B GIÁO D O I H C KINH T TP H CHÍ MINH TR N QUANG B O NG C A V N XÃ H TI P C N TÍN D NG C A CÁC H NƠNG THƠN VI T NAM LU N B GIÁO D O I H C KINH T TP H CHÍ MINH TR N QUANG B O NG C A V N XÃ H TI P C N TÍN D NG C A CÁC H NÔNG THÔN VI T NAM N : Kinh t phát tri n : 60310105 LU TH PGS TS NGUY N TR NG HOÀI i L ng lu h ng c a v n xã h nông thôn Vi n ti p c n tín d ng c a u c a tơi Ngo i tr nh ng tài li u tham kh c trích d n lu r ng toàn ph n hay nh ng ph n nh c a lu cs d nh n b ng c p c công b ho c nh Khơng có s n ph m/nghiên c u c c trích d Lu nh cn h c ho o khác , 2013 Tr n Quang B o c s d ng lu nh n b t k b ng c p t i ii L IC hoàn thành lu Quý Th n l c c a b n thân, chân thành c i h c Kinh t Thành ph H Chí Minh, nh t Quý Th y Cô Khoa Kinh t phát tri l a ch n nghiên c n th ki n th c n n t ng g i m v tài c bi t, g i l i bi Tr ng Hồi- n tình ch b n Th ng d n-PGS TS Nguy n ng ý ki ng viên tơi su t q trình th c hi n lu Cu i cùng, mu n c i thân yêu, b ki n t t nh t h tr tơi q trình tham gia khóa h c th c hi n lu t t nghi p u iii TÓM T T V n xã h c th a nh n giao d ch kinh t ti p c n tín d ng Nghiên c u nh m ki ph n c a v n xã h ki n nh m i quan h gi a thành n ti p c n tín d ng b i c nh nông thôn Vi t Nam Vi c c th c hi n thơng qua hai mơ hình h i v i mơ hình th nh t-mơ hình h i quy binary logistic-bi n ph thu c g m có hai bi u hi n ti p c c tín d ng th c khơng ti p c c tín d ng th iv i mơ hình th hai-mơ hình h i quy multinomial logistic-bi n ph thu c có ba bi u hi n ti p c n tín d ng th c, ti p c n tín d ng bán th c ti p c n tín d ng phi th c Trong c hai mơ hình, t p bi n gi i thích g m có ba thành ph n c a v n xã h i (m i quan h xã h i th c, m th c, ni m tin) y u t thu h i quy ch r ng m v i kh m nhân kh u h c Các k t qu i quan h xã h i th ng n p c n tín d ng th c, nh ng h h xã h i th c r th u kh c nhóm ti p c n tín d ng bán ng c thù c a th n ti p ng kê Ngoài ra, phát hi n c a nghiên c u cho th y, th hồn tồn khơng có i nh ng m c n tín d s n th ch p y u t c chi u i quan c nhóm ti p c n tín d ng th quan h phi th c ni ch i quan h phi ng m nh nh ng tín d ng nơng thơn Vi t Nam, tài c l i thu nh p c a h ng Các phát hi n gây ng c nhiên có th tính ng tín d ng nơng thơn Vi t Nam iv M CL C L i L ic ii Tóm t t iii M c l c iv Danh m vii Danh m c b ng vii Danh m c t vi t t t viii : GI I THI U CHUNG 1.1 Lý ch tài 1.2 M c tiêu nghiên c u 1.3 Ph m vi gi i h n nghiên c u u a nghiên c u 1.6 K t c u c a lu : LÝ THUY T V V N XÃ H I VÀ KHUNG PHÂN TÍCH 2.1 V n xã h i 2.1.1 Khái ni m thành ph n c a v n xã h i ng v n xã h i 10 2.2 Tài vi mơ 16 2.2.1 Các h i ti t ki m tín d ng xoay vòng 16 i cho vay 17 2.2.3 Tín d i 18 2.2.4 Cho vay theo nhóm 18 2.3 Lý thuy t v thông tin b t cân x ng ho 2.4 Th ng tín d ng 20 ng tín d ng nơng thơn ti p c n tín d ng 21 2.5 Các nghiên c u v m i quan h gi a v n xã h i ti p c n tín d ng 22 2.6 Khung khái ni ngh cho nghiên c u 28 v : U 31 3.1 Gi thuy t nghiên c u 31 3.2 Mơ hình nghiên c u 34 3.3 D li u