BỘ GIÁO DỤC VÀ ĐÀO TẠO NGÂN HÀNG NHÀ NƢỚC VIỆT NAM TRƢỜNG ĐẠI HỌC NGÂN HÀNG TP HỒ CHÍ MINH NGUYỄN MẠNH HÙNG PHÁT TRIỂN CÁC TỔ CHỨC TÀI CHÍNH VI MÔ TẠI VIỆT NAM LUẬN ÁN TIẾN SĨ TÀI CHÍNH NGÂN HÀNG TP H[.]
BỘ GIÁO DỤC VÀ ĐÀO TẠO NGÂN HÀNG NHÀ NƢỚC VIỆT NAM TRƢỜNG ĐẠI HỌC NGÂN HÀNG TP HỒ CHÍ MINH - NGUYỄN MẠNH HÙNG PHÁT TRIỂN CÁC TỔ CHỨC TÀI CHÍNH VI MƠ TẠI VIỆT NAM LUẬN ÁN TIẾN SĨ TÀI CHÍNH NGÂN HÀNG TP HỒ CHÍ MINH NĂM 2022 Tai ngay!!! Ban co the xoa dong chu nay!!! BỘ GIÁO DỤC VÀ ĐÀO TẠO NGÂN HÀNG NHÀ NƢỚC VIỆT NAM TRƢỜNG ĐẠI HỌC NGÂN HÀNG TP HỒ CHÍ MINH NGUYỄN MẠNH HÙNG PHÁT TRIỂN CÁC TỔ CHỨC TÀI CHÍNH VI MƠ TẠI VIỆT NAM LUẬN ÁN TIẾN SĨ Chuyên ngành: Tài Ngân hàng Mã số: 9.34.02.01 Ngƣời hƣớng dẫn khoa học: PGS TS HÀ QUANG ĐÀO TS NGUYỄN THẾ BÍNH TP HỒ CHÍ MINH NĂM 2022 i LỜI CAM ĐOAN Tên là: Nguyễn Mạnh Hùng Sinh ngày: 28/12/1988 tại: Hà Nội Là nghiên cứu sinh khóa XXI Trƣờng Đại học Ngân hàng TP HCM Chuyên ngành: Tài - Ngân hàng Mã số: 9.34.02.01 Đề tài nghiên cứu: Phát triển tổ chức tài vi mô Việt Nam Tôi xin cam đoan: Đây luận án thân trực tiếp thực dƣới hƣớng dẫn PGS.TS Hà Quang Đào TS Nguyễn Thế Bính; Nghiên cứu không trùng lặp với nghiên cứu khác đƣợc công bố trƣớc đây; Các số liệu thơng tin nghiên cứu đảm bảo tính xác, trung thực khách quan, theo quy định hành Tơi xin hồn tồn chịu trách nhiệm cam đoan TP HCM ngày tháng năm 2022 NGHIÊN CỨU SINH Nguyễn Mạnh Hùng ii LỜI CẢM ƠN Lời đầu tiên, xin gửi lời cảm ơn đến Ban Giám hiệu tồn thể q Thầy, Cơ, đội ngũ giảng viên, viên chức, ngƣời lao động, Viện Nghiên cứu Khoa học Công nghệ Ngân hàng, Khoa Sau đại học, phòng ban chức Trƣờng Đại học Ngân hàng Thành phố Hồ Chí Minh giúp đỡ, tạo điều kiện tốt cho tơi q trình học tập, nghiên cứu bảo vệ luận án theo quy định Tôi xin cảm ơn PGS TS Hà Quang Đào TS Nguyễn Thế Bính quan tâm động viên, hƣớng dẫn giúp đỡ tơi q trình nghiên cứu hoàn thành luận án Đồng thời bày tỏ lời cảm ơn chân thành đến tổ chức nhà khoa học có ý kiến nhận xét, góp ý tóm tắt luận án; gia đình bạn bè suốt trình học tập, nghiên cứu thực luận án Xin gửi lời cảm ơn đến tất ngƣời NGHIÊN CỨU SINH Nguyễn Mạnh Hùng iii TĨM TẮT LUẬN ÁN Luận án có mục tiêu tổng quát đề xuất giải pháp hàm ý sách phù hợp nhằm phát triển tổ chức tài vi mơ Việt Nam tƣơng lai Để đạt đƣợc mục tiêu, luận án hệ thống hóa sở lý thuyết tài vi mơ, tổ chức tài vi mơ, phát triển tổ chức tài vi mơ lý thuyết liên quan Từ đó, luận án phân tích thực trạng phát triển tổ chức tài vi mơ Việt Nam Bên cạnh đó, hoạt động tổ chức tài vi mơ bền vững tảng phát triển (Schreiner 2002; CGAP 2003), luận án phân tích đo lƣờng yếu tố tác động đến bền vững tổ chức tài vi mơ Luận án thu thập số liệu 27 tổ chức tài vi mô Việt Nam giai đoạn 2008 - 2019, có tổ chức thức 23 tổ chức bán thức Luận án sử dụng phƣơng pháp nghiên cứu định tính nhƣ so sánh, phân tích thống kê mô tả bên cạnh phƣơng pháp nghiên cứu định lƣợng Luận án sử dụng nhiều phƣơng pháp ƣớc lƣợng khác liệu bảng biến tƣơng tác để đánh giá khác mức độ tác động hai nhóm tổ chức tài vi mơ thức bán thức Kết nghiên cứu cho thấy hầu hết số tài tổ chức tài vi mơ nhìn chung đạt triển vọng, tổ chức thức có số tốt tổ chức bán thức Khi đánh giá yếu tố tác động đến bền vững tổ chức tài vi mơ giai đoạn 2008 - 2019, nghiên cứu biến tuổi không tác động biến có tác động với mức độ khác đến tổ chức Trong quy mơ tỷ suất lợi tức danh mục cho vay tác động chiều đến bền vững tổ chức tài vi mơ Ngồi biến cịn lại gồm: Tỷ lệ nợ/VCSH, kích cỡ khoản vay trung bình, tỷ lệ nợ hạn, tỷ lệ chi phí hoạt động, chi phí trung bình khách hàng vay hiệu suất nhân viên tác động ngƣợc chiều đến bền vững TCTCVM Đề tài có tính đánh giá tác động khác biệt yếu tố đến bền vững hai nhóm tổ chức tài vi mơ thức bán thức Trong biến LNTA, CPB, OER, PAR30, ALS có tác động mạnh đến bền vững tổ chức tài vi mơ nhóm