Climate change and income diversification in the mekong river delta a panel data analysis

UNIVERSITY OF ECONOMICS ERASMUS UNVERSITY ROTTERDAM HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES VIETNAM THE NETHERLANDS VIETNAM –THE NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS CLIMATE CHANGE AND INCOME DIVERSIFICATION IN THE MEKONG RIVER DELTA: A PANEL DATA ANALYSIS BY NGUYEN THI TUYET NGA MASTER OF ARTS IN DEVELOPMENT ECONOMICS HO CHI MINH CITY, December 2016 UNIVERSITY OF ECONOMICS HO CHIMINHCITY VIETNAM INSTITUTE OF SOCIAL STUDIES THE HAGUE THE NETHERLANDS VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS CLIMATE CHANGE AND INCOME DIVERSIFICATION IN THE MEKONG RIVER DELTA: A PANEL DATA ANALYSIS A thesis submitted in partial fulfilment of the requirements for the degree of MASTER OF ARTS IN DEVELOPMENT ECONOMICS By NGUYEN THI TUYET NGA Academic Supervisor: PHAM KHANH NAM HO CHI MINH CITY, December 2016 ACKNOWLEDGEMENT I would first like to thank my thesis supervisorDr Pham Khanh Nam of the Vietnam – The Netherlands Programme (VNP) at Ho Chi Minh City University of Economics He consistently allowed this paper to be my own work, but steered me in the right direction whenever he thought I needed it I would like to express my gratitude to the VNP officers who were involved in mythesis processby updating thesis schedule and providing good conditions for my research process Without their passionate participation, the thesis process could not have been successfully conducted Finally, thanks are also due to my classmates for providing me with unfailing support and continuous encouragement throughout my years of study and through the process of researching and writing this thesis This accomplishment would not have been possible without them Thank you Nguyen ThiTuyetNga Ho Chi Minh City, December 2016 Page i ABSTRACT The main objective of this study is to analyze the impact of climate change and other socio – economic determinants on the income diversification strategy in Mekong River Delta The data set is drawn from the VHLSS 2010,2012 and 2014, while climatic data including temperature and precipitation are extracted from the statistics website of The Ministry of Agriculture and Rural Development (MARD) Salinity data is collected from the Vietnam Institute of Meteorology, Hydrology and Climate Change.Findings of the study show that farmers have tendency to diversify their activities to reduce risk of crops failure when there is increasing temperature in dry season and precipitation in wet season However, it is recognized that those relationships are non-linear Diversification behavior is discovered not to be sensitive with the salinity intrusion and other climate variables Regarding socio-economic determinants, the household labor ratio and land area holding are found to be positively correlated with the income diversification, while educational qualification has the negative effect A male household head would more likely to diversify their activities to disperse risk of climate change than female head From this result, many policies are recommended in order to support farmers to access an effective diversification strategy, helping them to response actively to climate change consequences Page ii TABLE OF CONTENT Chapter Page Acknowledgement i Abstract ii Table of content iii List of tables v List of figures vi Introduction 1.1 Research problem 1.2 Research objective 1.3 Research scope 1.4 Thesis structure Literature review 2.1 Theoretical review 2.1.1 Climate change .4 2.1.2 Impact of climate change 