The Impact of Participatory Projects on Social Capital:

Transcription

1 The Impact of Participatory Projects on Social Capital: Evidence from Farmland Consolidation Projects in Japan Taisuke Takayama (Meikai University) Tomoaki Nakatani (Hokkaido University) October 22,

2 Contents 1. Introduction 2. Farmland Consolidation Projects and Social Capital 3. Data and Estimation Model 4. Empirical Results 5. Concluding Remarks 2

3 Introduction Participatory Development Projects (e.g. Community- Driven Development Projects) Beneficiaries participate through community development projects and local decentralization Advantages: targeting of the poor, reduced project costs, improved project maintenance, allocative efficiency, and accumulation of social capital (SC) Help overcome major obstacle to economic development 3

4 Introduction (cont.) The determinants of SC remain poorly understood (Gugerty and Kremer 2002, Miguel et al. 2006). It is difficult to make generalizations about institutional impact: World Bank 2002; Casey et al The effects of participatory development on SC in each project area have been mixed: Gugerty and Kremer 2002; World Bank 2002; Vajja and White 2008; Labonne and Chase 2011; Casey et al. 2012; Feigenberg et al Projects in developing countries Indicators of SC 4

5 Participatory Development Projects in Japan Farmland Consolidation Projects (FCPs) in paddy fields Improving labor and land productivity Nurturing the development of core farmers Farmland readjustment and infrastructure development Sources: Taisetsu LID 5

6 Participatory Development Projects in Japan (cont.) Implementation process for FCPs in rural communities Proposal from farmers in community Public project by the central or prefectural government Central Government or Prefecture Agreement by more than 2/3 of landowners Project area Agreement by more than 2/3 of landowners in the project area Rural community B Rural community C Decision of the project area Proposal by more than 15 beneficiary farmers Rural community A More than 15 beneficiary farmers Sources: MAFF 6

7 Research Objective Evaluate the impact of Farmland Consolidation Projects (FCPs), a type of participatory development project in Japan, on community-level bonding and bridging social capital by using propensity score Social Capital Bonding SC: the ease with which farmers within a community act collectively Bridging SC: the ease with which farmers act collectively with farmers or stakeholders in other communities 7

8 Contents 1. Introduction 2. Farmland Consolidation Projects and Social Capital 3. Data and Estimation Model 4. Empirical Results 5. Concluding Remarks 8

9 Positive Effects of FCPs on Bonding SC Before FCPs After FCPs Irrigation canal Public projects under landowners agreements Reshaping adjacent farmland Coordination among a large number of landowners Farm road Farm road Sources: AAO Accumulation of bonding SC through consensus-building in project areas during the project implementation process Nurturing core farmers in project areas Obligation for communities to hold community meetings to obtain consensus on future farmland use Resolution of fragmentation 9

10 Negative Effects of FCPs on Bonding SC FCPs Simplify irrigation canal and farm road maintenance Reduce opportunities for collective action Before FCPs After FCPs Irrigation canal Farm road Farm road It is not known which are greater: the positive or negative effects of FCPs on bonding social capital. 10

11 Positive Effects of FCPs on Bridging SC Central Government or Prefecture Agreement by more than 2/3 of landowners in the project area Rural community B Project area Rural community C Decision of the project area Proposal by more than 15 beneficiary farmers Rural community A More than 15 beneficiary farmers Horizontal and vertical networks Between communities Close relationships with relevant organizations Accumulation of bridging SC 11

12 Contents 1. Introduction 2. Farmland Consolidation Projects and Social Capital 3. Data and Estimation Model 4. Empirical Results 5. Concluding Remarks 12

13 Estimation Model Average Treatment Effect on the Treated (ATT) (1) Y i (0) Y i D i ATT E Y i (1) Yi (0) Di 1 (1) D 1 E Y (0) D 1 E Y i i :Social capital in community i if community i implements FCP :Social capital in community i if community i does not implement FCP :Dummy variable equal to 1 if community i implements an FCP and 0 otherwise Propensity Score Matching: PSM(Rosenbaum and Rubin 1983) ATT i 1) D 1, P( X ) E Y (0) D 0, P( X ) E Y i ( i i i i i i P ( X i ) : propensity score and is estimated via a probit or logit model 13

14 Our Data Community-level data World Census of Agriculture and Forestry 2000 and 1990, by the Ministry of Agriculture, Forestory, and Fisheries Information on agriculture and forestry at the prefecture, municipality, and rural community levels Unit of Observation: Rural Community FCPs are targeted at the rural community level Smallest unit of regional society in Japan Effects are reflected in the rural community-level agricultural data. 14