bán thức so với nhóm thức Tuy nhiên tổ chức tài vi mơ thức chịu tác động ngƣợc iv chiều PPR lớn so với tổ chức tài vi mơ bán thức Các biến cịn lại DER, PY có tác động nhƣ hai nhóm tổ chức Trên sở luận án đề xuất giải pháp nhằm phát triển hoạt động tổ chức tài vi mơ Việt Nam nhƣ tập trung vào việc tăng quy mô vốn chủ sở hữu; xây dựng sách lãi phí; biện pháp phịng ngừa rủi ro tín dụng; nâng cao lực cán bộ, nhân viên; ứng dụng công nghệ thông tin q trình hoạt động; thâm nhập cơng xóa đói giảm nghèo bên cạnh việc xây dựng tƣ thƣơng mại; thúc đẩy q trình thức hóa tổ chức tài vi mơ bán thức phát triển sản phẩm dịch vụ v ABSTRACT The general objective of the thesis has been that proposing appropriate solutions and policy implications would be implied the development of microfinance institutions (MFIs) in Vietnam in the future To achieve that goal, the thesis priorly systematized the theoretical basis of microfinance, MFIs, development of MFIs and related theories Since then, the thesis analyzed the current situation of development of MFIs in Vietnam Besides sustainability is the foundation of development in the operation of MFIs (Schreiner 2002; CGAP 2003) So the thesis has analyzed and measured factors which have affected on sustainability of MFIs The thesis has collected data of 27 microfinance institutions in Vietnam in the period 2008 - 2019, including formal organizations and 23 semi-formal organizations The thesis has used qualitative research methods such as comparison, descriptive statistical analysis, in addition to quantitative research methods The thesis has used many different estimation methods for panel data and interactive variables to evaluate the difference in the level of impact between two groups which include formal and semi-formal microfinance institutions Research results have showed that most of the financial indicators of MFIs have been generally favorable, in which formal institutions have had better indicators than semi-formal institutions When assessing the factors affecting the sustainability of MFIs in the period 2008 - 2019, the study has showed that the age variable has had no impact and variables have had different levels of impact on the subtainable institutions In which, the size and rate of return on the loan portfolio have a positive impact on the sustainability of microfinance institutions In addition, the remaining variables include: Debt/equity ratio, average loan size, delinquency ratio, operating expense ratio, average cost per borrower, and staff performance negatively affect the sustainability of microfinance institutions The new point could be known that the thesis has evaluated the different impacts of each factor on the sustainability of two groups which include formal and semi-formal MFIs In which LNTA, CPB, OER, PAR30, ALS variables have had stronger impacts on the sustainability of semi-formal MFIs than formal MFIs However, formal MFIs have been more negatively affected vi by PPR than semi-formal MFIs The remaining variables have been DER, PY which have been the same effect on the two groups of organizations On that basis, the thesis has also proposed solutions to develop the operation of MFIs in Vietnam such as focusing on increasing the size and equity; formulating policies on interest and fees; improving risk management capacity; improving the quality of staff; applying technology to reduce costs; penetrating deeply in poverty reduction and maintain a commercial mindset; promoting the formalization of semi-formal MFIs and development products and services vii DANH MỤC TỪ VIẾT TẮT Stt Ký hiệu Tiếng Việt ADB Ngân hàng phát triển Châu Á BTC Bộ Tài Tiếng Anh Asian Development Bank CEP FEM Tổ chức TCVM TNHH MTV cho ngƣời lao động nghèo tự tạo việc làm làm Mơ hình tác động cố định