EN中文

Journal of Resources and Ecology >

2024 , Vol. 15 >Issue 2: 293 - 303

DOI: https://doi.org/10.5814/j.issn.1674-764x.2024.02.005

Agroecology and Agricultural Development

Effect of China’s Taiwan Agricultural Investment in China’s Mainland: Based on the Model of VAR and VEC

  • LI Hangfei ,
  • YANG Lin , *
Expand
  • College of Tourism and Geography, Shaoguan University, Shaoguan, Guangdong 512005, China
*YANG Lin, E-mail:

LI Hangfei, E-mail:

Received date: 2022-10-13

  Accepted date: 2023-04-30

  Online published: 2024-03-14

Supported by

The National Natural Science Foundation of China(41771136)

The Philosophy and Social Science Planning Project of Guangdong(GD22CYJ26)

The Science and Technology Planning Project of Shaoguan(200811154532282)

Fold

Abstract

Based on the data from the 1991-2016 agricultural investment of China’s Taiwan in China’s Mainland and the agricultural GDP of the latter, through models of vector autoregressive (VAR) and vector error correction (VEC), the influences of China’s Taiwan agricultural investment on the development of agriculture in the eastern, central, and western regions of China are discussed. The results show a long-term equilibrium relationship between China’s Taiwan agricultural investment and agricultural development in China’s eastern, central, and western regions. In the long term, China’s Taiwan investment in agriculture in the eastern, central, and western regions of China have certain positive promoting effect on their agricultural development. However, there is an obvious regional diversity in investment effect: Impulse response and variance decomposition show that the positive effect from China’s Taiwan agricultural investment in China’s western region agricultural development is most significant, and it is significantly higher than that in the eastern region; its contribution to the central region’s agricultural development is little. VEC model analysis shows that in the short term, China’s Taiwan investment in agriculture has a significant positive effect on the agricultural development of China’s eastern region, but not on the agricultural development of the central and western regions.

Key words： China’s Taiwan; agricultural investment; VAR model; China’s Mainland; VEC model

Cite this article

LI Hangfei , YANG Lin . Effect of China’s Taiwan Agricultural Investment in China’s Mainland: Based on the Model of VAR and VEC[J]. Journal of Resources and Ecology, 2024 , 15(2) : 293 -303 . DOI: 10.5814/j.issn.1674-764x.2024.02.005

1 Introduction

Foreign Direct Investment (FDI) is one of the main methods of technology diffusion at the macroscale (Lin and Zeng, 2006). As an important technology diffusion channel, its spillover effect on host countries has attracted the attention of experts and scholars. Advanced technical knowledge spreads globally through FDI across national boundaries, changing the host country’s market structure through demonstration, learning, linkage, competition, and talent flow effects. FDI can improve the host country’s labour productivity and promote the rapid growth of the national economy (Liu and Wang, 2017). Consequently, governments worldwide have scrambled to adopt different preferential policies to attract FDI. Current FDI studies mainly focus on the economic and technological spillover effects of FDI on host countries (Zhao and Xue, 2017). The spillover effect of FDI on a host country can be divided into two categories: intra-industry spillover (also known as horizontal spillover) and inter-industry spillover (also known as vertical spillover, which can be further divided into forward- and backward-correlation spillovers). Studies have shown that the horizontal spillover effect of FDI is relatively significant and can effectively promote the economic development of host countries (Blomstrǒm and Person, 1983). Moreover, some scholars found that FDI had no significant spillover effect on the host country or negative effects, possibly because the technology gap between host country enterprises and investment enterprises was too large, or the host country enterprises could not effectively absorb the latest technology diffusion of FDI enterprises, leading to the negative effect of crowding-out being greater than the positive role of technology diffusion (Haddad and Harrison, 1993; Javorcik, 2004).
FDI enterprises are conducive to the rapid development of China’s economy (Xu et al., 2007), and their influence is more significant than that of fixed asset investments (Guo and Luo, 2009). Simultaneously, the vertical (backward- correlated) spillover effect generated by FDI is stronger than the horizontal spillover effect (Li and Xian, 2010). Scholars have drawn different conclusions regarding the spillover effect of FDI on China’s agricultural economy. Some scholars believe that FDI has a crowding-out effect on China’s agricultural development (Ma et al., 2013), which is not conducive to improving China’s agricultural production efficiency (Xu and Dong, 2011). Some studies have shown that FDI has a positive effect on China’s agricultural development (Zhou, 2014), and that some agricultural enterprises can benefit from FDI spillovers (Han and Wu, 2020), promote the improvement of agricultural productivity (Liu et al., 2018). Wei finds that the spillover effect of FDI on China’s agricultural economy showed regional differences in space, with negative spillover effect in the eastern and central regions and positive technology spillover effect in the western regions (Wei, 2015). Research of Wang shows that FDI in agriculture has a significant promoting effect on agricultural TFP and its subdivisions (Wang et al., 2019).
The development of China’s Taiwan is closely related to that of China’s Mainland, and the development of agriculture is a very important part of the former’s economic take-off. Owing to the influence and restriction of land resources, labour force, market and environment, China’s Taiwan agriculture needs to expand the market outside the island. The China’s Mainland and China’s Taiwan are originally from the same root, connected by one river, and have obvious advantages on their five borders. Coupled with the preferential policies of China’s Mainland to China’s Taiwan, China’s Mainland has become an ideal platform for agricultural investment in China’s Taiwan.
Presently, studies on the development of agriculture in China’s Mainland affected by China’s Taiwan are mainly reflected in the following aspects: First, the research area: due to its special geographical location, Fujian province has become a key area for China’s Taiwan businessmen to invest in agriculture, and the agricultural cooperation and development between Fujian and Taiwan has become the focus of scholars’ research (Li, 2015; Chen and Zhuang, 2016; Chen et al., 2020). Second, research methods: Qualitative analysis is more common, whereas quantitative discussion is less common. Scholars have discussed the diffusion mechanism of China’s Taiwan agriculture in China’s Mainland from the perspectives of relational economic geography theory (Li, 2020) and evolutionary economic geography theory (Li et al., 2021), and analysed the development of China’s Taiwan Farmers’ Pioneer Parks, which are cross-strait agricultural cooperation platforms (He et al., 2016; Gu and Li, 2017). Third, research perspective and content: 1) The spatial and temporal distribution characteristics (Zhou et al., 2012), evolution trend (Sun, 2021), location choice (Li, 2021), influencing factors (Xu et al., 2021) and spillover effect (Xie et al., 2018; Li, 2019) of China’s Taiwan agricultural investment in China’s Mainland were discussed on a macro level. 2) The characteristics, path and mechanism of China’s Taiwan agricultural technology diffusion in China’s Mainland were discussed from the micro-level of technology diffusion (Chen et al., 2019; Li and Wei, 2021), and scholars took the specific agricultural technologies such as Gaoshan Tea (Li et al., 2020a), Orchid (Chen et al., 2019; Li et al., 2020b) as the research object to discuss its influence on agricultural development in specific regions of China’s Mainland.
In conclusion, research results on the impact of FDI on agricultural economic development are abundant and can provide the basis for this study. However, there are relatively few researches on the impact of China’s Taiwan investment on agricultural development in China’s Mainland, and the existing researches mainly focus on qualitative discussion, while quantitative analysis is rare and only confirms the spillover effect of diffusion, without analysing the change of its spillover effect with time. There are few studies on the comparative analysis of regional differences in the effects of China’s Taiwan agricultural investment in China’s Mainland. There are huge differences in the natural conditions and social and economic development levels among different regions in China’s Mainland; therefore, the spillover effect of China’s Taiwan agricultural investment and its impact on the development of the agricultural economy are bound to be different.
Based on agricultural investment and GDP data from 1991 to 2016, this study uses vector autoregression (VAR) and vector error correction (VEC) models to explore the influence of China’s Taiwan agricultural investment on the agricultural development of China’s eastern, central and western regions, providing certain references for theoretical and practical research on agricultural investment. The results show that in the long run, China’s Taiwan agricultural investment can promote the development of agriculture in China’s eastern, central, and western parts to varying degrees. In the short term, China’s Taiwan agricultural investment has a positive effect on agricultural development in China’s eastern regions, but not on agricultural development in China’s central and western regions.