15 Indicators of FCP implementation Binary indicator of project implementation Equals 1 if the area of readjusted farmland increased and the ratio of readjusted farmland increased by more than a specific number of percentage points from 1990 to 2000; and 0 otherwise. Area of Readjusted Farmland Area of readjusted farmland in 1990 < Area of readjusted farmland in 2000 Ratio of Readjusted Farmland Farmland in Community In 1990 In 2000 < Area of readjusted farmland in

16 Indicators of FCP Implementation (cont.) Robustness check 4 Categories for projects implementation Variables equal 1 if the area of readjusted farmland increased and the ratio of readjusted farmland increased by Farmland in Community (1) More than 0 percentage points in 1990 in 2000 (2) More than 50 percentage points (3) More than 75 percentage points (4) 100 percentage points in 2000 in 2000 in 2000 Area of readjusted farmland in 2000 Area of readjusted farmland in 2000 Area of readjusted farmland in 2000 Area of readjusted farmland in

17 Indicators of SC SC: Intangible and difficult to quantify. Use proxy variables by applying Principal Component Analysis (PCA) (Fujiie et al. 2005) Variables that represent interactions Within the community Bonding SC Between communities and with other organizations Bridging SC Divide SC into agriculture- and non-agriculture-related Rural communities(smallest unit of society) have various groups and social relationships FCPs affects non-agriculture-related activities as well as agriculture-related activities 17

18 Indicators of Bonding SC Agriculture-related Number of meetings held by farmers Number of agriculture-related organizations for Young people Women The elderly Management method (All villagers = 4, only farmers = 3, employees = 2, not implemented = 1) Irrigation canals Farm roads Non-agriculture-related Number of non-agriculturerelated organizations for Young people Women The elderly Management method (All villagers = 4, only farmers = 3, employees = 2, not implemented = 1) Common facilities 18

19 Indicators of Bridging SC Agriculture-related Binary variable for whether the community provides programs that allow urban residents to Experience agriculture, forestry, and fisheries Directly sell agricultural products to urban residents Benefit from study-away opportinities in the community Non-agriculture-related Binary variable for whether the community provides programs that allow urban residents focusing on Traditional arts and crafts Traditional festivals 19

20 Sample Size (Treated and Untreated Communities) (1) More than 0% Ratio of Readjusted Farmland Increased by (2) More than 50% (3) More than 75% (4) 100% Treated 14,007 10,998 8,691 5,803 Untreated 34,190 34,190 34,190 34,190 Total 48,197 45,188 42,881 39,993 We only include rural communities(48,197 out of 140,122) whose ratios of readjusted farmland in 1990 are 0%. 20

21 Descriptive Statistics: Community Characteristics Variable Obs. year Mean S.D. Agricultural area (urban) 48, Agricultural area (intermediate) 48, Agricultural area (mountainous) 48, Distance to DID (0.5 to 1hr) 48, Distance to DID (more than 1hr) 48, Ratio of elderly farmers 48, Ratio of part-time farm households 48, Number of farm households 48, Gradient (flat) 48, Gradient (gentle) 48, Agricultural promotion area 48, Agricultural promotion area (farmland) 48, City planning area (urbanization promotion area) 48, City planning area (urbanization control area) 48, City planning area (not designated) 48, Social capital ('90) 48, Note: DID refers to a densely inhibited district (city/town/village). Social capital ('90) refers to the social capital of the community in

22 Descriptive Statistics: Specific Outcomes Variable Obs. year Mean S.D. Specific outcomes (agriculture-related bonding social capital) Irrigation management 36, Farm road management 36, Number of meetings 36, Number of agriculture-related organizations for young people 36, Number of agriculture-related organizations for women 36, Number of agriculture-related organizations for the elderly 36, Specific outcomes (non-agriculture-related bonding social capital) Common facilities management 33, Number of non-agriculture-related organizations for young people 36, Number of non-agriculture-related organizations for women 36, Number of non-agriculture-related organizations for the elderly 36, Specific outcomes (agriculture-related bridging social capital) Programs to experience agriculture, forestry, and fisheries 36, Direct sale of agricultural products 36, Programs for temporary transfers to rural communities 36, Specific outcomes (non-agriculture-related bridging social capital) Programs to experience traditional arts and crafts 36, Programs to experience traditional festivals 36,

23 Contents 1. Introduction 2. Farmland Consolidation Projects and Social Capital 3. Data and Estimation Model 4. Empirical Results 5. Concluding Remarks 23

