基于依赖视角的世界棉花贸易网络及形成机制研究

doi:10.16716/j.cnki.65-1030/f.2023.03.002

摘要/Abstract

摘要：

相较于传统网络指标，将“依赖”概念引入棉花贸易网络研究中可以更准确地揭示棉花贸易特征。文章基于新的贸易依赖指标构建1999—2019年世界棉花贸易依赖网络，利用时态随机指数图模型对20年来核心棉花贸易国家的进口、出口贸易依赖网络进行分析，结果显示：（1）棉花进口依赖度提高的国家主要分布于亚洲地区，棉花贸易出口依赖较进口依赖网络集中度更高，其网络中心主要集中在欧美地区；（2）中国棉花进口依赖度的大幅提高表现为依赖对象多元化和依赖程度加深，出口依赖度的大幅降低主要源于低依赖国家的脱离；（3）棉花贸易依赖关系的形成具有三元传递特征，出口依赖关系的建立具有偏好连接特征，少数具有出口依赖的国家能够通过在网络中建立的优势地位进一步促进依赖关系的形成；（4）两国接壤关系和自由贸易协定交互属性能显著促进棉花贸易依赖关系的形成，但经济发展水平对依赖关系形成的影响具有异质性，相较于低收入国家，中等收入国家更易形成进口依赖，高收入国家同时倾向于形成进口、出口双向依赖。今后可以从深化国际贸易合作、优化棉花产业布局等角度出发，改善中国棉花贸易对外依赖形势。

关键词: 棉花贸易网络, 贸易依赖, 时态随机指数图模型, 世界棉花贸易

Abstract:

Compared with the traditional network index, introducing the concept of "dependency" into the cotton trade network can reveal the characteristics of cotton trade more accurately. Based on the new trade dependence index, this paper constructs the 1999-2019 world cotton trade dependence network, analyzes the import and export trade dependence network of core trading countries in the past two decades, and uses temporal random index graph model to study. The results are as follows: (1) The countries with increased cotton import dependence are mainly distributed in Asia, and the network of export dependence is more concentrated than that of import dependence, and its network centers are mainly concentrated in Europe and America. (2) The rapid increase of China's cotton import dependence is reflected in the diversification of the dependent objects and the deepening of the dependence degree, and the sharp decline of export dependence is mainly due to the separation of low-dependent countries. (3) The formation of cotton trade dependence has the characteristics of ternary transmission, and the establishment of export dependence has the characteristics of "preference connection". A few countries with export dependence can further promote the formation of dependency through the dominant position established in the network. (4) Both the border relationship and the interactive nature of the free trade agreement between the two countries can significantly promote the formation of the cotton trade dependence relationship, but the economic level has a heterogeneous effect on the formation of the dependence relationship. Compared with low-income countries, middle-income countries are prone to import dependence, while high-income countries tend to form two-way import and export dependence. From the perspective of deepening international trade cooperation and optimizing the layout of the cotton industry, we can improve China's foreign dependence on cotton trade in the future.

Key words: the cotton trade network, trade dependence, time exponential random graph model, world cotton trade

中图分类号:

F746
F313

李敏, 曹城. 基于依赖视角的世界棉花贸易网络及形成机制研究[J]. 新疆财经, 2023, (3): 17-28.

LI Min, CAO Cheng. A Study of the World Cotton Trade Network and Its Formation Mechanism Based on Dependency Perspective[J]. Finance & Economics of Xinjiang, 2023, (3): 17-28.

