is the main source of the farm households' income, fluctuations in ll fu agricultural income stir up the farmers' balanced budgets and hence they have to collect oi m income off-farm income to smooth their living However, this effect is small nh at Besides, farm households' participation in non-farm activities is depended on risks in z z agricultural production The positive effect of this factor explains that farm households vb jm ht will tackle the difficulties in farm production by joining in non-farm work This implies that farmers are risk-averse and it is consistent with Section 2.2.2 k l.c gm Combining the above factors will help us to explain the diversification of income more om logically High ratio of labours but the scarcity of land for production leads to labour an Lu abundance Agricultural production is risky and so reduces the farm income Therefore, n ey t re but much idle employment, and this in tum makes them more favor in off-farm activi- va households will simultaneously face the problems of inadequate money for expenditures ties 73 Out of the explanatory variables, household size, non-labour income, access to credits, and age of household head insignificantly affect on farm households' participation in non-farm activities Larger household size though statistically induces working in non- ng hi farm activities, its insignificance can be resulted from the effect of higher number of ep dependents Non-labour income and borrowed loans from banks are considered as sub- w stitutes of non-farm income source in farm households It is statistical to keep farm n lo ad households out of non-farm work However, since there exist some other options for ju y th farm to borrow money, i.e., access to informal credits, this cause the insignificant ef- yi fects of these variables on farm households' decision to participate in non-farm work pl Finally, insignificant effect of age of households' heads on working off the farm of farm ua al households is because deciding on participating in non-farm work of individual mem- n va bers is depended strongly on themselves rather than the heads' decision n ll fu oi m This model, in general, set out the determinants of farm households' participation in at nh non-farm activities Labour is the most decisive factor inducing the diversification of z income out of the farm Besides, factors relevant to agricultural production, particularly z ht vb land size, farm income per capita and risks forced farm households to participate in non- k jm farm activities These effects are consistent with theoretical expectations gm In summary, this chapter analyzed determinants of farm households' income and income l.c diversification for the case of Tan Phu Thanh Village Analyzing farm households' port- om an Lu folios of income, the study shows that non-farm income occupies the larger share of to- tal income Households engaged in non-farm activities obtained higher income than t 74 ey households' income and income diversification in the village t re derived Importantly, labour factor plays the most decisive role in generating farm n va others did not The higher level of diversification is associated with the higher income CHAPTER6 CONCLUSIONS AND RECOMMENDATIONS ng hi ep w 6.1 CONCLUSIONS n lo ad The study has investigated different income portfolios of farm households based on the y th cross-sectional data on 217 farm households in the Tan Phu Thanh Village Farm ju yi households are classified into pure and non-pure farm households in which pure farm pl ua al households are those not participate in non-farm activities Theories of farm house- n hold economics and livelihood diversification are first reviewed to form a theoretical va n background for the study A log-linear model of farm households' income and a logit fu ll model of farm households' participation in non-farm activities are then specified to un- oi m derstand the determinants of farm households' income and income diversification at nh z Farm households' total income are devided into farm income, non-farm income and z vb non-labour income On-farm income accounts for only one-third of the households' total jm ht income (31.83 percent) Among on-farm activities, fruit production provides the highest k l.c gm share of income derived from agriculture (15.51 percent) In the mean while, paddy production contributes a small share of income from agriculture This implies that in- om vesting in fruit production will bring higher profits to farm households than paddy pro- an Lu duction Therefore, changing cropping pattern from rice-monoculture production to n va paddy-fruit production will