24 Probit Estimates for FCP Implementation Marginal effects S.E. Agricultural area (urban) *** Agricultural area (intermediate) *** Agricultural area (mountainous) *** Distance to DID (0.5 to 1hr) *** Distance to DID (more than 1hr) *** Gradient (flat) *** Gradient (gentle) *** Agricultural promotion area *** Agricultural promotion area (farmland) *** City planning area (Urbanization promotion area) *** City planning area (Urbanization control area) *** City planning area (not designated) *** Ratio of elderly farmers *** Ratio of part-time farm households *** Number of farm households *** Social capital ('90) Observations 48,197 LR chi 2 (16) 3, Log likelihood -27, Pseudo R

25 Impact of FCPs on Bonding SC: Propensity score estimates Before One-to-one NN Radius Kernel Agriculture-related bonding social capital *** *** *** *** Treated (more than 0%) vs Untreated (0.014) (0.023) (0.013) (0.014) 36, *** *** *** *** Treated (more than 50%) vs Untreated (0.014) (0.022) (0.016) (0.014) 33, *** *** *** *** Treated (more than 75%) vs Untreated (0.014) (0.026) (0.018) (0.015) 31, *** *** *** *** Treated (100%) vs Untreated (0.018) (0.026) (0.021) (0.019) 29,305 Non-agriculture-related bonding social capital Treated (more than 0%) vs Untreated *** (0.014) (0.023) (0.014) (0.015) *** Treated (more than 50%) vs Untreated (0.016) (0.023) (0.016) (0.017) ** Treated (more than 75%) vs Untreated (0.016) (0.023) (0.019) (0.017) Treated (100%) vs Untreated (0.019) (0.033) (0.021) (0.019) Obs. 36,492 33,874 31,887 29,305 25

26 Impact of FCPs on Bridging SC: Propensity score estimates Before One-to-one NN Radius Kernel Agriculture-related bridging social capital Treated (more than 0%) vs Untreated (0.011) (0.018) (0.012) (0.014) 36, Treated (more than 50%) vs Untreated (0.013) (0.021) (0.013) (0.013) 33, ** Treated (more than 75%) vs Untreated (0.015) (0.021) (0.014) (0.015) 31, Treated (100%) vs Untreated (0.018) (0.029) (0.020) (0.018) 29,305 Non-agriculture-related bridging social capital Treated (more than 0%) vs Untreated *** ** (0.015) (0.021) (0.013) (0.015) 36, *** ** *** *** Treated (more than 50%) vs Untreated (0.014) (0.022) (0.016) (0.014) 33, *** *** *** *** Treated (more than 75%) vs Untreated (0.015) (0.022) (0.017) (0.016) 31, *** *** *** *** Treated (100%) vs Untreated 29, (0.016) (0.027) (0.020) (0.016) Obs.

27 Impact of FCPs on Bonding SC (Specific Outcomes): Propensity score estimates Treated (more than 0%) vs Untreated Agriculture-related bonding social capital Irrigation management(all villagers = 4, only farmers = 3, employees = 2, not implemented = 1) Before One-to-one NN Radius Kernel *** *** *** *** (0.014) (0.019) (0.014) (0.015) Farm road management ** (0.018) (0.024) (0.014) (0.019) Number of meetings *** *** *** *** (0.072) (0.108) (0.081) (0.074) Number of agriculture-related organizations for young people Number of agriculture-related organizations for women Number of agriculture-related organizations for the elderly (0.002) (0.003) (0.002) (0.002) (0.005) (0.007) (0.005) (0.005) *** (0.002) (0.003) (0.002) (0.002) 27

28 Impact of FCPs on Bonding SC (Specific Outcomes): Propensity score estimates(cont.) Treated (more than 0%) vs Untreated Non-agriculture-related bonding social capital Management of common facilities Before One-to-one NN Radius Kernel *** *** *** *** (0.010) (0.015) (0.010) (0.011) Number of non-agriculture-related organizations for young people *** (0.006) (0.010) (0.007) (0.007) Number of non-agriculture-related organizations for women ** ** ** (0.007) (0.013) (0.007) (0.007) Number of non-agriculture-related organizations for the elderly *** ** ** (0.007) (0.012) (0.007) (0.007) 28

