适宜性人力资本与区域经济协调发展
作者简介:生延超(1978- ),男,河南南阳人,博士,教授,主要从事区域经济与产业经济研究。E-mail: 22714407@qq.com
收稿日期: 2017-10-16
要求修回日期: 2018-01-06
网络出版日期: 2018-04-20
基金资助
国家社会科学基金重大项目(15ZDB132);国家社会科学基金青年项目(13CJL066);湖南省科技计划研究项目(2014FJ6056);湖南省社科规划项目(12JD41)
Appropriate human capital and coordinated development of regional economy
Received date: 2017-10-16
Request revised date: 2018-01-06
Online published: 2018-04-20
Copyright
适宜性人力资本必须与区域产业结构、区域经济发展模式相匹配,才能最大程度地发挥不同层次人力资本的效能,以有效提升区域经济发展水平。从全国和区域层面出发,基于ESDA方法和空间计量分析方法,运用1998-2014年全国30个省市的面板数据,从初级、中级、高级人力资本、物质资本、劳动力、全要素生产率等要素视角探寻适宜性人力资本促进区域经济发展效能的空间差异,研究表明:① 人力资本与省域生产总值水平均存在显著的空间相关性和路径依赖性。② 在全国层面,初级、中级人力资本对区域经济增长作用不显著,高级人力资本则作用很明显。③ 在区域层面,东部及东北地区区域经济发展主要依赖高级人力资本,中部和西部初级人力资本贡献显著,中级人力资本在四个区域均不显著。这为区域人力资本培育和区域经济高速发展政策制定奠定了理论基础。
生延超 , 周玉姣 . 适宜性人力资本与区域经济协调发展[J]. 地理研究, 2018 , 37(4) : 797 -813 . DOI: 10.11821/dlyj201804013
Appropriate human capital (AHC) should match regional industrial structure and regional economic development model, because a matched AHC can maximize human capital effectiveness to enhance the level of regional economic development. Taking both national and regional levels into account, we use the panel data of 30 Chinese provinces and cities from 1998 to 2014. We introduce ESDA method and spatial econometric method for us to examine the spatial differences in AHC promoting regional economic development effectiveness from the perspectives of primary human capital, intermediate human capital, advanced human capital, material capital, labor force, total factor productivity and other factors. Our research shows that (1) there are significant spatial correlations and path dependence between human capital and regional GDP, (2) at the national level, the impact of primary and intermediate human capitals on regional economic growth is insignificant, whereas the impact of advanced human capital is obvious, and (3) at the regional level, in the eastern and northeastern regions, the economic development mainly relies on advanced human capital, in the central and western regions, the impact of primary human capital is significant, and in the four regions the impact of intermediate human capital is not significant. The findings lay a theoretical foundation for regional talent cultivation and policy making of regional economic development.
Tab. 1 Human capital Moran's I index and its Z values of China's 30 provinces表1 中国30个省域人力资本Moran's I指数及其Z值 |
| 时间 | Moran's I | sd | Z值 | P值 |
|---|---|---|---|---|
| 1998 | 0.333 | 0.1089 | 3.3636 | 0.004 |
| 2003 | 0.3341 | 0.1076 | 3.4145 | 0.003 |
| 2008 | 0.3286 | 0.1060 | 3.4142 | 0.005 |
| 2014 | 0.3784 | 0.1124 | 3.6628 | 0.004 |
Tab. 2 Spatial correlation patterns of human capital in various provinces of China from 1998 to 2014表2 1998-2014年有代表性年份各省域人力资本的空间相关模式②(② 由于采用地理邻接空间权值矩阵,海南空间相关性不显著,因而没有列出。) |
| 年份 | 第一象限(HH) | 第二象限(LH) | 第三象限(LL) | 第四象限(HL) |
|---|---|---|---|---|