m u nghiên c u 36 3.3.1 D li u 36 3.3.2 M u nghiên c u 38 3.4 Mô t ng bi n 39 3.4.1 Bi n ph thu c 39 3.4.2 Bi c l p 40 3.4.3 Bi n ki m soát 40 3.5 Phân tích x lý s li u 44 : K T QU NGHIÊN C U 47 m c a m u nghiên c u 47 m nhân kh u h c c mc ah 47 a kho n vay 48 4.1.3 Tình tr ng ti p c n tín d ng c a h nơng thơn 49 4.1.4 S li u th ng kê mô t v v n xã h i 52 4.2 M i quan h gi a bi n 54 n 54 4.2.2 Ki nh Pearson Chi-Square v m i quan h gi a bi c l p, bi n ki m soát, bi n ph thu c 57 4.3 Phân tích h i quy 58 4.3.1 Ti p c n tín d ng th c 58 4.3.2 Ti p c n tín d ng t lo i hình khác 62 4.4 Ki nh mơ hình 70 4.4.1 Ki nh gi thuy t nghiên c u 70 4.4.2 Ki nh m phù h p c a mơ hình t ng th 73 4.5 Th o lu n k t qu 74 5: K T LU N VÀ G I Ý CHÍNH SÁCH 80 vi 5.1 K t lu n 80 5.2 G i ý sách 82 5.3 Nh ng h n ch c a nghiên c ng nghiên c u ti p theo 83 TÀI LI U THAM KH O 84 PH L C 90 Ph l c 91 Ph l c 94 Ph l c 3a 96 Ph l c 3b 99 Ph l c 4a 107 Ph l c 4b 112 Ph l c 4c 118 vii DANH M 2.1 M i quan h gi a v n xã h i ti p c n tín d ng 29 3.1 Mơ hình phân tích 35 DANH M C B NG B ng 2.1 Hai lo i m i: phi th c th c B ng 2.2 Các ch s ng v n xã h i 11 B ng 2.3 Tóm t t nghiên c u v v n xã h i ti p c n tín d ng 24 B ng 3.1 Các t nh tham gia vào m u nghiên c u 38 B ng 3.2 Tóm t t bi n dùng nghiên c u 43 B m nhân kh u h c c B n 48 mc ah a kho n vay 49 B ng 4.3 Ti p c n tín d ng t lo i hình khác 50 B ng 4.4 S ng) theo t ng lo i tín d ng 51 B ng 4.5 S li u th ng kê mô t c a bi c l p 52 B ng 4.6 V n xã h i d a theo t nh 53 B ng 4.7 V n xã h i c B ng 4.8 Ma tr B ng 4.9 Ki a theo ngu n ti p c n tín d ng 54 a bi c l p bi n ki m soát 56 nh Pearson Chi-Square 57 B ng 4.10 H i quy binary logistic 60 B ng 4.11 H i quy multinomial logistic 67 viii DANH M C T VI T T T CIEM Central Institute for Economic Management CSXH Chính sách xã h i GDP Gross Domestic Product ILSSA Institute of Labor Science and Social Affairs IPO Initial Public Offering IPSARD Institute of Policy and Strategy for Agriculture and Rural Development n.d no date NGO Non-government organization NHTM i NN-PTNT Nông nghi p-Phát tri n nông thôn ROSCA Rotating Savings and Credit Association SPSS Statistical Package for the Social Sciences TCVM Tài vi mô VARHS Vietnam Access to Resources Household Survey VXH V n xã h i WVS World Values Survey ZINB Zero-inflated negative binominal 109 Case Processing Summary N Tiep can cac loai TD Marginal Percentage Tiep can TD chinh thuc 483 56.2% Tiep can TD ban chinh thuc 100 11.6% Tiep can TD phi chinh thuc 276 32.1% 859 100.0% Valid Missing Total 859 Subpopulation 115a a The dependent variable has only one value observed in 62 (53.9%) subpopulations Model Fitting Information Model Fitting Criteria Likelihood Ratio Tests -2 Log Model Likelihood Intercept Only 451.368 df Sig 468.767 Final Chi-Square 17.400 008 Pseudo R-Square Cox and Snell 020 Nagelkerke 024 McFadden 011 Likelihood Ratio Tests Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood of Reduced Effect Model Chi-Square df Sig Intercept 485.659 34.292 000 fl_net 464.647 13.280 001 inf_net 453.075 1.708 426 110 trust 454.674 3.306 191 The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model The reduced model is formed by omitting an effect from the final model The null hypothesis is that all parameters of that effect are 111 Parameter Estimates 95% Confidence Interval for Exp(B) Tiep can cac loai TDa Tiep can TD ban chinh thuc B Intercept Std Error Wald df Sig Exp(B) Lower Bound Upper Bound -3.167 657 23.194 000 fl_net 378 105 12.956 000 1.460 1.188 1.794 inf_net -.022 021 1.022 312 979 938 1.021 355 213 2.782 095 1.426 940 2.164 -.933 368 6.423 011 fl_net 104 073 2.024 155 1.110 961 1.281 inf_net -.011 012 770 380 989 966 1.013 088 121 527 468 1.092 862 1.383 trust Tiep can TD phi chinh thuc Intercept trust a The reference category is: Tiep can TD chinh thuc Classification Predicted Tiep can TD ban chinh Tiep can TD phi chinh Observed Tiep can TD chinh thuc Tiep can TD chinh thuc thuc thuc Percent Correct 483 0 100.0% Tiep can TD ban chinh thuc 99 1.0% Tiep can TD phi chinh thuc 276 0 0% 99.9% 1% 0% 56.3% Overall Percentage 112 Ph l c 4b- H i quy bi n ph thu c ca_typetheo bi trust bi n ki c l p fl_net, inf_net i outliers) NOMREG ca_type (BASE=FIRST ORDER=ASCENDING) BY h_head sex ethnic collat WITH fl_net inf_net trust age edu distance inc age_squ /CRITERIA CIN(95) DELTA(0) MXITER(100) MXSTEP(5) CHKSEP(20) LCONVERGE(0) PCONVERGE(0.000001) S INGULAR(0.00000001) /MODEL /STEPWISE=PIN(.05) POUT(0.1) MINEFFECT(0) RULE(SINGLE) ENTRYMETHOD(LR) REMOVALMETHOD(LR) /INTERCEPT=INCLUDE /PRINT=CLASSTABLE PARAMETER SUMMARY LRT CPS STEP MFI Nominal Regression Notes Output Created 08-May-2013 08:53:05 Comments Input Data E:\SCCA\Thesis\ca_type_without_outliers_new.sav Active Dataset filter_outliers Filter Weight Split File N of Rows in Working Data File Missing Value Handling Definition of Missing 849 User-defined missing values are treated as missing Cases Used Statistics are based on all cases with valid data for all variables in the model 113 Syntax NOMREG ca_type (BASE=FIRST ORDER=ASCENDING) BY h_head sex ethnic collat WITH fl_net inf_net trust age edu distance inc age_squ /CRITERIA CIN(95) DELTA(0) MXITER(100) MXSTEP(5) CHKSEP(20) LCONVERGE(0) PCONVERGE(0.000001) SINGULAR(0.00000001) /MODEL /STEPWISE=PIN(.05) POUT(0.1) MINEFFECT(0) RULE(SINGLE) ENTRYMETHOD(LR) REMOVALMETHOD(LR) /INTERCEPT=INCLUDE /PRINT=CLASSTABLE PARAMETER SUMMARY LRT CPS STEP MFI Resources Processor Time 00:00:00.749 Elapsed Time 00:00:00.710 [filter_outliers] E:\SC-CA\Thesis\ca_type_without_outliers_new.sav Warnings Unexpected singularities in the Hessian matrix are encountered This indicates that either some predictor variables should be excluded or some categories should be merged The NOMREG procedure continues despite the above warning(s) Subsequent results shown are based on the last iteration Validity of the