Fixed effects model FSS Bền vững tài Financial Self-Sustainablity ISS Bền vững thể chế Institutional SelfSustainablity MFI Tổ chức tài vi mơ Microfinance Institution NGO Tổ chức phi phủ Non government organization NHCSXH Ngân hàng sách xã hội 10 NHNN Ngân hàng Nhà nƣớc 11 NHTM Ngân hàng thƣơng mại 12 NHTW Ngân hàng trung ƣơng 13 OSS Bền vững hoạt động Operational Self-Sustainablity 14 PAR Tỷ lệ nợ hạn Portfolio at Risk 15 REM Mơ hình tác động ngẫu nhiên Random effects model 16 ROA Return on asset 17 ROE 18 SGMM Tỷ suất lợi nhuận tổng tài sản Tỷ suất lợi nhuận vốn chủ sở hữu Phƣơng pháp mô men tổng quát hệ thống 19 TCTCVM Tổ chức tài vi mơ Return on equity System Generalized method of moments MFIs - microfinance institutions viii Stt Ký hiệu Tiếng Việt 20 TCTD Tổ chức tín dụng 21 TCVM Tài vi mơ 22 TNHH Trách nhiệm hữu hạn 24 TYM Tổ chức TCVM TNHH Tình thƣơng 25 VCSH Vốn chủ sở hữu 26 WB Ngân hàng Thế giới Tiếng Anh World Bank 47 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 144.27 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roa Coef [95% Conf Interval] roa L1 .2801288 0496197 5.65 0.000 178134 3821237 age lnta der als par30 oer cpb ppr py par30formal _cons -.0570268 4103064 -.0645017 -3.53e-07 1.78547 -.8082941 9.46e-07 -.0018301 4795979 -8.194104 -3.387867 0727302 3172769 0328369 1.48e-07 4222758 1190602 4.87e-07 0013205 0769396 4.538456 6.298093 -0.78 1.29 -1.96 -2.39 4.23 -6.79 1.94 -1.39 6.23 -1.81 -0.54 0.440 0.207 0.060 0.024 0.000 0.000 0.063 0.178 0.000 0.083 0.595 -.2065257 -.2418655 -.131999 -6.56e-07 9174695 -1.053026 -5.60e-08 -.0045445 3214464 -17.52303 -16.33378 0924722 1.062478 0029955 -4.94e-08 2.65347 -.5635623 1.95e-06 0008843 6377495 1.134825 9.558049 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age lnta py der par30 als cpb ppr) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roa collapsed Instruments for levels equation Standard age lnta py der par30 als cpb ppr _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roa collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 1.95 but not weakened by many instruments.) overid restrictions: chi2(8) = 12.03 weakened by many instruments.) -1.97 -0.55 Pr > z = Pr > z = 0.048 0.581 Prob > chi2 = 0.983 Prob > chi2 = 0.150 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 11.99 Prob > Difference (null H = exogenous): chi2(1) = 0.04 Prob > iv(age lnta py der par30 als cpb ppr) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(8) = 12.03 Prob > chi2 = chi2 = 0.101 0.834 chi2 = chi2 = 0.150 48 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 52.17 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max roa Coef Std Err roa L1 .1590383 0755778 age lnta der als par30 oer cpb ppr py oerformal _cons -.0365617 -.6312788 -.0666723 -8.29e-08 3938241 -.3193833 -9.98e-08 002592 2417575 3753666 16.89111 1075554 1682413 0190709 8.86e-08 365664 0449659 3.36e-07 0016178 0509773 2180812 3.982434 t = = = = = 272 27 10.07 11 P>|t| [95% Conf Interval] 2.10 0.045 0036859 3143906 -0.34 -3.75 -3.50 -0.94 1.08 -7.10 -0.30 1.60 4.74 1.72 4.24 0.737 0.001 0.002 0.358 0.291 0.000 0.769 0.121 0.000 0.097 0.000 -.257645 -.9771038 -.105873 -2.65e-07 -.3578091 -.411812 -7.91e-07 -.0007334 1369721 -.0729056 8.7051 1845215 -.2854539 -.0274715 9.93e-08 1.145457 -.2269546 5.91e-07 0059174 3465429 8236389 25.07712 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L2.roa collapsed Instruments for levels equation Standard age lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L2.roa collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Pr > z = Pr > z = 0.060 0.218 Prob > chi2 = 0.911 Prob > chi2 = 0.394 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 8.16 Prob > chi2 = Difference (null H = exogenous): chi2(1) = 0.25 Prob > chi2 = 0.318 0.618 Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 3.35 but not weakened by many instruments.) overid restrictions: chi2(8) = 8.41 weakened by many instruments.) -1.88 -1.23 49 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 90.09 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max roa Coef Std Err roa L1 .1434665 0662794 age lnta der als par30 oer cpb ppr py cpbformal _cons -.4908565 4303471 -.040439 4.28e-07 1.175933 -.1805603 -2.96e-06 0061199 419888 2.91e-06 -11.45264 1963545 7226206 0388314 6.36e-07 5060408 3012016 6.72e-06 0034147 11725 7.38e-06 12.76052 t = = = = = 272 27 10.07 11 P>|t| [95% Conf Interval] 2.16 0.040 0072272 2797057 -2.50 0.60 -1.04 0.67 2.32 -0.60 -0.44 1.79 3.58 0.39 -0.90 0.019 0.557 0.307 0.507 0.028 0.554 0.663 0.085 0.001 0.697 0.378 -.8944689 -1.055021 -.1202581 -8.79e-07 1357511 -.7996891 -.0000168 -.0008991 1788772 -.0000123 -37.68226 -.0872441 1.915715 03938 1.74e-06 2.216115 4385684 0000108 0131389 6608988 0000181 14.77699 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roa collapsed Instruments for levels equation Standard lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roa collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 2.95 but not weakened by many instruments.) overid restrictions: chi2(8) = 8.38 weakened by many instruments.) -2.23 -0.85 Pr > z = Pr > z = 0.026 0.397 Prob > chi2 = 0.937 Prob > chi2 = 0.398 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 7.57 Prob > Difference (null H = exogenous): chi2(1) = 0.80 Prob > iv(lnta der als par30 oer cpb ppr py) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(8) = 8.38 Prob > chi2 = chi2 = 0.372 0.370 chi2 = chi2 = 0.398 50 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 127.14 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roa Coef [95% Conf Interval] roa L1 .2050416 0345004 5.94 0.000 134125 2759582 age lnta der als par30 oer cpb ppr py pprformal _cons -.3465167 8259196 -.0405915 1.26e-07 1.085795 -.3540135 -1.99e-07 0032081 3782783 -.0027218 -18.34219 1507364 4907955 0239035 1.74e-07 3831059 069611 1.81e-07 0026938 0453265 0096591 10.90158 -2.30 1.68 -1.70 0.72 2.83 -5.09 -1.10 1.19 8.35 -0.28 -1.68 0.030 0.104 0.101 0.475 0.009 0.000 0.280 0.244 0.000 0.780 0.104 -.6563598 -.1829249 -.0897258 -2.31e-07 2983091 -.4971009 -5.70e-07 -.0023291 2851084 -.0225763 -40.7507 -.0366736 1.834764 0085428 4.84e-07 1.87328 -.2109261 1.72e-07 0087453 4714481 0171326 4.066325 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roa collapsed Instruments for levels equation Standard lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roa collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 3.18 but not weakened by many instruments.) overid restrictions: chi2(8) = 8.88 weakened by many instruments.) -1.80 -1.36 Pr > z = Pr > z = 0.071 0.175 Prob > chi2 = 0.922 Prob > chi2 = 0.352 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 8.64 Prob > Difference (null H = exogenous): chi2(1) = 0.24 Prob > iv(lnta der als par30 oer cpb ppr py) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(8) = 8.88 Prob > chi2 = chi2 = 0.280 0.622 chi2 = chi2 = 0.352 51 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 47.24 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roa Coef [95% Conf Interval] roa L1 .1115874 0775332 1.44 0.162 -.0477844 2709592 age lnta der als par30 oer cpb ppr py pyformal _cons -.0815145 -.3702743 -.0664079 -7.70e-08 4004906 -.3089025 -1.95e-08 0027016 2286303 2038677 11.38836 1008591 2170244 0158419 9.80e-08 3318773 0513418 2.31e-07 0016838 0449802 1109057 4.836514 -0.81 -1.71 -4.19 -0.79 1.21 -6.02 -0.08 1.60 5.08 1.84 2.35 0.426 0.100 0.000 0.439 0.238 0.000 0.934 0.121 0.000 0.077 0.026 -.2888333 -.8163744 -.0989713 -2.78e-07 -.281693 -.4144371 -4.95e-07 -.0007596 