2.1.3 Adaptation of people to climate change .5 2.1.4 Income diversification 2.1.4.1 Definition and classification of income diversification .7 2.1.4.2 Motivations of income diversification 2.1.4.3 Income diversification measurements 2.2 Empirical review 10 2.2.1 Impact of temperature and precipitation variation 10 2.2.2 Impact of high salinity intrusion to income diversification 12 2.2.3 Impact of socio-economic characteristics on income diversification 14 Research methodology .16 3.1 Analytical framework 16 3.2 Methodology .17 3.2.1 Income diversification index 17 3.2.2 Model specification 18 3.2.3 Variable description 20 Page 3.3 Data sources 24 3.4 Salinity measurement 26 Result and discussion 28 4.1 Overview of the Mekong River Delta 28 4.1.1 Geographical position and natural conditions 28 4.1.2 Socio – economic conditions 28 4.1.3 Impact of climate change on the Mekong River Delta 29 4.2 Salinity intrusion in the Mekong River Delta .31 4.3 Descriptive statistics of variables 34 4.3.1 Dependent variable 34 4.3.2 Independent variables 35 4.4 Empirical results 41 4.4.1 Findings of the Poisson model 41 4.4.2 Findings of the Tobit model 44 4.4.3 Interpretation 45 Conclusion 50 5.1Conclusion 50 5.2Policy implications 51 5.3Research limitations and research directions 52 Reference 53 Appendix 57 Page LIST OF TABLES Table 3.1.Variable description 21 Table 4.1.Descriptive statistics 34 Table 4.2.Results of the panel Poisson model and panel Tobit model .42 LIST OF FIGURES Figure 3.1.Analytical framework .16 Figure 3.2.Salinity stations in Mekong River Delta 27 Figure 4.1.GDP share per sector in Mekong River Delta 29 Figure 4.2.Regional Division of Mekong River Delta .31 Figure 4.3.The salinity intrusion map of Mekong River Delta 32 Figure 4.4.Income shares of households in Mekong River Delta 35 Figure 4.5.Precipitation in Mekong River Delta 37 Figure 4.6.Temperature in Mekong River Delta 38 Figure 4.7.Salinity at stations in Long An and Ca Mau – Bac Lieu 40 Figure 4.8.Marginal effect of precipitation in wet season 44 Figure 4.9.Marginal effect of temperature in dry season 45 Figure A.1.Salinity at stations in Mekong River Delta 59 CHAPTER 1: INTRODUCTION 1.1 Research problem Climate change has been the most controversial issue in the world due to its significant impacts on many aspects of society and economy in which agriculture is the most vulnerable sector The variationof climate conditions is reflected through temperature rising, abnormal precipitation, droughts, or floods Thoseare the main reasons for insects, diseases, and crop failures (Zerihun, 2012) Climate change is generally harmful for crop production, but indeed, the impact is much diversified (IPCC, 2014).Specifically, higher temperature shortens the growing period of rice, leading to rice yield reduction; however, in some study the increased CO2 from the pollution has supported the photosynthesis process of some crops such as maize and wheatresulting in a better productivity of cereals In Japan, the increase of 1oC in the 20th century has resulted in the drop of wheat, vegetables, milk, and egg production In Russia, the potential production of major crops is acknowledged to fallby 50% on average due to the climate change.On the other side, climate change does notgive identical effects on agriculture sector in different areas in the world due to alternative natural conditions and specific socio-economic characteristics of each region All demographic properties and adaptive solutions of people in an area are the main factors, which determined the vulnerability