yield higher income for farm households households' total income Total income of the non-pure farm households is greater than 75 ey markably The data shows that non-farm income accounts for 64.33 percent of farm t re Participating in non-farm activities increases the farm households' total income re- two times that of the pure farm households on average This wide difference is resulted completely from amount of income derived from non-farm activities This is consistent with the result of econometric approach that if the pure farm households engage in non- ng farm work, their total income will increase 0.59 percent hi ep Farm households' income is strongly affected by labour ratio and area of cultivated land w n Out of the significant factors, labour ratio is the strongest positive effect of farm house- lo ad holds' income with the elasticity of 0.63 Similarly, farm households with larger land- y th holding will derive higher income Therefore, it is clear that labour and land are the two ju yi important inputs of farm production; and expanding area of cultivated land and/or grow- pl ua al ing in labour ratio will improve the farm income and households' total income as well n In addition, households with higher labour ratio can allocate their labour endowment in va n more activities, including non-farm activities and then increase their tot~l income ll fu m oi The most important finding is that diversification is positively associated with the farm nh households' total income Descriptive analysis illustrates the positive relationship be- at z tween the number of activities farm households engaged in and the total income de- z ht vb rived The econometric approach also highlights this significant effect This implies that k jm the more diversified the household, the higher income the household derived Conse- l.c gm quently, diversifying farm households' endowments into different activities is considered as an effective strategy for farm households to increase their income om an Lu Access to credit and difficulties in farm production are insignificant factors of farm households' income in this case It is even more interesting that risks in farm production ey t re ofhigher non-farm income, which is derived from participating in non-farm activities n va increases the farm households' total income This can be explained by the offset effect t 76 There is a number of factors influencing income diversification of farm households The findings show that labour ratio, land size, farm income and risks in agricultural production are significant factors affected on farm households' participation in non-farm activi- ng hi ties ep Labour ratio is the strongest positive factor inducing farm households to participate in w n non-farm activities This means that farm households with higher labour ratio are more lo ad likely to derive additional amounts of income from working in non-farm work The y th ju positive effect of labour ratio on both farm households' income and income diversifica- yi tion implies that given the recent land endowment, agricultural activities cannot absorb pl ua al full labour endowment of farm households Therefore, when labour ratio increases, farm n households' labours need to work off the farm n va fu ll The study also points out the negative effect of landholding on farm households' deci- m oi sion to participate in non-farm activities This means that households with higher area nh at of cultivated land are ready to work in farm activities only The positive effect of land- z holding on farm households' income is useful to explain for their choice of working in z vb farm work only Because of larger landowning, most of farm households' labour en- jm ht dowment is used on their farms and so, they are unlikely to work off the farm k gm l.c Besides, low level of agricultural income per capita is also a factor pushing farm house- om holds' labours to participate in non-farm activities Farm income is the principal source an Lu that farm households rely on for both food consumption and in-cash expenditures Due ey t re expected, they have to work in non-farm to balance their income n va to this necessity, whenever this amount of income is lower than that of farm households Lastly, farm households' taking up non-farm activities is partly affected by risks in agricultural production The result shows that farm households recently faced risks in their 77 production are more likely to engage in non-farm activities Since the marginal effect of risk on farm households' participation in non-farm activities is relatively strong, it is essential to include risk in to study