| 1998 | 山东、江苏、河南、河北、浙江、上海、福建、安徽、湖北、湖南、辽宁 | 北京、广西、江西、山西、天津 | 重庆、贵州、陕西、吉林、内蒙古、云南、黑龙江、甘肃、青海、宁夏、新疆 | 广东、四川 |
| 2003 | 山东、江苏、河南、河北、浙江、上海、福建、安徽、湖北 | 北京、广西、江西、山西、天津、湖南 | 重庆、贵州、陕西、吉林、内蒙古、云南、黑龙江、甘肃、青海、宁夏、新疆 | 广东、四川、辽宁 |
| 2008 | 山东、江苏、河南、河北、浙江、上海、福建、安徽、湖北、北京 | 广西、江西、山西、天津、湖南 | 重庆、贵州、陕西、吉林、内蒙古、云南、黑龙江、甘肃、青海、宁夏、新疆、辽宁 | 广东、四川 |
| 2014 | 山东、江苏、河南、河北、上海、福建、安徽、湖北、北京 | 广西、山西、天津、湖南、江西 | 重庆、贵州、陕西、吉林、内蒙古、云南、黑龙江、甘肃、青海、宁夏、新疆 | 广东、四川、辽宁 |
Fig. 1 LISA cluster chart of human capital图1 人力资本的LISA集群图 |
Fig. 2 LISA cluster chart of regional GDP图2 地区生产总值的LISA集聚图 |
Tab. 3 The spatial correlation testing results based on geometric distance space weight matrix表3 基于地理距离空间权重矩阵的空间相关性检验结果 |
| 变量 | 空间效应检验 | |||
|---|---|---|---|---|
| 最小二乘估计 | 空间固定效应 | 时间固定效应 | 时空固定效应 | |
| FE_rsqr2 | 0.9933 | 0.9892 | 0.9981 | |
| LM-lag | 130.8652*** | 263.2731*** | 3.6398* | 4.0528** |
| LM-error | 174.2979*** | 70.4367*** | 3.4887* | 2.5570 |
| Robust LM-lag | 80.2733*** | 194.3314*** | 2.6523 | 1.5339 |
| Robust LM-error | 123.7060*** | 1.4950 | 2.5012 | 0.0380 |
| R2 | 0.9743 | 0.9823 | 0.9896 | 0.6259 |
| LIK | 148.8044 | 492.8111 | 371.1544 | 812.5794 |
| LR test | 882.8500*** | 639.5366*** | ||
注:***、**、*分别表示在1%、5%、10%水平下通过显著性检验。 |
Tab. 4 SLM and SEM regression results based on geometric distance space weight matrix表4 基于地理距离空间权值矩阵的SLM和SEM回归结果③(③ 由于本文采用地理邻接矩阵,仅考虑省域与邻近省域之间的人力资本空间相关性,未考量跨省域之间的空间相关性,而拥有中级人力资本的劳动力是区域人力资本流动的主力,也可能导致中级人力资本对区域经济增长影响减小,需进行下一步研究。) |
| 变量 | 空间滞后模型 | 空间误差模型 | |||
|---|---|---|---|---|---|
| 时空固定效应 | 随机效应 | 时空固定效应 | 随机效应 | ||
| lnpri | 0.0069 | -0.0139* | 0.0060 | 0.0226 | |
| (0.6465) | (-1.7567) | (0.5929) | (1.3644) | ||
| lnint | -0.0267 | -0.0255 | -0.0235 | -0.0680*** | |
| (-1.5204) | (-1.4493) | (-1.3785) | (-2.8068) | ||
| lnadv | 0.03102** | 0.0359** | 0.0296** | 0.0517*** | |
| (2.4677) | (2.5097) | (2.3746) | (3.5996) | ||
| lnmc | 0.0493*** | 0.0810*** | 0.0487 | 0.0888*** | |
| (5.9812) | (8.1592) | (5.9808) | (8.8454) | ||
| lnlab | -0.0656** | 0.1232*** | -0.0654** | 0.1513*** | |
| (-1.9668) | (3.1596) | (-2.0210) | (3.7156) | ||
| lntfp | 0.8852*** | 0.9173*** | 0.8770*** | 0.8974*** | |
| (26.1095) | (22.7826) | (26.3365) | (21.4475) | ||
| 常数项 | - | -3.2078*** | - | 0.5391 | |
| - | (-9.6138) | - | (1.2845) | ||
| ρ | -0.0839** | 0.4669*** | - | - | |
| (-1.8656) | (23.1118) | - | - | ||
| λ | - | - | -0.1069*** | 0.9530*** | |
| - | - | (-1.7126) | (127.3991) | ||
| Teta | - | 0.0230*** | - | 113.5472*** | |
| - | (5.4785) | - | (4.3651) | ||
| Hausman test | - | -150.9005*** | - | -44.2174*** | |
| R2 | 0.9981 | 0.9970 | 0.9981 | 0.9971 | |
| LR test | 1646.2938*** | 1156.2360*** | 1645.4935*** | 1007.9686*** | |
| LIK | 814.4135 | -150.9005 | 814.0134 | 495.2509 | |
注:***、**、*分别表示在1%、5%、10%水平下通过显著性检验,()内为统计值。 |
Tab. 5 Spatial effect tests of regional human capital and regional economic growth model表5 分区域人力资本与区域经济增长模型空间效应检验 |