to climate change.In spite of diversified impacts,it is undeniable that climate change has severely affected food security all over the world Being the second biggestrice exporter in the world just after Thailand, Vietnam has 90% of exported production derived from the Mekong River Delta Located nearby the final branches of Mekong River before converging into the ocean, Mekong River Delta is the wide fertile area, which is appropriatefor rice paddy cultivation and is known as the biggest rice granary in Vietnam However, in recent years, Mekong River Deltaisseriously exposed to threat of climate change, which is clearly shown in high saline intrusion in coastal areas, droughts, and the shortage of fresh water in dry season, resulting in the restriction of arable land.In particular, according to the projected climate scenario in 2100 of the Ministry of Agriculture and Rural Development, if the sea level risesby 1meter, an approximate of 40% arable land could be sunk in salt water The yield shortfall has significantly caused the Page income loss for farmers, theincrease of poverty, and the social insecurity at the same time However,overcoming all difficulties of natural conditions, Mekong River Delta still keeps a stable development rate of production In order to deal with environmental challenge, farmers have applied many solutions to adapt with climate change and improve their lives In those solutions, income diversification is considered as an effective response to climate change (Smit et al., 2000; Bryan et al.,2011) Specifically, income diversification helps farmers to reduce the risk of crop failureand increase household’s total revenue (Zerihun, 2012;Haiwang et al., 2015).Income diversification process is understood as the way in which farmers participate in manyactivities to generate incomefor their households For example, household’s income sources could stem fromgrowing varieties of rice, fruits and other cereals; livestock breeding; aquaculture rearing; or non-farm activities Although income diversification is observed in various levels, researchers still concern aboutthe drivers of income diversification Several studies suggested that drivers are temperature, drought, salinity,price change, and institutional change Understanding separate channels that lead to the farmer’s behavior on diversifying income is important since it allows policy makers to know what to focus on in their policies for farmers Moreover, drivers of income diversification in the Mekong River Delta could be different from other areas and in the world where evidences could be found.Specific evidences for the Mekong River Delta are what policy makers need In Vietnam, income diversification process, which is considered as an effort to reduce thethreatens of climate change, is favorablyrecommended for farmers by Vietnamese Government.Besides,Government policies also relate to the improvement of physical infrastructure, financial subsidy, and the openness of agriculture market However, both uncertainties about determinants of income diversification and the response of farmers to climate changecould lead to the inefficiency or less efficiency of Government supporting policies Therefore, a research of climate change and income diversification could producereliable and sustainable evidences for policy makers about the impact of climate change on income diversification Based on those empirical findings, policy makers could implement policies, which are more efficient to support farmers in income diversification process gender | 1667806 