1361722 -.0241022 1.446762 1258044 0758257 -.0338444 1.24e-07 1.082674 -.2033678 4.56e-07 0061628 3210884 4318377 21.32996 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age L.lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L2.roa collapsed Instruments for levels equation Standard age L.lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L2.roa collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Pr > z = Pr > z = 0.087 0.189 Prob > chi2 = 0.884 Prob > chi2 = 0.384 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 8.07 Prob > chi2 = Difference (null H = exogenous): chi2(1) = 0.46 Prob > chi2 = 0.327 0.497 Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 3.68 but not weakened by many instruments.) overid restrictions: chi2(8) = 8.53 weakened by many instruments.) -1.71 -1.31 52 PHỤ LỤC 12 MỨC ĐỘ TÁC ĐỘNG CỦA CÁC BIẾN TƢƠNG TÁC ĐẾN BIẾN PHỤ THUỘC ROE MƠ HÌNH HỒI QUY SGMM Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 45.36 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roe Coef [95% Conf Interval] roe L1 -.0098031 0658043 -0.15 0.883 -.1450657 1254594 age lnta der als par30 oer cpb ppr py ageformal _cons 8719341 10.93006 1798124 -7.47e-07 4901451 -.9300022 -1.54e-07 -.0193963 9336667 -2.635751 -249.78 5713837 3.215629 1924353 5.28e-07 1.598001 3280806 8.76e-07 0079644 142971 1.112936 77.04199 1.53 3.40 0.93 -1.42 0.31 -2.83 -0.18 -2.44 6.53 -2.37 -3.24 0.139 0.002 0.359 0.169 0.761 0.009 0.862 0.022 0.000 0.026 0.003 -.3025619 4.320241 -.2157441 -1.83e-06 -2.794593 -1.604382 -1.95e-06 -.0357673 6397857 -4.923424 -408.1421 2.04643 17.53988 575369 3.37e-07 3.774884 -.2556229 1.65e-06 -.0030253 1.227548 -.3480769 -91.41792 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age L.lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L2.roe collapsed Instruments for levels equation Standard age L.lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L2.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Pr > z = Pr > z = 0.036 0.164 Prob > chi2 = 0.145 Prob > chi2 = 0.868 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 3.37 Prob > chi2 = Difference (null H = exogenous): chi2(1) = 0.51 Prob > chi2 = 0.849 0.475 Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 12.14 but not weakened by many instruments.) overid restrictions: chi2(8) = 3.88 weakened by many instruments.) -2.10 -1.39 53 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 9.29 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roe Coef [95% Conf Interval] roe L1 .079465 0542219 1.47 0.155 -.0319898 1909197 age lnta der als par30 oer cpb ppr py lntaformal _cons 6112908 9.772358 158095 -8.46e-07 0982826 -1.022883 -5.54e-07 -.019894 9676377 -1.84062 -218.4358 565359 3.611584 1992907 5.47e-07 1.935122 3388495 8.80e-07 00859 2318536 8094 85.07472 1.08 2.71 0.79 -1.54 0.05 -3.02 -0.63 -2.32 4.17 -2.27 -2.57 0.290 0.012 0.435 0.134 0.960 0.006 0.535 0.029 0.000 0.031 0.016 -.5508213 2.348641 -.251553 -1.97e-06 -3.879417 -1.719398 -2.36e-06 -.0375509 4910557 -3.504366 -393.3094 1.773403 17.19607 5677429 2.80e-07 4.075982 -.3263684 1.26e-06 -.0022371 1.44422 -.1768749 -43.56218 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age L.lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L2.roe collapsed Instruments for levels equation Standard age L.lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L2.