its effect on income diversification ng hi ep In conclusion, farm households' income diversification is affected by labour ratio, land size, agricultural income and difficulties in non-farm production Farm households' in- w n come is affected by labour ratio, land size, participating in non-farm activities and espe- lo ad cially level of diversification hnportantly, the main finding of the study is that diversi- y th ju fication increases significantly farm households' income Based on these findings, the yi following section suggests some recommendations for local authority in making strate- pl n ua al gies on the development of Chau Thanh A District va 6.2 POLICY RECOMMENDATIONS n ll fu oi m In the context of the limitation of cultivated land but labour abundance, diversifying in- at nh come-generating activities is a predominant strategy to absorb the labour and to secure z the farm households' subsistence income The diversification can be applied in both z jm ht vb farm and non-farm sectors k In agriculture, farm households can increase their total income by maximizing profits gm from diversified activities on their farm Since income from only paddy production l.c (even intensive cultivation) cannot enrich farm households, taking full advantage of om an Lu land use in generating more agricultural sources of income will improve their livings A part of their landowning should be used for fruits production In addition, farmers 78 ey paddy and vegetable, three crops of premature maize and beefs breeding, etc These sys- t re freshwater aquatic products (anabas, snake-heads, catfish, etc.), paddy and prawn, n va should apply some systems ofrotational crops on their farming land, such as: paddy and terns will increase farm households' income and keep land fertilized Diversifying such activities also mitigate risk and reduce the variation in farm households' income ng hi In the Mekong River Delta, since most of farm households are holding small areas of ep cultivated land (0.1-0.2 ha/person) and farm production is scattered and inefficient, it is w more effective to fuse small farms into a large farm to cultivate in combinations of pro- n lo duction With a large farm, co-farmers can be active in irrigation, technology, seeds, ad ju y th capital, etc and selling their products Therefore, the final profits per area of cultivated land in combinations of production will be higher than that in individual farm house- yi pl holds' production ua al n Besides, expanding non-farm activities is more convincible Since non-farm activities va n are considered as lower risk and higher revenue than farm activities, working in non- ll fu oi m farm can stabilize and increase income of farm households The local authority should nh stimulate the development of industrial establishments in the district The development at of foodstuff processing industry will not only absorb the agricultural outputs as its in- z z puts but also provides jobs for local labours Therefore, farm households' labours, espe- vb jm ht cially the idle labours in farm households can work for these manufactories and hence k increase their total income l.c gm In general, income diversification is necessary for farm households to increase their to- om tal income This study has already pointed out some effects of households' characteris- an Lu tics on income diversification of farm households However, the study still left some va other factors such as interternporal effects of these factors on diversification, effects of n study these effects in a further research 79 ey (i.e migration from rural to urban for seeking income, etc.) Therefore, it is better to t re macro-policies, etc as well as effects of income diversification on social management REFERENCES ng hi ep Abdulai, A and A CroleRees (200 1) 'Determinants of Income Diversification amongst Rural Households in Southern Mali', Working Paper Zurich: Swiss Federal In- w n stitute ofTechnology lo ad y th Ban Vat Gia Chinh Phu (Governmental Unit of Price Management), 2000 'Tai Lieu ve ju yi Kinh Te Trang Trai (Document of Farming Economics) Ho Chi Minh City: Ho pl n ua al Chi Minh Publisher n va Barrett, C.B and T Reardon (2001) 'Assets, Activities, and Income Diversification ll fu among African Agriculturalists: Some Practical Issues' Project Report to United oi m Stated Agency for International Development (USAID) nh at Barrett, C.B., Bezuneh, M and A Aboud (2001a) 'Income Diversification, Poverty z z Traps and Policy Shocks in Cote d'Ivoire and Kenya' Working Paper jm ht vb Barrett, C.B., Bezuneh, M., Clay, D.C and T Reardon (eds) (2000) 'Heterogeneous k l.c gm