2 Theory and hypothesis

The influence of FDI on the economic growth of host countries has been extensively studied. Scholars have discussed the relationship between FDI and economic growth mainly based on economic growth theory. The relationship between FDI and the economic growth of the host country, according to traditional economic growth theory, is shown in Fig. 1. The relationship between FDI and the host country’s economic growth according to the new economic growth theory is shown in Fig. 2.
Fig. 1 Traditional economic growth theory
Fig. 2 New economic growth theory
Therefore, FDI can increase the capital accumulation and formation rates in a host country. It can introduce new technologies, improve the technical levels of host countries, improve product performance, and increase the capital-to- output ratio. At the same time, FDI can induce and urge host countries to improve their microeconomic system, macroeconomic system, and market organisation to different degrees to improve the allocation rate of resources and promote economic growth.
Relevant studies have shown that the impact of FDI on the economic growth of the host country (region) is influenced by the difference in technical level between the host country (region) and the investing country (Haddad and Harrison, 1993; Javorcik, 2004), economic development level of host country (region) (Ma et al., 2016), domestic fixed asset investment and human capital (Han and Wu, 2020), marketization level (Zhang, 2007), amount of FDI (degree of agglomeration) (Yan, 2006), geographical distance (Keller, 2002) and other factors. In general, the host country (region) level of economic development, human capital, market, and the amount of FDI can promote the economic growth of the host country. Domestic investment in fixed assets and geographic distance play a negative correlation with the economic growth of the host country, and a certain degree of technology gap is beneficial to the technology diffusion of FDI and promotes economic development in host countries.
The economic development level of eastern China is significantly higher than that of the central and western regions; its agricultural foundation is relatively good, and its gap with China’s Taiwan agricultural technology level is relatively small, which is conducive to the absorption of China’s Taiwan advanced agricultural technology. Additionally, the economic development of the eastern region focuses on the tertiary and secondary industries, and capital investment is mostly concentrated in the service industry, while the investment proportion of agricultural fixed assets is relatively low (Table 1). As can be seen from Fig. 3, the agricultural investment of China’s Taiwan businessmen in China’s eastern region is much higher than that in the central and western regions. The input of agricultural investment of China’s Taiwan businessmen can alleviate the relative shortage of agricultural investment to some extent. In addition, 66.67% of the state-level agricultural cooperation pilot zones across the Taiwan Straits and 62.07% of the state-level China’s Taiwan Farmer Pioneering parks are located in the eastern region, providing an important platform for the spread of China’s Taiwan agricultural technologies on the China’s Mainland.
Table 1 Agricultural location quotient and agricultural fixed asset investment in the eastern provinces of China
Province Liaoning Beijing Tianjin Hebei Shandong Shanghai
Agricultural location quotient 1.135805 0.058794 0.143173 1.266402 0.842571 0.045173
Proportion of fixed assets investment in agriculture (%) 3.664557 1.317541 2.542282 5.591442 2.957974 0.060525
Province Jiangsu Zhejiang Fujian Guangdong Hainan China
Agricultural location quotient 0.612612 0.483604 0.953792 0.531295 2.720647 1
Proportion of fixed assets investment in agriculture (%) 0.969081 1.430189 3.839435 1.625406 1.461739 4.134692

Note: 1. Agricultural location quotient = (Provincial agricultural GDP/ Provincial GDP) / (National agricultural GDP/ National GDP). If the value is greater than one, agriculture has an industrial comparative advantage; if the value is less than one, there is no industrial comparative advantage. 2. Proportion of fixed investment in agriculture = (Fixed asset investment in agriculture, forestry, animal husbandry, and fishery)/fixed asset investment in society.