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Pr > z = Pr > z = 0.022 0.152 Prob > chi2 = 0.154 Prob > chi2 = 0.549 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 5.76 Prob > chi2 = Difference (null H = exogenous): chi2(1) = 1.13 Prob > chi2 = 0.569 0.288 Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 11.95 but not weakened by many instruments.) overid restrictions: chi2(8) = 6.88 weakened by many instruments.) -2.29 -1.43 54 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 27 F(11, 26) = 1247.94 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roe Coef [95% Conf Interval] roe L1 -.0521087 0502876 -1.04 0.310 -.1554762 0512589 age lnta der als par30 oer cpb ppr py derformal _cons 0727267 7983188 1927163 -2.79e-07 231108 -.8749119 -4.05e-07 -.0014165 8025311 7943447 -11.71895 2068969 9084684 0998783 2.76e-07 1.181205 2414723 5.39e-07 0019758 1497894 1494121 21.09206 0.35 0.88 1.93 -1.01 0.20 -3.62 -0.75 -0.72 5.36 5.32 -0.56 0.728 0.388 0.065 0.320 0.846 0.001 0.459 0.480 0.000 0.000 0.583 -.3525561 -1.069065 -.0125865 -8.45e-07 -2.196893 -1.371265 -1.51e-06 -.005478 4946346 4872237 -55.07431 4980095 2.665702 3980191 2.87e-07 2.659109 -.3785586 7.02e-07 0026449 1.110428 1.101466 31.6364 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age py der par30 als oer cpb ppr) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).(L3.roe L3.derformal) collapsed Instruments for levels equation Standard age py der par30 als oer cpb ppr _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.(L3.roe L3.derformal) collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(15) = 19.24 but not weakened by many instruments.) overid restrictions: chi2(15) = 6.98 weakened by many instruments.) -1.99 -1.39 Pr > z = Pr > z = 0.047 0.165 Prob > chi2 = 0.203 Prob > chi2 = 0.958 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(13) = 8.04 Prob > Difference (null H = exogenous): chi2(2) = -1.07 Prob > iv(age py der par30 als oer cpb ppr) Hansen test excluding group: chi2(7) = 2.20 Prob > Difference (null H = exogenous): chi2(8) = 4.77 Prob > chi2 = chi2 = 0.841 1.000 chi2 = chi2 = 0.948 0.782 55 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 60.92 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roe Coef [95% Conf Interval] roe L1 .3672434 049206 7.46 0.000 266099 4683878 age lnta der als par30 oer cpb ppr py alsformal _cons -.2011319 2.465128 1207691 -1.43e-07 5054909 -.7041422 -4.02e-07 -.0023905 6522168 -2.06e-07 -52.41366 3852794 2.271471 2197079 5.52e-07 1.610301 2095251 3.87e-07 0047683 2164403 2.56e-06 57.45475 -0.52 1.09 0.55 -0.26 0.31 -3.36 -1.04 -0.50 3.01 -0.08 -0.91 0.606 0.288 0.587 0.798 0.756 0.002 0.308 0.620 0.006 0.937 0.370 -.9930851 -2.203948 -.330847 -1.28e-06 -2.80453 -1.134827 -1.20e-06 -.0121919 2073175 -5.48e-06 -170.5136 5908214 7.134204 5723852 9.92e-07 3.815512 -.2734572 3.93e-07 0074109 1.097116 5.07e-06 65.68627 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age py der par30 als oer cpb ppr) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roe collapsed Instruments for levels equation Standard age py der par30 als oer cpb ppr _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 18.25 but not weakened by many instruments.) overid restrictions: chi2(8) = 7.20 weakened by many instruments.) -2.24 -1.34 Pr > z = Pr > z = 0.025 0.182 Prob > chi2 = 0.019 Prob > chi2 = 0.515 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 3.05 Prob > Difference (null H = exogenous): chi2(1) = 4.15 Prob > iv(age py der par30 als oer cpb ppr) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(8) = 7.20 Prob > chi2 = chi2 = 0.880 0.042 chi2 = chi2 = 0.515 56 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 21 F(11, 26) = 273.62 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max roe Coef Std Err roe L1 .360353 0358462 age lnta der als par30 oer cpb ppr py par30formal _cons -.2492724 1.78697 0465147 -6.68e-08 3.303444 -1.037552 -6.58e-07 -.0040645 8966095 -17.12588 -35.16905 1869076 6437112 0888771 3.01e-07 2.412146 2881067 5.88e-07 0036934 184285 18.49329 11.84644 t = = = = = 272 27 10.07 11 P>|t| [95% Conf Interval] 10.05 0.000 2866701 4340359 -1.33 2.78 0.52 -0.22 1.37 -3.60 -1.12 -1.10 4.87 -0.93 -2.97 0.194 0.010 0.605 0.826 0.183 0.001 0.273 0.281 0.000 0.363 0.006 -.6334665 4638033 -.1361748 -6.86e-07 -1.654792 -1.629764 -1.87e-06 -.0116564 5178063 -55.13938 -59.51976 1349217 3.110138 2292043 5.52e-07 8.26168 -.4453406 5.51e-07 0035274 1.275413 20.88763 -10.81834 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roe collapsed Instruments for levels equation Standard age lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(9) = 16.71 but not weakened by many instruments.) overid restrictions: chi2(9) = 7.82 weakened by many instruments.) -2.45 -1.00 Pr > z = Pr > z = 0.014 0.319 Prob > chi2 = 0.053 Prob > chi2 = 0.552 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(8) = 3.54 Prob > Difference (null H = exogenous): chi2(1) = 4.28 Prob > iv(age lnta der als par30 oer cpb ppr py) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(9) = 7.82 Prob > chi2 = chi2 = 0.896 0.039 chi2 = chi2 = 0.552 57 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 27 F(11, 26) = 290.74 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max Std Err t P>|t| = = = = = 272 27 10.07 11 roe Coef [95% Conf Interval] roe L1 -.0588122 0654427 -0.90 0.377 -.1933316 0757072 age lnta der als par30 oer cpb ppr py oerformal _cons -.2042768 -.9753239 0970529 -1.68e-07 -.798619 -.8066599 -5.94e-07 0040196 5395227 1.731694 33.40969 2694385 6507348 1057619 2.63e-07 1.076054 2317685 5.96e-07 0037502 1791412 2246934 15.62099 -0.76 -1.50 0.92 -0.64 -0.74 -3.48 -1.00 1.07 3.01 7.71 2.14 0.455 0.146 0.367 0.529 0.465 0.002 0.328 0.294 0.006 0.000 0.042 -.7581156 -2.312928 -.1203437 -7.09e-07 -3.010479 -1.283067 -1.82e-06 -.0036891 1712926 1.26983 1.300286 3495619 3622807 3144496 3.73e-07 1.413241 -.3302528 6.31e-07 0117282 9077527 2.193558 65.51909 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).(L4.roe L3.derformal) collapsed Instruments for levels equation Standard age lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.(L4.roe L3.derformal) collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(15) = 15.09 but not weakened by many instruments.) overid restrictions: chi2(15) = 6.02 weakened by many instruments.) -1.66 -1.37 Pr > z = Pr > z = 0.097 0.170 Prob > chi2 = 0.445 Prob > chi2 = 0.979 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(13) = 6.87 Prob > Difference (null H = exogenous): chi2(2) = -0.85 Prob > iv(age lnta der als par30 oer cpb ppr py) Hansen test excluding group: chi2(6) = 2.09 Prob > Difference (null H = exogenous): chi2(9) = 3.93 Prob > chi2 = chi2 = 0.909 1.000 chi2 = chi2 = 0.911 0.916 58 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 81.38 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max roe Coef Std Err roe L1 .3716001 0592529 age lnta der als par30 oer cpb ppr py cpbformal _cons 4771785 5.094358 2556558 -2.65e-06 2.682314 -2.210707 0000336 -.0089639 8560502 -.0000366 -109.9675 7679755 2.528701 1425093 1.59e-06 1.459968 8832331 0000181 0080652 3044212 0000207 48.6425 t = = = = = 272 27 10.07 11 P>|t| [95% Conf Interval] 6.27 0.000 2498039 4933962 0.62 2.01 1.79 -1.67 1.84 -2.50 1.86 -1.11 2.81 -1.77 -2.26 0.540 0.054 0.084 0.107 0.078 0.019 0.074 0.277 0.009 0.088 0.032 -1.101418 -.1034604 -.0372763 -5.92e-06 -.3186929 -4.026219 -3.56e-06 -.0255422 2303035 -.0000791 -209.9536 2.055775 10.29218 5485879 6.11e-07 5.68332 -.3951959 0000708 0076144 1.481797 5.84e-06 -9.981393 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roe collapsed Instruments for levels equation Standard lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 14.06 but not weakened by many instruments.) overid restrictions: chi2(8) = 7.14 weakened by many instruments.) -2.25 -1.46 Pr > z = Pr > z = 0.024 0.146 Prob > chi2 = 0.080 Prob > chi2 = 0.521 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 