0507943 3.28 0.001 0672255 2663357 age | 0012656 0015894 0.80 0.426 -.0018495 0043807 land_ha | 0380917 0051055 7.46 0.000 0280851 0480983 scaled_salinity | -.0035491 0032377 -1.10 0.273 -.0098949 0027968 dry_temp | 5.436609 8.158859 0.67 0.505 -10.55446 21.42768 wet_temp | -16.22262 14.00837 -1.16 0.247 -43.67852 11.23328 dry_precipitation | 0005638 0017785 0.32 0.751 -.0029221 0040496 wet_precipitation | 0080996 0028302 2.86 0.004 0025525 0136467 sqr_dry_temp | -.1024101 1506563 -0.68 0.497 -.397691 1928709 sqr_wet_temp | 2908038 2512028 1.16 0.247 -.2015447 7831522 sqr_dryprecipitation | -2.87e-06 5.07e-06 -0.57 0.572 -.0000128 7.07e-06 sqr_wetprecipitation | -.0000175 6.12e-06 -2.85 0.004 -.0000295 -5.47e-06 _cons | 153.7981 256.2877 0.60 0.548 -348.5166 656.1127 -+ -/lnalpha | -16.47073 284.1286 -573.3526 540.4112 -+ alpha | 7.03e-08 00002 9.9e-250 5.0e+234 -LR test of alpha=0: chibar2(01) = 0.00 Prob >= chibar2 = 1.000 est sto re xtpoisson diversity_index1 `control' scaled_salinity dry_temp wet_temp dry_precipitation wet_precipitation sqr_dry_temp sqr_wet_temp sqr_dr > yprecipitation sqr_wetprecipitation , fe note: groups (7 obs) dropped because of only one obs per group Iteration Iteration Iteration Iteration 0: 1: 2: 3: log log log log likelihood likelihood likelihood likelihood = = = = -809.66846 -807.99749 -807.99727 -807.99727 Conditional fixed-effects Poisson regression Group variable: id Number of obs = Number of groups = Obs per group: Log likelihood = -807.99727 Wald chi2(16) Prob > chi2 -diversity_index1 | Coef Std Err z -+ -hh_size | 0064639 0308044 0.21 hh_labor_ratio | 0523888 2197282 0.24 migration | 0300864 0482235 0.62 education | 0334917 1367977 0.24 gender | 0677677 3751553 0.18 age | 0014319 0360825 0.04 land_ha | 0149515 0136784 1.09 scaled_salinity | -.0038368 0202574 -0.19 dry_temp | 5.342075 9.54153 0.56 wet_temp | -3.154381 24.05328 -0.13 dry_precipitation | 0002433 0026125 0.09 wet_precipitation | 0026639 0040137 0.66 sqr_dry_temp | -.0986839 1764474 -0.56 sqr_wet_temp | 0585224 4325856 0.14 sqr_dryprecipitation | -8.24e-07 7.47e-06 -0.11 sqr_wetprecipitation | -5.41e-06 8.52e-06 -0.63 P>|z| 0.834 0.812 0.533 0.807 0.857 0.968 0.274 0.850 0.576 0.896 0.926 0.507 0.576 0.892 0.912 0.526 1,064 355 = avg = max = 3.0 = = 3.35 0.9996 [95% Conf Interval] -.0539115 -.3782706 -.0644299 -.234627 -.6675233 -.0692886 -.0118576 -.0435405 -13.35898 -50.29795 -.0048771 -.0052027 -.4445144 -.7893298 -.0000155 -.0000221 0668394 4830481 1246028 3016103 8030586 0721523 0417607 035867 24.04313 43.98919 0053636 0105306 2471467 9063745 0000138 0000113 est sto fe hausman fe re Note: the rank of the differenced variance matrix (11) does not equal the number of coefficients being tested (16); be sure this is what you expect, or there may be problems computing the test Examine the output of your estimators for anything unexpected and possibly consider scaling your variables so that the coefficients are on a similar scale Coefficients -| (b) (B) (b-B) | fe re Difference -+ sqrt(diag(V_b-V_B)) S.E hh_size | 0064639 0043159 002148 0275512 hh_labor_r~o | 0523888 1683662 -.1159774 2001915 migration | 0300864 0419248 -.0118383 031546 education | 0334917 -.0751008 1085924 135512 gender | 0677677 1667806 -.0990129 3717008 age | 0014319 0012656 0001663 0360475 land_ha | 0149515 0380917 -.0231402 0126898 scaled_sal~y | -.0038368 -.0035491 -.0002877 019997 dry_temp | 5.342075 5.436609 -.0945336 4.947101 wet_temp | -3.154381 -16.22262 13.06824 19.55316 dry_precip~n | 0002433 0005638 -.0003205 0019136 wet_precip~n | 0026639 