Fig. 3 Regions distribution of China’s Taiwan agricultural investment in China’s Mainland
The eight provinces in the central region are large agricultural provinces. The comparative advantages of their agricultural industries are evident. The number of agricultural investments in fixed assets is greater than that in the eastern and western regions: the average agricultural investment in the central region was 127.23114 billion yuan, which was higher than that in the western region (68.65448 billion yuan) and eastern region (58.50997 billion yuan). At the same time, the proportion of agricultural investment in fixed assets (Shanxi and Heilongjiang provinces were as high as 13.94% and 11.06%, respectively) was high in central China, while the proportion of China’s Taiwan agricultural investment in the central region accounted for only 7.54% of the total investment, which is slightly higher than that of the western region (5.51%) and far lower than that of the eastern region. Human capital and marketisation levels in the central region are also lower than those in the eastern region.
The level of economic development and agricultural foundations is weak in China’s western region, and the level of agricultural technology is low. Compared to the eastern and central regions of China, it has no advantages in terms of marketisation level, human capital, or geographical distance from China’s Taiwan. Although agriculture has a comparative advantage in industry (the agricultural location quotient in each western province is higher than 1), investment in agricultural fixed assets is relatively low, far lower than that of central China. However, investment from China’s Taiwan is essentially the same as that in central China. To a great extent, agricultural investment from China’s Taiwan can compensate for the shortage of capital inputs in the western region.
Considering the location of the east, central, and western regions of China’s Mainland, the following assumptions were made:
Hypothesis H1: China’s Taiwan investment in agriculture plays a significant role in promoting agricultural development in the eastern part of China’s Mainland.
Hypothesis H2: China’s Taiwan agricultural investment has no obvious effect on agricultural development in the central part of China’s Mainland.
Hypothesis H3: China’s Taiwan investment in agriculture has an obvious promoting effect on agricultural development in the western part of China’s Mainland.

3 Data sources and research methods

3.1 Research area division

The eastern region of China’s Mainland includes 11 provinces (cities): Liaoning, Hebei, Beijing, Tianjin, Shandong, Jiangsu, Shanghai, Zhejiang, Fujian, Guangdong, and Hainan. The central region comprises eight provinces: Heilongjiang, Jilin, Shanxi, Henan, Hubei, Anhui, Hunan, and Jiangxi. The western region includes Inner Mongolia, Shaanxi, Ningxia, Qinghai, Gansu, Xinjiang, Tibet, Yunnan, Guizhou, Chongqing, Sichuan, and Guangxi.

3.2 Data sources and processing

The data of China’s Taiwan agricultural investment in China’s Mainland came from the “monthly data statistics of investment in China’s Mainland in different regions and industries” of the Investment Review Committee of the Ministry of Economic Affairs of China’s Taiwan. The variables of agricultural economic development were expressed as the GDP of the primary industry, and the data came from the “Statistical Yearbook of China” (1992-2017) and “Statistical yearbooks of provinces (cities and districts)” (1992-2017). As the unit of agricultural investment data for China’s Taiwan is thousand US dollars, it is converted into RMB according to the current exchange rate between US dollars and RMB. To eliminate the influence of price factors, this study references related research (Ding and Ji, 2014; Liu et al., 2015; Zhu et al., 2018), the date of China’s Taiwan agricultural investment, and the first industrial GDP of China’s Mainland was processed using the consumer price index. Simultaneously, to ensure the reliability of the research, the natural log of the time-series data was taken to reduce the effects of different variances.

3.3 Research methods

The vector auto-regression (VAR) model is a multivariate data analysis method (Ding and Ji, 2014). Compared to structural econometric models, the VAR model does not have any a priori constraints on the variables in the system, which makes all current variables apply regression to the lagged terms of all variables. This model avoids several complicated issues such as the subjective division of endogenous and exogenous variables due to imperfect economic theories. The VAR method has a unique applicability for studying dynamic interrelationships between a series of variables. It is a mainstream model used to study macroeconomic economies. The general form of the model is as follows.
Y t = a + ∑ i = 1 p β i Y t − i + ε t
In the formula (1), a is the intercept column vector of VAR model, Yt is a homogenous and stationary linear stochastic process composed of (n×1) vectors, βi is a coefficient matrix of (n×n), Yt-i is the i order lag variable of Yt vector, and εt is a random disturbance term, i=1, 2,…, p. Based on the agricultural investment data of China’s Taiwan businessmen in the eastern, central and western regions of China’s Mainland and the agricultural GDP of the eastern, central and western regions of China’s Mainland, this paper establishes VAR models respectively to discuss the impact of China’s Taiwan agricultural investment on the agricultural economic development in the eastern, central and western regions of China’s Mainland.
Vector error correction (VEC) imposes cointegration constraints on vector autoregression (VAR), which can further explore the relationship between variables. In the short term, the VEC equation includes not only a stable long-term trend, but also short-term fluctuations. In the long run, the cointegration relationship reflects the intensity of the adjustment to the long-term equilibrium state when short-term fluctuations are unbalanced.