model fit is uncertain Case Processing Summary N Tiep can cac loai TD 56.8% 94 11.1% Tiep can TD phi chinh thuc 273 32.2% Khong la chu ho 200 23.6% La chu ho Gioi tinh 482 Tiep can TD ban chinh thuc Chu ho Tiep can TD chinh thuc Marginal Percentage 649 76.4% Nu 253 29.8% Nam 596 70.2% 114 Dan toc Dan toc khac 25.3% Dan toc Kinh 634 74.7% Khong the chap 553 65.1% Co the chap 296 34.9% 849 Tai san the chap 215 100.0% Valid Missing Total 849 Subpopulation 849a a The dependent variable has only one value observed in 849 (100.0%) subpopulations Model Fitting Information Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood Model Intercept Only 1039.963 df Sig 1578.963 Final Chi-Square 539.001 24 000 Pseudo R-Square Cox and Snell 470 Nagelkerke 557 McFadden 341 Likelihood Ratio Tests Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood of Reduced Effect Model Chi-Square df Sig Intercept 1.040E3a 000 fl_net 1049.374 9.411 009 inf_net 1042.390 2.427 297 trust 1042.260 2.297 317 age 1056.302 16.340 000 115 edu 1046.817 6.854 032 distance 1052.733 12.770 002 inc 1040.343 381 827 age_squ 1055.055 15.092 001 h_head 1042.146 2.184 336 sex 1047.103 7.140 028 ethnic 1061.617 21.654 000 collat 1475.846 435.883 000 The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model The reduced model is formed by omitting an effect from the final model The null hypothesis is that all parameters of that effect are a This reduced model is equivalent to the final model because omitting the effect does not increase the degrees of freedom 116 Parameter Estimates 95% Confidence Interval for Exp(B) Tiep can cac loai TDa Tiep can TD ban chinh thuc B Intercept Std Error Wald df Sig Exp(B) Lower Bound Upper Bound -24.066 2.356 104.373 000 fl_net 445 147 9.158 002 1.560 1.170 2.081 inf_net 007 029 062 803 1.007 952 1.065 trust 162 252 414 520 1.176 717 1.928 age 082 091 815 367 1.085 909 1.296 edu 005 045 013 910 1.005 919 1.099 -.049 023 4.688 030 952 911 995 inc 000 000 147 701 1.000 1.000 1.000 age_squ 000 001 490 484 999 997 1.001 [h_head=0] 282 379 553 457 1.326 630 2.789 [h_head=1] 0b [sex=0] 968 363 7.105 008 2.633 1.292 5.365 [sex=1] 0b [ethnic=0] -.257 344 559 455 773 394 1.517 [ethnic=1] 0b [collat=0] 19.749 000 377397743.763 377397743.763 377397743.763 [collat=1] 0b -15.329 1.566 95.877 000 fl_net 149 109 1.872 171 1.161 938 1.437 inf_net 029 023 1.560 212 1.029 984 1.076 distance Tiep can TD phi chinh thuc Intercept 117 trust -.160 171 874 350 852 610 1.191 age -.176 060 8.520 004 838 745 944 edu -.078 034 5.264 022 925 866 989 distance 003 003 747 387 1.003 997 1.009 inc 000 000 374 541 1.000 1.000 1.000 age_squ 002 001 7.936 005 1.002 1.001 1.003 [h_head=0] -.232 301 595 440 793 440 1.429 [h_head=1] 0b [sex=0] 314 284 1.221 269 1.368 785 2.386 [sex=1] b [ethnic=0] -1.183 269 19.290 000 306 181 519 [ethnic=1] 0b [collat=0] 20.744 000 1.021E9 1.021E9 1.021E9 [collat=1] b a The reference category is: Tiep can TD