0080996 -.0054356 0028459 sqr_dry_temp | -.0986839 -.1024101 0037262 0918496 sqr_wet_temp | 0585224 2908038 -.2322814 3521753 sqr_drypre~n | -8.24e-07 -2.87e-06 2.04e-06 5.48e-06 sqr_wetpre~n | -5.41e-06 -.0000175 0000121 5.93e-06 -b = consistent under Ho and Ha; obtained from xtpoisson B = inconsistent under Ha, efficient under Ho; obtained from xtpoisson Test: Ho: difference in coefficients not systematic chi2(11) = (b-B)'[(V_b-V_B)^(-1)](b-B) = 8.34 Prob>chi2 = 0.6822 ** Choose the random effect model: xtpoisson diversity_index1 `control' scaled_salinity dry_temp wet_temp dry_precipitation wet_precipitation sqr_dry_temp sqr_wet_temp sqr_dr > yprecipitation sqr_wetprecipitation, re cluster() Fitting Poisson model: Iteration Iteration Iteration Iteration 0: 1: 2: 3: log log log log likelihood likelihood likelihood likelihood = = = = -1655.2474 -1655.0338 -1655.0335 -1655.0335 Fitting full model: Iteration Iteration Iteration Iteration Iteration Iteration Iteration Iteration Iteration Iteration Iteration 0: 1: 2: 3: 4: 5: 6: 7: 8: 9: 10: log log likelihood = -1894.1417 log likelihood = -1665.3907 log likelihood = -1656.6608 log likelihood = -1655.4867 log likelihood = -1655.1319 log likelihood = -1655.0524 log likelihood = -1655.0379 log likelihood = -1655.0345 log likelihood = -1655.0337 log likelihood = -1655.0335 likelihood = -1655.0335 Random-effects Poisson regression Group variable: id Number of obs = Number of groups = Random effects u_i ~ Gamma Obs per group: Log likelihood = -1655.0335 Wald chi2(16) Prob > chi2 -diversity_index1 | Coef Std Err z -+ -hh_size | 0043159 0137782 0.31 hh_labor_ratio | 1683662 0905751 1.86 migration | 0419248 0364741 1.15 education | -.0751008 0187117 -4.01 gender | 1667806 0507943 3.28 age | 0012656 0015894 0.80 land_ha | 0380917 0051055 7.46 P>|z| 0.754 0.063 0.250 0.000 0.001 0.426 0.000 1,071 362 = avg = max = 3.0 = = 128.99 0.0000 [95% Conf Interval] -.0226888 -.0091578 -.0295631 -.111775 0672255 -.0018495 0280851 0313206 3458902 1134126 -.0384265 2663357 0043807 0480983 scaled_salinity dry_temp wet_temp dry_precipitation wet_precipitation sqr_dry_temp | | | | | | -.0035491 5.436609 -16.22262 0005638 0080996 -.1024101 0032377 8.158859 14.00837 0017785 0028302 1506563 -1.10 0.67 -1.16 0.32 2.86 -0.68 0.273 0.505 0.247 0.751 0.004 0.497 -.0098949 -10.55446 -43.67852 -.0029221 0025525 -.397691 0027968 21.42768 11.23328 0040496 0136467 1928709 sqr_wet_temp | 2908038 2512028 1.16 0.247 -.2015447 7831522 sqr_dryprecipitation | -2.87e-06 5.07e-06 -0.57 0.572 -.0000128 7.07e-06 sqr_wetprecipitation | -.0000175 6.12e-06 -2.85 0.004 -.0000295 -5.47e-06 _cons | 153.7981 256.2877 0.60 0.548 -348.5166 656.1127 -+ -/lnalpha | -16.47073 284.1286 -573.3526 540.4112 -+ alpha | 7.03e-08 00002 9.9e-250 5.0e+234 -LR test of alpha=0: chibar2(01) = 0.00 Prob >= chibar2 = 1.000 ** Test the significance of the variable: test `control' scaled_salinity dry_temp wet_temp sqr_dry_temp sqr_wet_temp sqr_dryprecipitation sqr_wet > precipitation ( 1) ( 2) ( 3) ( 4) ( 5) ( 6) ( 7) ( 8) ( 9) (10) (11) (12) (13) (14) (15) (16) dry_precipitation wet_precipitation [diversity_index1]hh_size = [diversity_index1]hh_labor_ratio = [diversity_index1]migration = [diversity_index1]education = [diversity_index1]gender = [diversity_index1]age = [diversity_index1]land_ha = [diversity_index1]scaled_salinity = [diversity_index1]dry_temp = [diversity_index1]wet_temp = [diversity_index1]dry_precipitation = [diversity_index1]wet_precipitation = [diversity_index1]sqr_dry_temp = [diversity_index1]sqr_wet_temp = [diversity_index1]sqr_dryprecipitation = [diversity_index1]sqr_wetprecipitation = chi2( 16) = 128.99 Prob > chi2 = 0.0000 test wet_precipitation sqr_wetprecipitation ( 1) [diversity_index1]wet_precipitation = ( 2) [diversity_index1]sqr_wetprecipitation = chi2( 2) = Prob > chi2 = 8.21 0.0165 ** Calculate Marginal effect: xtpoisson diversity_index1 scaled_salinity c.dry_temp##c.dry_temp c.wet_temp##c.wet_temp c.dry_precipitation##c.dry_precipitation c.wet_pre > cipitation##c.wet_precipitation `control', re cluster() Fitting Poisson model: Iteration Iteration Iteration Iteration 0: 1: 2: 3: log log log log likelihood likelihood likelihood likelihood = = = = -1655.2403 -1655.0268 -1655.0265 -1655.0265 Fitting full model: Iteration 0: Iteration 1: log likelihood = -1894.1399 log likelihood = -1665.3846 Iteration Iteration Iteration Iteration Iteration Iteration Iteration 2: 3: 4: 5: 6: 7: 8: log log log log log log log likelihood likelihood likelihood likelihood likelihood likelihood likelihood = = = = = = = -1656.6542 -1655.4798 -1655.1249 -1655.0454 -1655.031 -1655.0275 -1655.0267 Iteration 9: log likelihood = -1655.0265 Iteration 10: log likelihood = -1655.0265 Iteration 11: log likelihood = -1655.0265 Random-effects Poisson regression Group variable: id Number of obs = Number of groups = Random effects u_i ~ Gamma Obs per group: = avg = max = 3.0 = = 129.00 0.0000 Wald chi2(16) Prob > chi2 Log likelihood = -1655.0265 diversity_index1 | Coef Std Err + -scaled_salinity | -.0035484 0032375 dry_temp | 5.442579 8.160219 | c.dry_temp#c.dry_temp | -.1025196 1506811 | wet_temp | -16.27622 14.02314 | c.wet_temp#c.wet_temp | 2917645 2514663 | dry_precipitation | 0005633 0017784 | c.dry_precipitation#c.dry_precipitation | -2.86e-06 5.07e-06 | wet_precipitation | 0080959 0028312 | c.wet_precipitation#c.wet_precipitation | -.0000175 6.12e-06 | hh_size | 0043187 0137781 hh_labor_ratio | 168364 0905751 migration | 0419269 0364738 education | -.0750974 0187117 gender | 1667777 0507943 age | 0012656 0015894 land_ha | 0380908 0051054 1,071 362 z P>|z| [95% Conf Interval] -1.10 0.67 0.273 0.505 -.0098938 -10.55116 -0.68 0.496 -.3978492 19281 -1.16 0.246 -43.76107 11.20862 1.16 0.246 -.2011005 7846294 0.32 0.751 -.0029224 0040489 -0.57 0.572 -.0000128 7.07e-06 2.86 0.004 0025468 0136449 -2.85 0.004 -.0000295 -5.45e-06 0.31 1.86 1.15 -4.01 3.28 0.80 7.46 0.754 0.063 0.250 0.000 0.001 0.426 0.000 -.0226859 -.0091599 -.0295603 -.1117716 0672228 -.0018495 0280844 0313233 3458879 1134142 -.0384232 2663326 0043807 0480973 _cons | 154.4647 256.5812 0.60 0.547 -348.4252 657.3546 + -/lnalpha | -17.66624 255.5196 + -alpha | 2.13e-08 5.43e-06 LR test of alpha=0: chibar2(01) = 5.4e-05 Prob >= chibar2 = 0.497 -518.4755 483.143 6.7e-226 6.7e+209 ** Draw marginal effect graph of Precipitation in the wet season: marginsplot Variables that uniquely identify margins: wet_precipitation margins, dydx(wet_precipitation) at( wet_precipitation =(107.14(30)345.29)) Average marginal effects Model VCE : OIM Number of obs Expression : Linear prediction, predict() dy/dx w.r.t : wet_precipitation 1._at : wet_precip~n = 107.14 2._at : wet_precip~n = 137.14 3._at : wet_precip~n = 167.14 4._at : wet_precip~n = 197.14 = 002797 21.43631 1,071 5._at : wet_precip~n = 227.14 6._at : wet_precip~n = 257.14 7._at : wet_precip~n = 287.14 8._at : wet_precip~n = 317.14 | Delta-method | dy/dx Std Err z + -wet_precipitation | _at | | 004355 0015416 2.82 | 0033075 0011894 2.78 | 00226 0008503 2.66 | 0012125 0005489 2.21 | 000165 0003869 0.43 | -.0008825 0005178 -1.70 | -.00193 0008103 -2.38 | -.0029775 0011468 -2.60 - P>|z| 0.005 0.005 0.008 0.027 0.670 0.088 0.017 0.009 [95% Conf Interval] 0013335 0009762 0005934 0001366 -.0005934 -.0018975 -.0035182 -.0052252 0073764 0056387 0039265 0022884 0009234 0001325 -.0003418 -.0007298 // Tobit model: / is not a valid command name r(199); xttobit diversity_index2 scaled_salinity dry_temp wet_precipitation sqr_dry_temp sqr_wet_temp sqr_dryprecipitati > on sqr_wetprecipitation `control' , ul(1) re cluster() Obtaining wet_temp dry_precipitation starting values for full model: Iteration 0: Iteration 1: Iteration 2: Iteration 3: log log log log likelihood likelihood likelihood likelihood = = = = log log log log log likelihood likelihood likelihood likelihood likelihood = = = = = 164.51368 173.39111 173.56978 173.56995 Fitting full model: Iteration Iteration Iteration Iteration Iteration 0: 1: 2: 3: 4: -338.06347 -270.15085 -269.09481 -269.08819 -269.08818 Random-effects tobit regression Group variable: id Number of obs = Number of groups = Random effects u_i ~ Gaussian Obs per group: = avg = max = Integration method: mvaghermite Integration pts = Log likelihood = -269.08818 Wald chi2(16) Prob > chi2 -diversity_index2 | Coef Std Err z -+ -scaled_salinity | 0003955 0020019 0.20 dry_temp | -5.094096 2.920753 -1.74 wet_temp | -4.931524 6.094254 -0.81 dry_precipitation | -.0003053 0006958 -0.44 wet_precipitation | -.0022184 001155 -1.92 sqr_dry_temp | 0948993 0539556 1.76 sqr_wet_temp | 0878975 1094411 0.80 sqr_dryprecipitation | 1.14e-06 1.98e-06 0.58 sqr_wetprecipitation | 4.81e-06 2.46e-06 1.96 hh_size | -.0089625 0072813 -1.23 hh_labor_ratio | -.047354 0497081 -0.95 migration | -.0157082 0147405 -1.07 education | 0508484 0129138 3.94 gender | -.1413014 0345128 -4.09 age | -.0010515 001133 -0.93 land_ha | -.003443 0036456 -0.94 P>|z| 0.843 0.081 0.418 0.661 0.055 0.079 0.422 0.563 0.050 0.218 0.341 0.287 0.000 0.000 0.353 0.345 1,071 362 3.0 12 = = 47.59 0.0001 [95% Conf Interval] -.0035282 -10.81867 -16.87604 -.0016691 -.0044822 -.0108516 -.1266032 -2.73e-06 -6.47e-09 -.0232336 -.1447801 -.0445989 0255379 -.2089452 -.003272 -.0105882 0043193 6304745 7.012994 0010585 0000454 2006503 3023982 5.02e-06 9.63e-06 0053087 0500721 0131826 076159 -.0736576 0011691 0037023 _cons | 138.7191 107.0625 1.30 0.195 -71.11951 -+ -/sigma_u | 2459029 0138871 17.71 0.000 /sigma_e | 1954196 0060676 32.21 0.000 348.5577 2186847 1835274 273121 2073118 -+ rho | 5505751 6724636 -0 left-censored observations 775 uncensored observations 296 right-censored observations 6129133 0312161 ** Test the significance of the variable: test scaled_salinity dry_temp wet_temp dry_precipitation wet_precipitation sqr_dry_temp sqr_wet_temp sqr_dryprecipitation sqr_wetprecipitat > ion `control' ( 1) ( 2) ( 3) ( 4) ( 5) ( 6) ( 7) ( 8) ( 9) (10) (11) (12) (13) (14) (15) (16) [diversity_index2]scaled_salinity = [diversity_index2]dry_temp = [diversity_index2]wet_temp = [diversity_index2]dry_precipitation = [diversity_index2]wet_precipitation = [diversity_index2]sqr_dry_temp = [diversity_index2]sqr_wet_temp = [diversity_index2]sqr_dryprecipitation = [diversity_index2]sqr_wetprecipitation = [diversity_index2]hh_size = [diversity_index2]hh_labor_ratio = [diversity_index2]migration = [diversity_index2]education = [diversity_index2]gender = [diversity_index2]age = [diversity_index2]land_ha = chi2( 16) = Prob > chi2 = 47.59 0.0001 test dry_temp sqr_dry_temp ( 1) [diversity_index2]dry_temp = ( 2) [diversity_index2]sqr_dry_temp = chi2( 2) = Prob > chi2 = 7.05 0.0295 test wet_precipitation sqr_wetprecipitation ( 