29 Impact of FCPs on Bridging SC (Specific Outcomes): Propensity score estimates Treated (100%) vs Untreated Agriculture-related bridging social capital Programs to experience agriculture, forestry, and fisheries Before One-to-one NN Radius Kernel (0.001) (0.002) (0.002) (0.001) Direct sale of agricultural products ** *** *** Programs for temporary transfers to rural communities (0.002) (0.004) (0.003) (0.003) (0.001) (0.001) (0.001) (0.001) Non-agriculture-related bridging social capital Programs to experience traditional arts and crafts (0.002) (0.004) (0.002) (0.002) Programs to experience traditional festivals *** ** ** ** (0.003) (0.005) (0.004) (0.004) 29

30 1. Introduction Contents 2. Farmland Consolidation Projects and Social Capital 3. Data and Estimation Model 4. Empirical Results 5. Concluding Remarks 30

31 Concluding Remarks We examine the impact of Farmland Consolidation Projects on social capital in Japan. Main results Positive impact on agriculture-related bonding social capital Number of community meetings and management of common-pool resources Negative impact on non-agriculture-related bridging social capital Allowing for the direct sale of agricultural products and holding programs for city residents to experience festivals Promotion of agricultural activities and discouragement of nonagricultural activities 31

32 Definition of Variables Variable Characteristics Agricultural area (urban) Agricultural area (intermediate) Agricultural area (mountainous) Distance to DID (0.5 to 1hr) Distance to DID (more than 1hr) Definition Dummy: 1 if the agricultural area is classified as an urban area, 0 otherwise Dummy: 1 if the agricultural area is classified as a flat agricultural area, 0 otherwise Dummy : 1 if the agricultural area is classified as an intermediate agricultural area, 0 otherwise Dummy: 1 if the time distance to a densely inhibited district (city/town/village) is 0.5 to 1 hour, 0 otherwise Dummy: 1 if the time distance to a densely inhibited district (city/town/village) is more than 1 hour, 0 otherwise Ratio of elderly farmers Denominator = total population engaged in farming Ratio of part-time farm households Denominator = total number of farm households Number of farm households Total number of farm households Gradient (flat) Dummy: 1 if the gradient is smaller than 1/100, 0 otherwise Gradient (gentle) Dummy: 1 if the gradient is 1/100 to 1/20, 0 otherwise Agricultural promotion area Dummy: 1 if the community is in an agricultural promotion area, 0 otherwise Agricultural promotion area (farmland) Dummy: 1 if the community is in an agricultural promotion area and designated as a farmland area, 0 otherwise City planning area (Urbanization promotion area) Dummy: 1 if the city planning area is an "urbanization promotion area, 0 otherwise City planning area (Urbanization control area) Dummy: 1 if the city planning area is an "urbanization control area, 0 otherwise City planning area (not designated) Dummy: 1 if the community is in a city planning area but not designated as either an urbanization promotion area or an urbanization control area, 0 otherwise Social capital ('90) Social capital of the community in 1990 Readjustment dummy Dummy: 1 if the area and ratio of readjusted farmland increased between 1990 and 2000, 0 otherwise Specific outcomes (agriculture-related bonding social capital) Irrigation management All villagers = 4, only farmers = 3, employees = 2, not implemented = 1, nonexistent = 0 Farm road management All villagers = 4, only farmers = 3, employees = 2, not implemented = 1, nonexistent = 0 Number of meetings Total number of meetings held by farmers Number of agriculture-related organizations for youth Total number of agriculture-related organizations for young people Number of agriculture-related organizations for women Total number of agriculture-related organizations for women Number of agriculture-related organizations for the elderly Total number of agriculture-related organizations for the elderly Specific outcomes (non-agriculture-related bonding social capital) Common facilities management All villagers = 4, only farmers = 3, employees = 2, not implemented = 1, nonexistent = 0 Number of non-agriculture-related organizations for youth Total number of non-agriculture-related organizations for young people Number of non-agriculture-related organizations for women Total number of non-agriculture-related organizations for women Number of non-agriculture-related organizations for the elderly Total number of non-agriculture-related organizations for the elderly Specific outcomes (agriculture-related bridging social capital) Experience program for agriculture, forestry, and fisheries Dummy: 1 if an experience program related to agriculture, forestry, and fisheries is offered, 0 otherwise Direct sale of agricultural products Dummy: 1 if the direct sale of agricultural products is undertaken, 0 otherwise Program for temporary transfer to rural community Dummy: 1 if study trips to the rural community are offered, 0 otherwise Specific outcomes (non-agriculture-related bridging social capital) Program for experience of traditional arts and crafts Dummy: 1 if an exchange program related to traditional arts and crafts is offered, 0 otherwise Program for experience of events Dummy: 1 if an exchange program related to a traditional festival is offered, 0 otherwise 32