中国交通运输投资及其经济溢出效应时空演化——1997-2013年省级面板的实证
作者简介:林雄斌(1988- ),男,江西玉山人,博士研究生,研究方向为交通规划与政策。E-mail: linxiongbin@sz.pku.edu.cn
收稿日期: 2016-03-19
要求修回日期: 2016-07-21
网络出版日期: 2016-09-10
基金资助
国家自然科学基金项目(41371168)
深圳市海外高层次人才计划(KQCX20130628093909157)
The temporal-spatial impacts of transport investment on economic growth: An empirical analysis of provincial panel data, 1997-2013
Received date: 2016-03-19
Request revised date: 2016-07-21
Online published: 2016-09-10
Copyright
交通运输投资是区域规划和政策制定的重要内容,但交通运输投资对促进经济增长的时空效应仍存在较大争议。在总结交通成本、交通投资与经济增长关系的基础上,建立1997-2013年省级尺度的交通运输投资、劳动力和经济总量的面板模型,综合评估交通投资对经济增长的多尺度时空间效应。研究发现:① 全国层面上看,交通投资对经济增长具有显著的引导效应,其弹性系数为11.4%,但在不同地区这一弹性系数存在显著差异。总体上,在城市化水平和社会基础较好的东部和东北地区,交通运输投资对社会经济增长的作用更加明显。② 在时间尺度上,随着中国城市化水平的不断扩大,交通投资对经济增长的作用也逐渐增强。③ 在考虑交通投资和经济总量时间滞后性之后,在全国层面上,当期交通投资和前期交通运输投资对经济的增长作用程度大致相同,但相对中西部地区而言,东部地区当期交通投资对经济增长作用更加显著。考虑到交通运输投资的这种时空差异,应根据不同地区不同发展阶段的特点,采取谨慎地投资策略以强化对经济增长的引导作用。
林雄斌 , 杨家文 . 中国交通运输投资及其经济溢出效应时空演化——1997-2013年省级面板的实证[J]. 地理研究, 2016 , 35(9) : 1727 -1739 . DOI: 10.11821/dlyj201609011
During the past decades, governments and scholars have found that the investments in the transportation sectors play a dominant role in promoting economic development both at urban and regional scales. However, much of literature has not reached a consensus on the magnitude of the impacts of transportation investments, especially when temporal and spatial differences of transportation investment should be considered. In order to fill this theoretical gap, this paper initially summaries the literature on the relations between transport cost, transport investment and economic growth, and then develops statistical models to test these effects by using provincial panel data incorporating transportation investment, labor market and economic conditions in China from 1997 to 2013. Finally, it re-evaluates the comprehensive impacts of the transport investment on economy from the perspective of multi-level temporal and spatial effects. These empirical results indicate that: (1) The impact of transport investments on economic development is significantly positive at the national scale (elasticity coefficient is 11.4%), but the magnitude varies in different regions. One general trend is that the impact is more obvious in richer and urbanized regions, such as eastern and northeastern parts of China. (2) The impacts have become higher in the wake of high speed of urbanization and economic development in China. (3) In the time lag model, the effects of the current transportation investment almost play the same role as the previous investment after the time lag effect is taken into account, while this lag effect in eastern China is more significant than that in central and western China. Since large-scale transportation investments will continue to serve as one of crucial strategies for regional planning and development in China, one should carefully evaluate the potential effects of transportation investment when making transportation investment planning and policies. In conclusion, this study has considerable planning implications for spatial planning and policy-making decisions on transport investment.
Key words: transport investment; economic growth; regional policy; panel data
Tab. 1 Methods and results of the stationary test of panel data表1 面板数据平稳性检验方法和结果 |
| 变量单位根 | LLC检验 | IPS检验 | ADF检验 | PP检验 |
|---|---|---|---|---|
| lnEcon | 3.267 | 9.184 | 15.559 | 1.288 |
| △lnEcon | -6.740*** | -4.643*** | 112.001*** | 121.183*** |
| lnTI | -7.826*** | -1.823** | 84.397** | 36.374 |
| △lnTI | -40.969*** | -38.085*** | 655.722*** | 692.011*** |
| lnNTI | 0.610 | 6.122 | 10.609 | 9.800 |
| △lnNTI | -21.845*** | -16.692*** | 335.893*** | 364.857*** |
| lnLab | -12.233*** | -10.113*** | 217.109*** | 214.882*** |