Constraints, Incentives and Income Diversification Strategies in Rural Africa' Project Report to US AID om an Lu Barrett, C.B., Reardon, T and P Webb (2001 b) 'Non-farm Income Diversification and Household Livelihood Strategies in Rural Africa: Concepts, Dynamics and Pol- va n icy Implications' Working Paper Famine Ethiopia' Working Paper Tufts University 80 ey t re Block, S and P Webb (2002) 'The Dynamics of Livelihood Diversification in Post- Can Tho Statistical Office (CSO) (2002), Statistical Yearbook 2002, Can Tho City: Can Tho Statistical Office Press ng Chau Thanh A Statistical Office (2002), Statistical Yearbook 2002, Can Tho City: Can hi ep Tho Statistical Office Press w n Dang Kim Son (1998) 'Development of Agricultural Production Systems in the Mekong lo ad Delta', in Vo Tong Xuan and S Matsui (eds.) Development of Farming Systems ju y th in the Mekong Delta of Vietnam Ho Chi Minh City: Ho Chi Minh City Publish- yi ing House pl al n ua Ellis, F (1993) Peasant Economic: Farm Households and Agrarian Development Sec- n va ond edition Cambridge: Cambridge University Press fu ll Ellis, F (2000) Rural Livelihoods and Diversity in Developing Countries New York: oi m Oxford University Press at nh z Gujarati, D.M (1995), Basic Econometrics, 3rd edition McGraw-Hill International Edi- z vb tion jm ht k Hoddinott (1994) 'A Model of Migration and Remittances Applied to western Kenya' om l.c gm Oxford Economic Paper, Volume 57, No.1 Iwamoto, Izumi (2001) 'The Household Economy and the Diversification of Farming in an Lu Vietnam' in CHO Kenji and YAGI Hironori (eds) (200 1) Vietnamese Agricul- va ture Under Market-Oriented Economy Vietnam: Agricultural Publishing House, n t re pp 169-88 ey Kozel, V (1990) The Composition and Distribution of Income Washington, D.C: The World Bank 81 Larson and Mundlak (1997) 'On the Intersectoral migration of Agricultural Labour' Economic Development and Cultural Change, Vol 45, No.2 ng LeVan Phung, Tran Thi Tuyet and Tran Van Hung (1999) 'Changing Occupations in hi ep Rural Areas' (Thay doi nghe nghiep cac khu vue nong thon), in Haughton, D., Haughton, J., Bales, S., Truong Thi Kim Chuyen, and Nguyen Nguyet Nga w n lo Vietnamese Households -A Quantitative Analysis (Ho gia dinh Vietnam - Nhin ad qua phan tich dinh Iuong) Hanoi: The National Political Publisher ju y th yi Nguyen Van Sanh, Vo Tong Xuan and Tran Anh Phong (1998) 'History and Future of pl ua al Farming Systems in the Mekong Delta' in Vo Tong Xuan and S Matsui (eds.) n Development of Farming Systems in the Mekong Delta of Vietnam Ho Chi Minh va n City: Ho Chi Minh City Publishing House ll fu m Norton (1993) Introduction to Economics of Agricultural Development New York: oi at nh McGraw-Hill Inc z z Reardon, T (1997) 'Using Evidence of Household Income Diversification to Inform ht vb Study of the Rural Non-farm Labour Market in Africa,' World Development jm k 25(5): 735-748 l.c gm Reardon, T., Barrett, C., Kelly, V., and K Savadogo (1999) 'Policy Reforms and Sus- om tainable Agricultural Intensification in Africa' Development Policy Review 17 an Lu (4) December ber): pp 1172-76 82 ey tive,' American Journal of Agricultural Economics Volume 76, No (Decem- t re Farm investment in African Households: Adding the Capital Market Perspec- n va Reardon, T., Crawford, E and V Kelly (1994) 'Links between Non-farm Income and Reardon, T., Stamoulis, K., Balisacan, A., Cruz, M.E., Berdegue, J., and B Banks (1998) 'Rural Non-farm Income in Developing Countries,' Special Chapter in The State of Food and Agriculture 1998 Rome: Food and Agricultural organiza- ng hi tion of the United Nations ep Savadogo, K., Reardon, T and K Pietola (1998) 'Adoption of Improved Land-Use w n Technologies to Increase Food Security in Burkina Paso: Relating Animal Trac- lo ad tion, Productivity and Non-farm Income' Agricultural Systems 598 ju y th yi The People Committee of Chau Thanh A District (PCC) (2002a) Report on Socio- pl economic Development Strategy in the period of 2005 - 2010 (Bao cao Dinh ua al huong Phat trien Kinh te - Xa hoi giai an 2005 - 201 0) n n va ll fu The People Committee of Chau Thanh A District (PCC) (2002b) The Master Plan on oi m Socio-economic Development of Chau Thanh A District, Can Tho Province up to at nh 2010 (Quy Roach Tong The Phat Trien Kinh Te- Xa Hoi Huyen Chau Thanh A Tinh Can Tho Thoi Ky Den Nam 2010) z z vb jm ht Vo Quang Minh, Le Quang Tri and Ryuichi Yamada (2002) 'GIS and Multicriteria k Evaluation Techniques for Land Use Allocation: A Case Study of Tan Phu gm Thanh Village, Chau Thanh A District, Cantho Province', in JIRCAS-CTU- l.c CLRRI-SOFRI (2002) Development