4 Empirical analysis of the influence of China’sTaiwan agricultural investment on agricultural development in China’s Mainland

Since the transformation of agricultural investment into economic benefits is a long and relatively backward period, China’s Taiwan agricultural investment in China’s Mainland is represented by the cumulative value, and LNDBTW, LNZBTW and LNXBTW represent the cumulative value of China's Taiwan agricultural investment in the eastern, central, and western regions of China’s Mainland, respectively, after taking logarithms. LNDB, LNZB and LNXB represent the agricultural GDP of the eastern, central, and western regions of China’s Mainland, respectively, after taking logarithms. The cointegration test model, Granger test model, and vector auto-regression (VAR) model between LNDB and LNDBTW, LNZB and LNZBTW, LNXB (1997-2016) and LNXBTW (1997-2016) were constructed to investigate the impact of China’s Taiwan agricultural investment on the agricultural development in the eastern, central and western parts of China’s Mainland.

4.1 Stationarity test of variables

The ADF test method was used to conduct unit root tests for six variables: LNDB, LNDBTW, LNZB, LNZBTW, LNXB and LNXBTW. The results showed that the six time-series data were all non-stationary (95% confidence level). After the second-order difference, LNDB, LNDBTW, LNZB and LNZBTW were all stationary data at a significance level of 0.01; after the first-order difference, LNXB and LNXBTW were stationary data at a significance level of 0.01. The results of the stationarity tests are presented in Table 2.
Table 2 Unit root tests of variables
Inspection form (c, t, p) ADF Critical value (1%) Critical value (5%) Critical value (10%) Inspection conclusion
D2LNDB (c, t, 1) ‒5.750585 ‒4.440739 ‒3.632896 ‒3.254671 Smooth
D2LNDBTW (c, 0, 1) ‒6.198601 ‒3.857386 ‒3.040391 ‒2.660551 Smooth
D2LNZB (c, 0, 1) ‒5.422502 ‒3.769597 ‒3.004861 ‒2.642242 Smooth
D2LNZBTW (c, 0, 1) ‒14.67004 ‒3.769597 ‒3.004861 ‒2.642242 Smooth
DLNXB (c, 0, 0) ‒4.994655 ‒3.857386 ‒3.040391 ‒2.660551 Smooth
DLNXBTW (c, 0, 0) ‒4.048867 ‒3.857386 ‒3.040391 ‒2.660551 Smooth

Note: 1. D2LNDB, D2LNDBTW, D2LNZB and D2LNZBTW are the second-order differences in LNDB, LNDBTW, LNZB and LNZBTW, respectively; DLNXB and DLNXBTW are the first-order differences in LNXB and LNXBTW, respectively; 2. In the test form, c is the intercept term, t is the time trend term and p is the lag order.

4.2 Co-integration analysis

As mentioned above, LNDB and LNDBTW and LNZB and LNZBTW are second-order single integrations, LNXB and LNXBTW are first-order single integrations that can further test for co-integration. The cointegration equations between China’s Taiwan investment in agriculture in the eastern, central, and western regions of China’s Mainland and their agricultural growth are types (2), (3), and (4).

LNDB = 1.259708 × LNDBTW + 0.701125 + ε

LNZB = 0.839251 × LNZBTW + 6.868914 + ε

LNXB = 1.568386 × LNXBTW + 1.427074 + ε

Equations (2), (3), and (4) pass the significance test as a whole as well as all variables and constant terms. The elasticity coefficients between the agricultural GDP in the eastern, central, and western regions of China’s Mainland and China’s Taiwan investments are 1.259708, 0.839251, and 1.568386, respectively. For every unit increase in China’s Taiwan agricultural investment, the agricultural GDP of the eastern region of China’s Mainland increased by 1.259708 units, the agricultural GDP of the central region of China’s Mainland increased by 0.839251 units, and the agricultural GDP of the western region of China’s Mainland increased by 1.568386 units. The three cointegration equations show that China’s Taiwan agricultural investment has different promoting effects on agricultural growth in different regions of China’s Mainland.

4.3 Granger causality analysis

The cointegration analysis shows that there is a long-term equilibrium relationship between agricultural development in the eastern, central, and western regions of China’s Mainland and agricultural investment from China’s Taiwan. The granger causality test should be used to test whether this equilibrium relationship constitutes a causal relationship. It can be seen from Table 3 that LNXBTW is the granger cause of LNXB at the significance level of 0.05, indicating that China’s Taiwan investment in agriculture can effectively promote the growth of agricultural GDP in western China, which is consistent with the results of co-integration analysis. At a significance level of 0.1, LNDBTW is the granger reason for LNDB, indicating that China’s Taiwan investment in agriculture can promote the growth of agricultural GDP in the eastern part of China’s Mainland to a certain extent, which is consistent with the results of the co-integration analysis. The test results show that LNZBTW is not the granger reason of LNZB, indicating that China’s Taiwan investment in agriculture cannot effectively promote the growth of agricultural GDP in the central part of China’s Mainland, which is inconsistent with the results of co-integration analysis.
Table 3 Granger causality tests
Null hypothesis Lag length F statistic P values Inspection results
LNDBTW does not granger cause LNDB 4 3.01650 0.0579 Reject null hypothesis
LNZBTW does not granger cause LNZB 2 0.95515 0.4025 Accept null hypothesis
LNXBTW does not granger cause LNXB 1 6.12932 0.0249 Reject null hypothesis

Note: Hysteresis length is the optimal hysteresis length in the corresponding VAR model.