| △lnLab | -32.366*** | -24.396*** | 482.951*** | 789.531*** |
注:*、**和***分别为10%、5%和1%水平上达到显著。 |
Fig. 1 The density of transportation investment of each province of China in 1997, 2005 and 2013图1 1997年、2005年和2013年中国各省域交通运输投资的强度 |
Fig. 2 The sharing ratio of transportation investment of each province of China in 1997, 2005 and 2013图2 1997年、2005年和2013年中国各省域交通运输投资比例 |
Tab. 2 Results of spillover effects of transport investment both at national and regional levels表2 分地区交通投资对经济增长的带动效应 |
| 变量 | 模型1 | 模型2 | 模型3 | 模型4 | 模型5 |
|---|---|---|---|---|---|
| 区域汇总 | 东部地区 | 中部地区 | 西部地区 | 东北地区 | |
| C | 0.218* | 1.231*** | 1.579*** | 0.458*** | -0.175 |
| (1.740) | (6.402) | (5.098) | (3.546) | (-0.250) | |
| lnTI | 0.114*** | 0.162*** | 0.172*** | -0.040 | 0.224*** |
| (3.426) | (3.475) | (3.447) | (-0.956) | (3.588) | |
| lnNTI | 0.618*** | 0.509*** | 0.266*** | 0.508*** | 0.082 |
| (18.812) | (10.803) | (4.831) | (10.944) | (1.513) | |
| lnLab | 0.443*** | 0.415*** | 0.551*** | 0.599*** | 0.980*** |
| (17.503) | (11.654) | (10.561) | (21.090) | (8.581) | |
| 调整R2 | 0.928 | 0.946 | 0.967 | 0.965 | 0.952 |
| Log likelihood | -157.495 | -3.706 | 65.469 | 20.386 | 33.059 |
| F-statistic | 358.540*** | 155.827*** | 158.730*** | 298.896*** | 53.528*** |
| AIC | 0.674 | 0.279 | -0.892 | -0.004 | -0.512 |
| 样本量 | 527 | 170 | 102 | 204 | 51 |
注:*、**和***分别为10%、5%和1%水平上达到显著;括号内为t统计量。 |
Tab. 3 Results of spillover effects of transport investment at national levels in different periods表3 分时段交通投资对经济增长的带动效应 |
| 变量 | 模型6 | 模型7 | 模型8 |
|---|---|---|---|
| (1997-2001年) | (2002-2007年) | (2008-2013年) | |
| C | -0.530* | 0.153 | 0.496*** |
| (-1.890) | (0.970) | (2.642) | |
| lnTI | -0.036 | 0.048 | 0.401*** |
| (-0.601) | (0.978) | (7.491) | |
| lnNTI | 0.731*** | 0.880*** | 0.435*** |
| (10.609) | (18.764) | (8.996) | |
| lnLab | 0.588*** | 0.233 | 0.337*** |
| (14.252) | (6.184) | (6.954) | |
| 调整R2 | 0.858 | 0.941 | 0.932 |
| Log likelihood | -72.309 | -6.742 | -13.214 |
| F-statistic | 133.974*** | 370.508*** | 318.179*** |
| AIC | 1.036 | 0.169 | 0.239 |
| 样本量 | 155 | 186 | 186 |
注:*、**和***分别为10%、5%和1%水平上达到显著;括号内为t统计量。 |
Tab. 4 Results of spillover effects of transport investment accounting time lag impact表4 分地区考虑时间滞后性的交通投资对经济增长的引导效应 |
| 变量 | 模型9 | 模型10 | 模型11 | 模型12 | 模型13 |
|---|---|---|---|---|---|
| 区域汇总 | 东部地区 | 中部地区 | 西部地区 | 东北地区 | |
| C | 0.684*** | 1.030*** | 0.748*** | 0.321*** | 1.348* |
| 7.950 | 5.869 | 4.178 | 3.869 | 2.003 | |
| lnTI | 0.015*** | 0.031*** | 0.017** | 0.004 | 0.056 |
| 3.382 | 3.311 | 2.272 | 0.902 | 1.663 | |
| lnNTI | 0.072*** | 0.100*** | 0.063*** | 0.053*** | 0.066*** |
| 12.355 | 8.490 | 6.076 | 7.455 | 2.713 | |
| lnLab | 0.011 | 0.024 | 0.015 | 0.013* | -0.025 |
| 1.575 | 1.570 | 1.498 | 1.821 | -0.428 | |
| lnEconi-1 | 0.843*** | 0.775*** | 0.831*** | 0.903*** | 0.779*** |
| 68.014 | 30.070 | 32.999 | 68.441 | 10.420 | |
| lnTIi-1 | 0.015*** | -0.001 | 0.030*** | 0.019*** | 0.011 |
| 3.402 | -0.090 | 4.033 | 4.763 | 0.318 | |
| 调整R2 | 0.997 | 0.996 | 0.997 | 0.999 | 0.981 |
| Log likelihood | 691.859 | 203.288 | 170.434 | 347.321 | 46.196 |
| F-statistic | 5467.670*** | 2764.348*** | 2966.929*** | 10334.020*** | 345.719*** |
| AIC | -2.645 | -2.354 | -3.322 | -3.441 | -1.591 |
| 样本量 | 496 | 160 | 96 | 192 | 48 |
备注:*、**和***分别为10%、5%和1%水平上达到显著;括号内为t统计量。 |
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] |
[
|
| [28] |
|
| [29] |
[
|
| [30] |
|
| [31] |
|
| [32] |
|
| [33] |
|
| [34] |
|
| [35] |
|
| [36] |
|
| [37] |
|
| [38] |
[
|
| [39] |
[
|
| [40] |
[
|
| [41] |
[
|
| [42] |
|
| [43] |
[
|
| [44] |
[
|
| [45] |
[
|
| [46] |
|
| [47] |
|
| [48] |
|
| [49] |
[
|
| [50] |
|
| [51] |
[
|
| [52] |
[
|
| [53] |
[
|
| [54] |
[
|
| [55] |
[
|
/
| 〈 |
|
〉 |