3.86 Prob > Difference (null H = exogenous): chi2(1) = 3.28 Prob > iv(lnta der als par30 oer cpb ppr py) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(8) = 7.14 Prob > chi2 = chi2 = 0.796 0.070 chi2 = chi2 = 0.521 59 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 20 F(11, 26) = 141.98 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max roe Coef Std Err roe L1 .3191727 0434221 age lnta der als par30 oer cpb ppr py pprformal _cons -.2080207 3.339056 1380926 -2.01e-07 7599559 -.7128162 -1.07e-07 -.0014107 7394686 -.0269667 -73.69146 4830489 1.341027 1166199 3.86e-07 1.489015 206947 4.93e-07 0049579 1259391 0230153 31.65818 t = = = = = 272 27 10.07 11 P>|t| [95% Conf Interval] 7.35 0.000 2299173 4084281 -0.43 2.49 1.18 -0.52 0.51 -3.44 -0.22 -0.28 5.87 -1.17 -2.33 0.670 0.019 0.247 0.607 0.614 0.002 0.830 0.778 0.000 0.252 0.028 -1.200942 5825354 -.1016231 -9.94e-07 -2.300758 -1.138202 -1.12e-06 -.0116018 4805972 -.0742753 -138.7658 7849005 6.095576 3778083 5.92e-07 3.820669 -.2874306 9.07e-07 0087805 9983401 0203419 -8.617141 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(lnta der als par30 oer cpb ppr py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roe collapsed Instruments for levels equation Standard lnta der als par30 oer cpb ppr py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(8) = 18.59 but not weakened by many instruments.) overid restrictions: chi2(8) = 6.13 weakened by many instruments.) -2.25 -1.33 Pr > z = Pr > z = 0.025 0.184 Prob > chi2 = 0.017 Prob > chi2 = 0.633 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(7) = 3.05 Prob > Difference (null H = exogenous): chi2(1) = 3.08 Prob > iv(lnta der als par30 oer cpb ppr py) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(8) = 6.13 Prob > chi2 = chi2 = 0.880 0.079 chi2 = chi2 = 0.633 60 Dynamic panel-data estimation, two-step system GMM Group variable: id Time variable : year Number of instruments = 19 F(11, 26) = 19.14 Prob > F = 0.000 Number of obs Number of groups Obs per group: avg max roe Coef Std Err roe L1 .1502457 0654595 age lnta der als par30 oer cpb ppr py pyformal _cons -.9921651 11.27983 3821925 -2.71e-07 12.25731 -1.093962 1.14e-07 -.0087598 1.45344 -1.228765 -264.1652 4857177 4.332548 1406863 5.92e-07 4.76342 4958533 4.70e-07 0085588 3030533 7040462 100.2267 t = = = = = 272 27 10.07 11 P>|t| [95% Conf Interval] 2.30 0.030 0156917 2847997 -2.04 2.60 2.72 -0.46 2.57 -2.21 0.24 -1.02 4.80 -1.75 -2.64 0.051 0.015 0.012 0.650 0.016 0.036 0.810 0.316 0.000 0.093 0.014 -1.990572 2.374153 0930077 -1.49e-06 2.465964 -2.113203 -8.52e-07 -.0263527 8305047 -2.675952 -470.1841 0062419 20.18551 6713772 9.45e-07 22.04866 -.0747208 1.08e-06 008833 2.076375 2184231 -58.14627 Warning: Uncorrected two-step standard errors are unreliable Instruments for first differences equation Standard D.(age der als oer ppr cpb py) GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/11).L.roe collapsed Instruments for levels equation Standard age der als oer ppr cpb py _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.roe collapsed Arellano-Bond test for AR(1) in first differences: z = Arellano-Bond test for AR(2) in first differences: z = Sargan test of (Not robust, Hansen test of (Robust, but overid restrictions: chi2(7) = 11.08 but not weakened by many instruments.) overid restrictions: chi2(7) = 5.46 weakened by many instruments.) -2.24 0.21 Pr > z = Pr > z = 0.025 0.834 Prob > chi2 = 0.135 Prob > chi2 = 0.604 Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(6) = 4.77 Prob > Difference (null H = exogenous): chi2(1) = 0.69 Prob > iv(age der als oer ppr cpb py) Hansen test excluding group: chi2(0) = 0.00 Prob > Difference (null H = exogenous): chi2(7) = 5.46 Prob > chi2 = chi2 = 0.573 0.407 chi2 = chi2 = 0.604