chinh thuc b This parameter is set to zero because it is redundant 0 118 Classification Predicted Tiep can TD chinh Tiep can TD ban Tiep can TD phi thuc chinh thuc chinh thuc Observed Tiep can TD chinh thuc Percent Correct 351 125 72.8% Tiep can TD ban chinh thuc 18 17 59 18.1% Tiep can TD phi chinh thuc 53 11 209 76.6% 49.7% 4.0% 46.3% 68.0% Overall Percentage Ph l c 4c- H i quy bi n ph thu c ca_typetheo bi c l p fl_net, inf_net trust bi n ki m soát GET FILE='E:\SC-CA\Thesis\SC-CA.sav' DATASET NAME DataSet1 WINDOW=FRONT NOMREG ca_type (BASE=FIRST ORDER=ASCENDING) BY h_head sex ethnic collat WITH fl_net inf_net trust age edu distance inc age_squ /CRITERIA CIN(95) DELTA(0) MXITER(100) MXSTEP(5) CHKSEP(20) LCONVERGE(0) PCONVERGE(0.000001) S INGULAR(0.00000001) /MODEL /STEPWISE=PIN(.05) POUT(0.1) MINEFFECT(0) RULE(SINGLE) ENTRYMETHOD(LR) REMOVALMETHOD(LR) /INTERCEPT=INCLUDE /PRINT=CLASSTABLE FIT PARAMETER SUMMARY LRT CPS STEP MFI Nominal Regression Notes Output Created 08-May-2013 08:59:27 Comments Input Data E:\SC-CA\Thesis\SC-CA.sav Active Dataset DataSet1 Filter Weight Split File N of Rows in Working Data File Missing Value Handling Definition of Missing 859 User-defined missing values are treated as missing Cases Used Statistics are based on all cases with valid data for all variables in the model 119 Syntax NOMREG ca_type (BASE=FIRST ORDER=ASCENDING) BY h_head sex ethnic collat WITH fl_net inf_net trust age edu distance inc age_squ /CRITERIA CIN(95) DELTA(0) MXITER(100) MXSTEP(5) CHKSEP(20) LCONVERGE(0) PCONVERGE(0.000001) SINGULAR(0.00000001) /MODEL /STEPWISE=PIN(.05) POUT(0.1) MINEFFECT(0) RULE(SINGLE) ENTRYMETHOD(LR) REMOVALMETHOD(LR) /INTERCEPT=INCLUDE /PRINT=CLASSTABLE FIT PARAMETER SUMMARY LRT CPS STEP MFI Resources Processor Time 00:00:00.468 Elapsed Time 00:00:00.396 [DataSet1] E:\SC-CA\Thesis\SC-CA.sav Warnings There are 1718 (66.7%) cells (i.e., dependent variable levels by subpopulations) with zero frequencies Case Processing Summary N Tiep can cac loai TD Marginal Percentage 11.6% 276 32.1% Khong la chu ho 203 23.6% 656 76.4% Nu 256 29.8% Nam 603 70.2% Dan toc khac 216 25.1% Dan toc Kinh Tai san the chap 100 La chu ho Dan toc 56.2% Tiep can TD phi chinh thuc Gioi tinh 483 Tiep can TD ban chinh thuc Chu ho Tiep can TD chinh thuc 643 74.9% Khong the chap 555 64.6% 120 Co the chap 304 859 Valid Missing 35.4% 100.0% Total 859 Subpopulation 859a a The dependent variable has only one value observed in 859 (100.0%) subpopulations Model Fitting Information Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood Model Chi-Square Intercept Only 1148.636 Sig 1613.020 Final df 464.384 24 000 Goodness-of-Fit Chi-Square df Sig Pearson 2109.208 1692 000 Deviance 1148.636 1692 1.000 Pseudo R-Square Cox and Snell 418 Nagelkerke 493 McFadden 288 Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood of Reduced Effect Model Chi-Square df Sig Intercept 1.149E3a 000 fl_net 1157.527 8.891 012 inf_net 