图/表 7

图1

图2

图3

图4

表1

表2

表3

参考文献 27

[1]	BENEDICTIS L D, TAJOLI L. The world trade network[J]. World economy, 2011(8):1417-1454.
[2]	王健, 董俊哲, 陈浩, 等. 全球棉花进出口贸易分析及展望[J]. 棉纺织技术, 2018(3):81-84.
[3]	孙洁. 棉花价格波动对我国纺织品出口贸易的影响及其对策[J]. 价格月刊, 2015(2):66-69.
[4]	GRANOVETTER M S. Economic action and social structure:the problem of embeddedness[J]. Administrative science quarterly, 1984(19):481-510.
[5]	ARTHUR W B, DURLAUF S N, LANR D A. The economy as an evolving complex system II[M]. Reading: Addison-Wesley Publishing Company, 1997.
[6]	JACKSON M O. An overview of social networks and economic applications[J]. Handbook of social economics, 2011(1):511-585.
[7]	程中海, 冯梅. 基于动态复杂网络的世界棉花贸易时空分异特征与贸易格局分析[J]. 国际经贸探索, 2017(10):36-50.
[8]	刘婷婷, 张蕙杰, 康永兴, 等. 社会网络视角下的全球棉花贸易格局分析[J]. 世界农业, 2022(4):26-36.
[9]	BURT R S. The social structure of competition[M]. Boston: Harvard University Press, 1992.
[10]	FRANKEL J A. Regional trading blocs in the world economic system[M]. Washington DC: Institute for International Economics, 1997.
[11]	HAVENS R M, BALASSA B. The theory of economic integration[J]. Southern economic journal, 1962(1):47.
[12]	HIRSCH J E, MEHO L I. On the relationship between the ISI impact factor and the number of citations received by papers[J]. Journal of the american society for information science and technology, 2004(10):32-41.
[13]	刘林青, 闫小斐, 杨理斯, 等. 国际贸易依赖网络的演化及内生机制研究[J]. 中国工业经济, 2021(2):98-116.
[14]	段文奇, 刘宝全, 季建华. 国际贸易网络拓扑结构的演化[J]. 系统工程理论与实践, 2008(10):71-75+81. doi: 10.12011/1000-6788(2008)10-71
[15]	RAUCH J E. Business and social networks in international trade[J]. Journal of economic literature, 2001(4):1177-1203.
[16]	戴卓. 国际贸易网络结构的决定因素及特征研究:以中国东盟自由贸易区为例[J]. 国际贸易问题, 2012(12):72-83.
[17]	FAGIOLO G, REYES J, SCHIAVO S. The evolution of the world trade web:a weighted-network analysis[J]. Journal of evolutionary economics, 2010(4):479-514.
[18]	刘华军, 杜广杰. 中国雾霾污染的空间关联研究[J]. 统计研究, 2018(4):3-15.
[19]	唐晓彬, 崔茂生. “一带一路”货物贸易网络结构动态变化及其影响机制[J]. 财经研究, 2020(7):138-153.
[20]	CAO D P, LI H, WANG G B, et al. Dynamics of project-based collaborative networks for BIM implementation:analysis based on stochastic actor-oriented models[J]. Journal of management in engineering, 2017(3):1-12.
[21]	FINGER K, LUX T. Network formation in the interbank money market:an application of the actor-oriented model[J]. Social networks, 2017(48):237-249.
[22]	RONBINS G, SNIJDERS T, WANG P, et al. Recent developments in exponential random graph (p^*) models for social networks[J]. Social networks, 2007(2):192-215.
[23]	刘林青, 陈紫若. 中国优势产业组合的动态演化机制研究:基于TERGM的实证分析[J]. 科技进步与对策, 2020(11):70-78.
[24]	许和连, 孙天阳, 成丽红. “一带一路”高端制造业贸易格局及影响因素研究:基于复杂网络的指数随机图分析[J]. 财贸经济, 2015(12):74-88.
[25]	孙忆, 孙宇辰. 自由贸易协定能提升国家间亲密度吗?:基于中国周边FTA的实证分析[J]. 世界经济与政治, 2017(4):129-154+160.
[26]	汪亚楠, 王海成, 张夏. 自由贸易协定对贸易政策不确定性的影响研究[J]. 宏观经济研究, 2021(7):26-37.
[27]	LEIFELD P, CRANMERS S J, DESMARAIS B A. Temporal exponential random graph models with btergm:estimation and bootstrap confidence intervals[J]. Journal of statistical software, 2018(6):1-36.

变量属性	变量符号	变量名称	变量解释
行动者属性	Sender	发送者效应	具有某种属性的节点是否具有链出的倾向
	Receiver	接收者效应	具有某种属性的节点是否具有链入的倾向
	Homphily	同配性	具有相同属性的节点是否具有彼此连接的倾向
	Dcov	连续性主效应	节点连续性属性的值对网络连接的影响，即马太效应
网络协变量效应	Edgecov	网络外生效应	其他网络中存在的关系是否更容易产生贸易关系
内生结构效应	Edges	边	网络密度的衡量，一般为负值
	Mutual	互惠性	网络节点相互链出关系的倾向
	Gwesp	传递性闭合	网络中3个节点形成闭合的倾向，即当i对j发出关系、j对k发出关系时，i对k发出关系的趋势
	Gwodegree	扩张性	网络中大多数节点是否具有相似的扩张型星型结构
	Gwidegree	聚敛性	网络中大多数节点是否具有相似的聚敛型星型结构
时间依赖结构效应	Memory	稳定性	上一期网络中的关系是否会影响下一期网络关系的形成
	Delrecip	延迟性	当j在t=1时向i发送关系，是否会导致在t=2时从i连接到j
	Timecov	时间—边交互效应	随着时间的推移，某协变量在解释网络时是否变得更为重要

变量	模型一	模型二	模型三
Sender（GDP_Mid）	-0.649^*** [-0.822，-0.464]	-0.683^*** [-0.806，-0.503]	-0.532^*** [-0.841，-0.153]
Sender（GDP_High）	0.376^*** [-0.096，0.879]	0.203^*** [-0.165，0.521]	0.103^*** [-0.405，0.543]
Receiver（GDP_Mid）	0.583^*** [0.214，0.774]	0.587^*** [0.325，0.738]	0.457^*** [0.125，0.605]
Receiver（GDP_High）	1.318^*** [1.145，1.489]	1.199^*** [1.000，1.472]	0.952^*** [0.808，1.039]
Dcov（Population）	0.002^*** [0.001，0.001]	0.001^*** [0.001，0.001]	0.001^*** [0.001，0.001]
Edges	-4.986^*** [-5.480，-4.525]	-5.098^*** [-5.568，-4.410]	-4.045^*** [-4.344，-3.848]
Mutual	0.402^*** [0.189，0.611]	0.251^*** [-0.009，0.492]	-0.002^*** [-0.138，0.038]
Gwesp		0.781^*** [0.504，1.024]	0.773^*** [0.580，1.016]
Gwodegree		-2.691^*** [-3.057，-2.494]	-2.580^*** [-2.840，-2.071]
Gwidegree		0.532^*** [-0.123，0.777]	0.957^*** [0.798，1.313]
Edgecov（RTA）	22.230^*** [21.946，22.641]	22.203^*** [21.825，22.666]	23.076^*** [21.588，23.379]
Edgecov（Border）	0.850^*** [0.398，1.331]	0.806^*** [0.377，1.206]	0.118^*** [-0.285，0.472]
Time（Memory）			1.227^*** [1.030，1.423]
Time（Delrecip）			0.004^*** [-0.126，0.435]
Timecov（RTA）			10.163^*** [4.498，10.776]