4.4 Empirical analysis of vector auto-regression (VAR) model

4.4.1 Lag selection

Three VAR models between LNDB and LNDBTW, LNZB and LNZBTW, LNXB and LNXBTW were constructed, and the optimal hysteresis periods of the three models were determined using the hysteresis length standard in the hysteresis structure. The lag period of the VAR model between LNDB and LNDBTW is shown in Table 4, and the optimal lag period of the model is four. Similarly, the optimal lag periods of the VAR model between LNZB and LNZBTW and between LNXB and LNXBTW are two and one, respectively.
Table 4 VAR lag order selection criteria
Lag LOGL LR FPE AIC SC HQ
0 ‒1.010715 NA 0.004508 0.273701 0.372887 0.297067
1 74.44040 130.3247 6.83e‒06 ‒6.221855 ‒5.924298 ‒6.151759
2 80.27560 9.018038 5.85e‒06 ‒6.388691 ‒5.892763 ‒6.271865
3 84.92929 6.345931 5.68e‒06 ‒6.448117 ‒5.753817 ‒6.284561
4 100.5206 18.42614* 2.10e‒06* ‒7.501876* ‒6.609204* ‒7.291589*

Note: * represents the lag order of the model selected according to the corresponding criteria; NA represents no data.

4.4.2 Model stability test

All characteristic roots of the three VAR models were within a unit circle (Fig. 4, and the small figures a, b, and c respectively represent the unit root of the three VAR models between LNDB and LNDBTW, LNZB and LNZBTW, LNXB and LNXBTW). Therefore, the three VAR models, namely VAR(4), VAR(2), and VAR(1), are stable, meet the theoretical requirements, and can be further analysed by impulse response and variance decomposition.
Fig. 4 The stability test of VAR model

Note: The scale indicates the size of the feature root in the VAR model.

4.4.3 Analysis of impulse response function

The VAR model is a non-theoretical model that uses an impulse response function to analyse the impact of the standard deviation of the random disturbance term on endogenous variables and each variable’s corresponding response degree after receiving the corresponding impact (Fig. 5). The small figures a, b, and c in Fig. 5 respectively represent the impulse response function of agricultural GDP in the eastern, central, and western regions of China’s Mainland to the impact of agricultural investment from China’s Taiwan. The solid line is the impulse response function curve, and the dashed line is the plus or minus double standard deviation band.
Fig. 5 Response of LNDB, LNZB, LNXB to LNDBTW, LNZBTW, LNXBTW, respectively

Note: The solid line is the impulse response function curve, and the dashed line is the plus or minus double standard deviation band.

Figure 5 showes that the response degree of agricultural GDP in the eastern, central, and western regions of China’s Mainland to the impact of agricultural investment from China’s Taiwan varies greatly. Giving a standard deviation impact on agricultural investment come from China’s Taiwan, although there are a certain positive effect on agricultural GDP in the eastern, central and western regions in China’s Mainland, and this positive effect will begin to appear in the second year after the impact, the effect difference is obvious: the positive effect on western region in China’s Mainland is most significant, and significantly higher than that of eastern and central regions in China’s Mainland, which achieved peak in the fourth year (0.045), then held steady state (Fig. 5c), which is consistent with co-integration analysis and granger test variance analysis (Fig. 6c); the positive effect on eastern region in China’s Mainland is significant, which achieved peak in the ninth year (0.026), then showed a trend of convergence (Fig. 5a), which is consistent with co-integration analysis and granger test, variance analysis (Fig. 6a); the positive effect on central region in China’s Mainland is very weak, which is consistent with the results of the granger test and the variance analysis (Fig. 6b).
Fig. 6 Variance decomposition of LNDB, LNZB, LNXB

4.4.4 Analysis of variance decomposition

To further explore the relative effects of China’s Taiwan agricultural investment on agricultural development in the eastern, central, and western regions of China’s Mainland, variance decomposition was carried out for LNDB, LNZB, and LNXB respectively based on the established 3 VAR models. In Fig. 6, the horizontal axis represents the response time, that is, the number of lagged periods (unit: years), and the vertical axis represents the variance ratio. Fig. 6 (a, b, c) shows the variance decomposition diagram of agricultural GDP in eastern (LNDB), central (LNZB), and western (LNXB) China.
Figure 6 showes that: 1) The variance contribution rate of China’s Taiwan agricultural investment to agricultural GDP in western China was the largest (Fig. 6c). It gradually increased from 0% in Phase 1 to 42% in Phase 15 and then remained stable, indicating that the impact of China’s Taiwan agricultural investment on agricultural economics in western Mainland China gradually strengthened and was sustained. 2) The variance contribution rate of the China’s Taiwan investment in agriculture to agricultural GDP of eastern China was lower than that of western China, which raised gradually from 0% in phase 1 to 12% in phase 15, and then remained stable, suggesting that the influence of China’s Taiwan investment in agriculture on agricultural economic growth in eastern Mainland China also gradually increased and continuous, but its strength was lower than that of the western region in Mainland China; 3) China’s Taiwan business investment in agriculture has little contribution to the variance of agricultural GDP in the central region of the China’s Mainland (Fig. 6b), and it has little promotion effect on the agricultural economic growth in the central region of the China’s Mainland. These results agree with the conclusions of the pretextual cointegration analysis, granger test, and impulse response function analysis.
The above analysis verified the three hypotheses put forward in the second part of the paper, namely hypothesis H1: China’s Taiwan investment in agriculture has a significant promoting effect on agricultural development in the eastern part of China’s Mainland; hypothesis H2: China’s Taiwan businessmen’s agricultural investment has no or no obvious effect on agricultural development in the central part of m China’s Mainland; and hypothesis H3: China’s Taiwan investment in agriculture has an obvious promoting effect on agricultural development in the western part of China’s Mainland.