1149.837 1.201 549 trust 1151.578 2.942 230 age 1159.829 11.193 004 121 edu 1153.951 5.315 070 distance 1154.059 5.423 066 inc 1149.064 428 807 age_squ 1158.759 10.123 006 h_head 1150.104 1.468 480 sex 1153.809 5.174 075 ethnic 1168.157 19.521 000 collat 1530.386 381.750 000 The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model The reduced model is formed by omitting an effect from the final model The null hypothesis is that all parameters of that effect are a This reduced model is equivalent to the final model because omitting the effect does not increase the degrees of freedom 122 Parameter Estimates 95% Confidence Interval for Exp(B) Std Tiep can cac loai TD a B Tiep can TD ban Intercept chinh thuc Lower Error Wald df Sig Exp(B) Bound Upper Bound -8.040 2.111 14.501 000 fl_net 397 134 8.719 003 1.487 1.143 1.935 inf_net -.009 024 146 702 991 945 1.039 trust 292 239 1.496 221 1.339 839 2.139 age 083 077 1.163 281 1.087 934 1.265 edu 017 043 158 691 1.017 935 1.106 -.015 011 1.687 194 985 963 1.008 inc 000 000 153 695 1.000 1.000 1.000 age_squ 000 001 978 323 999 998 1.001 [h_head=0] 350 359 950 330 1.420 702 2.871 [h_head=1] 0b [sex=0] 768 340 5.119 024 2.156 1.108 4.194 [sex=1] 0b [ethnic=0] -.271 330 676 411 763 400 1.455 [ethnic=1] 0b [collat=0] 3.450 485 50.606 000 31.488 12.173 81.453 [collat=1] 0b -1.311 1.368 918 338 distance Tiep can TD phi Intercept chinh thuc fl_net 121 104 1.349 245 1.129 920 1.384 inf_net 012 018 438 508 1.012 977 1.048 trust -.085 162 276 599 918 668 1.262 age -.121 048 6.462 011 886 807 973 edu -.062 033 3.614 057 940 882 1.002 distance 003 003 704 401 1.003 997 1.009 inc 000 000 130 719 1.000 1.000 1.000 age_squ 001 000 6.043 014 1.001 1.000 1.002 [h_head=0] -.057 291 038 846 945 535 1.670 [h_head=1] 0b [sex=0] 200 272 542 462 1.222 717 2.081 [sex=1] 0b 123 [ethnic=0] -1.106 262 17.797 000 331 198 553 [ethnic=1] 0b [collat=0] 5.310 611 75.508 000 202.384 61.095 670.417 [collat=1] 0b a The reference category is: Tiep can TD chinh thuc b This parameter is set to zero because it is redundant Classification Predicted Tiep can TD chinh Observed Tiep can TD chinh thuc Tiep can TD ban Tiep can TD phi thuc chinh thuc chinh thuc Percent Correct 352 127 72.9% Tiep can TD ban chinh thuc 25 12 63 12.0% Tiep can TD phi chinh thuc 53 216 78.3% 50.1% 2.7% 47.3% 67.5% Overall Percentage ... NG C A V N XÃ H TI P C N TÍN D NG C A CÁC H NƠNG THƠN VI T NAM N : Kinh t phát tri n : 60310105 LU TH PGS TS NGUY N TR NG HOÀI i L ng lu h ng c a v n xã h nông thôn Vi n ti p c n tín d ng c a... VXH ti p c n tín d ng p lý thuy t v VXH, lý thuy t v tài vi mơ lý thuy t thông tin b t cân x ng ho ng tín d ng Ph n hai trình bày th ng tín d ng vi c ti p c n lo i hình tín d ng nông thôn Vi t Nam... -Gi i tính -Tu i -Ch h h cv n -Thu nh p -Dân t c - Ti p c n tín d ng Tín d ng bán th c -H i Ph n -H i Nông dân -H i c u chi n binh c a NGO -Các h i tín d ng khác Ngu n: xu t c a tác gi t Tín d