1) [diversity_index2]wet_precipitation = ( 2) [diversity_index2]sqr_wetprecipitation = chi2( 2) = Prob > chi2 = 3.86 0.1448 ** Calculate Marginal effect: xttobit diversity_index2 scaled_salinity c.dry_temp##c.dry_temp c.dry_precipitation##c.dry_precipitation c.wet_preci > pitation##c.wet_precipitation `control', re cluster() Obtaining starting values for full model: Iteration Iteration Iteration Iteration 0: 1: 2: 3: Fitting full model: log log log log likelihood likelihood likelihood likelihood = = = = 164.51055 173.38895 173.56764 173.56781 c.wet_temp##c.wet_temp Iteration 0: Iteration 1: log likelihood = log likelihood = Random-effects tobit regression Group variable: id 173.56781 173.56781 Number of obs = Number of groups = 1,071 362 Random effects u_i ~ Gaussian Obs per group: Integration method: mvaghermite Integration pts = Log likelihood = 173.56781 Wald chi2(16) Prob > chi2 = avg = max = 3.0 12 = = 41.25 0.0005 + scaled_salinity |.0004745 0.733 -.0022467 0031957 dry_temp | -3.97197 2.255179 -1.76 0.078 | c.dry_temp#c.dry_temp | 0739636 0416519 1.78 0.076 | wet_temp | -3.129728 4.598307 -0.68 0.496 | c.wet_temp#c.wet_temp | 0556771 0825489 0.67 0.500 | dry_precipitation | -.0002633 0005265 -0.50 0.617 | c.dry_precipitation#c.dry_precipitation | 9.19e-07 1.49e-06 0.62 0.538 | wet_precipitation | -.0014003 0008433 -1.66 0.097 | c.wet_precipitation#c.wet_precipitation | 3.04e-06 1.80e-06 1.69 0.091 | hh_size | -.0065029 0053121 -1.22 0.221 hh_labor_ratio | -.0536531 0355123 -1.51 0.131 migration | -.0161259 0116518 -1.38 0.166 education | 0291604 0085039 3.43 0.001 gender | -.0845108 0228651 -3.70 0.000 age | -.0005967 0007589 -0.79 0.432 land_ha | -.0002609 0027889 -0.09 0.925 _cons | 98.29724 80.98127 1.21 0.225 -60.42313 257.0176 + -/sigma_u | 1595978 0085215 18.73 0.000 /sigma_e | 1643946 0044296 37.11 0.000 + -rho | 485198 032434 - 0013884 0.34 -8.392039 4481002 -.0076726 1555999 -12.14224 5.882788 -.1061158 21747 -.0012952 0007686 -2.01e-06 3.84e-06 -.003053 0002525 -4.87e-07 6.57e-06 -.0169145 -.1232558 -.038963 0124931 -.1293257 -.0020842 -.0057271 0039087 0159497 0067112 0458276 -.0396959 0008907 0052054 142896 1557128 1762996 1730763 4220835 5486863 left-censored observations 1,071 uncensored observations right-censored observations **Draw marginal effect graph of Temperature in the dry season: marginsplot Variables that uniquely identify margins: dry_temp margins, dydx(dry_temp) predict(e(.,1)) at(dry_temp=(26.4(.1)27.9)) Average marginal effects Model VCE : OIM Number of obs = Expression dry_temp : E(diversity_index2|diversity_index2|z| 0.238 0.294 0.379 0.522 0.787 0.736 0.203 0.027 0.009 0.009 0.011 0.014 0.015 0.016 0.016 0.015 [95% Conf Interval] -.1209977 -.1025422 -.0831692 -.0632891 -.0434547 -.0245665 -.0083015 0029693 0087655 0112655 0123699 0130307 0137232 0147232 0162143 018326 0300793 0310264 0316299 0321343 0329283 0347512 0391177 0482368 0624252 0793382 0969079 1140116 1300035 1444385 156969 1673114 ... on income diversification 2.1 Theoretical review 2.1.1 Climate change Climate change is defined as: ? ?A change in the state of the climate that can be identified by changes in the mean and/ or the. .. apanel data analysis for 362 households in the Mekong River Delta of Vietnam, which is a coastal areaseverelyaffected by the climate change Most of households participate in the agricultural activities,... selecting plants for the coastal area According to the study, barley and wheat are higher salinity tolerant than rice and corn Cotton and sugar beet have higher degree of saline than bean, pea and