4.5 Vector error correction (VEC) model analysis

The analysis of this study confirmed that there exists a long-term equilibrium relationship between China’s Taiwan agricultural investment and agricultural development in the eastern, central, and western regions of China’s Mainland; however, in the short term, this equilibrium relationship may be broken, and the relationship between variables may be unbalanced. Therefore, to further research the short-term dynamic relationship between China’s Taiwan agricultural investment and the growth of agriculture in the eastern, central, and western regions of China’s Mainland, it is necessary to establish a VEC based on the VAR model. The regression equation of VEC model (the unit roots of VEC model are less than or equal to 1, which pass the test) is as follows (ECMt is error correction):
The eastern region in China’s Mainland:

D(LNDBt)=‒0.101058×ECMt‒1+0.464069×D(LNDBt‒1)‒

0.119821×D(LNDBt‒2)+0.279823×

D(LNDBt‒3)+0.122802×D(LNDBTWt‒1)+

0.030759×D(LNDBTWt‒2)+0.22718×

D(LNDBTWt‒3) ‒0.003010

ECMt=LNDBt‒1.259708×LNDBTWt‒0.701125

R2= 0.602820, AIC=‒7.326158, SC=‒6.433487

The central region in China’s Mainland:

D(LNZBt) =‒0.048417×ECMt‒1 +0.325053×D(LNZBt‒1)+

0.031958×D(LNZBTWt‒1)+0.032099

ECMt= LNZBt‒0.839251×LNZBTWt‒6.868914

R2= 0.158449, AIC=‒5.280502, SC=‒4.789646

The western region in China’s Mainland:

D(LNXBt)=‒0.397962×ECMt‒1‒0.158504×D(LNXBt‒1)+

0.119645×D(LNXBTWt‒1)+0.078769

ECMt= LNXBt‒1.568386×LNXBTWt‒1.427074

R2= 0.296788, AIC=‒5.030768, SC=‒4.536117

For the eastern region of China's Mainland, the error correction mechanism is a negative feedback process (the coefficient of ECMt is negative and passes the significance test). When the short-term fluctuation deviated from the long-term equilibrium, LNDB was adjusted to the equilibrium state at a correction rate of 10.11% (the coefficient of ECMt). From the perspective of the regression coefficient, the short-term fluctuation of LNDB is mainly determined by self-variation (lagged phase 1) and LNDBTW (lagged phase 1 and lagged phase 2; the T value passes the significance test). In addition, R2=0.602820, AIC=‒7.326158, SC= ‒6.433487, which indicates that the model has a good fitting degree; in the short term, the error correction effect is relatively obvious, and China’s Taiwan investment has a significant positive effect on agricultural development in the eastern region of China’s Mainland.
Similarly, in the central region of China’s Mainland, the error-correction mechanism is also a negative feedback process. If the short-term fluctuation deviates from the long-term equilibrium, LNZB adjusts to the equilibrium state at a correction rate of 4.84%. From the perspective of the regression coefficient, the short-term fluctuation of LNZB is mainly determined by its own changes (lagged phase 1). However, the VEC equation and regression coefficients fail to pass the significance test at the level of 0.05, indicating that the error correction effect is not obvious in the short term and that the effect of China’s Taiwan investment on agricultural development in the central region of China’s Mainland is not significant.
In the western region of China’s Mainland, the error correction mechanism is also a negative feedback process. If the short-term fluctuation deviates from the long-term equilibrium, LNXB will adjust to the equilibrium state at a correction rate of 39.80%, which is larger than that in the eastern and central regions of China’s Mainland. From the perspective of the regression coefficient, the short-term fluctuation in LNXB was mainly determined by self-variation (lagged phase 1) and LNXBTW (lagged phase 1). However, the VEC equation and regression coefficients fail to pass the significance test at the level of 0.05, indicating that the error correction effect is not obvious in the short term and that the effect of China’s Taiwan investment on agricultural development in the western region of China’s Mainland is not significant.