| 区域 | 检验 | 统计量 | P值 |
|---|---|---|---|
| 东部 | LM-lag | 11.5863 | 0.001 |
| LM-error | 11.4267 | 0.001 | |
| Robust LM-lag | 1.4966 | 0.221 | |
| Robust LM-error | 1.3370 | 0.248 | |
| 中部 | LM-lag | 2.2231 | 0.136 |
| LM-error | 1.5443 | 0.214 | |
| Robust LM-lag | 13.1752 | 0.000 | |
| Robust LM-error | 12.4964 | 0.000 | |
| 西部 | LM-lag | 6.8045 | 0.009 |
| LM-error | 2.7567 | 0.097 | |
| Robust LM-lag | 13.6932 | 0.000 | |
| Robust LM-error | 9.6454 | 0.002 | |
| 东北部 | LM-lag | 1.5596 | 0.212 |
| LM-error | 4.6544 | 0.031 | |
| Robust LM-lag | 7.2425 | 0.007 | |
| Robust LM-error | 10.3373 | 0.001 |
Tab. 6 The estimation of SLM model for regional economic growth by regional human capital表6 分区域人力资本作用于区域经济增长的SLM模型估计结果 |
| 区域 | 东部 | 中部 | 西部 | 东北部 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| feF | reF | feF | reF | feF | reF | feF | reF | ||||
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | ||||
| lnpri | -0.0155 | -0.0506** | 0.0077*** | 0.0177 | 0.0982** | -0.2087*** | 0.1042 | -0.1431** | |||
| (-1.1708) | (-2.1173) | (0.2615) | (0.5034) | (2.4198) | (-3.123) | (1.4617) | (-2.351) | ||||
| lnint | -0.0205 | -0.0739 | 0.0403 | -0.2795*** | -0.1064** | -0.6759*** | -0.2039** | 0.1452 | |||
| (-0.7518) | (-1.4514) | (1.0680) | (-3.1679) | (-1.966) | (-5.085) | (-2.192) | (1.1240) | ||||
| lnadv | 0.0527*** | 0.1161 | -0.0151 | 0.2632*** | -0.0115 | 0.8169*** | 0.0623* | -0.0061 | |||
| (2.7518) | (3.056) | (-0.6918) | (3.9785) | (-0.361) | (7.0355) | (1.8947) | (-0.0620) | ||||
| lnmc | 0.0510*** | 0.0737 | 0.0065* | 0.2141*** | -0.0037 | 0.1292* | 0.4603*** | 0.5446*** | |||
| (4.2763) | (2.870) | (0.4659) | (4.4907) | (-0.218) | (1.7580) | (11.165) | (6.5448) | ||||
| lnlab | 0.2170*** | 0.8642 | -0.6139*** | 0.6993*** | -0.0068 | -0.0199 | 0.7304*** | 0.2340 | |||
| (4.2652) | (14.3871) | (-6.3901) | (4.1051) | (-0.500) | (-0.334) | (3.3003) | (1.4033) | ||||
| lntfp | 0.7910*** | 1.4595 | 0.6337*** | 1.8679*** | 0.1745*** | 1.1517*** | 0.1755 | 1.2869*** | |||
| (12.5645) | (21.1350) | (11.4677) | (12.1899) | (4.5416) | (7.5424) | (1.3980) | (4.5917) | ||||
| 常数项 | - | -8.5116*** | - | -8.1923*** | - | 3.5783*** | - | -3.6811** | |||
| - | (-16.0990) | - | (-5.8474) | - | (2.9552) | - | (-3.4005) | ||||
| ρ | -0.0589*** | 0.0809*** | -0.2360*** | -0.2360** | -0.2360** | -0.2360 | -0.2360** | -0.2360*** | |||
| (-0.9021) | (3.6905) | (-0.2360) | (-3.5458) | (-2.426) | (-3.419) | (-3.702) | (-2.8672) | ||||
| Teta | - | 0.1788*** | - | 0.1327** | - | 0.0850*** | - | 0.5300* | |||
| - | (3.2065) | - | (2.4684) | - | (3.3271) | - | (1.9347) | ||||
| Hausman test | - | -389.5587*** | - | -199.3102*** | - | 207.369*** | - | -21.5759*** | |||
| R2 | 0.9986 | 0.9932 | 0.9989 | 0.9823 | 0.9944 | 0.8886 | 0.9990 | 0.9824 | |||
| LR test | 553.1089*** | 248.4976*** | - | - | - | - | - | - | |||
| LIK | 294.8083 | 142.5026 | - | - | - | - | - | - | |||