of New Technologies and Their Practice for om Sustainable Farming Systems in The Mekong Delta pp 302-15 Proceedings of an Lu the 2002 annual Workshop of JIRCAS Mekong Delta Project Can Tho: Can Tho va n University ey t re 83 APPENDIX Appendix 1: OLS Regression Results of The Linear Model of Farm Households' ng hi income (Equation 3.3) ep reg income land labratio divlevel hhtype credit risk if income F R-squared Adj R-squared Root MSE lo ad \' ju y th 212 13.43 0.0000 0.2822 0.2612 9875.6 yi pl income I Coef Std Err t [95% Conf Interval) P>ltl -+ -land 2513.942 1489.218 69 0.093 -422.206 5450.09 labratio 16664.84 3445.924 4.84 0.000 9870.844 23458.83 divlevel -7998.82 3299.608 -2.42 0.016 -14504.34 -1493.302 hhtype 8191.448 2045.701 4.00 0.000 4158.137 12224.76 credit -1971 99 1507.594 -1.31 0.192 -4944.368 1000.388 risk 25.76354 1546.969 0.02 0.987 -3024.247 3075.774 cons 4777.586 3684.123 30 0.196 -2486.044 12041.22 n ua al n va ll fu oi m hettest nh at Cook-Weisberg test for heteroskedasticity using fitted values of income Ho: Constant variance chi2(1} 6.95 Prob > chi2 0.0084 z z ht vb jm ovtest k Ramsey RESET test using powers of the fitted values of income Ho: model has no omitted variables F(3, 202) 0.36 Prob > F = 0.7812 om l.c gm an Lu n va ey t re 84 Appendix 2: OLS Regression Results of The Log-Linear Model of Farm Households' income (Equation 3.3) ng reg lnincome lnland lnlbratio lndivlevel hhtype credit risk if lnincome >7.5(*) hi df ss MS Source I -+ ep Model Residual I I 28.6865518 56.2558895 203 w Number of obs F( 6, 203) Prob > F R-squared Adj R-squared Root MSE 4.78109196 277122609 -+ I lnincome I 84.9424413 209 406423164 n Total 210 17.25 0.0000 0.3377 0.3181 52642 lo Coef Std Err P>ltl t [95% Conf Interval) ad -+ ju y th lnland lnlbratio lndivlevel hhtype credit risk cons yi pl 1012844 6336913 -.2333736 5680249 -.1005477 0348446 9.47448 0499973 1180181 0836186 1169201 0812322 0845483 1542244 2.03 5.37 -2.79 4.86 -1.24 0.41 61.43 0.044 0.000 0.006 0.000 0.217 0.681 0.000 0027039 4009927 -.3982459 3374912 -.2607148 -.1318608 9.170393 199865 8663899 -.0685012 7985585 0596194 2015501 9.778567 al (*) using if to drop some outliners in the sample n ua Note: n va hettest Cook-Weisberg test for heteroskedasticity using fitted values of lnincome Ho: Constant variance chi2(1) 0.00 Prob > chi2 0.9958 ll fu m oi ovtest Ramsey RESET test using powers of the fitted values of lnincome Ho: model has no omitted variables F(3, 200) 0.26 Prob > F = 0.8533 at nh z coef z corr, lndivlevel vb lnland lnlbratio hhtype credit risk cons Variable VIF 1/VIF Mean VIF I 1.14 85 ey -+ t re 0.849343 0.832815 0.928125 0.811028 0.989933 0.863774 n 1.18 1.20 08 23 01 1.16 1.0000 va lnland lnlbratio lndivlevel hhtype credit risk 1.0000 -0.3308 an Lu I -+ 1.0000 -0.0180 -0.1234 om vif 1.0000 -0.0395 -0.0852 -0.5881 l.c 1.0000 0.0345 -0.0401 0.1867 0.2925 gm 1.0000 -0.0719 -0.3935 0.0669 -0.0483 -0.5585 k 1.0000 -0.1087 1.1277 0.2176 -0.0699 -0.2788 -0.2137 jm lnland lnlbratio lnherfind hhtype credit risk cons ht -+ - Appendix 3: Logit Regression Results of Farm Households' Participation in Nonfarm Activities (Equation 3.4) ng hi legit hhtype age labratio nlbinc credit hhsize risk pcfinc land if pcfinc chi2 Pseudo R2 Logit estimates w n Log likelihood= -70.169123 212 46.35 0.0000 0.2483 lo ad -hhtype I Coef Std Err z [95% Conf Interval] P>lzl y th -+ ju yi -.0157555 5.658177 -.0000753 - 0803913 0710516 1.120119 -.0006289 -.9435.935 -.4125802 pl n ua al age labratio nlbinc credit hhsize risk pcfinc land cons 016181 1.279547 000064 4902822 1580709 478606 0002497 4477035 309376 -0.97 4.42 -1.18 -0.16 0.45 2.34 -2.52 -2.11 -0.32 -.0474697 3.15031 -.0002008 -1.041327 -.2387617 1820684 - 0011184 -1.821076 -2.97891 0.330 0.000 0.240 0.870 0.653 0.019 0.012 0.035 0.753 0159588 8.166044 0000502 8805442 3808649 2.058169 -.0001394 - 0661109 2.153749 va n lfit fu ll Logistic model for hhtype, goodness-of-fit test oi m 212 212 204.77 0.4520 at nh number of observations number of covariate patterns Pearson chi2(203} Prob > chi2 z z lstat vb I Total 24 10 1 198 14 -+ + Total 178 34 I 212 om Pr ( + 1-D} Pr( -1 D) Pr(-DI +} Pr( Dl -} 70.59% 2.25% 12.12% 28.57% - rate for true D + rate for classified + - rate for classified - 86.79% Correctly classified 86 ey False False False False t re 97.75% 29.41% 87.88% 71.43% n Pr( +I D) Pr( -1-D} Pr( Dl +} Pr(-DI -} va Sensitivity Specificity Positive predictive value Negative predictive value an Lu Classified + if predicted Pr(D} >= True D defined as hhtype -= + rate for true -D l.c 174 + gm -+ + - k True -D -D jm Classified I ht Logistic model for hhtype