5 Discussion and conclusions

This study discusses the impact of China’s Taiwan agricultural investment on agricultural economic development in the eastern, central, and western regions of China’s Mainland. The conclusions are as follows:
(1) There is a long-term equilibrium relationship between China’s Taiwan agricultural investment and agricultural development of the eastern, central and western regions in China’s Mainland. In the long term, China’s Taiwan investment in agriculture has different degrees of promoting function on agricultural development in the eastern, central, and western regions of China’s Mainland, but there are obvious regional differences The results of impulse response and variance decomposition analysis show that it has the strongest promoting effect on agricultural development in western China, and has a certain promoting effect on agricultural development in eastern China, but has little promoting effect on agricultural development in central China.
(2) VEC model analysis shows that in the short term, China’s Taiwan investment in agriculture has a positive effect on the development of agriculture in eastern China, but has no positive effect on the development of agriculture in China’s central and western regions.
The two sides of the Taiwan Strait are closely linked by mountains and rivers, and the “five edge advantages” are evident. Compared with general agricultural FDI, the agricultural investment of China’s Taiwan in China’s Mainland has its particularity. The perspective of this study enriches theoretical research on FDI in agriculture to a certain extent. At the same time, the research methods and conclusions provide a reference and guidance for China’s Mainland regions on using China’s Taiwan agricultural investment to promote regional agricultural development, which has practical significance. In summary, the promoting effect of China’s Taiwan agricultural investment has the characteristics of long-term and regional differences. When making use of agricultural investment, local governments should proceed from long-term interests, bearing in mind that they should act in haste and seek success. They should formulate reasonable policies to attract investment in accordance with the actual situation such as “different regions, different periods and different industries”, strive to improve the technological diffusion effect of capital, increase farmers’ income and promote rural revitalisation.
This study analyses the impact of China’s Taiwan agricultural investment based on time-series data from 1991 to 2016 using the cointegration test, granger test, vector autoregressive (VAR) model, and vector error correction (VEC) model. The research conclusion is basically consistent with existing research results, that is, FDI has a certain promoting effect on agricultural economic development, but due to the particularity of China’s Taiwan’s investment, its promoting effect on agricultural development in different regions of China’s Mainland is not the same. The research conclusions have guiding significance in theory and practice. At the same time, there are some shortcomings in this paper. For example, China’s Taiwan agricultural investment is spread throughout China’s Mainland. Considering provincial administrative regions or key municipal administrative regions as the research objects and basic units, the research results are of more practical significance and better reflect the characteristics and effects of China’s Taiwan agricultural investment. In addition to the level of economic development, the proportion of agricultural fixed asset input, the comparative advantage of the agricultural industry, agricultural foundation, etc., the factors affecting the efficiency of China’s Taiwan agricultural investment may also be related to the specific industrial types of China’s Taiwan agricultural investment, such as agriculture, forestry, animal husbandry, fishery, and agricultural manufacturing. At the same time, owing to the availability of data, the study did not consider the situation after 2017, which will have an impact on the research results. On the one hand, in follow-up research, the scale of the basic research unit can be reduced to make the research conclusion more consistent with the actual situation. On the other hand, the investment behaviour and effect of China’s Taiwan businessmen on a certain type of agriculture or even a certain agricultural variety such as orchids in a certain region can be studied from the perspective of agricultural industry segmentation.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Blomstrǒm M, Person H. 1983. Foreign direct investment and spillover efficiency in an underdevelopment economy: Evidence from the Mexican manufacturing industry. World Development, 11(6): 493-501.

DOI

[2]
Chen J, Wei S Q, Chen S L. 2019. Spatial and temporal diffusion of China’s Taiwanese agricultural technology: A case study in Zhangpu County, Fujian Province. Scientia Geographica Sinica, 39(6): 957-966. (in Chinese)

DOI

[3]
Chen J, Wei S Q, Yan X Y. 2020. Technology spillover mechanism and influencing factors of China’s Taiwanese agricultural investment in Fujian: Based on China’s Taiwan funded phalaenopsis enterprise in Zhangzhou. Taiwan Research Journal, (1): 79-89. (in Chinese)

[4]
Chen Y L, Zhuang P F. 2016. Dynamic relationship between Taiwan businessmen’s investment in Fujian agriculture industry and agricultural products trade between Fujian and Taiwan. Journal of Fujian University of Technology, 14(1): 62-66. (in Chinese)

[5]
Ding Z L, Ji C J. 2014. An empirical study on China’s import, export, real exchange rate and economic growth based on VAR model. International Trade Issues, (12): 91-101. (in Chinese)

[6]
Gu X H, Li G H. 2017. Suggestions on speeding up the development of China’s Taiwan Farmers Pioneer Parks: Taking Qingliu Taiwan Farmers Pioneer Park of Fujian as an example. Fujian Agricultural Science and Technology, (8): 60-64. (in Chinese)

[7]
Guo X B, Luo Z. 2009. The impact of foreign investment characteristics on China’s economic growth. Economic Research, 44(5): 52-65. (in Chinese)

[8]
Haddad M, Harrison A. 1993. Are there positive spillovers from direct investment? Evidence from panel data for Morocco. Journal of Development Economics, 42(1): 51-74.

DOI

[9]
Han Y, Wu L P. 2020. Decomposition of innovation spillover effect of FDI on Chinese agricultural enterprises: Threshold regression analysis based on absorption capacity. International Trade Issues, (8): 132-146. (in Chinese)

[10]
He H, Hou Y, Dang X M, et al. 2016. A comparative study of Hainan and Fujian provinces on China’s Taiwan agricultural exchange and cooperation: Taking experimental area of cross-straits agriculture cooperation of Hainan and China’s Taiwan Farmers Pioneer Park of Fujian as an example. Chinese Journal of Tropical Agriculture, 36(1): 72-75. (in Chinese)

[11]
Javorcik B. 2004. Does foreign direct investment increase the productivity of local firms? In search of spillovers through backward linkages. American Economic Review, 94(3): 605-627.