注:feF为固定效应,reF随机效应,***、**、*分别表示在1%、5%、10%水平通过显著性检验,()内为统计值。 |
Tab. 7 The estimation of SEM model for regional economic growth by regional human capital表7 分区域人力资本作用于区域经济增长的SEM模型估计结果 |
| 区域 | 东部 | 中部 | 西部 | 东北部 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| feF | reF | feF | reF | feF | reF | feF | reF | ||||
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | ||||
| lnpri | -0.0127 | -0.0466** | 0.0085 | 0.0141 | 0.0943** | 0.0855* | 0.1146* | -0.088 | |||
| (-1.0936) | (-1.9457) | (0.3000) | (0.6053) | (6.6141) | (1.6882) | (1.7971) | (-1.4541) | ||||
| lnint | -0.0026 | -0.0835* | 0.0519 | -0.1882*** | -0.1038** | -0.0854 | -0.2125** | 0.0568 | |||
| (-0.1067) | (-1.7196) | (1.4564) | (-3.0514) | (2.2899) | (-1.533) | (-2.484) | (0.7405) | ||||
| lnadv | 0.0405** | 0.1240*** | 0.0519 | 0.1745*** | -0.0044 | 0.0012 | 0.0698** | 0.0304 | |||
| (2.3007) | (3.5493) | (-1.0346) | (-3.0514) | (-1.996) | (0.0428) | (2.3371) | (0.8926) | ||||
| lnmc | 0.0434*** | 0.0680*** | 0.0274** | 0.2160*** | 0.0117 | 0.0335** | 0.5908*** | 0.4888*** | |||
| (3.7945) | (3.0196) | (2.0961) | (6.1396) | (-0.1480) | (2.2477) | (19.262) | (15.7769) | ||||
| lnlab | 0.1534*** | 0.7008*** | -0.5788*** | 0.7705*** | -0.0185 | -0.0364*** | 0.8516*** | 0.7088*** | |||
| (3.1440) | (9.9385) | (-6.7309) | (7.4859) | (-1.5050) | (-3.178) | (4.4124) | (7.3906) | ||||
| lntfp | 0.7918*** | 1.5539*** | 0.6114*** | 1.4487*** | 0.1192*** | 0.0498 | 0.3316*** | 0.6515*** | |||
| (13.4603) | (25.3003) | (12.8890) | (18.9505) | (3.3893) | (1.4908) | (3.2401) | (5.6682) | ||||
| 常数项 | - | -7.9195*** | - | -8.1743*** | - | 7.6718*** | - | -4.2121*** | |||
| - | (-14.5409) | - | (-9.2174) | - | (18.422) | - | (-4.7451) | ||||
| λ | -0.0589*** | 0.3334*** | -0.3149*** | -0.2607** | 0.2339*** | 0.9681*** | -0.4019*** | 0.6862*** | |||
| (-0.9021) | (4.5358) | (-2.8679) | (-2.6654) | (2.7642) | (180.05) | (-4.292) | (12.5372) | ||||
| Teta | - | 20.7440** | - | 2.3050** | - | 93.4932** | - | 0.5304 | |||
| - | (2.4473) | - | (2.4629) | - | (2.4725) | - | (0.9963) | ||||
| Hausman test | - | -1544.137*** | - | -849.9903*** | - | -61.164*** | - | -92.4725*** | |||
| R2 | 0.9985 | 0.9932 | 0.9990 | 0.9890 | 0.9948 | 0.9941 | 0.9991 | 0.9967 | |||
| LR test | 568.9462*** | 260.4066*** | - | - | - | - | - | - | |||
| LIK | 302.7269 | 148.4571 | - | - | 216.437 | 100.3692 | 128.8873 | - | |||
注:feF为固定效应,reF随机效应,***、**、*分别表示在1%、5%、10%水平通过显著性检验,()内为统计值。 |
The authors have declared that no competing interests exist.
| [1] |
|
| [2] |
|
| [3] |
|
| [4] |
[
|
| [5] |
[
|
| [6] |
[
|
| [7] |
[
|
| [8] |
[
|
| [9] |
[
|
| [10] |
[
|
| [11] |
[
|
| [12] |
[
|
| [13] |
[
|
| [14] |
[
|
| [15] |
[
|
| [16] |
[
|
| [17] |
[
|
| [18] |
[
|
| [19] |
[
|
| [20] |
[
|
| [21] |
[
|
| [22] |
[
|
| [23] |
[
|
| [24] |
|
| [25] |
[
|
| [26] |
|
| [27] |
|
/
| 〈 |
|
〉 |