DOI

[12]
Keller W. 2002. Geographic localization of international technology diffusion. American Economic Review, 92(1): 120-142. (in Chinese)

DOI

[13]
Li F, Wei S Q, You X J. 2020a. Process and mechanism of embeddedness of the China’s Taiwan high mountain tea industry in Mainland China from the perspective of relations: A vase study of Yongfu Town, Zhangping City. Tropical Geography, 40(6): 1026-1038. (in Chinese)

[14]
Li F, Wei S Q, You X J. 2021. Agglomeration and evolution of China’s Taiwan high mountain tea industry in Mainland China from the perspective of evolutionary economic geography: A case study in Yongfu Town, Zhangping City. World Regional Studies, 30(2): 344-354. (in Chinese)

[15]
Li H F. 2019. Analysis on the relationship between China’s Taiwan agricultural investment and China’s Mainland agricultural development based on VAR and VEC models. Agricultural Exploration in Taiwan, (5): 7-14. (in Chinese)

[16]
Li H F. 2020. Analysis on the characteristics of social network of China’s Taiwan orchid technology diffusion from the perspective of relational economic geography. Economic Forum, (11): 101-105. (in Chinese)

[17]
Li H F. 2021. Study on location selection of agricultural direct investment of China’s Taiwan in China’s Mainland: Based on ways of grey relation and spatial autocorrelation. Journal of Southwest China Normal University (Natural Science Edition), 46(9): 81-88. (in Chinese)

[18]
Li H F, Wei S Q. 2021. Analysis on the diffusion characteristics of China’s Taiwan’s orchid technology in less developed areas in the Mainland. Journal of Fujian Normal University (Natural Science Edition), 37(4): 69-76. (in Chinese)

[19]
Li H F, Wei S Q, Chen S L, et al. 2020b. Characteristics of social network of China’s Taiwan’s orchid technology diffusion in Guangdong-Taiwan agricultural cooperative pilot zone in Shaoguan City, Guangdong Province. Geographical Research, 39(11): 2479-2492. (in Chinese)

[20]
Li J W, Xian G M. 2010. Spillover effect analysis of foreign direct investment through backward correlation approach. International Trade Issues, (4): 73-79. (in Chinese)

[21]
Li Y S. 2015. On innovation of China’s Taiwanese invested agricultural industrial clusters: Taking Fujian Province as a case. Journal of Fujian Agriculture and Forestry University (Philosophy and Social Sciences), 18(6): 16-19. (in Chinese)

[22]
Lin L, Zeng G. 2006. Research on technology diffusion and global technology duality. World Economic Research, (4): 23-29. (in Chinese)

[23]
Liu N J, Han Y J, Wang P P. 2018. Has FDI increased the productivity of agricultural firms in China? Evidence from panel data on 99801 agricultural firms. China Rural Economy, (4): 90-105. (in Chinese)

[24]
Liu Y B, Zhou Y F, Wang X Z, et al. 2015. Research on threshold effect of FDI on economic development from the perspective of market integration: A case study of Yangtze River Economic Belt. Exploration of Economic Issues, (6): 118-125. (in Chinese)

[25]
Liu Y F, Wang Z. 2017. Literature review on international technology spillovers based on foreign direct investment channels. Scientific Decision-Making, (2): 76-96. (in Chinese)

[26]
Ma S Z, Chen Y, Wang X X. 2013. Dynamic impact of agricultural FDI on China’s food security from the perspective of seed industry R&D capability. Management World, (7): 71-79. (in Chinese)

[27]
Ma W, Wang C P, Li X. 2016. Technology spillover effect of agricultural FDI: Empirical analysis based on panel threshold model of 27 provinces. Economic Geography, 36(7): 167-172. (in Chinese)

[28]
Sun Z H. 2021. A study on favorable factor of the spread of China’s Taiwan’s agricultural investment to the China’s west. Straits Technology and Industry, (1): 1-3. (in Chinese)

[29]
Wang Y F, Zhang Y, Liao M. 2019. The impact of foreign direct investment on agricultural total factor productivity: Mechanism of action and empirical evidence. Contemporary Economic Research, (6): 74-86. (in Chinese)

[30]
Wei K L. 2015. Regional differences in technical efficiency of agricultural production based on FDI spillover effect. Social Scientists, (10): 60-65. (in Chinese)

[31]
Xie G E, Xu Y J, Yang F M. 2018. Trade effect analysis on China’s Taiwan’s agricultural investment to China’s Mainland. Journal of East China University of Science and Technology (Social Science Edition), (3): 65-77, 116. (in Chinese)

[32]
Xu B W, Dong W. 2011. Research on the influence of technology spillover of foreign direct investment on agricultural product processing industry. Taiwan Agricultural Exploration, (1): 19-22. (in Chinese)

[33]
Xu H L, Wei Y Q, Lai M Y. 2007. Research on back link spillover effect of foreign direct investment. Management World, (4): 24-31. (in Chinese)

[34]
Xu J, Su Y M, Zhuang P F. 2021. Analysis on the motivation of the locational selection of Chinese Taiwan agricultural investment in the Chinese mainland. Taiwan Agricultural Research, (4): 16-21. (in Chinese)

[35]
Yan B. 2006. Analysis of spillover effect and related influencing factors in foreign direct investment industry. Economic Review, (1): 41-46. (in Chinese)

[36]
Zhang H Y. 2007. Technology diffusion of foreign capital and China’s economic growth:Theory, demonstration and policy. Chengdu, China: Bashu Publishing House. (in Chinese)

[37]
Zhao F, Xue Q. 2017. Spillover effects of FDI on economic development of host countries: Literature review. Technical Economics, 36(2): 123-132. (in Chinese)

[38]
Zhou Q, Wang H P, Zheng Z. 2012. Analysis of agriculture-related investment of China’s Taiwan businessmen in Mainland China. Taiwan Agricultural Exploration, (5): 6-9. (in Chinese)

[39]
Zhou Z Z. 2014. A study on the productivity effect of FDI in China’s agriculture. Diss., Wuhan, China: Wuhan University. (in Chinese)

[40]
Zhu Z M, Zhu J, Fang P. 2018. Comparative analysis of the impact of FDI and domestic investment on carbon emission intensity in Jiangsu Province. Journal of Nanjing University of Technology (Social Science Edition), 17(3): 77-86. (in Chinese)

Options
Outlines

/

〈 〉